Benchmarking Telecommunication and Access in the ...

SIBIS IST–2000-26276

Statistical Indicators Benchmarking the Information Society

Benchmarking Telecommunication and Access in the Information Society by

March 2003

Benchmarking Telecommunications and Access in the Information Society

2

This publication has been prepared by Technopolis Ltd., Brighton (UK) in the context of the IST-26276-SIBIS project (“SIBIS Statistical Indicators Benchmarking the Information Society”) in co-operation with the other partners in the project.

All publications of the SIBIS project – including this report – are available in electronic format on the Internet at:

www.sibis-eu.org

Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use which might be made of the following information. The views expressed are those of the authors and do not necessarily reflect those of the European Commission. Nothing in this report implies or expresses a warranty of any kind. Results from this report should only be used as guidelines as part of an overall strategy.

For further information, contact:

empirica GmbH

Oxfordstr. 2

D-53111 Bonn

E-mail: [email protected]

A great deal of additional information on the European Union is available on the Internet. It can be accessed though the Europa server: http://europa.eu.int

© SIBIS project and European Communities, 2003

Reproduction is authorised provided the source is acknowledged.

Printed in Germany.


3

TABLE OF CONTENTS

FOREWORD...............................................................................................................................4

PREFACE....................................................................................................................................5

1 Executive summary..............................................................................................................7 1.1 Introduction to the topic .................................................................................................7 1.2 Results of the data ........................................................................................................8 1.3 Description of future work ............................................................................................ 13 1.4 Conclusions ................................................................................................................ 13

2 Introduction ........................................................................................................................16 2.1 Topic area definition .................................................................................................... 16 2.2 The approach.............................................................................................................. 18 2.3 Overview of the report ................................................................................................. 19

3 Identification of the indicator framework and hierarchy.....................................................21 3.1 Traditional indicators for telecommunications and access .............................................. 21 3.2 New SIBIS indicators selection..................................................................................... 22

4 Analysis of Data .................................................................................................................26 4.1 Analysis of indicators about citizens and society............................................................ 26

5 The use of compound indices............................................................................................43 5.1 The data..................................................................................................................... 44 5.2 Snapshot indicators and country groupings – factor and cluster analysis......................... 46

6 Further developments: remaining gaps in the statistical coverage of the topic................52 6.1 The existing gap on regional level data indicators for T&A.............................................. 52 6.2 Some methodological problems when collecting information on T&A with telephone survey

methodologies ............................................................................................................ 58

7 Conclusions........................................................................................................................60

8 ANNEX 1 – References .....................................................................................................62

9 ANNEX 2 - Methodology of the survey..............................................................................63 9.1 General Population Survey (GPS) ................................................................................ 63 9.2 Decision Maker Survey (DMS) ..................................................................................... 70 9.3 Questionnaires ............................................................................................................ 78


4

FOREWORD For some years now statistical indicators on Information Society have been central in the policy making process. This has been best demonstrated through the benchmarking exercise of eEurope 2002 Action Plan, and its further inclusion as a key activity in eEurope 2005. Having recognised this need and driven by the difficulties in obtaining reliable and appropriate statistics, the IST programme supported a pan-European research effort during Framework Programme 5. The prime objective has been to develop and make available methodologies, tools and new statistical indicators which can help remedy the deficit in this field. It is in this context that the SIBIS project was launched (IST-26276, Statistical Indicators Benchmarking the Information Society”, www.sibis-eu.org). This document presents some of the project’s main findings so far. There are at least two main reasons that make this document interesting. First, it is one of the few original attempts to have a coherent and comprehensive approach in measuring the Information Society. As such it is expected to stimulate further debate and research among the professional statistical community, leading to an improved statistical competence in Europe. Second, it provides a unique single source of data on real time which supports many of the new IST research areas, at the launch of Framework Programme 6. Building on the original SIBIS research, in particular on the results of the indicator surveys, the project has produced 9 reports, selected from those addressed by e-Europe. The SIBIS work attracts further interest since it also supports the e-Europe 2005 initiative. SIBIS is carrying out an evaluation and a benchmarking of the eEurope 2005 initiative for the 15 EC Member States and the 10 Accession countries which will become available later in 2003. Both the reports and benchmarking results can be obtained from the SIBIS web site. The publication of the SIBIS project results is a timely and direct contribution to benchmark progress on key issues of the information society in general and the e-Europe initiative in particular.

Thanassis Chrissafis [email protected] DG INFSO-C6


5

PREFACE This report is a main deliverable of the SIBIS project (Statistical Indicators Benchmarking the Information Society), funded by the European Commission under the “Information Society Technology” Programme (1998-2002). The overall goal of SIBIS is to develop and pilot indicators for monitoring progress towards the Information Society, taking account of the “e-Europe action lines”. SIBIS focuses on nine topics of interest, i.e. Telecommunications and Access, Internet for R&D, Security and Trust, Education, Work and Skills, Social Inclusion, e-Commerce, e-Government and e- Health. Within the SIBIS project two surveys (a General Population Survey and a Decision Makers Survey – businesses), were conducted between March and May 2002 covering the nine eEurope topics. This report describes the outcomes with respect to the topic of “Telecommunications and Access”. The indicators presented are largely taken from a the General Population Survey which covered consumers and, therefore, this topic should be looked at in the context of informing interested parties on the access, usage and impact of technologies on the general population, not on businesses. The report is organised in eight chapters. The first three chapters give the reader an overview of the main outcomes (Executive Summary), the context (Introduction) and the indicators developed (Identification of the Indicator Framework and Hierarchy). The core of the report is the analysis of indicators, provided in Chapter 4. The indicators are presented as Access indicators, Usage indicators and Impact indicators of ICT technologies including Internet, Broadband and Mobile Telephony. The indicators are analysed in a number of ways, for example, by age groups, EU15 vs. US, and straightforward country comparisons, giving a variety of interesting results. Finally, this chapter also includes innovative compound indicators aimed at measuring impact of ICT across different clusters of countries and emerging paths of users migrating to faster connections than dial-up. The final chapters summarise the outcomes of the study as well as areas where future research might be necessary. The main audience should be policy makers, statistical offices at all levels (national, e.g. CBS, Statistisches Bundesamt, Statistics Finland etc., and supranational, e.g. Eurostat, OECD), industry leaders and researchers in the domain and those involved and interested in benchmarking the domain throughout Europe and the world. These institutes should consider the questions posed and the subsequent indicators developed by SIBIS a valuable input for their yearly surveys. The project includes a series of workshops with such institutions in the countries represented by the SIBIS consortium. The report is also of interest to the European Commission (in particular DG INFSO) and to government official and regulator bodies and ICT Observatories. For each of the nine topics a separate SIBIS report (WP2) was issued in 2001. The WP2 report was aimed at setting the scene on the topic, defining the gaps in the statistical coverage and suggesting innovative indicators to be developed through the subsequent survey. The current report, although an independent document, is an interim report. The final version for all SIBIS reports will be issued in April 2003. SIBIS is led by Empirica (Bonn, Germany), and includes the following project partners: RAND Europe (Leiden, The Netherlands), Technopolis Ltd. (Brighton, UK), Databank Consulting (Milan, Italy), Danish Technological Institute (Taastrup, Denmark), Work


6

Research Centre Ltd. (Dublin, Ireland), Fachhochschule Solothurn Nordwestschweitz (Olten, Switzerland). Technopolis is an International research and consulting organisation focusing on science, technology and innovation policy. Technopolis works for regional and national government bodies, international organisations as well as private companies. We are based in the UK, Sweden, Netherlands, France and Austria. Our work spans most scientific disciplines, technologies and industrial sectors. Our strong emphasis is on implementation and providing advice about 'how to do it better'. Depending on the needs of individual projects, we work alone or together with other international policy and technology research organisations. This report has been peer-reviewed in accordance with our quality assurance standards and may, therefore, be represented as a Technopolis product. For more information about Technopolis or this document, please visit our web site (www.technopolis.co.uk), or contact: Rebecca Alden (Senior Consultant) [email protected] Maria del Mar Negreiro (Consultant) [email protected]


7

1 Executive summary

1.1 Introduction to the topic The topic of telecommunications and access is a physical cornerstone of the information society and is both wide ranging and ‘horizontal’ in nature. In many ways the topic can be considered as the fundamental ‘enabler’ – it allows the other eEurope domains to ‘happen’. Telecommunications and access does not stop with infrastructure. There are already great changes and innovations taking place in accessing Internet technologies, mobile telephony, multimedia and other emerging technologies, all of which contribute to the current state-of-play in Europe. Although in many cases the Internet is becoming pervasive, access and usage patterns do vary according to socio-economic and geographic factors. Mobile telephony is also an exceptionally fast-growing sector. For the purposes of indicator development in SIBIS the term 'telecommunications' has been interpreted very broadly to include all the networks (cable, mobile, Internet, as well as copper wire) over which all types of information (voice, data, sound, image) is carried. So, although the main focus is on telephony networks, computer networks, the Internet, cable (TV as well as telephony), and wireless forms of transmission are also included. Overall, a more accurate descriptor in these circumstances would be ‘communications networks’. ‘Access’ is another loose defined topic. It can be defined formally as ‘the ability to retrieve data, graphics, sound, text etc whether on-line’. Translated into the context of eEurope, this topic covers the wide range of devices by which users access electronic ‘information’ – e.g. computers, telephones, multimedia kiosks, televisions and other hybrid devices. In defining the statistical boundaries of the SIBIS study for the topic telecommunications and access, the work undertaken has taken into consideration the fact that fixed telecommunications networks have been in existence for over 100 years and there is a long history of indicators associated with measuring progress in this area. The main areas of measurement have traditionally been in measuring the size and growth of the market and different technologies as well as being used as an aid to predicting revenues, profits, universality and potentiality. The process of evolving measurement and indicators was simplified by the fact that telecommunications in most countries was developed by a state monopoly. With just one provider it was easier to measure output. Newer networks have not been subject to such long-term scrutiny. Most of the underpinning technologies for these networks have only been developed in the last 25 years and also identifying the potential user base has become more difficult. No longer are telephone lines supplied to just two groups - business and residential customers. Today an array of network products and services are targeted at multiple niche user groups who will consume more than one service. Measurement of this is further complicated by the increase in the speed of adoption. Technologies can be adopted and disposed of in less than a decade. Witness the usage of WAP which will be replaced by the advanced second and third generation mobile telephony standards. Any tracking measurement over a medium to long term must make allowance for the possibility of technology redundancy by seeking to measure user experience rather than the technology. This investigation and indicator development focuses on issues of access and usage, which at their most basic can be described in a binary form. Firstly, users either have access to a


8

particular service or not. Secondly, there are also a potentially limited number of ways in which services can be used. Although some issues of impact have been explored through SIBIS, the study does not expand to examine the technical quality or perceived value of such services. For example, it does not attempt to analyse the percentage of time a connection to the Internet is working at 100% of its capacity or whether a user considers one ISP's content to give better value for money than another. This is a potentially very interesting extension to the work currently being tackled under this project and certainly an important area to examine. However, due to constraints in piloting indicators, it was considered more pressing to establish indicators that inform the baseline at this early stage. More sophisticated and elaborate indicators such as compound indicators are also rarely available due to the problems associated with combining methodologies. Similarly, although basic indicators exist for newer technologies, (and these are becoming more common), in many instances they are also often not comparable, nor yet ready to meet the challenge of emerging topics of interest. Examples of the latter include the ability to robustly measure VOIP (Voice over Internet Protocol). This is because of the rapid cycle of technology adoption. The same issues apply to access mechanisms. Although telephone (fixed, mobile) and television ownership rates are well known, there is less information on the extent to which newer forms of access mechanism are available or used. New channels include digital TV, Internet-enabled phones, and interactive TV. Emerging channels will be dominated by the next generation of mobile products. Finally, this topic area does not cover ‘content’; neither the provision nor the use of it. Although tremendously interesting, it is an entire domain in itself, and would increase the necessary research to facilitate indicator development by about 100%. Nor does this topic include ICT market size, productivity issues and matters related to quality of service.

1.2 Results of the data A large number of innovative telecommunications and access indicators were identified within the work of the SIBIS project and a proportion of those were piloted in the SIBIS survey in order to test, to a certain extent, their usefulness and validity. A small number of the piloted indicators are not innovative but provide a filtering role in order to ask questions on more interesting and sophisticated indicators. The report also presents some of the more basic indicators in order to provide a better context for the latter. Also as mentioned above, the limited size of the survey meant that a number of innovative indicators identified as part of the SIBIS work could not be piloted. Some of these are nevertheless discussed in the section on gaps. The indicators are presented as Access indicators, Usage indicators and Impact indicators. The indicators are analysed in a number of ways, for example, by age groups, EU15 vs. US and straightforward country comparisons giving a variety of interesting results.

1.2.1 Access indicators Share of Internet users with broadband access This indicator is a baseline indicator and a fundamental measure of the impact broadband access has had on the population of Europe. The figures presented in the report show the share of at-home Internet users in Europe who have a broadband connection at home,


9

including cable modem, DSL, and alternative broadband technologies such as wireless. The results of the survey indicate that while there is significant variation across countries, the overall level of penetration of broadband is low. An average of about 7% of at-home users respondents have such a connection, While behind the US, with just over 16%1, the average does include some Member States that have very high levels of broadband. For example, Belgium which benefits from a dense urbanised geography where access to telecoms networks such as cable is relatively ubiquitous. In other Member States, the midband technology of ISDN is popular. In particular, Germany has an ISDN penetration of 17% among the online population. This may be due to an aggressive marketing of ISDN services to consumer customers of Deutsche Telekom since the late 1990s. In contrast, the UK has a rate of only approximately 2% as ISDN was primarily marketed as a business product and had high associated installation costs. The lesson from midband is that consumer can and do migrate to better access technologies. It is also worth contrasting its penetration level with the measure of those without any Internet access. Across the EU15 nations, 55.7% of the European population have no Internet access at home at all. The Netherlands had the lowest number of people not connected to the Internet at 27%, while Greece had the highest number of people not connected with about 85%. Respondents' access methods were also analysed on the basis of their age. While much is expected of broadband, it still has a long adoption cycle ahead. Thus narrowband is still the predominant bandwidth for Internet connections across Europe. However, given that dial-up uptake is stagnating in the US, while this does not necessarily mean death of dial-up access, it highlights the high levels of broadband uptake that the US is currently registering in comparison to the EU 15. Dial up subscribers are less prevalent in the US than last year2. Similarly in the UK , where broadband at home penetration rates has increased two fold since August last year, narrowband users are also stagnating. Take-up of Internet consumer market narrowband (metered/unmetered) and broadband3

Aug 02 Nov 02 Feb03 UK homes connected to the Internet

42% 42% 45%

UK homes connected to the Internet using narrowband

84% 80% 79%

UK homes connected to the Internet using narrowband (unmetered)

38% 35% 38%

UK homes connected to the Internet using narrowband (metered|)

29% 28% 25%

UK homes connected to the Internet using narrowband (unsure whether metered or unmetered)

18% 16% 6%

UK homes connected to the Internet using broadband

7% 9% 13%

1 Current US penetration BB rates at home stands at over 20% according to ITU data, February 2003 2 Morgan Stanley , March 2003 3 Oftel, Monitoring Internet studies, March, 2003.


10

As governments encourage the roll-out of broadband because of its associated benefits this indicator will be an important surrogate measure for the impact of policy. Migrators to a faster connection than dial-up Understanding the migration path from dial-up connection to broadband access is important if measures to encourage its adoption are to be better formulated. The US experience indicates that long experienced narrowband users migrate to it because of the better quality of access and the faster and persistent connections associated with broadband. Likewise, the closer the narrowband and broadband gap is in relation to availability and low connection prices, the more likely users are to migrate from narrowband. This measure is important for informing policy that encourages broadband usage.

Access to the Internet At Home This indicator supplies useful data about the context and nature of Internet usage. At-home usage is seen as the natural locus for effective use of being ‘online’. The combination of privacy and consistent access from the same PC means that users are more willing to make use of applications such as e-commerce, interacting with organisations that hold personal information about them, including government and health providers for example. While at-home usage is generally established as a norm, the at-home location is not as endemic as might be expected. In contrast to the 50% plus of respondents have such access in Northern nations there are many Internet users in Europe who do not have this privilege.

Figure1: Internet usage by location

Internet usage by location(in % of population)

0

10

20

30

40

50

60

70

80

90

100

DK S

NL

FIN

UK

IRL D A L

EU

-15 B I E F

EL P

CH

US

A

occasional usage (less than once a month)usage only somewhere elseusage only at workusage only at home

usage at home and at work

Base: respondents with Internet access at home who have used the Internet at least once in the last 4 weeks (N=4,631). EU15 results weighted by EU15 population (N=3,838). Source: SIBIS 2002, GPS Internet Drop-Outs This indicator attempts to measure the level of users dropping out of using the Internet at home. As such, it can be used to detect possible failures in Internet provision. A high figure could indicate structural problems within the Internet sector, such as too high access costs or a lack of appropriate applications and content. The highest reported drop out rate was Sweden at 13.7%. This is interesting, as Sweden is a mature Internet nation. Users have had time to travel along the full user life cycle and have been exposed to a variety of alternative


11

access contexts. There could be a variety of reasons for this which would be interesting to explore in detail. For example, it could also be because employees are now accessing the Internet from the workplace and have therefore dropped their at-home connections.

1.2.2 Usage indicators

Figure 2: Online tenure

Experience of online usage(in % of population)

0

10

20

30

40

50

60

70

80

90

100

DK S

FIN NL

UK A

IRL L B

EU

-15 D I P E F

EL

CH

US

2 years and more 1 year - 2 years 6-12 months < 6 months Don't know non-Internet users

Question: When did you use the Internet for the first time? Base: Respondents who accessed the Internet at least once in the last 12 months (N=6,905), EU15 results weighted by EU 15 population (N=5,828)Source: © SIBIS 2002, GPS Column percentage. Internet users according to on-line tenure This indicator measures the Internet tenure of Internet users, or the length of time since the first Internet use. It is clear from the EU15 average of 61% of Internet users having two or more year's tenure, that exposure to, and experience of the Internet encourages broadband adoption since it correlates with the countries that have a wider penetration of broadband connections. This relation has been further explored in the ‘migrators from dial-up to a faster connection’ compound indicator. Degree of e -mail networking intensity E-mail is often considered the 'killer app', a popular reason for people to go online. While volume measures of e-mails sent generate very large numbers, understanding the nature of a user's email network and their intensity of e-mail usage provides an indicator that has a higher utility. Respondents were asked to indicate their degree of networking among friends and relatives and the degree of usage of e-mail networks with friends and relatives. E-mail penetration among friends and family was highest in Scandinavian countries where over half of the respondents’ friends and families have an e -mail address. The US also had a higher rate of email penetration than the EU average. Alternative devices for accessing the Internet Assessing the usage levels of different methods of accessing the Internet is an important indicator as currently there is a renaissance in this area. While access via TV and dedicated


12

devices were experimented with in the mid-Nineties, the newer methods are more robust and potentially more appealing to consumers. TV-set orientated devices such as digital TV (DTV) set-top boxes and games consoles today have network links. GSM networks now allow palmtop computers and WAP phones to access the Internet. Rather than replace PC-based Internet access these methods and devices could result in bi-modal usage by consumers. Though it is also important to explore whether different usages are intrinsic to the use of different technologies. These measurements are particularly useful to look at by age group as younger groups are the predominant bimodal users. Despite the wide spread marketing of DTV in some markets it is not being heavily used to go online. Only 9.9% of those respondents (who have used an alternative Internet access method, indicated that they used DTV for it. Degree of mobile networking intensity: Degree of mobile ownership networking among friends and relatives A natural corollary of the higher degree of mobile phone ownership in Europe is a high networking intensity. Over 60% of respondents’ relatives and friends owned a mobile phone. Again, this indicator is a basis for exploring the usage of mobile phone services. The secondary indicator confirms that communication applications, such as SMS, are by far the most important driver of phone usage, especially among those under 25.

Figure 3: Mobile data use according to age groups

Mobile usage according to age groups in EU-15(in % of each age group)

0

10

20

30

40

50

60

70

80

up to 24 25 to 49 50 to 64 65 and more

mobile Internet user only SMS user neither SMS nor mobile Internet (incl. don't know) no mobile phone

Source: SIBIS 2002, GPS, Base: all respondents weighted by EU15-population (N=10,306) Over 80% of under 25s use SMS, and their usage (see fig.11b) includes not only communication purposes, but also services such as sports, news, ringtones, downloads, and purchasing the likes of tickets is also relatively stronger than among those in older groups.

1.2.3 Impact indicators While it is very feasible to count the instances of occurrence associated with Telecommunications and Access technologies it is important to understand the impact they


13

have on citizens too. However, impact indicators have been more difficult to operationalise. Survey questions seeking responses about impacts are fielded only to current Internet users. Therefore, they fail the plausibility test. The technology must have had an impact as it is being used. However, some impact indicators have emerged that are useful. In particular, consumer perceptions of broadband. The majority of broadband users perceive it as better than expected

1.3 Description of future work No survey is perfect and no bundle of indicators complete. There are topics which constitute large subjects in themselves and are the focus of other existing studies. These include new IPv6 and e-content. There are indicators that because of the survey methodology are problematic to gather data for. It is difficult to interview those under 16 by telephone for example. It is also difficult to measure quality of service as users may not be exposed to multiple services that would allow them to make comparative qualitative judgements, most have just one ISP or mobile network provider Likewise some very innovative indicators could not be tracked on a Pan-European level anymore since there were not enough representative samples and they would not be transferable at this stage to major European surveys such as Eurostat. The usage indicators piloted give only a partial picture of usage. Co-ordinating them with surveys concerning e-content (or with panel measurement data methodologies) would close a gap and allow a fuller picture to emerge of what consumers are doing online. Also there are problems constructing indicators on impact from the SIBIS dataset as only responses from current users are collected. A more difficult gap to fill is the lack of regional level data. While SIBIS indicators can be classified by nation, cutting the data by region is more difficult. As much EU assistance is delivered at regional level and many regional programmes contain an Information Society ‘priority’ it is increasingly important to measure access and usage at this level. The current turmoil associated with third generation mobile telecoms networks should not dilute efforts to fill a gap associated with them. The development of 3G will eventually lead to new data delivery mechanisms across wide areas. They will be an important infrastructure for the Information Society in years to come, especially as a means of overcoming the digital divide. The usage of mobile services indicator currently gives only a partial insight into this topic, but could easily be expanded when a representative sample of 3G users can be assembled. Studying the mobile networks is desirable, but given the constraints on how many questions can be fielded to respondents it was decided to concentrate on questions relating to the telecommunications and access networks that are currently more widely adopted.

1.4 Conclusions Analysing how Europe is doing in the area of telecommunications and access is complex. It is clear from the data that one size does not fit all in attempting to measure the access, usage and impact of telecommunications and access across the Member States and the


14

other participating countries. From an analysis of the data from the General Population Survey (GPS) and from the compound indicators a number of differences have been identified across the participating European Countries and between the Europe Union and the US. Benchmarking data is often presented separately for what are in fact related indicators. To add value to such approaches, the results of the SIBIS surveys have been analysed and presented in a series of integrated "snapshots" that enable easy understanding of inter-related indicators and patterns across Europe. Such snapshots have been prepared for mobile telephony, Internet usage and migration to higher bandwidth connections by home users. They show that trends are not always linear and that both individual countries and clusters of countries can show differing development and migration paths depending on national circumstances. The ‘Internet snapshot’ indicates three Internet clusters, with some overlap with and some divergence from the mobile clusters. For the Internet snapshot Greece, Portugal, Italy and Spain join France as laggards, and the Netherlands, Denmark and Switzerland join the forerunners. As might be expected, there is a clear positive association between PC usage, home access to the Internet and recent/regular Internet usage The mobile snapshot shows the expected strong positive association between individual mobile phone ownership and extent of ownership amongst one's family and friends. At the time of the SIBIS survey, France and the US clearly lagged behind on both counts 4. Although there is some tendency for countries with greater mobile penetration to have more usage of SMS by mobile phone owners there is enough divergence to suggest that other factors also play a role. Some countries in the intermediate cluster (e.g. Belgium and Switzerland), have SMS usage levels as high as those portrayed in cluster 1. Generally the percentage of SMS users is over two times higher in the EU-15 than it is in the US. Moreover, in some countries such as Finland mobile owners are four times more likely to be SMS users than their US counterparts. The ‘migration from dial-up to a faster connection snapshot’ suggests that for most countries there is a "traditional" migratory path. More regular usage of the Internet is associated with an increased proportion of users with longer (2 or more years) usage, and a trend towards an increased proportion who have migrated to a faster connection in the forerunner countries. They seek a better online experience, e.g. quicker downloads or always on connections. The clustering of the US, Sweden, Finland and Denmark illustrates this in the cluster of bubbles portrayed above. The more experienced users are, the more likely they are to upgrade to higher bandwidth and, therefore, to being realistically able to use all the benefits and services within the Information Society. As a general rule, there are clear signs that the propensity of Internet users to do various things online increases with increasing levels of Internet penetration in a country increasing tenure and increasing take up of broadband technologies. In addition a number of methodological lessons have also been learnt: It remains important to go on counting instances It remains extremely important, in the light of these results, to go on counting instances in some of the European countries involved in the survey. However, the next step of measurement is where SIBIS ‘adds value’ as there are numerous pan-European surveys already in existence which are counting. SIBIS has the opportunity to suggest sets of

4 Although the mobile phone market is very dynamic and France has seen strong growth in mobile ownership

since the SIBIS survey levels are still low in comparison to other countries.


15

indicators which help provide a picture of the ‘how and why?’. In spite of this need, it must be noted that identifying usage patterns and impact is much more difficult. The evolution of the usage behaviours also means that the classifications have to be reassessed regularly to make sure that they are not out of date. As user numbers increase, so does the depth of content available to them. This is an exciting area of development, which can give great insight into the use of technologies in Europe. A number of flexible indicators could be chosen and used to test the levels of penetration of technologies in each country before piloting sophisticated indicators The differences across the Member States and Switzerland are high and vary for each technology area. This means that the current approach of asking exactly the same set of questions in each country leads to some country data, particularly in the cases of usage and impact indicators, being statistically insignificant. A way of creating surveys, that are comparable, but do not ask obsolete questions, is to devise a system of standard flexible indicators which could be drawn on to create a set which best work in a country specific situation. A set of penetration indicators could also be identified which would be asked across the board in order to judge when some countries start to have significant samples of a chosen technology in use so that more sophisticated indicators could be introduced. Measure also experience not only technology As mentioned above, the shortening of technology product life cycles makes any tracking measurement problematic. This problem is compounded by the fact that user definitions and perceptions of technology vary across countries. Therefore, over the medium and long term, measuring experience, measuring consumers’ satisfaction levels, insulates indicators from changing technology and its varying nomenclature. It is a complementary more qualitative factor that has to be taken into account.


16

2 Introduction

2.1 Topic area definition Problem description The topic of telecommunications and access is a physical cornerstone of the information society. It covers both the physical networks over which information is carried, as well as the products and services used to access those networks. It is also ‘horizontal’ as it cuts across many of the other priority action lines of the eEurope initiative. In many ways the topic can be considered as the fundamental ‘enabler’ – it allows the other eEurope domains to ‘happen’. Focusing on the physical networks as a whole, achieving a reality of eventual universal access and actual use of potential converging technologies and content is an extremely complex topic. It relies on a number of technological, environmental and socio-economic conditions in Europe to aid its development. It is currently being hampered by a number of factors including the narrowband legacy, lack of appropriate content and competition5 to name, but a few. In moving towards universal access, broadband is probably the most important enabling technological development of the current time, and it is, therefore, imperative to measure who has access to it, and what it is being used for. Also, although convergence is almost universally accepted as eventually certain to happen, this meta-process is still in its infancy. Its nascent nature is still difficult to predict so the impact on businesses and consumers on it is also hard to measure. There are also the issues of the digital divide, which encompasses problems associated with, the availability of telecommunications infrastructure, associated costs, linguistics, culture and other socio-economic conditions. The digital divide, universal access and the convergence of technologies are the large hurdles that must be overcome before Europe will truly be a digital society. Telecommunications and access does not stop with infrastructure, there are already great changes and innovations taking place in accessing Internet technologies, mobile telephony, multimedia and other emerging technologies, all of which contribute to the current state-of-play in Europe. Although in many cases the Internet is becoming pervasive, access and usage patterns do vary according to socio-economic and geographic factors. Mobile telephony is also an exceptionally fast-growing sector. Understanding their meta-processes, the hurdles and how technologies are adopted and used requires measurement. While some indicators do exist to allow measurement there are important gaps in this existing data. The SIBIS telecommunications and access indicators should go someway towards helping understand the factors and actions that will influence current progress towards an eEurope.

5 Even though it is two years since Member States were required by EU law to open their local exchanges to

enable competing operators to provide DSL latest figures from February 2003 the European Competitive Telecommunications Association (ECTA) revels that still less than 5% of the nearly 200 million telephone lines in the EU are equipped for DSL. Moreover, over 80%current DSL exchanges are incumbents ones


17

2.1.1 Framework for assessing the area For the purposes of this study the term 'telecommunications' has been interpreted very broadly to include all the networks (cable, mobile, Internet, as well as copper wire), over which all types information (voice, data, sound, image) is carried. So, although the main focus is on telephony networks, computer networks, the Internet, cable (TV as well as telephony), wireless forms of transmission are also included. Overall, a more accurate descriptor in these circumstances would be ‘communication networks’. ‘Access’ is another loose defined topic. It can be defined formally as ‘the ability to retrieve data, graphics, sound, text etc whether on-line or off-line’. Translated into the context of eEurope this topic covers the wide range of devices by which users access electronic ‘information’ – e.g. computers, telephones, multimedia kiosks, televisions and other hybrid devices. In terms of defining the statistical boundaries within which this study is conducted, fixed telecommunications networks have been in existence for over 100 years, so there has been plenty of time for statisticians, users and the industry to have developed indicators. These indicators typically measure the size and growth of the market and different technologies as well as being used as an aid to predict revenues, profits, universality and potentiality. The process of evolving measurement and indicators was simplified by the fact that telecommunications in most countries was developed by a state monopoly. With just one provider it was easier to measure output. However, newer networks – competitive fibre-optic networks, wireless, the Internet and cable, – have not been subjected to such long-term scrutiny. Most of the underpinning technologies for these networks have only been developed in the last 25 years. The networks are also provided by multiple organisations often cooperating and competing with each other at the same time. Monitoring output becomes a much more difficult process. Identifying the potential user base has also become more difficult. No longer are telephone lines supplied to just two groups - business and residential customers. Today an array of network products and services are targeted at multiple niche users groups who will consume more than one service. As the physical communication infrastructure underpinning these new networks is expensive to develop and deploy it requires large amounts of capital to be invested. The demands of the capital market have accelerated the need for 'up-to-date' data about market coverage and penetration of these newer networks. While previously, the telecommunications industry was used to gather, analyse and publish data on an annual basis, the financiers of the new networks were seeking data on a quarterly basis in order to comply with financial reporting requirements of capital markets, in particular the US. Thus while the data collection cycle was compressed, the amount of data produced increased. The growth imperative associated with the Internet boom also led to measures being abused. An Internet service provider (ISP) can count the number of users it has in different ways: the number of users; the number of accounts; the number of active accounts in the last three months; and the number of accounts including those where the user has requested cancellation, but which have not been processed. Indeed, some governments also used such definitional gymnastics to define an 'online citizen' as someone living within three kilometres of a public Internet access point. Indeed, although basic indicators for public switched


18

telephone networks (PSTNs) are widely available, commonly collected data from different sources can still be conflicting. More sophisticated and elaborate indicators such as compound indicators are rarely available due to the problems associated with combining methodologies. Similarly, although basic indicators exist for newer technologies, (and these are becoming more common) in many instances they are also often not comparable, nor yet ready to meet the challenge of emerging topics of interest. This is because of the rapid cycle of technology adoption. New network and access technologies are introduced and are measured before the market has had time to accept or reject it as valid technology. For example, VOIP is a technology that was first mooted in the mid-1990s, but even today it is a still a poorly-adopted technology. However, the lack of a large enough user base, that is statistically representative, does not deter some from modelling and projecting the user behaviour and comparing it with other access technologies. Such comparisons are not reliable bases. The same issues apply to access mechanisms. Although telephone (fixed, mobile) and television ownership rates are well known, there is less information on the extent to which newer forms of access mechanism are available or used. New channels include digital TV, Internet-enabled phones, and interactive TV. Emerging channels will include the new generation of 3G products.

2.2 The approach As there is already a lot of statistical data in the public domain, the study takes a dual approach. The SIBIS project has researched, among others6, the topic of telecommunications and access and identified a selection of current knowledge gaps that cannot be plugged on a pan-European levels with public domain sources. A General Population Survey, covering all 15 Member States, Switzerland and the US commissioned by SIBIS in April 2002 was used to pilot test a number of the new indicators in the general population and collect baseline data to fill these gaps7. It was not possible to pilot all the identified indicators in the survey, but a large number have nevertheless been included. The indicators that have been collected through the SIBIS survey include a range of availability, access, usage and impact indicators. This data can be custom analysed and thus, elicit primary data that is not easily available elsewhere. In addition, the study looks at combining existing indicators and measures that are in the public domain in order to examine the usefulness of new compound indicators8. Constructing compound indicators can, however, be methodologically risky. For example, it is quite likely that the base indicators have not been calculated on the same basis, or using the same sample. There is also an issue of timeframe in such a fast moving areas where combining current data with data produced even 6 months ago may not be useful. However, compound indicators can be very rewarding, useful and interesting. While they may not irrefutably demonstrate a cause and effect relationship, they can be indicative of something, or a propensity to do something – and can often provoke further questions; the answers to which can be particularly illuminating.

6 EGovernment, eHealth, eEducation, eCommerce, Work Employment and Skills, Internet for R&D, Social

Inclusion, Trust and Security 7 This GPS included questions from all 9 topic areas. 8 This may also include new SIBIS indicators


19

2.2.1 Identification of the stakeholders and their interactions In order to give a framework to this study it is important to identify who might potentially be interested in these new indicators and also use them in some way in existing surveys. One of the main aims of SIBIS is to be used as a test bed for innovative indicator development that can be picked up by official bodies. The project is not only aiming to provide results of a pan-European survey, but pilot these new indicators to test their validity so that they can be used elsewhere. The project is providing a complementary resource, which can be exploited by relevant stakeholders across Europe. The main groups of stakeholders involved in indicator generation and use in the field of telecommunications and access are:

Table 1: Stakeholders and Indicators Indicator Generation Indicator Use Statistical Agencies – National and European, ICT Observatories,-National and European

Industry – Telecommunication operators, equipment manufacturers

Policy bodies Telecommunication Regulators Analysts and Consultants Policy bodies Industry Analysts and Consultants Regulators Users, consumers, watchdogs and consumer

groups Data Publishers – IDC, Private enterprises and public institutions Measurement companies (such as RIPE or Netwizards)

Private sector, public institutions, and individuals (citizens).

There is a lot of formal interaction between certain groups – especially between regulators and industry (statutory obligations) and regulators and policy-makers (a key channel for policy making). Given the economic strength of the telecoms sector and the perceived European lead in 3G telecoms, there can also be strong relationships between analysts, consultants, data publishers and the industry. In many instances, information transfer is quite open and transparent. Indeed, some measurement and analyst firms are respected as reliable independent data providers that have inform policy within both public regulators and private corporations over a number of decades.

2.3 Overview of the report This report presents the research on telecommunications and access indicators carried out by the SIBIS project, and the results of testing many of these indicators in the SIBIS General Population Survey (GPS) carried out in April 2002 in the EU 15 Member States, Switzerland and USA. Chapter 3 defines the scope of the analysed topic and the main issues concerning statistical measurements in this field. A description of main stakeholders and their interactions provides the background of the research. Chapter 4 presents the conceptual framework used to develop SIBIS indicators and discusses the motivation of their development and their contribution to the understanding of telecommunications and access development. Chapter 5 and 6 presents the survey results and also presents some combined compound indicators to look at the validity of combining different questions to illicit further results and


20

analyses. The analysis is conducted to determine whether the various indicators implicit in the survey questions are closely correlated and can be combined to form compound indicators. Chapter 7 discusses some of the gaps still remaining in the topic of telecommunications and access and some of the interesting indicators already identified that were not piloted in the General Population Survey Chapter 8 outlines the main conclusions resulting from research


21

3 Identification of the indicator framework and hierarchy A comprehensive review of existing statistical indicators, and the problems associated with collecting data has already been dealt with extensively prior to the production of this report in WP1 and WP29. This chapter deals solely with the choice of piloted telecommunications and access statistical indicators, which are useful and relevant for citizens, business and the rest of society.

3.1 Traditional indicators for telecommunications and access Part of the work leading up to the piloting of indicators in the SIBIS General Population Survey (GPS)10 included the collection and description of over 500 relevant indicators which were considered to be useful and/or frequently cited to measure the Information Society, particularly relating to telecommunications and access. Access, Use and Impact indicators are the main focus of the analysis undertaken through the SIBIS GPS survey. This report also includes a small selection of existing indicators, which come from external and Internationally well-recognised sources which have been chosen to complement the work presented. As stated previously in the project, there is no inclusion or subsequent analysis of indicators of market size or of digital content, which constitutes a topic that should be studied in its own right. Also some policy issues do not lend themselves particularly well to statistical analysis and are, therefore, not included under the scope of SIBIS. These include, for example, the speed of unbundling the local loop or discussions on radio spectrum frequency or policies regarding the sale of 3G/wireless Internet licences. This is because whilst it is always possible to arrive at a ‘number’ of something, the real value lies in the contextual, explanatory information.

3.1.1 Why access indicators are currently more popular than usage/impact indicators

The Information Society relies fundamentally on device user bases. Unless PCs, PDAs, phone or kiosks are purchased, installed and used, no network effect can occur. Traditionally the IT manufacturing industry has relied on market indicators associated with shipments. It is simpler to count boxes at a finite number of manufacturers than to identify the many organisations and homes where multiple users may actually use them. On the software and service side of IT a similar mentality has prevailed. Licences and accounts have been easier to count than what people do with a product and service. In the telecommunications and access universe this mentality has been adopted to measure ISP accounts, number of phone lines and mobile phone handset sales. They are relatively simple proxies for access. Some user surveys can be conducted to identify a multiple-user figure and from this a measure of access can be made. 9 For more information, refer to the website www.sibis -eu.org 10 The SIBIS Decision Makers Survey took place in parallel but was more extensively used for other topic areas

such as eCommerce


22

Identifying usage patterns and impacts is much more difficult. Usage must first be observed over a time period. It then needs to be classified. Then the classifications must be applied to the observed data. This is a time-consuming process. The evolution of usage behaviour means that classifications can become increasingly out of date or too broad. As user numbers increase so does the depth of content available to them. Re-classification may be required to sub-divide definitions. Comparing classifications over time can be problematic as these changes must be back calculated to compare like with like. Sometimes this is just not possible if a statistically representative sample is required.

3.2 New SIBIS indicators selection Because the area of telecommunications and access is so diverse and so many indicators already exist, it is difficult to highlight just one or two areas to investigate. Because of this it is also tempting to create numerous compound indicators, to combine two or more items of existing information, rather than creating new indicators to deepen knowledge and understanding of the subject. However, as telecommunications and access is such a fast changing area, if only the compound approach is adopted there is a risk of omitting key data, such as the adoption rate of new technologies or new access mechanisms. This type of information requires old-fashioned methodology (such as counting the instances), but applied to new items. Therefore, the study pursues both approaches: the creation new innovative indicators and the testing of compound indicators. In order to follow a systematic approach the indicators are organised in a hierarchical way, grouping them on different levels and relating them to sub-topics and issues. Thus, in order to address the highlighted difficulties, the report suggests using three general areas or domains for defining new indicators: � Access to new technologies � Usage of new technologies � Impact of new technologies An earlier Topic Analysis report (WP2.1), provided a framework for the development of the indicators by breaking down the topic into a structure of several sub-topics and issues related to these sub-topics. To further refine the presentation of these indicators, the following “hierarchical trees” were developed with several levels of branches, which presents a picture of indicators that relate to one (or maybe several) of these branches. It became clear during the project that with regard to the costs of data gathering, an indicator is more efficient the higher it is located in the hierarchy. But, on the other hand, indicators further down the hierarchy tend to be more precise and may allow more flexibility in composing indices. A tree has the following levels: � Topic level. Access, Usage, and Impact of new technologies. � Sub-topic level. The technologies piloted in SIBIS are mainly Internet (all connection

methods), Broadband, and Mobile telephony. � Constructs. These are terms for theoretical concepts which can not be measured

directly (e.g. satisfaction), but have to be operationalised via indicators that are capable of measuring key aspects of the construct.

� Construct dimensions. Sometimes, constructs need to be broken down into different dimensions to make operationalisation easier (e.g., the notion of ‘satisfaction’ might contain four dimensions: cost, quality, reliability and speed)


23

� Indicators. These are the indicators we might use to measure the construct dimension � Sub-Indicators. These are more detailed indicators which address a particular sub-

domain or aspect Hence indicators which are being developing under the scope of SIBIS are grouped into: Broadband/High Speed Networks – this is probably the single most important enabling technological development of the current time and it is, therefore, imperative to measure who has access to it and what it is being used for. The indicators here include cost, availability, take-up and consumption of broadband technologies, usage, quality of service provision and regulation.

Internet – although in many cases the Internet is becoming pervasive, access and usage patterns do vary according to socio-economic-geographic factors. Proposed indicators include how the Internet is being accessed (location, bandwidth and multiple platforms users), purposes for which the Internet is used, quality of service provision.

Mobile – as an exceptionally fast-growing sector, mobile telephony is important. Indicators are shown for cost and awareness of cost, access, use and benefits of use, satisfaction with mobile telephony, access to, and use of 3G telephony. Overleaf is a diagram including all T&A questions piloted in the SIBIS surveys.

Figure 4: Overall Framework of Telecommunications and Access indicators piloted in the SIBIS Survey

Telecom m unications andAccess

Usage

A1 1, A1 2

ImpactAccess

Broadband Internet Mobi le Broadband Internet Mobile I nternetBroadb and Internet

A2 3, A2 6,A2 7

A1 8, B1 -5 A30A13 A3 , A4,A6, A7, A 8,

A9, A10 ,A15, D1

A1 , A5, A 6,A1 1, A 14,

A19, A2 0 Notinclud ed

A1 – A30 correspond to the consumers survey questions piloted in the GPS specifically for telecommunications and access included in Work Package 3. Where others are mentioned

Mobile


24

(B1-5 and D1), these are questions piloted as part of other SIBIS topics are also relevant to this topic11. Below is a table including a selected list of T&A indicators, which was created by looking at the most interesting indicators which were considered to be of primary importance for tracking the development of telecommunications and access on a Pan-European level. Out of the 18 indicators proposed below, 15 have been created using data directly piloted in SIBIS GPS. Additionally, 1 indicator ('impact of broadband technologies'), has been analysed using external data commissioned by Oftel in the UK.

Table 2: Indicators of relevance for SIBIS Selected New Telecommunications and Access indicators relevant for the topic

No. Indicator Name Domain Sub-Domain Cross-Tabs

Piloted in the survey

1 Share of at home Internet users according to type of bandwith (broadband, midband and narrowband

SIBIS Broadband By country and age groups

X

2 Degree of Broadband technologies take up SIBIS GPS Broadband By country

X

3 Migrators to faster than dial-up connections Access Broadband By country

X

4 Share of Internet users who access the Internet from different locations -- (a) at home (b) at the workplace (c) at an educational institution (d) at free PIAP (e) at commercial PIAP

Access Internet By country and age groups

X

5 Intensity of online usage Access Internet By country

X

Internet drop-outs - Share of persons who used to have Internet access at home, and do not have it anymore

Access Internet By country

X

7 Degree of Internet access through mobile phones

Access Mobile By country and age groups

X

8 Users according to online tenure (share of users according to length of time since first use of the Internet)

Usage Internet By country and age groups

X

9 Degree of e-mail networking intensity Usage Internet By country and age groups

X

10 T&A Degree of multi-device online use. (a) Digital TV (b) PDA/ palmtop (c) mobile phone (d) others

Usage Internet By country and age groups

X

11 Degree of mobile networking intensity: Degree of mobile ownership networking among friends and relatives

Usage Mobile By country

X

11 There is potential overlap with all the other SIBIS topics. Telecommunications and access forms a baseline set

of indicators that can help with the analysis of numerous other indicators that are being developed.


25

No. Indicator Name Domain Sub-Domain Cross-Tabs

Piloted in the survey

12 Degree of SMS use according to different purposes (communication, transactions, downloads, news subscriptions)

Usage Mobile By country and age groups

X

13 Use of mobile phones abroad Usage Mobile By country

Not piloted

14 Consumer perceptions of broadband

Impacts Broadband By country

Not piloted

15 Barriers to Internet usage Impacts Internet By country

X

16 Compound Indicators: Internet Snapshot Impacts Internet Cluster by countries

X

17 Compound Indicators: Mobile Snapshot Impacts Mobile Cluster by countries

X

18 Compound Indicators: Migration Snapshot Impacts Broadband Cluster by countries

X


26

4 Analysis of Data A large number of telecommunications and access indicators were piloted in the SIBIS survey. Although these were still a subset of the number that was originally presented to go forward in previous SIBIS work, they nevertheless constitute a large quantity of data to analyse. This chapter presents the indicators developed in SIBIS and piloted in the GPS that are considered to be of primary interest (presented in table 2). A number of the un-piloted indicators will be presented in section 5 when looking at ‘gaps’ and topics for future work, and in the forthcoming SIBIS indicator handbook, where all relevant indicators piloted and not piloted will be presented. Here we are concentrating our work on the indicators already piloted.

4.1 Analysis of indicators about citizens and society

4.1.1 Access indicators Degree of broadband technologies take-up Although this is a relatively straightforward indicator, it is fundamental to measure the share of Internet users who have access to broadband. Several other indicators stem from it and, therefore, it was necessary to pilot it in the SIBIS survey. One of the principal features of broadband in Europe is its diversity. Therefore, the question had to reflect the many broadband access methods available across the participating countries. Hence respondents to the SIBIS questionnaire were quizzed about the type of Internet connection they access at home. Possible responses were; dial up modem, ISDN, satellite, cable modem, xDSL, leased line, multiplex (T1/T3) or other. It is no surprise that dial up modems were the most popular method of at-home connection. This is classified as narrowband as it less than 64Kbps. Further classification of the responses according to bandwidth is a contentious process as definitions of bandwidth still vary. The minimum bandwidth rate suggested for classification as broadband is >128Kbps, though some classify >2Mbps as the threshold. The threshold is important because some definitions of broadband include ISDN access that has an upper bandwidth of 128Kbps. In this assessment, ISDN has been classified as midband and the definition of broadband as >2Mbps has been adopted, in accordance with the practice of Oftel, the UK telecoms regulator. Broadband services are currently primarily provided through DSL and cable modem services. Alternative technologies such as fixed wireless access, electric powerlines and satellites, are anticipated, but have not been fully brought to market yet. This may be because of the introduction of cheap-rate ADSL over the last year and a higher penetration of other broadband technologies such as cable operators which currently have a large base of subscribers in the UK, making it one of the most competitive markets in Europe in terms of existing broadband infrastructures. Competitiveness in the broadband market in Europe is still generally low. Most users upgrading to broadband are subscribing to incumbents deals. Even though it is two years since Member States were required by EU law to open their local exchanges to enable competing operators to provide DSL latest figures from February 2003 from the European Competitive Telecommunications Association (ECTA) revels that still less than 5% of the nearly 200 million telephone lines in the EU are equipped for DSL. Moreover, over 80% of current DSL exchanges are incumbents ones.


27

According to ITU12, one of the most interesting statistics used for comparison is the percent of Internet subscribers that are broadband users. The results of a recent ITU study indicate that broadband users almost always started by using a narrowband Internet connection. Thus, Internet users can be considered to be the population from which broadband users are drawn. The ratio of broadband users to total Internet users is the fraction of the relevant population that has adopted the innovation of broadband.

Table 3: Share of Internet users according to type of connection at-home by country

BB

broadband (e.g. DSL)

MB only

"mid"band (ISDN)

NB Only

narrowband (dial-up modem)

only an access type

not mentioned

DK No Access at home

% % % % % % AUSTRIA 11.06 6.48 17.96 1.33 10.26 52.91 BELGIUM 15.57 0.84 10.92 1.86 10.90 59.91 DENMARK 15.70 7.09 29.94 0.40 11.09 35.77 FINLAND 6.46 7.89 26.22 0.55 16.75 42.14 FRANCE 5.45 0.28 16.20 0.14 8.04 69.89 GERMANY 8.90 17.64 12.30 0.10 10.65 50.41 GREECE 1.53 1.76 8.38 0.17 4.97 83.19 IRELAND 1.87 2.97 25.77 1.29 20.69 47.41 ITALY 2.94 1.23 23.16 0.11 13.11 59.45 LUXEMBOURG 2.59 19.24 16.99 0.55 10.19 50.44 NETHERLANDS 15.12 13.20 34.30 0.10 9.69 27.59 PORTUGAL 3.31 1.17 10.60 0.38 5.97 78.57 SPAIN 5.76 0.48 13.81 0.21 10.00 69.72 SWEDEN 15.44 2.48 33.05 0.31 15.09 33.64 SWITZERLAND 9.53 10.17 19.72 1.04 19.84 39.70 UK 9.82 2.36 20.80 0.30 25.04 41.67 USA 16.57 0.10 34.56 0.35 11.88 36.54

EU15 7.50 5.68 17.96 0.26 12.73 55.87

BB= Broadband (includes cable, leased line, xDSL, T1 and T3, and Internet via Satellite technologies)

MB= Mid Band (ISDN)

NB = Narrowband

Other access type not mention (e.g. mobile)

Source: © SIBIS 2002, GPS

Base: all respondents (N=11,832), weighted; EU15 results weighted by EU15 population (N=10,306). The above figures illustrate the level of broadband enabled penetration among at-home users in European Member States, Switzerland and the US. According to the data, in the European Member States (EU15), approximately 7% of at-home users’ connections have broadband. A new report by Forrester Research confirms SIBIS figures. According to this study, by the end of 2002, the number of Western Europeans with broadband Internet access at home had climbed to six million, or seven percent. This is a low figure if compared to penetration of BB technologies in countries outside the EU, such as Japan or the US. In 2001, eMarketer research, estimated that broadband was adopted by less than 6% of Japanese households, but it found that by the end of 2002 broadband was in 15.5% of

12 ITU Internet Reports 2002: Internet for a Mobile Generation.


28

Japanese households13 which is just shy of the US at 16% last spring. While behind the US, which has around 16%14, there are individual countries in the EU that seem to have equally high levels of broadband penetration. For example, Belgium which benefits from a dense urbanised geography where access to telecoms networks such as cable is relatively easy. Similarly, Austria is high and is dominated by the Vienna urban area which is also highly cabled. In other Member States the midband technology of ISDN is popular. Germany for example has an ISDN penetration level of over 17% among the at-home Internet connections. This may be due to an aggressive marketing of ISDN services to consumer customers of Deutsche Telekom since the late 1990s. In contrast, the UK has an ISDN penetration of around 2%. This may be because ISDN was primarily marketed as a business product and had high associated installation costs. The lesson from midband is that consumer can and do migrate to better access technologies.

Table 4: Percentage share of respondents by access type at home according to age group

Age groups (according to Eurostat classification) up to 24 25 to 49 50 to 64 65 and more

broadband (e.g. DSL) Frequency 203.24 421.15 120.04 28.70 Column% 12.31 9.17 5.43 1.56

only "mid"band (ISDN) Frequency 91.43 378.66 100.10 11.09 Column% 5.54 8.24 4.53 0.60

only narrowband (dial-up modem)

Frequency 392.72 1079.90 300.56 74.41

Column% 23.79 23.51 13.61 4.05 only an access type not mentioned

Frequency 3.85 13.88 6.30 2.86

Column% 0.23 0.30 0.29 0.16 DK Frequency 231.52 627.70 342.67 107.96

Column% 14.03 13.67 15.51 5.87 no access at home Frequency 727.83 2071.70 1339.29 1614.30

Column% 44.09 45.11 60.63 87.77

Source: © SIBIS 2002, GPS

Base: all respondents weighted by EU15-population (N=10,306) When looking at age group comparisons, it is important to note that the majority of SIBIS respondents older than 50 years old do not have an Internet connection at home. This is particularly true for respondents older than 56, where 87% respondents do not access it from home. On the contrary, the majority of SIBIS respondents younger than 50 years old have an Internet connection at home.

13 Ministry of Public Managem ent Home Affairs, Post and Telecommunications, Japan, January 2003 14 Currently and according to ITU (December 2002) the number of USA households connected via BB at home

surpasses 20%


29

Figure 5: Type of Internet connections at home by age groups

At home internet connections according to age groups in EU-15(in % of population)

0

5

10

15

20

25


broadband (e.g. DSL) only "mid"band (ISDN) only narrowband (dial-up modem)only an access type not mentioned DK

Base: all respondents weighted by EU15-population (N=10,306) Source: © SIBIS 2002, GPS

Hence looking at Internet connections at home according to age groups, SIBIS results have portrayed that people between 24 and 49 years old show higher adoptions rates of both high speed and slow connections at-home. This correlates well with other indicators. For example, the young are also more likely to migrate from low-speed to higher speed connections and are traditionally described as early technology adopters. It is also the young who tend to be more interested in downloading digital media.

Migrators to fast-speed connections The analysis of the number of at-home Internet users who have migrated from slow dial-up connections to a faster one (be that a broadband or midband connection), is of interest for a variety of reasons. It could indicate a certain level of sophistication in usage or perhaps indicate which countries have cheaper higher-speed connections. It is a good indicator to combine with other indicators including the cost of different connections, pricing packages, usage patterns and the availability of plug-and-go connections.


30

Table 5: Percentage of migrators from a slower connection to a higher speed connection by country

Country Migrator (%) Non-migrator (%) AUSTRIA 41.6 57.1 BELGIUM 48.4 46.0 DENMARK 82.0 17.1 FINLAND 63.3 36.7 FRANCE 54.9 43.0 GERMANY 43.7 55.6 GREECE 27.2 72.8 IRELAND 39.8 60.2 ITALY 55.4 32.0 LUXEMBOURG 44.9 51.2 NETHERLANDS 64.8 34.2 PORTUGAL 68.8 27.2 SPAIN 75.3 24.7 SWEDEN 83.9 16.1 SWITZERLAND 44.4 55.6 UK 43.1 56.1 EU15 51.1 47.3 USA 79.3 19.9

Question: At home, did you have a connection before which was slower than your current one? Base: respondents with Internet fast connections at home (midband and broadband), who used to subscribe before to a slower service (N=1,621), weighted. EU15 results weighted by EU15 population (N=1,376). Source: © SIBIS 2002, GPS. Column percentage. The figures suggest that migration to a faster connection than dial-up is highest in the US, Sweden, Spain, Finland, the Netherlands and Denmark. The distribution of non-migrators, i.e. those who have not in the past had a slower connection and, therefore, have most likely signed up straight away to a midband or broadband connection, appears to be concentrated in the UK, Switzerland, Luxembourg, Ireland, and Austria. The next step in indicator development in this area would be to explore the satisfaction levels of people with connections, why they migrate and what barriers are still present. Access to the Internet from multiple places This indicator measures how long respondents spend, and from where they access the Internet. The nature of this indicator is to move beyond counting instances to measuring time factors. The data collected from this indicator can be presented in a number of ways. It is clear from the charts than in the most mature European Internet markets (DK, S, N, FIN, UK and CH), as in the USA there is a large amount of multi-context users who access the Internet from both at home and at work locations. Whereas in the other countries Internet users access the Internet generally from only one location, whether at home or at work. SIBIS also shows data that confirms most users access the Internet principally from home. However, although the reference is to access the Internet at home, the data also shows that there are established patterns of ‘bimodal usage’ as many users appear to access the Internet from more than one location. In the US as in Scandinavian countries, the UK and the Netherlands there is a large proportion of these ‘bimodal users’ who access the Internet both at home and at work.


31

Figure 6: Internet users accessing the Internet from different locations

Internet usage by location(in % of population)

0

10

20

30

40

50

60

70

80

90

100

DK S

NL

FIN

UK

IRL D A L

EU

-15 B I E F

EL P

CH

US

A

occasional usage (less than once a month)

usage only somewhere elseusage only at workusage only at homeusage at home and at work

Question: Have you accessed the Internet in the last four weeks from… Base: respondents with Internet access at home who have us ed the Internet at least once in the last 4 weeks (N=4,631). EU15 results weighted by EU15 population (N=3,838) Source: SIBIS 2002, GPS The indicator also looks at whether at-home users access the Internet from other locations outside of the home and work, and, if so, how much time was spent accessing ‘elsewhere’ in a typical week. The choices were: 1 At-school 2 In public places with free internet access 3 In public places with paid Internet access 4 Another place Response rates were low and, therefore, it is difficult to draw accurate conclusions for each category. Such nomadic access is usually associated with life events such as studentship, vacations or long term travel abroad, i.e. periods of life when there is no long-term residence. Here it is important to highlight the fact that telephone survey methodologies do not include responses from people younger than 16 years old, and, therefore, at-school responses were low. This might have affected responses for ‘In public places with free Internet access’ and ‘In public places with paid Internet access’. Different access locations can lead to different on-line activities and there are potentially many disadvantages when compared to at-home access. In addition to being able to access information at-home at any time, a person may be less likely to access, for example, personal health or financial information from a library or other public facility. A recent study in the USA15 shows that ICT use by individuals is strongly affected by use in the workplace. Not only are individuals who work in certain occupational areas more likely to use a computer and the Internet, but exposure in the workplace makes an individual substantially more likely to use a computer and the Internet. For example, 78% of 15 US Department of Commerce, 2002


32

households in which at least one family member uses a computer at work have a computer at home. In contrast, only 36% of households in which no family member uses a computer at work have a computer at home.

Figure 7: Intensity of online usage in different locations

Intensity of online usage(in % of population)

010

20

3040

50

60

7080

90

100

DK

UK S

NL D B

EU

-15 A L

FIN E

IRL I P F

EL

CH

US

High (over 6 hrs/week) Medium (between 1 and 5 hrs/week) Low (less than 1 hrs/week)occ. user (not asked) non-Internet user

Question: How much time do you spend in a typical week on using the Internet at home? Base: respondents with Internet access at home who have used the Internet at least once in the last 4 weeks (N=4,631). EU15 results weighted by EU15 population (N=3,838). Source: SIBIS 2002, GPS With regards to the intensity of online usage, the time spent on-line shows differing patterns across all the European countries surveyed. Scandinavian countries and the UK show a similar pattern to the US where the majority of users spent on-line for up to 5 hours per week. On the contrary, the EU15 average, and notably in Southern countries, particularly in Spain, the majority of users do spend lesser time using the Internet. The USA shows a high level of at-home use, but have the advantage of free local calls, encouraging more frequent connection to the Internet. In the United States, use in all types of locations is rising. Similarly, a potential emerging area for measuring an alternative Internet location is Wireless LANS which are being increasingly deployed for providing Internet access in specific locations. Commercial Wi-Fi 'hot' spots are springing up across a number of European countries at locations such as railway stations, airports, business parks and coffee shops. There are already several thousand hot spots across the US and numbers are expected to grow rapidly in Europe. As wireless 'hot' spots in Europe become more prevalent, it would be interesting to ask about hotspots levels of use in public places such as airports, restaurants, etc where access has been provided. This needs to be distinguished from other free PIAPS.


33

Figure 8: Broadband access and intensity of online use

Broadband access and intensity of online usage(in % of population)

A

B DK

FINF

D

EL IRLI L

NL

P

E

S

CH UK

US

EU-15

0

5

10

15

20

0 5 10 15 20 25 30 35

Intensive users (6 hrs and more per week)

Source: Technopolis analysis of SIBIS GPS 2002 The figure portrays a clear association between having a broadband at home connection at home and intensity of Internet use. SIBIS shows how in countries with higher broadband penetration rates at home, users tend to spend longer online per session when compared with narrowband users. This indicates not only that broadband encourages users to use the Internet more, but also that broadband technologies allow them to get a greater and more effective online experience than narrowband users. Those who move to higher bandwidth seek a better online experience, for example, through quicker downloads or always on connection that enable better exploitation of the potential benefits of the Information Society

Internet Drop Outs This indicator looks at the percentage of people in the EU Member States, Switzerland and the US who used to have Internet access at home and have now stopped their connections. There could be a number of reasons for this, for example, cost, ‘Internet fatigue’, faster access at work. It could also be the case that the respondent falls in a transition phase and has cancelled a narrowband connection, and is waiting for a faster one to be installed. This is not a definite drop-put, but a migrator waiting to get a faster connection. Hence what is needed in future work is some indication of why people drop out (cost, lack of interest, using elsewhere) and whether the change is temporary or not. In any case, if this indicator returns high levels it is worth investigating. The highest number of Internet dropouts is found in the USA with 20.2% of respondents saying they used to have a connection at home but do not anymore. In Europe, the countries with the highest drop out rates are the UK, Sweden, and Denmark, all early adopting nations where users have had time to experience a full cycle of Internet use.


34

Table 6: Percentage of Internet dropouts by country

Country Yes No AUSTRIA 8.1 91.9 BELGIUM 6.6 93.4 DENMARK 9.5 90.5 FINLAND 6.3 93.7 FRANCE 3.3 96.7 GERMANY 3.9 96.1 GREECE 3.8 96.2 IRELAND 6.9 93.1 ITALY 2.8 97.2 LUXEMBOURG 3.8 96.2 NETHERLANDS 7.8 92.2 PORTUGAL 3.4 96.5 SPAIN 4.7 94.7 SWEDEN 13.7 86.3 SWITZERLAND 6.8 93.2 UK 12.2 87.5 EU15 5.09 94.7 USA 20.2 79.8 Question: Did you once have Internet access in your home? Base: Share of respondents with formerly Internet access at home (N=6,266), weighted. EU15 results weighted by EU15 population (N=5,635). Source: © SIBIS 2002, GPS. Column percentage.


35

4.1.2 Usage indicators Users according to on-line tenure This indicator measures share of users according to length of time; tenure, since first use of the Internet. The SIBIS survey classifies tenure into three bands; less than six months since first Internet usage; more than six months, but less than two years; and more than two years. According to the SIBIS results, ‘on-line tenure’ is highest in the US, Scandinavian countries, Germany and UK, and is generally lower in the Mediterranean countries. These Northern European countries have more experience of usage and also more mature Internet markets. There are also other issues of culture and language. Therefore, it can be expected that Northern European countries have higher rates for Internet penetration rates, e-commerce transactions and e-skills aptitudes. It would be rewarding to compare broadband subscribers with on-line tenure, as well as on-line tenure to different age groups, regions and gender issues. Such indicators could be very useful for tracking the digital divide. Following these guidelines SIBIS has included among the Telecoms and Access compound indicators one exploring the correlation between online tenure, migrators to a higher-speed connection, and Internet usage among the at-home population. Thus this indicator portrayed how on-line tenure is an important indicator because depending on how experienced users are with the Internet, the more likely they are to use the Internet effectively and widely. It is also more likely that the more experienced users are, the more willing they are to migrate to a broadband connection.

Table 7: On-line tenure by country

First use of Internet Country < 6 months

ago 6-12 months

ago 1-2 years ago 2 years +

ago AUSTRIA 10.4 6. 4 20.6 62.3 BELGIUM 8.1 8.7 22.3 60.6 DENMARK 1.4 2.9 12.0 83.5 FINLAND 1.4 4.2 7.9 86.5 FRANCE 15.4 12.5 27.9 43.8 GERMANY 5.8 11.9 35.4 46.6 GREECE 9.6 16.6 25.6 47.2 IRELAND 5.3 10.2 28.0 55.7 ITALY 7.2 9.9 24.5 57.6 LUXEMBOURG 18.0 2.7 20.4 59.0 NETHERLANDS 3.3 8.2 25.1 63.4 PORTUGAL 9.2 7.0 17.4 64.4 SPAIN 8.0 12.0 33.3 46.3 SWEDEN 1.8 3.9 11.5 82.8 SWITZERLAND 9.6 8.4 22.3 56.9 UK 4.7 12.5 22.8 59.2 EU15 7.0 10.8 26.9 54.8 USA 3.3 3.4 14.4 78.8

Question: When did you use the Internet for the first time? Base: Respondents who accessed the Internet at least once in the last 12 months (N=6,905), EU15 results weighted by EU 15 population (N=5,828) Source: © SIBIS 2002, GPS Column percentage. Currently, the measurement used for ‘early adopters users’ in SIBIS is over 2 years, a measure that is already taking in a large proportion of the respondents. In the US for example, the majority of users (78%) have two years or more experience. For future work, it


36

would be interesting to change this indicator and class early adopters as those who have had perhaps 5 years or more experience of using the Internet. This would give more leverage to explore and distinguish the usage patterns between the experienced and the non-experienced.

Figure 9: On-line tenure according to country

Experience of online usage(in % of population)

0

10

20

30

40

50

60

70

80

90

100

DK S

FIN NL

UK A

IRL L B

EU

-15 D I P E F

EL

CH

US

2 years and more 1 year - 2 years 6-12 months < 6 months Don't know non-Internet users

Question: When did you use the Internet for the first time? Base: all Internet users (incl. users, who did not use it in the last 4 weeks), weighted (N=6,905). EU15 results weighted by EU15 population (N=5,828). Source: © SIBIS 2002, GPS Degree of e -mail networking intensity For this indicator, respondents were asked to indicate their degree of networking among friends and relatives and the degree of usage of e-mail networks with friends and relatives. E-mail remains a basic tool for interacting on-line and the measurement of e-mail penetration is important. It indicates a good level of Internet literacy and is fundamental to the completion of many on-line processes from registration to transaction. It is already been tracked by Oftel in the UK and by the Ministry of Labour in Finland. E-mail penetration among friends and family was highest in Scandinavian countries and the US, where over half of the respondents’ friends and family have an e-mail address. According to the OECD 2002 ICT Outlook report, in most countries, regular home users use the Internet most frequently for e-mail and information search. In Italy for example, the respective response rates are 63% and 60%. In spite of these high rates it is important to continue tracking this networking intensity among friends and family, especially considering that the project will soon include survey results from a number of candidate countries. It is interesting to note the US networking intensity among friends and family is twice as high as the EU average. One element which SIBIS did not track and which could have been interesting to explore is user usage of multiple e-mail addresses, i.e. how many of the respondents have more than one active e-mail address in order to distinguish between at-home and at-work e-mail networking and usage.


37

Cutting the data by different age groups reveals the majority of e-mail users are within the younger age group 16-25. This figure would have probably been higher if the age group of people younger than 16 years old had been included. Lower rates were recorded among those over 50, but the total figures for e-mail network intensity show that while around 40% of all SIBIS respondents communicated with friends or relatives through e-mail, over 50% claimed not to use e-mail at all. Analysis of PC-ownership and e-mail usage indicates that 15% of PC-owners users do not communicate through e-mail.

Table 8: E-mail usage by age groups (EU15) Age

E-Mail use in last 4 weeks


Total

Usage 54 48.5 27.5 6.4 41.0 No usage (but PC-Users) 24 18.2 13.1 6.1 15.3

No usage (PC-non-users) 21.8 33.3 59.2 87.3 43.5

DK 1 0.1 2 - 0.1 Total 100 100 100 100 100 Question: Have you used or received any email messages for work or for private purposes during the last four weeks? Base: all respondents in EU15 (N=10,306); weighted by EU15 population Source: SIBIS 2002, GPS. Column percentage.

Figure 10. E-mail use according to different age groups:

E-mail users according to age groups in EU-15(in % of population)

0

10

20

30

40

50

60

70

80

90

100


E-mail users No usage (but PC-users) No usage (PC-non-users) don't know

Question: Have you used or received any email messages for work or for private purposes during the last four weeks?

Base: all respondents in EU15 (N=10,306); weighted by EU15 population Source: SIBIS 2002, GPS


38

Users accessing the Internet through other devices than a PC/ MAC This question asked the respondents if in the last 4 weeks they had used any other devices to access the Internet or on-line services, and if so, had they used: � Digital TV � PDA/Palmtop � WAP/GPRS � Games console � Other It is important to study the degree to which different devices are being used for Internet connections. In Europe, due to the current high penetration of mobile phones, the data collected in SIBIS suggests emergent patterns of bimodal usage. However, the sample of respondents presented below is too low to judge country patterns. This work is best analysed as this stage through looking at the differences between the US and the EU average and differences of usage among different age groups.

Figure 11: USA vs. EU15 other connections devices than a PC

Users accesing the Internet at home from alternative devices other than a PC

0

10

20

30

40

50

60

70

80

90

100

Other accessmethod: Digital

TV

Other accessmethod:

PDA/palmtop

Other accessmethod: mobilephone w. WAP

Other accessmethod: game

console

Other accessmethod: other

Other accessmethod: DK

%

USA

EU15

Question: In the last four weeks did you access the Internet in any other way than via PC or Mac at least once? Which devices did you use for that: Did you use ...(multiple response)? Base: Internet-users who access the Internet also by other devices than PC or Mac, EU15 results weighted by EU15 population (N=340). USA results weighted (N=55). Multiple responses permitted Source: SIBIS 2002, GPS In the US at-home users are consistently accessing the Internet through more platforms than in the EU. This is particularly true for PDA/Palmtop, game consoles and digital television access. Digital television access is twice as high in the US than in the EU. Likewise access through PDA/Palmtop devices was four times higher. The results change for mobile phones where the EU shows higher rates of mobile telephone Internet access than the US. The ‘other’ figure appears to be quite high. There could be any number of reasons for this, including semantic confusion, for example, not counting a laptop as a PC.


39

If considering age groups SIBIS has noted that users younger than 25 include the highest rates of early adopters and multi-device users. Particularly for devices such as mobile, digital TV or game consoles. It has to be highlighted at this stage that because the data was gathered using telephony surveys, which do not allow respondents younger than 16 years old to be interviewed, early technology adopters could not be included in the response. Some devices figures such as those for game console users are very likely to be under represented. Degree of mobile networking intensity: Degree of mobile ownership networking among friends and relatives This indicator measures the degree of ownership among friends and relatives. It is interesting to highlight that over 60% of the respondents’ relatives and friends owned a mobile phone in the EU, whereas in the US mobile intensity network among friends and family only reached 40%. Likewise, when looking at the different age groups, people younger than 25 years old accounted for the highest penetration rates, since they stated that over 80% own a mobile phone and over 75% of their family and friends are also mobile phone owners. It is also very interesting to look at the degree of penetration of SMS use since this age group again recorded the highest usage, with 70% of mobile owners below 25 years old using SMS. Degree of SMS mobile data services use for communication, transactions, downloads, news subscriptions

Figure12: Mobile phone owners according to different usage: EU 15, US and Switzerland, April 2002

Mobile usage and usage of SMS and Internet via mobile(in % of population)

0

10

20

30

40

50

60

70

80

90

100

B

DK D EL E F

IRL I L

NL A P

FIN S

UK

EU

-15

CH

US

mobile Internet user only SMS user neither nor, but mobile phone owner

Base: all respondents (N=11,832), weighted; EU15 results weighted by EU15 population (N=10,306). Given that mobile telephony penetration is generally high (more than 65 %) across all EU 15 countries, in order to explore mobile phone usage SIBIS data has been broken by different age groups according to the degree of SMS usage in the last four weeks. Results have portrayed that SMSs are generally popular across mobile phone owners younger than 50 years old. This is specially true for mobile phone owners younger than 25, since the majority of them ( 80%) have used SMSs in the last four weeks.


40

Out of all of the usage options, the most popular use of SMS is for communication, and this is a common pattern among all age groups. Looking at the other types of services that are available via SMS, it would be interesting to further explore the use of SMS for downloads, particularly by age groups. For example, are the young more likely to download ringtones and logos than other age groups? It would also be interesting to look at the types of services used in the area of WAP technology to see how different age groups are using the technology, particularly as the under 25s are the main target user base for future multiple connections methods.

Figure 13: Degree of SMS mobile data services used (communication, transactions, downloads, news subscriptions)

SMSs use according to age groups in EU-15(in % of mobile owners)

0

10

20

30

40

50

60

70

80

90

100

Up to 24 25-49 50-64 65+ Total

Communicate with people Paying for: purchases, tickets or similar

Paying for: downloads, ringtones Receiving subscription services

Question: Have you, in the last four weeks, used SMS messages for…? Base: all mobile user who have used SMS in the last four weeks , weighted by EU15 population (N=4,443) Source: SIBIS 2002, GPS; When looking at different mobile usage patterns according to age groups, it is important to highlight that SIBIS respondents younger than 25 years old are the most active SMS users. In addition, they access the Internet though their mobile phones to a larger degree than other age groups, though this figure still remains below 10%. Subsequently, it is not a surprise that young mobile users are generally also mobile data users, whether using only SMS or also Internet data services. Whereas in other age groups there is a higher propensity of mobile users to use only voice calls and not mobile data (SMS or mobile Internet ). Moreover, a big proportion of respondents older than 50 years do not own a mobile phone for personal use. Hence, for future indicators tracking the proportion of bimodal users accessing the Internet through both mobile phones and computers, it seems that age will play a role when defining future mobile Internet user profiles, being among the main explanatory demographic factors.


41

Figure 14: Mobile phones use according to age groups

Mobile usage according to age groups in EU-15(in % of each age group)

0

10

20

30

40

50

60

70

80


mobile Internet user only SMS user neither SMS nor mobile Internet (incl. don't know) no mobile phone

Base: all respondents weighted by EU15-population (N=10,306),

Source: SIBIS 2002, GPS

4.1.3 Impact Indicators Consumer perceptions of broadband Broadband access to the Internet is often talked about in revered terms by consumers. Many wish to use it, but the reality is, as witnessed by SIBIS's own indicator, that few have access to it. However, it is important to be able to measure the perceived benefits of it and the impact it has on consumers. A comparative benchmarking survey16 in major nations to understand consumer perceptions of broadband has shown that fewer than 20% consider broadband to be 'worse than expected’. In the UK and Italy over 40% of respondents considered it to be better than expected.

Barriers to using the Internet As Europe moves towards universal access to the Internet and investment in public Internet access points and basic training schemes become more prevalent, it is necessary to compile sets indicators such ‘barriers to usage’. If the Internet is available everywhere, why are some people not using it? What are the barriers and inhibitors to adopt broadband technologies?, and why are some people dropping out from it?. From the data, it is clear that personal preference not to use the Internet is one of the largest barriers. Nearly 59% of the survey respondents who did not use the Internet agreed with the statement that 'it was not something for me' (a general sentiment that is a ‘catch all’ which could include opinions that also agree with other statements in the survey). A more specific barrier was a lack of ICT skills. Other specific barriers were that the Internet is perceived as

16 International Benchmarking study commissioned by Oftel, UK, March 2002.


42

expensive and time wasting. It is interesting to note that a lack of suitable content was only cited by around 22%. It would appear that the content gap has been closed.

Table 9: Barriers to Internet Usage: percentage of respondents agreeing or disagreeing with statements describing why they do not use the Internet (include frequency columns)

Barriers to Internet usage Agree completely Agree

somewhat Do not agree

DK TOTAL

It requires advanced computer skills

30.21 27.58 25.30 16.91 100

It is not easy enough to get access to

15.32 20.39 36.83 27.46 100

It is too time consuming

23.38 24.20 25.47 26.95 100

It is too expensive to use

22.45 21.15 20.49 35.90 100

It lacks useful or interesting information

10.76 11.31 50.55 27.38 100

It is not something for me

40.99 16.58 31.74 10.69 100

Please tell me for each statement whether you agree completely, agree somewhat or do not agree. The Internet…

Base: Respondents who have not accessed the Internet in the last 4 weeks, weighted by EU15 population (N=5,321).

Source: © SIBIS 2002, GPS. Column percentage.

Figure 15: Barriers to Internet usage: percentage of respondents agreeing or disagreeing with statements describing why they do not use the Internet

0 10 20 30 40 50 60 70 80 90 100

It requires adv ancedcomp uter skills

It is not easy enough toget access to

I T i s too time consuming

I t i s too expensive to use

It lacks useful ori nteresting inform ation

It is not som ething forme

Percentage

Agree completely Agree somewhat Do not agree Don't know Please tell me for each statement about the Internet whether you agree completely, agree somewhat or do not agree. Base: Respondents who have not accessed the Internet in the last 4 weeks, weighted by EU15-population (N=5,321). Source: © SIBIS 2002, GPS.


43

5 The use of compound indices Consultation with the UK Office of National Statistics confirmed that it is acceptable to construct compound indices, providing: � The data is from recognised and acknowledged sources � That is represents sample sizes/time periods which are sufficiently large/close in order for

the analysis to be meaningful Firstly, one way of creating compound indices is to use the concept of eReadiness/digital divide which is already the subject of several studies, including from the World Bank and the OECD. These often concentrate on developing countries and access to basic ICT. Under the terms of this study it could be instructive to look at patterns of access to emerging technologies across the EU and whether there is a correlation with say wealth, educational or employment status. In addition, a second way of structuring compound indicators is to go back to the conclusions of the research work for SIBIS. This initial work showed that there are already many indicators measuring the existence of something, fewer measuring the use of that technology and fewest measuring the impact or benefit of that technology. Thus, combining them together can give us a ‘snapshot’ index. Thirdly, the way the individual indicators have been grouped in the previous section could also provide compound indicators – just by asking all the questions under one of the headings. Since, as stated above, e-readiness indices have already been developed elsewhere, a second approach to compound indicators construction can also be undertaken. This is through combining variables together which have shown to be relevant to provide a ‘snapshot’ index. Variables included in this analysis have been derived from the SIBIS GPS data results, including innovative indicators regarding use and adoption of new technologies. Following the methodological guidelines provided by the UK Statistical Office, compound indicators have been constructed using exclusively SIBIS data. Originally, there was an intention to develop indices combining SIBIS data with external sources, but either the data provided was not as up to date as SIBIS, or were not sufficiently large in scope, covering only a few countries. Although these indicators, at this stage tend to produce results which are fairly obvious, they are statistically sound and this approach can eventually be extended to combine telecommunications and access indicators with other SIBIS indicators developed under the other topic areas (such as social inclusion, e-commerce, work, employment and skills for example). Moreover, they will gain in importance once they are combined in the second phase of the project on a EU25 Pan-European basis. No other available source right now can provide such a broad snapshot across European countries. In order to combine the variables and study their interactions, a multivariate factor analysis of the data was carried out. The principal advantage of this approach is that it does not rely solely on subjective judgement (it does not involve weighting of variables), but it is primarily dependent on groupings of statistically similar entities. This confers a degree of objective validity to the procedure which is at least beneficial as a starting point, although there is certainly a place for using judgement as to the utility of the results obtained, and for use of ‘manual’ modifications as appropriate.


44

The purpose of this methodology is to: � Improve understanding of the relationships between various aspects of ICT access and

usage � Determine whether the various indicators implicit in the survey questions are closely

correlated and can be combined to form ‘compound indicators’ � To assess whether the survey responses contain information which can be used to

‘explain’ variations in ICT usage in terms of, for example, differences in individuals’ social and economic attributes

� To test for statistically significant differences between countries, and whether there is evidence for distinct ‘clusters’ of nations with similar ICT characteristics within each cluster, but significant differences between clusters.

5.1 The data Use has been made of the full SIBIS survey data set, containing over 11,000 records of the responses of individuals from the 15 member states, Switzerland and the USA. Numerous modifications were done to the coding of the data to make it suitable for analysis. In addition, some questions were unsuitable for inclusion on the grounds that they were only put to a biased sample of respondents. For example, questions relating to the perceived degree of ease/difficulty of use of the Internet were only asked if the interviewee had indicated that they were a user. In addition, records with important omissions, such as ‘don’t knows’, were omitted. These modifications left a data set comprising of some 30 variables and nearly 5,000 records, very adequate for EU-wide multivariate factor analysis. The variables related mainly to PC, e-mail and Internet access and usage, use of mobile telephones, and socio-economic and demographic status.

5.1.1 Compound Indicators Internet Snapshot The first compound to explore comprises Internet access and usage (for both narrowband and broadband connections). The following four original variables are the main components of this factor:

Table 10: Internet usage snapshot Indicator Source

Description

SIBIS GPS Percentage using a PC in the previous 4 weeks

SIBIS GPS Percentage with access to the Internet at home

SIBIS GPS Percentage using the Internet in the previous 4 weeks


45

5.1.2 Mobile snapshot A second compound indicator to explore comprises mobile telephone ownership and what might be termed ‘basic’ patterns of usage and communication that do not involve the Mobile Internet. The following three original variables are the main components of this factor:

Table 11: Mobile usage snapshot Indicator Source Description

SIBIS GPS Ownership of a mobile telephone

SIBIS GPS Having a high proportion of family/friends owning mobiles

SIBIS GPS High use of SMS messages for communication with other people.

5.1.3 Migration to faster Internet Access snapshot (T&A 18) A third compound indicator comprises at home active users with more than two years experience since first use of the Internet who used to have a slow at-home Internet connection before and have now migrated to a faster connection than dial-up. The following three original variables are the main components of this compound:

Table 12: Migration to faster Internet Access snapshot Indicator Source Description

SIBIS GPS Internet At-Home users in the last four weeks

SIBIS GPS Having more than two years online tenure since first use of the Internet

SIBIS GPS Having migrated to a faster connection than dial-up

5.1.4 Multiple regression as a means of examining linkages between ICT usage and other standard variable

Additionally, and in the interest of examining the extent to which ICT usage can be associated with an individuals social and economic circumstances, different demographic variables, such as ‘Terminal education age’, ‘Occupational status’, and ‘age group’, have been selected as independent variables to conduct a Multiple Regression statistical data analysis. Hence, these three variables will be used in an attempt to ‘explain’ plausible variations in different dependent variables related to access, use and impact of Internet and mobile technologies, such as: � use of a PC in the previous 4 weeks � sending or receiving an e-mail in the previous 4 weeks � Time spent using the Internet at home


46

� Time spent using the Internet at work � ownership of a mobile telephone � Use of mobile to access the Internet

5.2 Snapshot indicators and country groupings – factor and cluster analysis

The resulting dataset was subject to factor analysis, designed to look for a small number of higher-level ‘factors’ consisting of combinations of strongly correlated groups of variables from the original dataset. In this way the original set of variables may be simplified and the results may suggest a usable set of ‘compound’ indicators made up of combinations of ‘raw’ measures. The principal advantage of this approach is that it does not rely solely on subjective judgement, but is primarily dependent on groupings of statistically similar entities. This confers a degree of objective validity to the procedure which is at least beneficial as a starting point, although there is certainly a place for using judgement as to the utility of the results obtained, and for use of ‘manual’ modifications as appropriate. Having identified a set of factors in this way, a ‘cluster analysis’ was performed on each to identify groups of countries with similar characteristics with respect to that factor. In general, cluster analysis is used to group members of a population (of, for example, animals, companies or countries) into classes with similar attributes, and hence to define useful groupings. In the current application, we seek to identify groups of countries where the pattern of (for example) ‘IT usage’ (one of the identified factors, see below), is similar within each group. This is done by looking, for each pair of countries, at statistical differences (or ‘distances’) between the average values of the variables comprising the new factor, and grouping those countries where the differences are relatively small. A cluster analysis was then carried out to examine how countries in the survey compared with respect to this factor. As shown below, three very clear and distinct clusters of countries emerged, which might be classified as high, medium and low PC impact. Factor 1: Internet usage snapshot The Internet snapshot explores Internet access and usage (for both narrowband and broadband connections). The data shows that access to and use of PCs and the Internet is typically twice as prevalent in the ‘cluster 1’ defined countries as in the ‘cluster 3’ defined countries. The three variables which form this snapshot (percentage using a PC in the previous 4 weeks, percentage with access to the Internet at home. Percentage using the Internet in the last 4 weeks) appear to offer a useful compound indicator of penetration level. Thus, Cluster 1, which includes highly mature Internet markets such as the US and Scandinavian nations, presents a more dynamic trend than the other two clusters of less developed Internet markets. Additional analysis indicates that most respondents who have used a PC in the last four weeks also have an at-home Internet connection. They have also accessed the Internet in the last four weeks and most of them also have Internet at home. In Cluster 2, and especially in Cluster 3, PC usage is at a higher level than Internet usage ‘within the last four weeks’. It is no surprise that it follows that the measure of Internet at home connections is even lower than that. This factor implies that in Mediterranean countries many PC users do not access the Internet in the same time period. It also suggests that many people who access the Internet do not have an Internet connection at home.


47

This composite indicator is useful in that it highlights some usage patterns. Interpreted as an index it could be transferred to the SIBIS+ candidate countries to track the degree of Internet impact.

Figure 16: Internet usage snapshot

0

10

20

30

40

50

60

70

80

0 10 20 30 40 50 60 70 80

Has used a PC in last 4 weeks (%)

Has used Internet in last 4 weeks (%)

ELP

I

FE

B

DA

IRL

UK FIN

DK

CH

US

NLS

Cluster 3Low Impact

Cluster 2Medium Impact

Cluster 1High Impact

Note: Luxembourg is hidden behind Germany

LEU-15

Key: The size of the 'has used Internet in last 4 weeks' circles represents the percentage of the sample not the volume Note: Luxembourg is hidden behind Germany. Source: Technopolis analysis of SIBIS 2002, GPS Membership of the three clusters is as follows.

Cluster 1: high impact Cluster 2: medium impact Cluster 3 low impact Denmark Austria France Finland Belgium Greece Netherlands Germany Italy Sweden Ireland Portugal Switzerland Luxembourg Spain US UK Factor 2: The Mobile Telephony snapshot A second group of clusters to emerge comprises mobile telephone ownership and what might be termed ‘basic’ patterns of usage, which do not involve the Internet. The most important contributory variables in this case are: � Ownership of a mobile telephone � A high proportion of family/friends owning mobiles


48

� High use of SMS for communication with other people. A cluster analysis on these constituents produced the following:

Figure 17: Mobile Telephony Snapshot

50

60

70

80

90

100

50 60 70 80 90

Mobile phone ownership (%)

Mor

e th

an 7

5% o

f frie

nds

and

fam

ilyal

so o

wn

mob

ile p

hone

s (%

)

High SMS usage (%)

US20

F42

EL53

E58

P56

B70

CH69

DK48

D58

I61

UK63

NL38

IRL72 FIN

79

L53

A66

S53

Cluster 2

Cluster 3

Cluster 1

EU-1557

Source: Technopolis analysis of SIBIS 2002, GPS In this case, membership of clusters falls as follows: Cluster 1 Cluster 2 Cluster 3 Denmark France Austria Greece USA Belgium Italy Finland Luxembourg Germany Netherlands Ireland Portugal Switzerland Spain UK Sweden Regarding ownership of a mobile by respondents and the prevalence of a high proportion of their family and friends owning mobiles, there is little to distinguish Clusters 1 and 3, as can be seen in the figure above. However, the prevalence of mobile ownership in Cluster 2 – comprising France and the USA – is clearly much lower. The main variable distinguishing members of the three clusters is SMS communication with other people. It is over twice as high in Cluster 3 countries as in Cluster 2 countries. Hence, it is clear that although mobile penetration is currently quite high in most Western countries, differences in usage patterns between countries are occurring, as well as divergences in the use of data mobile services for communication. Another potentially interesting factor analysis for future work would combine the central themes of Internet and mobile telephony usage. Had the data sets been large enough to be statistically significant the following variables could have been combined:


49

� Proportion of those using the Internet who have accessed it in a way other than via a PC � Proportion of those owning a mobile with WAP or 2.5G capability who have used it for

viewing WAP pages or web pages, or for e-mail, in the previous four weeks � Proportion of those with such a mobile who used it for making purchases via the Internet � Proportion of those with a mobile who have used SMS messages for receiving financial

information, sports results or other subscription services. This factor could perhaps be described as representing ‘sophisticated’ usage of a mobile, and as such may form a potentially useful compound indicator. Care in interpretation would be required on account of the highly ‘filtered’ or ‘conditional’ nature of some of the questions in this group. For these reasons, specification of the countries contained in each cluster is likely to mislead rather than inform at this stage. However, it is an important indicator to include in future work. Factor 3: Migration to a faster Internet connection snapshot Traditionally two factors have influenced when Internet users migrate to a faster connection. Based on experiences of the US and Scandinavian markets, it has been noted that once a majority of a total population has Internet access there is a migration of users with tenure, commonly defined as those with two years or more Internet experience, to faster connections. They seek a better online experience e.g. quicker downloads, always on connections. The clustering of the US, Sweden, Finland and Denmark illustrates this in the figure below. Not only do they have relatively high percentages of regular and tenured Internet users, but the size of the group with faster connections than dial-up is large too. Above 60% of SIBIS active at-home Internet users (those who have accessed the Internet at least once in the last four weeks). In comparison, there is a clear second cluster of countries where this migration level is lower. The exception here seems to be the Netherlands which registers similar migration levels, as can be seen in the bubble size, despite having lower amount of users in the last four weeks with more than two years tenure than the first cluster of countries.

Figure 18: Migration to faster access

20

30

40

50

60

70

80

90

100

20 30 40 50 60 70 80 90 100

Population who have accessed the Internet in last four weeks (%)

Pop

ulat

ion

who

hav

e ha

d ac

cess

to th

e In

tern

et fo

r 2 y

ears

+

Users who have migrated to a faster than dial-up connection (%)

EL5

P14

E15

F10

I 6

B20

EU-1515

D 23

L20

A15

IRL4 CH

14

UK9

NL 25

US21

S23

FIN16 DK

29

Typical migration path

Note: Ireland is hidden behind Luxembourg

Source: Technopolis analysis of SIBIS 2002, GPS


50

The more experienced users are, the more likely they are to upgrade to higher bandwidth and, therefore, to being realistically able to use all the benefits and services within the Information Society. Effects of circumstances on IT usage – multiple regression Multiple regression enables the testing of relationships between a ‘dependent’ variable and a set of ‘independent’ variables thought likely to influence it. The intention is to explain variations in the dependent variable in terms of variations in the independent variables – the statistical robustness of the overall regression can be assessed, as can the strength of the influence of each individual independent variable. In this application, the interest is in exploring the extent to which variations in questionnaire results on aspects of telecommunications and access between individuals can be ‘explained’ by differences in social and economic circumstances. The following were chosen as independent variables due to their presence in the survey data: � Terminal education age � Occupational status � Household Income A fourth socio-economic variable, ‘Social Grade’, was not used as it is defined in terms of education and occupation, and would lead to severe multi-colinearity problems. So far, these three variables have been used in an attempt to ‘explain’ variations in the following six dependent variables in turn: � Use of a PC in the previous four weeks � Sending or receiving an e-mail in the previous four weeks � Time spent using the Internet at home � Time spent using the Internet at work � Ownership of a mobile telephone � Use of mobile to access the Internet Some summary results are provided below.

Table 13: Multiple regression results Dep. Variable Terminal

Education Age Occupational Status

Household income

Coefficient t Coefficient t Coefficient t Use of PC 1.22 18.7 0.02 16.6 0.06 9.33 Use of e-mail 0.13 43.4 0.03 19.2 0.05 7.76 Time on Internet at home

0.21 10.6 0.04 8.81 0.13 7.0

Time on Internet at work

0.28 14.1 0.10 22.2 0.09 5.1

Ownership of mobile

0.02 2.8 .004 2.9 0.05 8.82

Internet access from mobile

-.002 -0.56 0.02 2.7 -0 -0.09

Source: Technopolis analysis of SIBIS 2002, GPS For the first four regressions, coverage use of a PC e-mail and Internet, all the variables (judging by the t-statistics) are significantly different from zero, and all act in the same direction – more education, higher occupational status and high household income all imply


51

greater use of each of the facilities. Associations with the ownership of a mobile, however, are much weaker, and barely significant. Association with the use of a mobile to access the Internet is totally insignificant.


52

6 Further developments: remaining gaps in the statistical coverage of the topic

This chapter describes on the one hand the 'gaps', i.e. those areas within the topic where official statistics currently fail to provide adequate measures for the issues discussed previously in SIBIS. On the other hand, it also comments on some methodological questions which could be complemented and discussed in further work to be conducted after SIBIS. It generally shows: • Where and why the SIBIS project has not developed indicators in certain areas • And how and why these indicators should be collected in future work Therefore the main aim of this chapter is to both fill some existing gaps and introduce some suggestions for future work on the topic area. Some of the indicators which could not be piloted in the GPS are nevertheless explained and specified in the forthcoming Indicator handbook, so it would be possible for relevant stakeholder to collect them in future work. Therefore, some of these considerations will be further discussed, and illustrated with existing and recent complementary data from external sources. Given this, it seems very relevant to highlight future work that should be done to cover existing gaps in telecommunications and access indicators, particularly those that can assist in mapping the digital divide. While it has only taken approximately five years for Internet penetration of households to rise from 5% to 50% it may take longer to achieve universal access. Evidence from Scandinavia and the US indicates that the penetration levels start to plateau at around 65% of the population. Prior to this point it is imperative to understand and measure the digital divide. Also it is becoming increasingly important to improve understanding of existing devices in order to follow progress on how they can narrow the digital divide. It is also important to understand which demographic or social groups are under-using or not using new technologies. As universal services such as education, social support, and medicine are increasingly being delivered, at least in part, online, it is highly probable that a new type of social exclusion will occur. Therefore, identifying availability of technologies by region, uptake, access device, and usage patterns and analysing them by income, education and social measures is a way forward. But qualitative regional surveys about barriers to technology access will also be required if solutions are to be developed.

6.1 The existing gap on regional level data indicators for T&A. There is a data black hole at sub-national levels within Europe. This is due to two reasons: 1. Definitions of regions vary. 2. Cost of surveying. Variations in regional definitions are problematic for the collection of indicator data at a sub-national level. Disparities occur at a number of levels. While Eurostat's definitions and hierarchy of regions is well established it does not always concur with regions as defined in policy. Many of the European regional development and social fund programmes are being targeted at sub-regional levels. So while a programme may be called 'East of Scotland' its policies are targeted at a ward level across multiple Eurostat regions in the East of Scotland. Such postcoding is effective at targeting the delivery


53

of development aid but it creates immense difficulties for measuring the impact of the Information Society within the communities. While it is said that telecommunications is a borderless technology, the regional impact of networks is important to understand. European Union structural funds are being spent on improving regional access and use of these networks. Such spending is justified on the basis of the impact on citizens and enterprises in those regions. Therefore, it must be open to measurement in order to assess the impact. The importance of understanding the impact of networks on regions is highlighted by preliminary research by DG Regional Policies that indicates that Information Society actions within Structural Funds programmes account for an estimated expenditure of EUR 15,849 billion. Official definitions of regions, such as those used by Eurostat, can and do vary from national definitions of regions and, therefore, existing measurements of the availability of technologies. Many measures of media penetration and usage are based on amalgamations of regions or on the geographic coverage of transmission stations. Such broadcast areas are of particular interest to the commercial sector so have been extensively used to map new technology penetration. Custom surveying of a population within such specific regions is possible, but it is expensive. Plus, the demand for statistically representative sampling can be frustrated simply because a region has a small population and a small uptake of a technology. Until a technology is widely adopted the user base will be too small to accurately measure in order to state that it is in an early-adopter phase. Collecting profile information at postcode level is a possible solution. Aggregating the information allows the creation of regions to match other definitions but this is a data- and time consuming approach that is best deployed over a long period before representative results can be analysed. 3G - An imperative to track Third generation mobile phone technologies, (3G) always grab headlines, for good reasons and for bad. In the current economic climate the headlines are associated with the revision of rollout plans across Europe, as 3G license-holders try to conserve cash. These are signs of a new realism entering the mobile sector. The news that availability of a full 3G mobile telecommunications network will not be available in Sweden for another four years, could signify delay of one of the key elements of the information society roll out in the EU. Sweden was expected to be one of the first European countries to make 3G services available throughout the country and had opted providently for a joint collaborative plan that suited the country best rather than auction to the highest bidder. However, one of the four license holders in the country, Orange, has asked for a three year delay in the original schedule that would have seen more than 99 per cent of the population covered by the end of 2003. The new date by which Sweden should have the proposed 3G coverage is the end of 2006. Additionally, there has also been a request from Orange to reduce the level of coverage from 8.86 million people in Sweden to 8.3 million. The move is attributed to the economic difficulties that have been experienced by telecommunications operators, who are now looking to reduce spending in the 3G area. As well as the request made to the Swedish telecoms authority for postponement, Orange has also announced that it will cut back spending at its German venture, Mobilcom. The Spanish telecommunications company Telefonica, has also said that it is pulling out of spending any money on non-Spanish 3G developments.


54

Table 14: European 3G licences rollout schedules EU Countries Licences Average cost of

licence(¤m) Coverage obligations (share of adult population)

Germany 6 8,468 25% by Dec 2003; 50% by 2005 UK 5 7,193 80% by end 2007 France 2 619 25% by Jul 2003;80% by 2009 Netherlands 5 537 60% by end 2007 Belgium 3 150 30% by Jul 2003; 85% by 2006 Italy 5 134 Regional capitals within 30 months Spain 4 134 Cities > 250,000 citizens by Aug 2002 Denmark 4 128 30% by end 2004; 80% by end 2008 Austria 6 117 25% by Dec 2003; 50% by Dec 2005 Portugal 4 100 20% by end 2002; 60% by end 2007 Switzerland 4 34 20% by end 2002; 50% by end 2004 Norway 2 25 80% by end 2004 Sweden 4 nil* 99.98%by end 2004 Finland 4 nil* None Source: Goldman Sachs, August 2002 However, one of the four license holders in the country, Orange, has asked for a three year delay in the original schedule that would have seen more than 99 per cent of the population covered by the end of 2003. The new date by which Sweden should have the proposed 3G coverage is the end of 2006. Additionally, there has also been a request from Orange to reduce the level of coverage from 8.86 million people in Sweden to 8.3 million. The move is attributed to the economic difficulties that have been experienced by telecommunications operators, who are now looking to reduce spending in the 3G area. As well as the request made to the Swedish telecoms authority for postponement, Orange has also announced that it will cut back spending at its German venture, Mobilcom. The Spanish telecommunications company Telefonica, has also said that it is pulling out of spending any money on non-Spanish 3G developments. Although only one of four license holders in Sweden, its move could still also affect the others, as the plans for establishing the 3G infrastructure in rural areas in the country is a collaborative efforts between all four license holders. Thus this could have an effect in Sweden as in other EU countries, of increasing the digital divide for 3G technologies. Thus, should those seeking to develop indicators for the Information Society also delay the measurement of these technologies and their impact? Existing technology is old. GSM's underlying software platform was designed nearly 15 years ago. Its replacement while delayed will be inevitable. Applications based on 3G will change consumer behaviour in a similar manner to the way broadband Internet access changed narrowband Internet users. 3G allows a persistent connection to be established between a users’ device and a networked information source. This 'always-on' functionality allows a deeper level of interaction to be designed. For example, it allows the tracking of location and the provision of location-sensitive information. If a suitable price regime is established it could also encourage widespread ongoing use of digital content rather than sporadic connection to sources. For all the above reasons it is important to track 3G. Although the shift to 3G is a radical one, in most cases entailing the construction of new networks, the transition from 2G to 2.5G services is likely to be accompanied by a more significant conceptual shift than that from 2.5G to 3G, introducing the concept of “always-on”


55

mobile communications. This is already beginning to bring the Internet and mobile communications onto a more common ground. The shift to 3G mobile will take this concept even further, but will essentially involve convergence and interoperability. Although mobile data services are already available on 2G platforms through WAP, i-mode and short messaging service (SMS), it is through the advent of 2.5 and 3G that users will begin to fully reap the benefits of the mobile Internet, through high-speed communications and multimedia applications. What we have learned from the success and failures of 2G technologies is that person-to-person messaging, simple interfaces and timely content delivery will be the key to future service development and revenue generation. A mere simulation of the fixed-line Internet experience will not compel users to take up mobile data services. The development of an adequate payment system for mobile devices is also crucial: for “always-on” applications, per-minute billing may have to give way to volume-based billing. As it has happened with Internet subscription packages, there will be a need to bring flat-rate connection packages for allowing consumers to fully exploit 3G services and content possibilities. The combination of mobile and Internet technologies—for instance in the form of SMS messaging—is already transforming the way people interact and the way business is done. Some 24 billion SMS messages were sent worldwide in the first quarter of 2002. Messaging services have also brought information technology closer to groups that have traditionally had limited access to it, such as children and the deaf community. High-speed data services combined with additional functionality, such as location technologies and improved security, will further enhance the user experience. On a technical level, the viability of future 3G services will rely on continued efforts towards the interoperability of radio interfaces, the evolution to an IP-based core network and the harmonisation of formats for content delivery. At the service level, convergence between the fixed and mobile Internet is already happening, through services such as mobile instant messaging and fixed-line SMS. This interoperability will eventually encompass complementary and alternative network technologies, such as wireless LANs, short-range connectivity technologies, fixed broadband networks, etc. Regulators and industry players alike need to realise that there are a number of different options for providing mobile Internet services, and that 3G services must be considered in their global context. In the short term, the key imperative for tracking 3G is the issue of the digital divide. While absolute penetration levels will remain low for a number of years the geographic penetration will vary significantly across Europe and within nations. 3G operators will target rollout in populated areas where there is likely to be demand for roaming access to digital networks, for example, in urban centres and travel locations such as airports and stations. The digital divide will be starkly illustrated by this rollout pattern. In particular, it will heavily emphasise the digital divide between rural and urban areas. It is ironic that while rural areas have been traditionally denied access to broadband networks, because of the limitation of physical cable networks, they will be denied the benefits of wireless technologies, which could overcome those limitations. The emergence of ‘digital deserts’ needs to be mapped. A practical example occurs in Spain. Telefonica is restricting its 3G rollout to cities with more than 250,000 people. Thus, for instance Extremadura, a region with a capital city of 100,000 people, will be totally excluded from 3G benefits. Need to track the deployment of alternative Broadband technologies where infrastructures have not arrived Generally, when looking at the penetration of different BB technologies, it is possible to highlight how there is a divergence in the penetration and level of competition of BB technologies in different countries. In fact by 2005, approximately 8.6 million potential residential and SOHO users in Europe will still not have access to high-speed cable modem or Digital Subscriber Line (DSL) services, according to analysis by McKinsey & Company.


56

Since there is a lack of infrastructures in deprived and isolated regions across the EU, it is important to track what policy measures can be implemented to solve these issues. One of the best European markets in terms of consumer's choice is the UK, where DSL and cable technologies are widely available. Moreover, for rural areas where broadband land infrastructures have not developed the government is starting to roll our broadband wireless policies. Traditionally there have been some linked problems in the provision of BB by satellite, since it was a technology not affordable for residential consumers and business compared to existing competitive BB technologies such cable or DSL. There was a belief that if ADSL roll-out is not achieved in rural areas, no competitive access arrangements will be available for the foreseeable future. However, this perspective has been challenged and can be questioned though different case studies in regional areas which have shown how satellite can be initially expensive (as it happened with ADSL in the beginning) but can be affordable as there is increasing take-up and use from consumer/business groups. For instance some recent wireless successful cases studies in 2002 include Northern Ireland and Wales.

6.1.1 The Imperative to measure subscription and connection rates Past evidence indicates the importance of monitoring the cost of access to digital networks and content. The popular adoption of the Internet in both Europe and the US did not pass the 10% threshold until acceptable price points and payment plans were in place. In the US, the 1996 move to flat-rate Internet access, so called 'eat-as-much-as-you-like' tariff structures, spurred rapid adoption of the Internet. AOL's move from metered to flat rate pricing allowed it to pass the 2 million users milepost. In Europe, adoption was limited by metered access to not only online services, but also to the actual telephony network. Not until 2000, when flat rate local call charges were popularly used by ISPs, did adoption start to accelerate significantly. Therefore, it can be expected that the existence of metered pricing structures associated with new technologies, be it WAP, 3G or public access points, will hold back adoption of these. Within the SIBIS GPS a question regarding cost of using Internet has not been included since it has been noted by our experts that telephone surveys are not the best methodology for doing that. Industry interviews would be best way of getting that information, but it is highly complicated to do this due to the large number of ISPs and different types of unmetered and metered types of offers across the EU. This is maybe one of the most difficult areas to benchmark currently in Europe due to the broad diversity of subscription connection packages to the internet in the different EU countries for both narrowband and broadband technologies. It is important to track how many experience narrowband users upgrade to broadband. As it is necessary to track how Internet at home penetration rates evolve according to narrowband/metered, narrowband/unmetered and broadband households. A comprehensive trend portraying an example of this set of indicators is included below,


57

Table 15: Take-up of Internet consumer market narrowband (metered/unmetered) and broadband17 Aug 02 Nov 02 Feb03 UK homes connected to the Internet 42% 42% 45% UK homes connected to the Internet using narrowband 84% 80% 79%

UK homes connected to the Internet using narrowband (unmetered)

38% 35% 38%

UK homes connected to the Internet using narrowband (metered|)

29% 28% 25%

UK homes connected to the Internet using narrowband (unsure whether metered or unmetered)

18% 16% 6%

UK homes connected to the Internet using broadband 7% 9% 13%

Among the established eEurope priorities, broadband adoption is included at the top. Broadband indicators relating to penetration and costs are very important, as broadband is seen as gateway for the full development of e-commerce and other online services. For those Internet users who just go online to access their e-mail and do a bit of web surfing, the relationship between the cost of dial-up and broadband is important. For the millions of Internet users in this category there is little economic incentive, as of late, to trade up to broadband and pay significantly more. However, a very positive aspect to highlight here is that Oftel’s18 study, which included data on consumer perceptions of broadband, portrayed how the majority of users who migrate to broadband cannot contemplate returning to narrowband. The majority of users upgrading to broadband are satisfied with the quality of service they receive in comparison to narrowband.

6.1.2 Further problems when tracking multi-platform and multi-context usage Developments in broadband, and increasing penetration of alternative access devices, will bring increasing difficulties when it comes to tracking usage across platforms and contexts. As new networks are created, the open environment of the Internet must become a lessor part of the digital universe. Encouraging owners and operators of the new private networks to allow access for metering, and thus measuring, is vital. Because PC-based online interaction with digital content is an established universe of information possibilities it will remain the dominant digital medium. Therefore, measurement of how alternative access technologies are used in conjunction with it is important. Finding out what users do on their mobiles, PDAs and through their set-top boxes must be placed in the context of the online PC environment, rather than being perceived a 'battle of the platforms'.

17 Oftel, Monitoring Internet studies, March, 2003. 18 Survey on consumer perception studies of broadband, 2002


58

Multi-context usage is more difficult to measure. Until there is majority adoption of digital wallets, IDs or other user authentication technologies, this will remain a difficult to measure as authoritatively identifying the user will be much more difficult than identifying a particular access device.

6.2 Some methodological problems when collecting information on T&A with telephone survey methodologies

6.2.1 RDD telephone survey Random digit dial telephone surveys use an application that randomly dials telephone numbers to ensure that the sample of respondents is truly random and or statistically representative. This method does, however, have problems. Market research industry codes of conduct forbid surveyors from interviewing children under 16 without consent of their parents. This can skew the responses to questions related to children's topics. In the topic in question, young users are a relevant group for of those interested in ICT to track. In the case of SIBIS, many users of games consoles will be under 16 so while they may be using such devices to access the Internet they are unlikely to be surveyed about this if contacted by telephone. Likewise, the multi-context access (question GPS A9) could not track users accessing the Internet from primary and secondary education institutions. Those younger than 16 years old were not monitored in this study, despite many of them being early and fast adopters of new technologies. Another point to be made is that if children under the age of 16 could have been included in some way, it would have been possible to track some of the e-Learning action plans objectives of promoting European Youth in a Digital Age. This is something which must be consider when developing indicators in future work. Alternative methodologies to telephone surveys must be applied in order to track children and teenagers use of ICTs. For instance, specialised children's panels can be assembled in order to solicit the opinions of children.

6.2.2 Some final methodological questions related to e -content gaps: Service and applications usage

Although SIBIS is not looking at developing e-content indicators, it is important to remark at this stage some indicators which could be added to easily track some indicators on e-content. While it is possible to identify the top sites in terms of traffic, visitors to a site, simple metrics tell only part of the story. Time spent per visits is important, as is the number of return visits. Consumers have the capacity to prioritise and favour just a handful of sites. Such sites, that are subject to regular visitation and deep usage, do not have exclusive mindshare but their consistent usage makes them useful indicators of usage trends. Understanding which ones are patronised by particular groups is important. Usage intensity can be calculated by relating the time spent per visit to the frequency of return visit. However, it is extremely important that the nature of a site is understood when interpreting usage intensity data. A tax return site that is receiving frequent short visits may be fulfilling an education role by reaching lots of people. But, if the online tax return form within it takes 30 minutes to complete and only 5% of visitors are spending 30 minutes plus then that service is not being used properly.


59

Gathering such metrics can be done using panels that deploy metering applications to participating users' PCs, but the restricted sample size, measured in the thousands, means that only very popular sites can be analysed. At-work usage metrics cannot be studied and compared consistently across Europe yet as only a few such panels exist in certain markets. If a site has less than 5% of total traffic in a sector the resulting data may not be reliable. To analyse less popular sites in this ways requires the aggregation of a much larger sample. Using surveys that place the meter at ISP servers can allow this to happen and allows reliability to extend to data gather from sites with 1% traffic level. With regards to browsers and applications, SIBIS only concentrates on piloting email services usage in the GPS. Besides email, in future work it would be useful to track other popular applications that are widely used such as peer-to-peer applications or instant messengers. Likewise, tracking ‘killer applications’ for emerging technologies, such as broadband or WAP is important since its absence it is one of the commonly mentioned inhibitors for not adapting to those technologies. Hence, it seems particularly important to track user’s favourite content services in relation to broadband penetration and mobile-PC bimodal usage developments. For instance, the EITO study currently includes an indicator that records the three most widely used Internet applications.


60

7 Conclusions Telecommunications and access is a topic area which underpins the advancement of Europe in the Information Society. The study itself helps to confirm a number of existing assumptions and explore differences across the EU and the US in access, usage and impact as well as giving an opportunity to test a set of new indicators on a pan-European level. The work also contributes to measuring the progress in adopting the New Regulatory Framework in Electronic communications, networks and services19 by tracking the penetration of broadband across Europe. From an analysis of the data from the GPS and from the compound indicators a number of differences have been identified across the participating European Countries and between Europe and the US. Analysing how Europe is doing in the area of telecommunications and access is complex. It is clear from the data that one size does not fit all in attempting to measure the access, usage and impact across the Member States and the other participant countries. These country comparisons will be explored further once the second phase of SIBIS, SIBIS+, which will include NAS countries is implemented, giving EU25 data. The 'Migrators to faster connections than dial-up snapshot' has portrayed how two factors have influenced when Internet users migrate to a faster connection. Based on the experiences of the US and Scandinavian markets, it has been noted that in the majority of active at-home online population, there is a migration of users with tenure, commonly defined as those with two years or more Internet experience, to faster connections. They seek a better online experience e.g. quicker downloads, always on connections. As well as the compound indicators, across the other indicators tested in the SIBIS survey, there is evidence of national differences and great disparities across Europe. There are some large differences in Internet usage between Northern European Countries and Mediterranean countries but similar adoption rates, and much higher mobile penetration and mobile data usage, particularly SMS. A number of methodological lessons have also been learned as a consequence of this study. It remains important to go on counting instances It remains extremely important, in the light of these results, to go on counting instances in some of the European countries involved in the survey. However, the next step of measurement is where SIBIS adds value as there are numerous pan-European surveys already in existence which are counting, SIBIS has the opportunity to suggest sets of indicators which help provide a picture of the ‘how and why?’ In spite of this need, it must be noted that identifying usage patterns and impacts is much more difficult. Usage must first be observed over a time period and then needs to be classified. These classifications then need to be reapplied to the observed data. The evolution of the usage behaviours also means that the classifications have to be reassessed regularly to make sure that they are not out of date. As user numbers increase, so does the depth of content available to them. This is an exciting area of development, which can give great insight into the use of technologies in Europe, especially given that the second phase of SIBIS will be looking at

19 COM(2000)239


61

these indicators over the NAS countries, providing one of the first comprehensive pictures of the state of telecomms and access indicators in the EU25. A number of flexible indicators could be chosen and used to test the levels of penetration of technologies in each country before piloting sophisticated indicators The differences across the Member States and Switzerland are high and vary for each technology area. This means that the current approach of asking exactly the same set of questions in each country leads to some country data, particularly in the cases of usage and impact indicators, being statistically insignificant. A way of creating surveys that are comparable, but do not ask obsolete questions, is to devise a system of standard flexible indicators which could be drawn on to create a set which best work in a country specific situation. A set of penetration indicators could also be identified which would be asked across the board in order to judge when some countries start to have significant samples of a chosen technology in use so that more sophisticated indicators could be introduced. In fact, one of the main conclusion embodied in SIBIS is that it is more important given the current technological disparities in penetrations and prices in Europe to measure experience rather than technology. As highlighted in SIBIS research, the shortening of technology product life cycles makes any tracking measurement problematic. This problem is compounded by users’ definitions and perceptions of technology that vary across countries. Therefore, over the medium and long term, measuring experience, measuring consumers’ satisfaction levels seems mandatory on a pan-European level. There is much more work that could be done. There are still many gaps that have been identified in this report and also there is much that can still be done with the results that have already come out of the survey. Some future work in SIBIS+ will look at comparing some of the indicators proposed in this report on an EU25 basis. This will make SIBIS one of the most comprehensive resources in the development of IS indicators and benchmarking comparisons in Europe.


62

8 ANNEX 1 – References EITO - European Information Technologies Observatory (2001-2002). European Commission (2000), Measuring Information Society. A Eurobarometer survey carried out by INRA (Europe). European Commission (2001), The Digital Divide - A Research Perspective. Report to the G8 Digital Opportunities Task Force. Eurostat (2001), Telecoms Outlook. 2002 European Commission, Brussels. Goldman Sachs, August 2002, Mobile notices, European 3G rollouts ITU-Internet for a mobile Generation, 2002 Nielsen/Netratings Broadband Study Q2 2002 OECD (2001), Understanding the Digital Divide. OECD, Paris. OECD (2002), ICT outlook, OECD, Paris. Oftel -Consumer perceptions of broadband services and International Benchmarking, 2002 Ovum, Consulting Insight: Broadband Advisor: 'Delivering on the EU Commission broadband vision', April 2003,


63

9 ANNEX 2 - Methodology of the survey

9.1 General Population Survey (GPS)

9.1.1 Outline of the study The survey was conducted in April-May 2002 in all 15 EU Member States plus Switzerland and the USA, using computer-aided telephone interviews. The survey was co-ordinated and executed by INRA, Germany. The population for this study was all persons aged 15 and over living in private households in the respective countries and speaking the respective national language(s). Subjects discussed included ownership and use of ICT equipment, use of the Internet and e-commerce activities, competence in the use of new media, questions on health and the Internet, the Internet and security concerns, e-government, telework, mobile work and other new ways of working, as well as further education and satisfaction with working conditions. 11,832 interviews were successfully completed. The average interview length per country varied between 10 and 20 minutes.

9.1.2 Methodology

Subject of study Topics of this survey were statements on interviewees’ ownership and use of ICT equipment, use of the Internet and e-commerce activities, competence in the use of new media, questions on health and the Internet, the Internet and security concerns, e-government, telework, mobile work forms, as well as further education and satisfaction with working conditions.

Study concept The study was conceived and executed as a cross national study. The co-ordination was carried out by INRA Deutschland GmbH, Mölln, on behalf of the client. The study consisted of two parts, a trial in Germany with a subsequent main survey in all participating countries.

Overall responsibility and co-ordination

INRA Deutschland GmbH, Mölln

Countries and executing institutes

Belgium: INRA Belgium Kroonlaan 159-165 Avenue de la Couronne 1050 Brussels Denmark: Gallup A/S Sundkrogsgade 10 2100 Copenhagen Germany: INRA Germany GmbH Papenkamp 2-6 23879 Mölln Finland: Taloustutkimus Oy Lemuntie 9 00510 Helsinki France: BVA B.P. 59 78222 Viroflay Cedex Greece: MEMRB – K.E.M.E 24 Ippodamou St. 11635 Athens Great Britain: BMRB International Saunders House, 53 The Mall, Ealing London W5 3TE Ireland: Lansdowne Market Research Ltd. 49 St. Stephens Green Dublin 2 Italy: INRA Demoskopea


64

Via Salaria, 290; Via Rubicone 41 00199 Roma Luxembourg: ILReS. S.A. 46, Rue du Cimetière 1338 Luxembourg / Bonnevoie Netherlands: NIPO Grote Bickersstraat 74 1013 ks Amsterdam Austria: Spectra Brucknerstr. 3-4/5 4020 Linz Portugal: METRIS Av. Eng. Arantes e Oliviera, No. 3-2 1900 Lisboa Sweden: GfK Sverige AB Box 401 22100 Lund Switzerland: Link Institut Spannortstrasse 7/9 6000 Luzern Spain: INRA España S.A. Calle Alberto Aguilera 7-5º 28015 Madrid USA: I.C.R 605 West Street Media, Pennsylvania 19063-2620

Survey methodology The study was carried out as a telephone survey (Computer Assisted Telephone Interview – C.A.T.I) in all countries.

Population The population for this study is all persons aged 15 and over living in private households in the respective countries and speaking the respective national language(s). Switzerland: Here the survey was carried out in both the German and French speaking parts of Switzerland. USA: The population includes English speaking people in the 48 continental federal states of the USA (excluding Alaska and Hawaii). Finland: Finnish speaking population.

Random sampling and selection process

Belgium: 3-stage selection process based on the INFO BEL telephone directory. Addresses in 10-fold translation, random sampling of households, selection of the target person via a birthday key.

Denmark: Geographically stratified sample based on telephone directories. Telephone numbers are generated so as to also include unlisted numbers. The selection of the target person results from a birthday key. Germany: Within the ADM telephone sampling system a representative, multistage random sample is drawn for each survey area. The selection data is based on the batch of all registered fixed network telephone numbers. Master numbers are formed by detaching the two final digits. Through the generation of new final digits from 00 to 99 number blocks are produced which contain listed as well as non listed numbers. As far as possible, business numbers were removed from this sampling frame. All telephone numbers were given an area code number, either the original or according to the known distribution in the number block. In this way numbers can be sorted regionally, thus increasing the precision of the sample. By means of the relation between sample size and distribution of households per regional cell an allocation table is produced which determines the number of samples to be drawn for each cell. Assuming a random starting point, all areas are processed with a fixed step width through set stages until the number to be selected from which areas for each cell has been determined. Subsequently the numbers are drawn randomly in a second selection stage. The selection of the target person results from a birthday key. Finland: Geographically and socio-demographicaly stratified random


65

sample based on information from official statistics. The selection of target households takes place at random. The selection of target persons results from a birthday key. France: Geographically and socio-demographically stratified random sample of 8000 starter addresses based on France Telecom directories. The selection of target persons results from a birthday key. Greece : Multistage stratified random sampling. The geographical stratification takes place on the basis of NUTS 1, NUTS 2 and location size. The size of each unit is determined on the basis of official statistics. The selection of the target person results from a birthday key.

Great Britain: The sample is based on a draw data-file. Through the generation of new final digits a sample frame is established which contains listed and unlisted as well as so far non-existent numbers. The selection of target households takes place at random. The selection of the target person is via quota. Ireland: Geographically sorted random sample based on the ”Eircom” telephone directory. Additional telephone numbers are also generated in order to include unlisted numbers. The selection of the target person results from a birthday key.

Italy: Geographically and socio-demographically stratified random sample. 705 sample points result from the stratification. Additional telephone numbers are generated in order to also include unlisted numbers. The selection of the household results from Random Digit Dialling and the selection of the target person results from a birthday key. Luxembourg: The sample is based on a draw data-file. Through the generation of new final digits from 00 to 99 a sample frame is established which contains listed and unlisted as well as so far non-existent numbers. The selection of target households takes place at random. The selection of the target person results from a birthday key. Netherlands: Geographically stratified random sample. The geographical sorting is based on postcode areas. Target person selection takes place through an algorithm, which selects the interviewee on the basis of age and gender of people living in the household. Austria: Geographically stratified random sample. The selection of the target household takes place through RDD (Random Digit Dialling). The selection of the target person results from a birthday key. Portugal: Geographically and socio-demographically stratified random sample. 200 sample points result from the stratification. The selection of households takes place via Random Digit Dialling, the selection of target persons via a birthday key.

Sweden: Geographically and socio-demographically stratified random sample. 200 sample points result from the stratification. The selection of households takes place via Random Digit Dialling, the selection of target persons via a birthday key. Switzerland: Geographically stratified random sample based on postcodes. Each postcode represents a sample cell. The selection of households takes place via Random Digit Dialling and the selection of the target person via an algorithm which selects the interviewee at random on the basis of a list of household members.

Spain: Geographically stratified random sample based on NUTS2 areas. 148 randomly selected sample points result from the sorting. Within these sample points addresses of target households are randomly drawn. Selection of target persons results from a birthday key. After about two thirds of the fieldwork the screening was targeted towards male members of the household due to a disproportionate number of female interviewees.

USA: Geographically stratified random sample based on the MSG-Genesys sampling process. The selection of households takes place via Random Digit Dialling and the selection of the target persons via a birthday


66

key. After the 758th interview the screening was targeted towards male members of the household due to a disproportionate number of female interviewees.

Survey period The interviews were carried out in the following period: 04.03. -18.05.2002

Interviews undertaken Total: 11,832

Belgium 16.0 min Luxembourg 16.2 min

Denmark 18.1 min Netherlands 18.4 min Germany 17.5 min Austria 15.8 min Finland 17.3 min Portugal 12.1 min

France 12.0 min Sweden 20.2 min Greece 10.2 min Switzerland 19.0 min Great Britain 18.0 min Spain 12.5 min

Ireland 17.7 min USA 18.3 min

Average interview length

Italy 14.0 min

Interviewers used Total: 632

Additional comments to the data set

Belgium: In order to improve the sample, an additional 85 interviews were carried out in some cells. Finland: In order to improve the sample, an additional 169 interviews were carried out in some cells. Netherlands: In order to improve the sample, an additional 30 interviews were carried out in some cells. Switzerland: In Switzerland respondents were not asked to deduct tax from income (Z19), as that is not the norm there.

Data supply One labelled SPSS data set of the main survey of all interviews.


67

Field report and outcomes B DK D FIN F EL UK IRL I L NL AT P S CH E USA

Method C.A.T.I.

1 gross sample (utilised addresses) 4506 3154 9999 2621 7300 5022 11392 3890 12006 8764 3640 4669 1403 5177 2327 6494 18162

1.1. non-contacts – thereof: 311 242 1701 40 3401 2346 139 1111 4436 5023 803 193 91 455 638 1239 4192

1.1.1 unobtainable 0 235 1202 0 2342 2077 123 654 4436 3748 522 124 43 113 638 644 3656

1.1.2 engaged 3 7 436 0 57 206 1 316 0 705 164 8 32 55 0 5 536

1.1.3 answer phone, fax, modem 308 0 63 40 1002 63 15 141 0 570 117 61 16 287 0 590 0

1.1.4 other 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1.2 sample neutral non-response – thereof: 1874 1917 4492 984 511 1022 5088 1051 2659 1316 805 2322 410 2808 322 1095 8789

1.2.1 invalid telephone numbers 955 1516 3760 97 60 529 4308 498 1657 790 652 858 334 2297 230 398 5725

1.2.2 not in the population 472 202 41 782 374 176 119 405 364 0 153 1248 47 16 0 164 478

1.2.3 business numbers 300 82 285 12 27 220 437 0 340 455 0 75 15 193 0 434 1331

1.2.4 other 147 117 406 93 50 97 224 148 298 71 0 141 14 302 92 99 1255

2 net sample – thereof: 2321 995 3806 1597 3388 1654 6165 1728 4911 2425 2032 2154 902 1914 1367 4160 5181

2.1 refusal 1470 468 2451 912 2231 747 5012 1134 3592 1000 1248 1609 364 1246 529 2255 3198

2.2 termination 114 0 87 0 30 0 80 11 201 0 0 1 6 19 0 115 143

2.3 target person contacted but interview impossible – thereof:

152 26 267 16 127 402 73 83 118 925 254 44 32 146 316 775 836

2.3.1 possible appointment outside field time 0 23 14 1 23 9 26 14 106 763 208 7 6 30 80 321 156

2.3.2 appointments to continue interview outside field time

152 0 200 0 104 295 47 65 12 17 11 34 18 24 194 179 669

2.3.3 other 0 3 53 15 0 98 0 4 0 145 35 3 8 92 42 275 11

2.4 complete interviews 585 501 1001 669 1000 505 1000 500 1000 500 530 500 500 503 522 1015 1004

3 exhaustion rate (%) (2.4/(2.1+2.2+2.4)) 27.0% 51.7% 28.3% 42.3% 30.7% 40.3% 16.4% 30.4% 20.9% 33.3% 29.8% 23.7% 57.5% 28.5% 49.7% 30.0% 23.1%


68

9.1.3 Weighting

1. Transformation from household sample to person sample:

As only one person per household is interviewed, the described sample procedure provides a household sample, i.e. each household of the base population has the same likelihood of being in the sample but not each person. With the weighting stage of the transformation the equal likelihood of households is replaced mathematically by the equal likelihood of the individuals. To this end, each data set is multiplied by the amount of people in the household aged 15 or over. This number is subsequently divided by the average household size in order to obtain the actual case number. 2. Adjustment of unweighted sample structure to the official statistic:

Because random samples are not evenly distributed across all population strata, the distribution of unweighted samples regularly and systematically deviate from the population distribution from official statistics. Through the mathematical weighting the sample distribution is adjusted to the official statistics. The national weighting factor (P10) which results from the iterative weighting was included in the data material. To this end the following criteria are used in the respective countries. Austria: age, gender, region; Belgium: age, gender, region, locality size; Denmark: age, gender, region; Germany: age, gender, region, locality size; Greece : age, gender, locality size; Finland: age, gender, region; France: age, gender, region, locality size; Ireland: age, gender, region; Italy: age, gender, region, locality size; Luxembourg: age, gender, region, locality size; Netherlands: age, gender, region; Portugal: age, gender, region, locality size; Sweden: age, gender, region; Switzerland: age, gender, region; Spain: age, gender, region, locality size; UK: age, gender, region; USA: age, gender, region, locality size. 3. Adjustment of weighted sample structure to the EU15-member states population:

This weighting factor was necessary to calculate total figures according to the whole population of the European Union member states. Furthermore it is useful to compare the EU with the US. Population sizes of each member state are weighted to reduce the distortion based on the sample sizes in each country. The different country-specific weighting factors are the following: Austria 0.44 Italy 1.63 Belgium 0.48 Luxembourg 0.02 Denmark 0.29 Netherlands 0.80 Germany 2.29 Portugal 0.55 Greece 0.59 Spain 1.09 Finland 0.21 Sweden 0.48 France 1.56 United Kingdom 1.57

Ireland 0.20 Switzerland, USA none


69

9.1.4 Sample characteristics and effect of weighting Total EU15

un

-w

eigh

ted

wei

ghte

d

% u

n-

wei

ghte

d

%

wei

ghte

d

un

-w

eigh

ted

wei

ghte

d

% u

n-

wei

ghte

d

%

wei

ghte

d

Total sample 11832 11832 100.0 100.0 11832 10306 100.0 100.0 Country Austria 500 500 4.2 4.2 - - - - Belgium 585 585 4.9 4.9 - - - - Denmark 501 501 4.2 4.2 - - - - Finland 669 669 5.7 5.7 - - - - France 1000 1000 8.5 8.5 - - - - Germany 1001 1001 8.5 8.5 - - - - Greece 505 505 4.3 4.3 - - - - Ireland 500 500 4.2 4.2 - - - - Italy 1000 1000 8.5 8.5 - - - - Luxembourg 500 500 4.2 4.2 - - - - Netherlands 530 530 4.5 4.5 - - - - Portugal 500 500 4.2 4.2 - - - - Spain 1015 1015 8.6 8.6 - - - - Sweden 500 500 4.2 4.2 - - - - Switzerland 522 522 4.4 4.4 - - - - UK 1000 1000 8.5 8.5 - - - - USA 1004 1004 8.5 8.5 - - - - EU15 - - - - 10306 10306 87.1 100.0 Age groups up to 24 1964 2019 16.6 17.1 1731 1651 16.8 16.0 25 to 49 5511 5309 46.6 44.9 4817 4593 46.7 44.6 50 to 64 2515 2495 21.3 21.1 2191 2209 21.3 21.4 65 and more 1833 2000 15.5 16.9 1558 1839 15.1 17.8 don’t know 9 9 0.1 0.1 9 14 0.1 0.1 Terminal education age up to 13 695 717 5.9 6.1 693 728 6.7 7.1 14 715 742 6.0 6.3 701 881 6.8 8.5 15 to16 1794 1750 15.2 14.8 1641 1820 15.9 17.7 17 to 20 3587 3515 30.3 29.7 2997 2937 29.1 28.5 21 and more 3266 3275 27.6 27.7 2743 2495 26.5 24.2 still studying 1687 1751 14.3 14.8 1463 1372 14.2 13.3 don’t know 88 81 0.7 0.7 77 73 0.7 0.7 Household type one person household 2006 1611 17.0 13.6 1682 1408 16.3 13.7 household with kids aged under 6 1723 1754 14.6 14.8 1451 1440 14.1 14.0 household with kids aged 6+ 2970 3152 25.1 26.6 2653 2655 25.7 25.8 two person household without kids

5063 5240 42.8 44.3 4467 4768 43.3 46.3

no answer on household size 70 75 0.6 0.6 53 35 0.5 0.3

Household income (according to national household income quartiles by Eurobarometer)

First quartile (lowest income) 1774 1580 15.0 13.4 1548 1299 15.0 12.6 Second quartile 2132 2084 18.0 17.6 1878 1764 18.2 17.1 Third quartile 2536 2521 21.4 21.3 2214 2087 21.5 20.3 Fourth quartile (highest income) 2968 3102 25.1 26.2 2502 2725 24.3 26.4 don’t know 1214 1295 10.3 10.9 993 995 9.6 9.7 refusal 1208 1249 10.2 10.6 1171 1436 11.4 13.9 Employment status paid employment 4966 4853 42.0 41.0 4291 4133 41.6 40.1 self-employed 935 941 7.9 8.0 809 799 7.8 7.8


70

Total EU15

un

-w

eigh

ted

wei

ghte

d

% u

n-

wei

ghte

d

%

wei

ghte

d

un

-w

eigh

ted

wei

ghte

d

% u

n-

wei

ghte

d

%

wei

ghte

d

unemployed/ temporarily not working 701 683 5.9 5.8 621 631 6.0 6.1

in education 1687 1751 14.3 14.8 1463 1372 14.2 13.3 retired or other not working 3441 3510 29.1 29.7 3034 3292 29.4 31.9 don’t know 102 94 0.9 0.8 88 80 0.9 0.8 Social grade (ESOMAR classification) unskilled manual workers and other less well educated w orkers/ employees

1332 1318 11.3 11.1 1238 1323 12.0 12.8

skilled workers and non-manual employees

1525 1445 12.9 12.2 1316 1287 12.8 12.5

well educated non-manual and skilled workers

1434 1402 12.1 11.8 1254 1121 12.2 10.9

managers and professionals 1577 1586 13.3 13.4 1265 1167 12.3 11.3 not specified 5964 6081 50.4 51.4 5233 5408 50.8 52.5

9.2 Decision Maker Survey (DMS)

9.2.1 Outline of the study The survey was conducted in March-May 2002 in seven EU Member States using computer-aided telephone interviews. The survey was co-ordinated and executed by INRA, Germany. The population for this study is defined as all establishments belonging to four aggregated industry sectors in the seven Member States. The interview was conducted with IT responsible persons in companies across all sectors of the economy. Subjects discussed included ownership and use of ICT equipment, use of the Internet and e-commerce and e-business activities, e-business security, e-government, web-site accessibility and ICT in research and development. 3,139 interviews were successfully completed. The average interview length per country varied between 14 and 18 minutes.

9.2.2 Methodology

Subject of study • Basic ICT: use of ICT and e-business technologies • e-commerce • e-business security • e-government • web-site accessibility • research and development • establishment demography

Study concept The study was conceived and executed as a cross national study. The co-ordination was carried out by INRA Deutschland GmbH. The study consisted of 2 parts, a trial in Germany with a subsequent main survey in all participating countries.

Overall responsibility and co-ordination

INRA Deutschland GmbH, Mölln

Countries and executing institutes

Germany: INRA Germany GmbH Papenkamp 2-6 23879 Mölln


71

Finland: Taloustutkimus Oy Lemuntie 9 00510 Helsinki France: BVA 101 avenue du General Leclerc 78222 Viroflay Cedex Greece: MEMRB – K.E.M.E 24 Ippodamou St. 11635 Athens

Survey methodology The study was carried out as a telephone survey (Computer Assisted Telephone Interview – C.A.T.I) in all countries.

Population The population for this study are establishments (in each respective country) in the four sectors:

• Manufacturing, Construction, Primary Sector

• Distribution, Catering, Transport & Communication

• Financial & Business Services

• Public administration, education, health, other personal and social services

Target person at the establishment was the person who is responsible for or significantly involved in decisions in the area of IT/ DP. In larger establishments/ organisations the head or another executive of the IT/ DP department. In smaller establishments/ organisations also the owner/ proprietor or managing director/ board member.

Random sampling and selection process

General: The sample was set up according to given industry and size class quota. Accordingly a stratified random sample was drawn from the universe, allowing for the relevant industries within the four aggregated sectors. Drawing the sample was organised locally by the national executing institutes. Germany: The sample was drawn from the Heins und Partner Business Pool. Heins und Partner have created a high quality business pool based on the available address inventories consisting of about 3.4 m data sets that have undergone comprehensive validation. For every enterprise comprehensive additional information is available, including corporate structure and branch office structure (220,000 branch offices) and is continuously being updated. The sample was drawn from the establishment file, which results from the transformation of enterprises into establishments and appending branch offices to the headquarters. Finland: The sample was taken from the so called "Blue Book - Salesleads database" which is edited by Helsinki Media Company Oy (Sanoma Magasines Finland). This data base contains of about 170,000 data sets and is being updated every two months. France: The sample was drawn from the “INSEE Siren file” (the national office of statistics). INSEE, as a public organisation, is responsible for gathering all economic and social data in France. These data sets are being updated every two months. Greece : The sample was drawn from the address inventory of ICAP (major establishment data base for Greece and member of the European Association of Directory and database Publishers). The data base is being updated every 18 months and also contains public sector addresses. Additionally, public sector addresses were taken from the national telephone inventory.


72

Great Britain: The sample was drawn from "BT’s Business Database". This is a representative data base of all establishments in the UK having a telephone number (including addresses by BT, Mercury, cable and about 92 further telecom carriers). The data base consists of about 1.6 m addresses and is being updated every two months. Italy: The sample was drawn from Dun & Bradstreet's data base. This data base is considered to be the most reliable source for Italy. Spain: The sample was drawn from Schober's data base. This data base is the most voluminous record as regards number of establishments for Spain.

Survey period The interviews were carried out in the following period: 21.03. -15.05.2002

Interviews undertaken

Total: 3,139

Germany 16.0 min Great Britain 16.2 min

Finland 16.4 min Italy 18.2 min France 14.1 min Spain 16.4 min

Average interview length

Greece 15.1 min

Interviewers used Total: 212 Quality control All interviewers were instructed about the questionnaire before the

beginning of field work. Field work was permanently controlled by supervisors. Because of computer aided realisation of interviews systematic errors of data gathering can be excluded. Furthermore the proper realisation of interviews was monitored according to institute standards. Following is the share of interviews monitored (by telephone):

Germany: 10% Finland: 5% France: 10% Greece: 20%

Great Britain: 8% Italy: 10% Spain: 30%

Additional comments to the data set

Question A8: Turnover indication in national currencies were translated in Euro except for UK. Some indications seem to be very high, others very low. National institutes have re-examined and verified statements by calling the respondents again and reassured turnover answers were as stated.

Problems encountered

In all countries fulfilling the quota for the largest establishments was difficult (mainly 500+ / partly also 200-499 employees). In these establishments on the one hand the availability of target persons is significantly lower, on the other hand are these target persons "over-researched” (which in part results in a general interdiction to take part in surveys). Due to this in France it was necessary to adapt the quota in order to achieve the number interviews aimed at (i.e. interviews - as far as possible - carried out in establishments of the next smaller size class).

Data supply One labelled SPSS-data set of the main survey of all interviews.


73

Field report and outcomes

D FIN F EL UK I E 1 Sample (gross), i.e. number dialled at least once 4917 1923 8061 1728 8726 10846 8489 1.1 Telephone number does not exist 787 47 598 43 416 1160 808 1.2 Not an establishment (i.e. private household, etc.) 46 15 0 2 0 0 235 1.3 Fax machine/ Modem 81 0 152 31 0 0 519 1.4 Quota completed, therefore address not used 0 849 1599 2 2659 848 1397 1.5 No target person in establishment 858 226 1261 35 1766 822 2043 1.6 Language problems 0 15 0 0 0 0 10 1.7 SUM (1.1+1.2+1.3+1.4+1.5+1.6) 1753 1152 3610 113 4841 2830 5012 2 Net sample (1 minus 1.7) 3164 771 4451 1615 3885 8016 3477

2.1 Nobody picks up phone (and max. contacts not yet exhausted) 325 2 326 229 32 804 18

2.2 Line busy, engaged 45 0 31 235 2 1852 9 2.3 Answering machine 111 4 82 15 0 0 482

2.4 Contact person refuses (i.e. refusal at reception, switchboard) 436 228 912 38 1354 1056 1022

2.5 Target person refuses 1044 204 1569 107 1672 1410 896 2.6 no appointment during fieldwork period possible 33 14 356 36 176 680 203 2.7 open appointment 604 4 642 644 52 1668 111 2.8 target person is ill/ cannot follow the interview 1 3 18 0 0 0 18 2.9 Interview abandoned 53 1 14 4 97 34 102 2.10 Interview error, cannot be used 0 5 0 6 0 0 109 2.11 SUM (2.1+2.2+2.3+2.4+2.5+2.6+2.7+2.8+2.9+2.10) 2652 465 3950 1314 3385 7504 2970 2.12 SUCCESSFUL INTERVIEWS 512 306 501 301 500 512 507

3 Completion Rate (2.12 / (2.11+2.12)), in % 16.18 39.69 11.25 18.63 12.87 6.38 14.58

Target and actual numbers of interviews F D I E UK FIN EL Quota Group required - achieved - required - achieved -

1 - 9 30 33 30 34 33 32 18 18 17 10 - 49 35 36 36 37 35 35 21 21 22 50 - 199 35 38 37 40 35 35 21 21 25 200 - 499 40 44 41 43 41 40 24 28 22 500+ 15 9 14 13 15 15 9 9 6

I Manufacturing, construction, primary sector

Sum 155 160 158 167 159 157 93 97 92 1 - 9 45 50 47 45 46 45 27 28 27 10 - 49 40 42 41 41 43 40 24 24 25 50 - 199 30 28 31 26 30 30 18 18 18 200 - 499 15 19 15 16 15 15 9 5 9 500+ 10 5 10 8 10 10 6 5 6

II Distribution, catering, transport and communication

Sum 140 144 144 136 144 140 84 80 85 1 - 9 30 32 30 34 30 30 18 16 17 10 - 49 20 19 21 23 21 20 12 14 11 50 - 199 10 13 10 17 10 10 6 6 8 200 - 499 10 13 10 6 10 10 6 7 6 500+ 10 8 9 4 7 8 6 6 6

III Financial and business services

Sum 80 85 80 84 78 78 48 49 48 1 - 9 20 20 24 19 20 20 12 13 13 10 - 49 25 29 25 26 25 25 15 16 16 50 - 199 30 22 30 34 30 30 18 18 18 200 - 499 35 32 35 31 35 35 21 23 20 500+ 15 9 16 15 16 15 9 10 9

IV Public administration, education, health, other personal and social services

Sum 125 112 130 125 126 125 75 80 76 Total 500 501 512 512 507 500 300 306 301

9.2.3 Weighting For the SIBIS DMS a sample stratified by sector/ size cells was used which ensured that in each sector, establishments from all size classes were sampled. In order to be able to raise figures to national level, some form of weighting is required which adequately reflects the structure and distribution of establishments (or related variables) in the universe of the


74

respective country (and, by implication, EU15). All presentation of SIBIS results indicates clearly which of these weighting schemes was used.

9.2.3.1 Original weight Within each country, the interviews were split according to a quota plan which guaranteed that the sample is not dominated by micro and small companies. The quotas roughly reflect the distribution of employment over sector and establishment size bands in the EU, and derive from research into establishment sampling frames undertaken for previous studies by Infratest and GfK in the course of ECaTT. They represent best estimates, but do not take account of country differences. The quota scheme looks as follows:

empirica SUGGESTED QUOTAS: Sectors (aggregated) X Size

1- 9 10 - 49 50 - 199 200 - 499 500+ Total % of total abs

% of total

% of total

% of total

% of total

% of total

Quota I 6% 30 7% 35 7% 35 8% 40 3% 15 31% 155

1 Mining, Energy

2 Manufacturing3 Construction

Quota II 9% 45 8% 40 6% 30 3% 15 2% 10 28% 140

4 Distribution

5 Hotels, Restaurants6 Transport, Communication

Quota III 6% 30 4% 20 2% 10 2% 10 2% 10 16% 80

7 Banking, Insurance

8 Business Services

Quota IV 4% 20 5% 25 6% 30 7% 35 3% 15 25% 125

9 Public Administration

10 Education

11 Health and Social Work

12 Other personal or social services

Total 25% 125 24% 120 21% 105 20% 100 10% 50 100% 500

Financial & Business Servicesincludes:

Public administration, education, health, other personal & social servicesincludes:

Manufacturing, Construction, Primary Sector,includes:

Distribution, Catering, Transport & Communicationincludes:

(The absolute numbers refer to countries with n=500) Weighting was used in cases where the quotas could not be reached exactly in line with this quota plan (mostly due to the limited absolute number of establishments in the two biggest size classes). Note that because of the use of a single quota plan for all countries, country differences in the distribution of employment over establishment size bands which occur in reality are not reflected in the data. This is due the lack of available data on the distribution of employment across establishments size bands in almost all EU Member States, and constitutes a considerable problem. This weight is, therefore, not used for presenting SIBIS results.

9.2.3.2 Weighting by employment The data available on the distribution of employment over establishment size bands is very limited for most EU Member States. SIBIS used data from a variety of sources, including � BT database (United Kingdom) � ISTAT Industry and Services Intermediate Census – latest available, 1996 (Italy)


75

� National Statistical Service of Greece - latest available, 1995 (Greece) � SIREN (France) � Tilstokeskus Official Statistics (Finland) � Heins + Partner B-Pool (Germany) � Schober Business Pool (Spain) and adjusted using data from the DG Enterprise/ Eurostat SME Database (latest available, 1997), to estimate the establishment/ employment structure for each country in the sample. The table below shows the resulting establishment size structure per country. Country

D E EL F FIN I UK EU7

1 to 9 23% 23% 59% 17% 13% 38% 14% 23%

10 - 49 19% 28% 16% 22% 16% 22% 31% 24%

50 - 199 21% 21% 8% 21% 19% 14% 26% 20%

200 - 499 13% 9% 6% 14% 16% 7% 13% 12%

Est

ablis

hmen

t si

ze b

and

500 and more 25% 18% 10% 25% 37% 19% 17% 21%

Total Column % 100% 100% 100% 100% 100% 100% 100% 100%

Using this weight, the weighted sample for each country therefore reflects employee distribution between the five establishment size bands within that country. This means that a data reference of, for example, ”20% of all establishments in country A” should be understood to mean ”establishments accounting for 20% of all employees in country A”.

9.2.3.3 Weighting by employment for EU7 averages Additionally another weighting factor was created to calculate average figures for all countries in the sample (which together represent roughly 82% percentage of total EU employment). Each country is represented in this weight according to its share in the total employment of the 7 EU countries in which the survey was conducted.

9.2.4 Sample characteristics and effect of weighting Total

unweighted weighted by employment*

% unweighted % weighted by employment*

Total sample 3139 3139 100.0 100.0 Country Finland 306 306 9.7 9.7 France 501 501 16.0 16.0 Germany 512 512 16.3 16.3 Greece 301 301 9.6 9.6 Italy 512 512 16.3 16.3 Spain 507 507 16.2 16.2 UK 500 500 15.9 15.9 Total

unweighted weighted by employment

(EU7) % unweighted

% weighted by employment

(EU7) Number of staff at site up to 9 803 713 25.6 22.7


76

Total

unweighted weighted by employment* % unweighted % weighted by

employment* 10 to 49 769 746 24.5 23.8 50 to 199 668 648 21.3 20.6 200 to 499 626 364 19.9 11.6 500 and more 273 668 8.7 21.3 Industry Sector primary: manufacturing, energy, mining, construction

990 989 31.5 31.5

secondary: distribution, catering, communication and transport

873 878 27.8 28.0

third: financial and business services

502 501 16.0 15.9

fourth: public administration, health, education, other social/ personal

774 772 24.7 24.6

Type of organisation headquarter of international operating organisation 267 348 8.5 11.1

headquarter of organisation only operating in country 607 536 19.3 17.1

division/ branch of international operating organisation 256 290 8.2 9.2

division/ branc h of organisation only operating in the respective country

309 312 9.8 9.9

other type 76 40 2.4 1.3 only one establishment 1617 1608 51.5 51.2 don’t know 7 6 0.2 0.2


77

9.3 Questionnaires

9.3.1 Questionnaire for the General Population Survey (GPS) Structure of the questionnaire: Module IN: Introduction and screening � Age � Educational attainment � Employment status � Occupation � Type of organisation � Main working place Module A: Basic ICT equipment access and use � Use of computer � Use of e-mail � Internet access and use � Methods of Internet access � Effects of Internet use � Barriers to using the Internet � Access to mobile phone � Mobile data services � Effects of mobile phone use Module B: E-commerce and other uses of the Internet � Online activities � Barriers to buying online Module D: Skills � Internet user experience and know-how Module L: e-Health � Use of online health information � Perception regarding the trust placed in online health information provider � Rationale for health info search Module J: Security � Security concerns � Reporting of security violations � Security-related awareness and behaviour Module K: e-Government � Preference for e-Government services � e-Government experience � Barriers to e-Government Module E: Telework � Home-based telework � Intensity of home-based teleworking � Duration of telework: � Financing of tele-workplace


78

� Interest in telework: � Perceived feasibility � Effects of telework Module F: Mobile work � Mobile work (Intensity): � Mobile telework Module G: Tele-cooperation/Tele-collaboration � Co-operation with external contacts using ICTs � e-Lancing Module H: Outcomes of work � Work-family balance � Job quality � Job satisfaction Module C: Educational attainment and lifelong learning � Company-provided training � Training provided by other organisations � Self-directed learning � Modes of training (use of eLearning) Module Z: Standard demography � Household size � Disability � Income


No

Branching

Question Answer categories

79

Module IN: Introduction and Screener questions GPS INTRO TEXT

ALL

Hello my name is ... calling for ...

We are presently conducting a scientific survey for the European Union in fifteen countries. I would like to talk to the person in your household, that is at least 15 years old, and whose birthday is up next.

[INTERVIEWER: IF NECESSARY] To topic of this survey is the internet and the work life.

[INTERVIEWER: IF NECESSARY] Your answers will be held strictly confidential and will be used only for scientific purposes.

[INTERVIEWER: IF NECESSARY] Your participation is very important to us, because you have been selected through a statistical procedure that will result in a typical selection of people in [COUNTRY]

[PROMPT: The interview will last about 15 minutes]

IN1

ALL

Would you please tell me in which year you were born? |_1_|_9_|__|__|

[DK]

PROGRAMMING: IF respondent born after 1986 END INTERVIEW!

IN2

ALL

Have you finished your full-time education or are you still studying?

(1) finished education already

(2) Is still studying

(3) DK

IN3

IF IN2=1

At what age did you finish full-time education?

[PROMPT: HOW OLD WHERE YOU WHEN YOU STOPPED FULL-TIME EDUCATION]

|__|__| years

[DK]

Transition X1

IF IN2=1

I would like to ask you a few questions regarding your employment situation.

IN4

IF IN2=1

At present are you in paid work either as an employee, civil servant or as self-employed?

(1) yes

(2) no

(3) DK

IN5a

IF IN4=1

Do you have one job or more than one job at present? (1) only one job

(2) more than one job

(3) DK

IN5b

IF IN5a=2,3

How many hours per week do you normally work, including paid overtime and taking all your jobs together?

|__|__|__|

[DK]

Transition X2

IF IN5a=2

For answering the following questions, please consider only your main job, i.e. the job you spend most of your working time on.

IN6

IF IN4=1

And are you ... [in your main job]

[INTERVIEWER: Read out answer categories]

(1) self-employed

(2) in paid employment (including civil servants)

(3) DK

IN7

IF IN4=2,3

And are you ...


(1) temporarily not working, e.g. because of unemployment, paternal leave or illness

(2) retired

(3) not working, because you are responsible for ordinary shopping and looking after the home.

(4) DK

IN8

IF IN6=1

What kind of work do you do? Are you a ...


(1) Professional (eg doctor, lawyer, accountant, architect)

(2) Farmer, fisherman

(3) Business proprietor, owner of company/shop, craftsmen, other self-employed person

(4) DK


No

Branching


80

IN9

IF IN8=3

How many employees do you have? |__|__|__|__|__|__|

[DK]

IN10

IF IN4=1

[In your main job,] Are you working full-time or part-time? (1) full-time

(2) part-time

(3) DK

IN11

IF IN4=1

How many hours per week do you normally work in your main job, [PROGRAMMER: Skip the following if IN6=1] including paid overtime?

|__|__|__|

[DK]

[PROGRAMMER: INCLUDE CHECK WITH IN5B]

IN12

IF IN6=2,3

Are you employed ...


(1) on an unlimited permanent contract

(2) on a fixed term contract

(3) on a temporary employment agency contract

(4) on apprenticeship or other training scheme

(5) other

(6) DK

IN13

IF IN7=2,3,4

Would you like to be in paid work? (1) yes

(2) no

(3) DK

IN14

IF IN8=3 or IN6=2

What kind of work do you do? Are you ...


(1) working mainly at a desk

(2) not working at a desk, but travelling (salesmen, driver, ...),

(3) not working at a desk, but in a service job (retail shop, restaurant, ...)

(4) doing some other kind of work

(5) DK

IN15

IF IN6=2

What position do you hold?


(1) Employed professional (employed lawyer, medical practitioner, accountant, architect etc.),

(2) Management

(3) Other non-manual employee

(4) Manual worker

(5) DK

IN16

IF IN15=2

And which of the following best describes your position?


(1) General management, director or top management (managing directors, director general, other director)

(2) Middle management, other management (department head, junior manager, teacher, technician)

(3) DK

IN17

IF IN15=4

And which of the following best describes your position?


(1) Supervisor

(2) Skilled manual worker

(3) Other (unskilled) manual worker, servant

(4) DK

IN18

IF IN15=2

How many employees you are responsible for? |__|__|__|__|__|__|

[DK]


No

Branching


81

IN19

IF IN6=2

For what kind of organisation do you work?


(1) a private firm or business or a limited company

(2) in the public sector or in a charity, voluntary organisation or trust

[PROMPT - DO NOT READ: (2) includes public companies, local or central government, civil service, armed forces, council, schools, universities or other grant funded education establishments, public authorities, charities, voluntary organisations]

(3) DK

IN20

IF IN6=2

How many employees work in the company/organisation for which you work?


(1) <10

(2) 10-49

(3) 50-249

(4) 250 and more

(5) DK

IN21

IF IN4=1

Do you work mainly ...


(1) in your own home

(2) in the same grounds or buildings as your home

(3) in different places using home as a base (e.g. travelling salesman, free insurance agent etc.)

(4) somewhere quite separate from home

(5) DK

Module A: Basic ICT equipment access and use GPS Transition A

ALL

Now we would like to ask you a few questions about computers and the Internet

A1

ALL

Have you used a PC, Mac or any other computer, for work or for private purposes - in the last four weeks?

(1) yes

(2) no

(3) DK

A3

IF A1=1

Have you sent or received any e-mail messages, for work or for private purposes, during the last four weeks?

(1) yes

(2) no

(3) DK

A4a

IF A3=1

How many of your friends and relatives have their own email address?


(1) all or almost all

(2) about three quarters

(3) about half

(4) about one quarter

(5) only few or no-one

(6) DK

A4b

IF A4a<5

And with how many of your friends and relatives do you communicate regularly via email?




(3) about half



(6) DK

A5

ALL

Do you have access to the Internet in your home? (1) yes

(2) no

(3) DK

A6

IF A5=2

Did you once have Internet access in your home? (1) yes

(2) no

(3) DK


No

Branching


82

A7

ALL

Have you used the Internet at least once in the last four weeks, at home, at school or work or at any other place?

(1) yes

(2) no

(3) DK

A8

IF A7=2,3

Have you used it in the last 12 months at least once? (1) yes

(2) no

(3) DK

A9

FOR (a):

IF A7=1 and A5=1

FOR (b)-(f):

IF A7=1

How much time do you spend in a typical week on using the Internet ... [item]

[INTERVIEWER: Read out answer categories for the first 2 items]

(a) at home?

(b) at the workplace?

(c) at school, university or another educational institution?

(d) at a public place where Internet access is free?

(e) at an Internet café or other place where you have to pay for access?

(f) at another place not mentioned yet

FOR EACH

(1) none

(2) less than 1 hour

(3) between 1 and 5 hours



(6) more than 20 hours

(7) DK

A10

IF A7=1 or A8=1

When did you use the Internet for the first time?


(1) < 6 months ago

(2) 6 - 12 months ago

(3) 1 year - 2 years ago

(4) 2 years + ago

(5) DK

A11a

IF A5=1

Do you know what technical method you use at home to connect to the Internet?

(1) yes

(2) no

(3) NA

A11b

IF A11a=1,3

I will read to you a number of methods to access the Internet. Which of these do you use at home?

[INTERVIEWER: Read out and code those that apply]

MULTIPLE ANSWERS

(1) Dial-up with modem

(2) Cable Modem

(3) Leased line

(4) xDSL

(5) ISDN

(6) T1 or T3 line [TRANSLATOR: Digital Multiplex connection]

(7) Internet access via satellite

(8) Other not mentioned (e.g. mobile)

(9) DK

A12

IF A11b=2,3,4,5,6,7

At home, did you have a connection before which was slower than your current one?

(1) yes

(2) no

(3) DK

A13

IF A12=1

Since moving to this faster type of connection, has the amount of time you spend online per week decreased, increased or remained roughly the same?

(1) Decreased

(2) Increased

(3) Remained roughly the same

(4) DK

A14

IF A7=1

In the last four weeks, have you accessed the Internet in any other way than via PC or Mac, at least once?

(1) yes

(2) no

(3) DK


No

Branching


83

A15

IF A14=1

Which devices did you use for that: Did you use ...

[INTERVIEWER: Read out and code those that apply]

MULTIPLE ANSWERS

(1) Digital TV*,

(2) a PDA or palmtop,

(3) a mobile phone with WAP or 2.5G** capability,

(4) a game console

(5) other

(6) DK

[* TRANSLATOR: Make sure that you take local brand names and colloquial terms into account]

** TRANSLATOR: Use term used in your country (e.g. Germany: GPRS)]

A18

IF A7=2,3

Now I will read to you a list of statements about the Internet.

Please tell me for each statement whether you agree completely, agree somewhat or do not agree.

The Internet ... [item]. Do you ...

(a) requires advanced computer skills,

(b) is not easy enough to get access to,

(c) is too time consuming,

(d) is too expensive to use,

(e) lacks useful or interesting information

(f) is not something for me

FOR EACH

(1) agree completely

(2) agree somewhat

(3) or do you not agree

(4) DK

A19

ALL

Do you have a mobile phone for your own personal use? (1) yes

(2) no

(3) DK

A20

ALL

How many of your friends and relatives have a mobile phone for their personal use?




(3) about half



(6) DK

A23

IF A19=1 and A15~=3

Have you used your mobile phone to view webpages or WAP pages, or to read your email, at least once in the last 4 weeks?

[TRANSLATORS: Confusion with SMS* to be avoided!]

(1) yes

(2) no

(3) DK

A26

IF A23=1

Have you used your mobile phone at least once in the last 12 months to make any purchases in the Internet, to download online information you are charged for or to make online payments?

(1) yes

(2) no

(3) DK

A27

IF A19=1

Have you, in the last four weeks, used SMS* messages for ...

(a) communication with other people?

(b) paying for purchases, admission tickets or something similar?

(c) paying for downloads such as ringing tones?

(d) receiving financial information, sport results or other subscription services?

[* TRANSLATOR: Check if another term is more common in your country]

FOR EACH

(1) yes

(2) no

(3) DK


No

Branching


84

A30

IF A19=1

(For (d) and (e)I: IF A19=1 and (A8=1 or A7=1) and IN4=1)

Now, think about what your everyday life would be like if you didn’t have a mobile phone. Please tell me how much you agree that if you didn’t have a mobile phone (ITEM). Would you say that you …


(a) you would often not be able to contact your friends and family, or be reached by them

(b) you would be less exposed to dangerous electromagnetic radiation

(c) you would be more helpless in case of emergencies

(d) you would not receive some of the information you need for your job

(e) you would have less exchange with some of your business contacts

(f) you would have less fun

FOR EACH:


(2) agree somewhat

(3) do not agree

(4) DK

Module B: E-commerce and other uses of the Internet GPS Transition B

IF A8=1 or A7=1

Now I would like to ask you a few questions about the Internet.

PROGRAMMING: B1 to B2: for each item in B1=1 ask directly B2, then go to next item in B1

B1

IF A8=1 or A7=1

You can use the Internet for many purposes. I´m going to read you a list of things you can do online and ask you whether you have done this online for your private purposes. For your private purposes, have you used it in the last 12 months...

(a) to find information about a product or service

(b) to order a product or service

(c) to conduct online-banking or to buy financial products

(d) to search for any health-related information

(e) to look for a job

FOR EACH

(1) yes

(2) no

(3) DK

B2

IF B1=1 and A7=1

[FOR EACH B1 ITEM] Have you done so in the last four weeks? (1) yes

(2) no

(3) DK

B5

IF A7=1

(For (c) and (d): IF A7=1 and IN4=1)

Many people in this country still do not have access to the Internet yet. Now please imagine our country were without the Internet for one month. What would it mean for your everyday life?

Please tell me how much you agree that if our country were without the Internet for a month you would (ITEM). Would you say that you would …


(a) be less well informed as a consumer

(b) feel socially excluded

(c) not receive some of the information you need for your job

(d) have less communication with some of your contacts at work / your business contacts

(e) have less contact with some of your friends

(f) have less fun

FOR EACH:


(2) agree somewhat

(3) do not agree

(4) DK


No

Branching


85

Module D: Skills GPS D1

IF A7=1 or A8=1

[Do not ask item (h) in UK, IRL, USA]

I would like to ask you a few questions about your skills in using the Internet. How confident would you feel... [item]

Please tell me whether you feel..


(a) using a search engine (such as Google or Yahoo) to find information on the Internet [TRANSLATORS: List two most widely used search engine brands in your country20]

(b) identifying the source of information provided on the Internet (c) using e-mail to communicate with others

(d) using Internet chat-rooms to contact other people

(e) using the Internet to make telephone calls

(f) creating a personal web / Internet page

(g) downloading and installing software onto a computer

[PROGRAMMING: Do not ask item (h) in UK, IRL, USA]

(h) understanding the content of websites written in English

FOR EACH

(1) very confident

(2) fairly confident

(3) not confident

(4) Do not know what this means [DO NOT READ OUT]

(5) DK

Module L: e-Health GPS Transition L

IF B1(d)=1

You said before, that you have used the Internet to search for health-related information:

L1

IF B1(d)=1

Have you been able to find health related information on the Internet? (1) yes

(2) no

(3) DK

L2

IF L1=1

Was the information suitable for your needs? (1) yes

(2) no

(3) DK

L3

IF L2=1

Websites with health related information are available in many languages.

When you searched, did you find Websites in your mother tongue sufficient or did you have to expand your search and consult sites in other languages, or did you even have to rely solely on sites in other languages?

(1) Websites in mother tongue were sufficient

(2) Had to expand my search and consult websites in other languages too

(3) Had to rely solely on websites in other languages

(4) DK

L4

IF B1(d)=1

And for what reasons did you search health-related information on the Internet?

Did you search health-related information on the Internet to ...[item]

(a) seek a second opinion on your own, a family member’s, or a friend’s medical diagnosis?

(b) be better informed on your general health?

(c) gather additional information since you care for an ill person or a person with a disability?

FOR EACH

(1) yes

(2) no

(3) DK

20 For example, check http://www.jupitermmxi.com/europelanding.html


No

Branching


86

L5

IF B1(d)=1

How trustworthy would you consider each of the following providers of health-related information:

[Item] : Are those ...


(a) Universities and other non-profit organisations active in the health sector / the health field

(b) pharmaceutical companies

(c) private health insurance providers

(d) patient advocacy and self-help groups

(e) hospitals

(f) professional medical associations

FOR EACH

(1) very trustworthy

(2) fairly trustworthy

(3) not trustworthy

(4) DK

Module J: Security GPS Transition J

IF A7=1

Now the topic is internet security.

J1

IF A7=1

How concerned are you about .[item]: Are you ...


(a) data security on the Internet, i.e. the loss or manipulation of your data?

(b) privacy and confidentiality on the Internet, i.e. personal information about you being misused by third parties?

FOR EACH

(1) very concerned

(2) somewhat concerned

(3) not concerned

(4) DK

J2

IF J1(a)=1,2 or J1(b)=1,2

Are these concerns stopping you from using the Internet to buy goods or services online: often, sometimes, or never?

(1) often

(2) sometimes

(3) never

(4) DK

J3

IF A7=1

Would you report violations of your on-line security, privacy and confidentiality to a third independent party, for example a public agency created for this task?


(1) yes, v ery likely

(2) maybe

(3) no

(4) DK J4

IF J3=1,2,3

Would it be easier for you to do so if you could do it anonymously? (1) yes

(2) no

(3) DK

J5

IF A7=1 & (B1(b)=1 or B1(c)=1)

How often are you aware of security features of websites when you use the Internet to buy online: often, sometimes or never?

(1) often

(2) sometimes

(3) never

(4) DK

J6

IF A7=1 & (B1(b)=1 or B1(c)=1)

And how often do you take security features of websites into account when deciding about whether to buy online: often, sometimes or never?

(1) often

(2) sometimes

(3) never

(4) DK

Module K: e-Government GPS Transition K

IF A7=1

Now I would like to ask you a few questions about the contact to government agencies through the Internet.

PROGRAMMING: K1 to K3: for each item in K1=1 ask directly K2, If K2=1 ask directly K3, then go to next item in K1


No

Branching


87

K1

IF A7=1

Here is a list of activities that require citizens to get in touch with public administration.

For each activity, please answer whether you would prefer to use the Internet or prefer to use the traditional way, that is face-to-face, by postal mail, fax or phone:

[INTERVIEWER: Repeat answer categories for the first 2 items]

(a) Tax declaration / filing your income tax return

(b) Use of job search services of public employment service

(c) Request for passport, driver's licence, birth certificates or other personal documents

(d) Car registration

(e) Declaration to the police, e.g. in case of reporting theft

(f) Searches for books in public libraries

(g) Announcement of change of address

FOR EACH

(1) Internet

(2) traditional way

(3) do not use this service [DO NOT READ OUT]

(4) DK

K2

IF K1=1

FOR EACH

Is it possible to use the Internet for this in the area you live?

FOR EACH

(1) yes

(2) no

(3) DK

K3

IF K2=1

FOR EACH

Have you ever tried using the Internet for this?

FOR EACH

(1) yes

(2) no

(3) DK

K4

IF A7=1

For each of the following statements about online services of public administration, please indicate whether you agree. Public services on the Internet ...[item].


(a) are not useful enough

(b) are faster than the traditional way

(c) require that you install special equipment or software

(d) reduce the number of mistakes public authorities make

(e) do not seem as safe as using the traditional way

(f) make it possible to deal with the authorities at more convenient times

(g) make it possible to deal with the authorities at more convenient locations, e.g. from home or from the workplace

(h) are difficult to use


(2) agree somewhat

(3) do not agree

(4) DK

Module E: Telework GPS Transition E

IF IN4=1 or IN13=1 or IN7=1

Now let’s talk about another topic:

With the help of telephone, fax and computer, many types of work can be done from home. If work results are transferred electronically, this is sometimes called telework.

E1

IF IN4=1

Do you presently telework from home, for at least some of your working time?

(1) yes

(2) no

(3) DK

E2

IF E1=2,3

Have you teleworked on a regular basis before, in the last five years? (1) yes

(2) no

(3) DK

E3

IF E2=1

Did you spend, on average, at least one full working day a week at home when you were teleworking?

(1) yes

(2) no

(3) DK


No

Branching


88

E4

IF E1=1

Do you spend, on average, at least one full working day a week teleworking from home?

(1) yes

(2) no

(3) DK

E5

IF E1=1

You indicated before that you work on average [PROGRAMMER: Insert result from IN5b, if blank insert result from IN11] hours per week. How many of these do you spend at home in a typical week?

|__|__|__|

[DK]

[PROGRAMMER: Insert check with IN5b or IN11]

E7

IF E1=1 and IN6=2

Has the equipment you use for teleworking at home been mainly, not mainly but partly, or not at all been paid for by your employer?

(1) mainly paid for by employer

(2) not mainly, but part ly paid for by employer

(3) not at all paid for by employer

(4) DK

E8

IF IN7=1 or IN13=1 or (E1=2,3 or E4=2,3)

If it was offered to you, how interested would you be in ... [item]. Would you be ...


(a) doing almost all your work teleworking at home

(b) telework where you did not spend all your working time, but at least one full working day per week at home

(c) work in an office provided near your home which would allow you to reduce commuting?

FOR EACH

(1) very interested

(2) somewhat interested

(3) not interested

(4) DK

E9a

IF E1=2,3 or E4=2,3

Would you say that your job is feasible for telework, under the assumption that you spend at least one full working day per week at home?

(1) yes

(2) no

(3) DK

E9b

IF E9a=2 and IN6=2

What are the main reasons why you consider your current job not to be feasible for telework? Is it because ...

[INTERVIEWER: Read out answer categories and code all that apply]

MULTIPLE ANSWERS

(1) your company does not permit telework?

(2) your superior does not approve of telework?

(3) your job requires face-to-face contact with customers, colleagues or other persons

(4) your job requires access to machines or other things which cannot be accessed from home

(5) Other reasons (DO NOT READ OUT)

(6) DK

E10

IF E1=1

For what reasons did you start teleworking? Please indicate for each of the following aspects how important it was for your decision to start teleworking. [item] Was this ... for you.


(a) I needed a more peaceful working environment

(b) I want to participate more in family life

(c) I want to be closer to clients or customers

(d) I need to look after a child or an other person who needs care

(e) My company asked me to start teleworking

(f) I want to reduce commuting

(g) I wanted to have more flexibility in how to organise my work

(1) very important

(2) somewhat important

(3) not important

(4) DK


No

Branching


89

E11

IF E1=1

Most working people are not allowed to work from home. Please consider you would not be allowed to telework from home, for whatever reasons.

What would that mean for your ability to do your job? Would it mean that you...[item]. Do you ...


(a) could not be in paid work at all

(b) could not do your job as well as with telework

(c) would have to look for another job which is located closer to your home

(d) would have to reduce your working hours per week

FOR EACH:


(2) agree somewhat

(3) do not agree

(4) DK

Module F: Mobile work GPS Transition F

IF IN4=1

Now let’s talk about the topic of mobile working.

F1

IF IN4=1

In the last four weeks, have you spent any of your working time away from your home and from your main place of work, e.g. on business trips, in the field, travelling or on customer’s premises?

(1) yes

(2) no

(3) DK

F2

IF F1=1

You indicated before that you work on average [PROGRAMMER: Insert result from IN5b, or if blank result from IN11] hours per week. How many of these do you spend away from home and your main place of work?

|__|__|__|

[DK]

[PROGRAMMER: Insert check with IN5b or IN11]

F3

IF F2>5

In the last four weeks, have you used online computer connections when travelling? By this I mean have you accessed the Internet for business purposes, or electronically transferred data to colleagues?

(1) yes

(2) no

(3) DK

F4

IF F3=1

For what purpose did you use these online connections? Have you used these to ...

(a) access the Internet

(b) send or read e-mails

(c) connect to your company's internal computer system

FOR EACH:

(1) yes

(2) no

(3) DK

F5

IF F3=1

Where did you use an online computer connection? Have you used it in the last four weeks at ...

(a) a hotel, conference site or similar location?

(b) another company's premises?

(c) an Internet café or an other commercial teleservice center?

(d) or on the move, using a mobile device for data transfer?

FOR EACH:

(1) yes

(2) no

(3) DK

Module G: Tele-cooperation/Tele-collaboration GPS Transition G

IF IN4=1 and (A1=1 or A7=1)

And how about the use of telecommunication technology at your work place:

G1

IF IN4=1 and A1=1

When you communicate with external contacts, do you sometimes use e-mail, video conference or electronic data transfer? [PROGRAMMER: skip the following if IN6=1] By external persons we mean customers, clients, suppliers, other business contacts, but also colleagues working at other locations of the same company.

(1) yes

(2) no

(3) DK

G2

IF G1=1

In a typical week, how often do you ...[item] for these external contacts?


(a) use e-mail

(b) use video-conferencing

(c) use e-mail attachments or other electronic data transfer

FOR EACH

(1) 10 or more times a day,

(2) at least once a day,

(3) at least once a week

(4) less often than once a week

(5) never

(6) DK


No

Branching


90

G4

IF IN6=1 and A7=1

I would like to know about the role the Internet plays in your business.

Do you sometimes attract new business through the Internet or via e-mail?

(1) yes

(2) no

(3) DK

G5

IF IN6=1 and A7=1

Do you sometimes deliver work results to your clients or customers through the Internet or via e-mail?

(1) yes

(2) no

(3) DK

G6

IF G4=1 and G5=1

Does it sometimes happen that you communicate with clients or customers exclusively by electronic means, i.e. via Internet, e-mail, phone or fax and without meeting face-to-face?

(1) yes

(2) no

(3) DK

Module H: Outcomes of work GPS Transition H

IF IN4=1

I would like to ask you a few more questions about your work.

H1

IF IN4=1

Please tell me for each of the following, how often you experience this. How often do you .. [item]?


(a) Find your work stressful

(b) Come home from work exhausted

(c) Find your job prevents you from giving the time you want to your partner or family

(d) Feel too tired after work to enjoy the things you would like to do at home

(e) Find your partner/family gets fed up with the pressure of your job

FOR EACH

(1) often

(2) sometimes

(3) never

(4) does not apply [DO NOT READ]

(5) DK

H2

IF IN6=2,3

In your current work arrangement, do you agree with the following statements about your job? [item] Do you ...


(a) I have a lot of say over what happens in my job

(b) I need to keep learning new things continuously

(c) I have concerns about whether my job is secure

(d) I have a high income

(e) I can adapt my starting & finishing times to my personal preferences

(f) I can adapt the number of weekly working hours to my personal preferences

FOR EACH:

(1) strongly agree

(2) somewhat agree

(3) disagree

(4) DK

H3

IF IN4=1

On the whole, are you very satisfied, somewhat satisfied, neither satisfied nor dissatisfied, somewhat dissatisfied or very dissatisfied with your job / your main job?

(1) very satisfied

(2) somewhat satisfied

(3) neither satisfied nor dissatisfied

(4) somewhat dissatisfied

(5) very dissatisfied

(6) DK

Module C: Educational attainment and lifelong learning GPS Transiti on C

IF IN4=1 or IN13=1 or IN7=1

Now I would like to ask you a few questions about training and learning.

C2

IF IN6=2,3

Did you participate in some kind of work-related training activities that were provided either by your company or by an other organisation, in the last four weeks?

(1) yes

(2) no

(3) DK

C9b

IF IN7=1 or IN6=1

Did you participate in some kind of training activities with the aim of preparing you for a future job, in the last four weeks?

(1) yes

(2) no

(3) DK


No

Branching


91

C14a

IF IN4=1

Apart from the training that may have been provided by others, did you engage in some kind of self-directed learning related to your work, in the last four weeks?

(1) yes

(2) no

(3) DK

C14b

IF IN7=1 or IN6=1

Apart from the training that may have been provided by others, did you engage in some kind of self-directed learning which was aimed at preparing you for a future job, in the last four weeks?

(1) yes

(2) no

(3) DK

C18

IF A1=1 and (C2=1 or C9b=1 or C14a=1 or C14b=1)

Did you use, in the course of your training and learning in the last four weeks, electronic learning materials such as learning programmes on CD-ROM, in company -internal computer systems or on the Internet?

(1) yes

(2) no

(3) DK

C19

IF C18=1

What did you use? Did you use

(a) CD-ROMs or other so-called offline media such as diskettes, audio or video tapes etc.?

(b) online learning materials provided on the internal computer system of your organisation or through the Internet

FOR EACH

(1) yes

(2) no

(3) DK

C20

IF IN2=2 and A1=1

Did you use, in the course of your studies in the last four weeks, electronic learning material such as learning programmes on CD-ROM, on the internal computer system of your school/university or through the Internet?

(1) yes

(2) no

(3) DK

C21

IF C20=1

What did you use? Did you use

(a) CD-ROMs or other so-called offline media such as diskettes, audio or video tapes etc.?

(b) online learning material provided on the internal computer system of your school/university or through the Internet?

FOR EACH

(1) yes

(2) no

(3) DK

Module Z: Standard demography GPS Finally we would like to ask you a few more questions for statistical

purposes:

Z17

ALL

How many people live in your household, yourself included? |__|__|

[DK]

Z18a

IF Z17>1

How old is the youngest? |__|__|

[DK]

Z18b

IF Z17>1

How many are 15 years and older? |__|__|

[DK]

[PROGRAMMER: Build in check with Z17 and Z18a]

Z14

ALL

Do you have any long-standing illness, disability or infirmity that limits your activities in any way? By long-standing I mean anything that has troubled you over a period of time or that is likely to affect you for a period of time.

(1) yes

(2) no

(3) DK

Z19

ALL

We also need some information about the income of this household to be able to analyse this survey.

What is your household's monthly net income (after tax)? Please count the total wages and salaries per month of all members of this household; all pensions and social security benefits; child allowances and any other income like rents etc.

[ADD IF NECESSARY: Of course, your answer (as all other answers in this interview) will be treated confidentially and referring back to you or your household will be impossible.]

Is it less or more than <income 1>, <income 2> or <income 3>.

(1) less than <income 1>

(2) <income 1> to less than <income 2>

(3<income 2> to less than <income 3>.

(4 <income 3> or more

(5) DK

(6) Refusal


No

Branching


92

Z20

ALL

Looking back over the last three years, has your household income increased, decreased, or remained roughly the same?

(1) increased

(2) decreased

(3) remained roughly the same

(4) DK

(5) Refusal

Z21

ALL

Gender

[INTERVIEWER: Ask only if in doubt]

(1) male

(2) female

Data provided by survey organisation Categories

P0 Survey Number 101438

P1 Country Code |__|__|

P2 Interview Number |__|__|__|__|

P3 Date of Interview: Day |__|__|,

Month |__|__|

P4 Time of the beginning of the interview (USE 24 HOUR CLOCK): Hour |__|__|,

Minute |__|__|

P5 Number of minutes the interview lastet |__|__|__|

P6 Size of locality |__|__|

P7 Region |__|__|

P8a Postal Code / Area code

must be convertible into NUTS 2 regions

|__|__|__|__|__|__|__|__|

P8b NUTS 2 regions |__|__|__|__|__|__|__|__|

P9 Interviewer Number |__|__|__|__|__|

P10 Weighting Factor |__| . |__|__|__|__|__|

P11 Language of interview (Luxembourg, Belgium, Finland, Switzerland) |__|


93

9.3.2 Questionnaire for the Decision Maker Survey (DMS) Structure of the questionnaire: � Introduction and Screener Section Module A: Basic characteristics � Type of organisation � Number of staff (employees) � Turnover Module B: Module B: Basic ICTs take-up and intensity of use (e-Business) � e-Mail � Internet � Intranet � EDI � Video-conferencing � Call-centre � Staff access to ICTs Module C: e-Commerce � Website/ Internet presence � Online sales � Barriers to e-commerce (selling) � Benefits from / Outcomes of e-commerce � Online procurement � Barriers to online procurement � Benefits from/ Outcomes of online procurement � Online supply chain integration � e-Marketplaces Module D: e-Business security � Security breaches � Information security strategy � Barriers to security � Security provisions Module F: e-Government � Use of e-Government services � Barriers to e-Government Module G: Website accessibility � Design for all” / ”universal design” principle awareness Module E: R&D � R&D staff � Computer staff in R&D unit(s) � IT staff providing computer services to R&D � Outsourced computer services for R&D � Vacancies in IT for R&D


No

Branching


94

Introduction and Screener Section DMS Database/address information: Categories

A11

ALL

Main business activity

PROGRAMMER: Copy from database

NACE code (2-digit level)

I__I__I

1 Mining, Energy (includes NACE 10 - 14/ 40, 41)

2 Manufacturing (includes NACE 15 - 37)

3 Construction (includes NACE 45)

4 Distribution (includes NACE 50, 51, 52)

5 Hotels, Restaurants (includes NACE 55)

6 Transport, Communication (includes NACE 60, 61,62, 63, 64)

7 Banking, Insurance (includes NACE 65, 66, 67)

8 Business Services (includes NACE 70, 71, 72, 73, 74 [except: 74.13])

9 Public Administration (includes NACE 75 [except 75.2])

10 Education (includes NACE 80)

11 Health and Social Work (includes NACE 85)

12 Other personal or social services (includes NACE 90, 91, 92, 93)

A12

ALL

Establishment/ size (if available)

PROGRAMMER: Copy from database

According to database

a) OPEN (if available)

I__I__I__I__I__I__I 6-digit numerical

[1] not available from database

and

b) in categories, i.e.

(1) 0 - 9

(2) 10 - 49

(3) 50 - 199

(4) 200 - 499

(5) 500+

(6) not available from database


No

Branching


95

S1 (INTRO)

ALL

At reception/switchboard:

Good morning/good afternoon. My name is ... . I am calling for ... [name of institute].

We are presently conducting a scientific survey in several European countries. The topic is the use of information and communications technologies. I would like to talk to the person who is responsible for DP/IT decisions at your location.

INT.: NOTE:

THIS SHOULD BE THE HEAD OF THE DP/IT DPT. OR A SENIOR PERSON IN THE DP/IT DPT. IN SMALLER FIRMS IT CAN ALSO BE THE MANAGING DIRECTOR, THE GENERAL MANAGER OR THE OWNER.

INT.: ADD, IF NECESSARY:

Your participation is very important to us, because your firm has been selected through a statistical procedure that will result in a typical selection of firms in [COUNTRY]


The interview will last approx. 15 minutes

(1) put through to target person Ù CONTINUE

(2) target person currently unavailable Ù MAKE APPOINTMENT FOR CALLBACK

(3) no such person Ù TERMINATE

(4) refusal to participate Ù END

S2 (INTRO)

ALL

At target person:

Good morning/good afternoon. My name is ... . I am calling for ... [name of institute].

We are presently conducting a scientific survey in several European countries. The topic is the use of information and communications technologies. We are talking to people who are responsible for DP/IT decisions at their respective locations.

Can I just check: Would you be the right person to talk to at your location and can we do the interview now?


Your participation is very important to us, because your firm has been selected through a statistical procedure that will result in a typical selection of firms in [COUNTRY]


The interview will last approx. 15 minutes

(1) yes, interview now Ù CONTINUE

(2) yes but no time at the moment Ù MAKE APPOINTMENT FOR CALLBACK

(3) no, other person responsible at this location Ù ASK TO BE PUT THROUGH TO THAT PERSON , RESPECTIVELY ASK FOR CONTACT DETAILS. AT NEW TARGET PERSON START AGAIN WITH QUESTION S2

(4) no, other person responsible at another location Ù TERMINATE

(5) refusal to participate Ù TERMINATE

A13

ALL

Function of target person

What is your position in your establishment? What of the following is the most appropriate?

INT.: READ OUT. SINGLE ANSWER.

(1) Owner/Proprieter

(2) Managing Director/Board Member

(3) Head of Establishment/Site

(4) Head of IT/DP

(5) Other senior member of IT/DP Department

(6) Other Ù TERMINATE

Module A: Basic characteristics DMS Transition A

ALL Let us start with some general questions about your establishment.


No

Branching


96

A2

ALL

Does your organisation have only one establishment, or has it more than one establishment?

By establishment we mean a single indentifiable unit at a particular address.

[TRANSLATOR: Be very careful to identify a correct translation for "establishment"]

(1) only one establishment

(2) more than one establishment

(3) DK

A4

IF A2=2

How many employees does your organisation have in total in [country], including yourself?

INT.: IF "DK" SAY:

If you do not know it exactly, can you give me an estimate?


[DK]

A5

ALL

And how many employees work for your organisation AT THIS ESTABLISHMENT, including yourself?

INT.: IF "DK" SAY: If you do not know it exactly, can you give me an estimate?

PROGR.: CHECK:

IF A2=(2), Answer in A5 MUST be < Answer in A4!

IF NOT RE-ASK A4 / A5


[DK] Ù TERMINATE INTERVIEW

PROGR.: CHECK QUOTA (according to answer in A5)

1 up to 9 employees Ù QUOTA

2 10 - 49 employees Ù QUOTA



5 500+employees Ù QUOTA

IF "DK" TO QUESTIONS A5

A3

IF A2=2

Is your establishment ...?

INT.: READ OUT ALL ANSWER CATEGORIES. SINGLE ANSWER.

(1) the headquarters of an internationally operating organisation

(2) the headquarters of an organisation that only operates in this country

(3) a division or branch operation of an internationally operating organisation

(4) a division or branch operation of an organisation that only operates in this country

(5) other [INT.: DO NOT READ]

(6) DK


No

Branching


97

A8

ALL

EXCEPT IF A11 (NACE Code) = 75, 80, 85

Please indicate your establishment's turnover in the last financial year.

INT.: IF "DK", SAY:

If you do not know it exactly, can you give me a rough estimate?

INT.: PLEASE TRY TO GET AT LEAST AN ESTIMATE.

INDICATE IF ANSWER IS GIVEN IN EURO OR IN PREVIOUS NATIONAL CURRENCY (/UK: RESP. OR IN GBP)

(1) Turnover given IN EURO

(2) Turnover given IN PREVIOUS NATIONAL CURRENCY (UK: Always use GBP)

(3) DK, no answer to turnover

Turnover given:

I__I__I__I.I__I__I__I.I__I__I__I.I__I__I__I

12-digit numerical

A9

ALL

EXCEPT IF A11 (NACE-Code) =75, 80, 85

Has the turnover of your establishment increased, decreased or roughly stayed the same when comparing the last financial year with the year before?

(1) increased

(2) decreased

(3) roughly stayed the same

(4) DK

Module B: Basic ICTs take-up and intensity of use (e-Business) DMS Transition B

ALL

Now we would like to ask you some questions about the use of Information and Communications Technologies in your establishment.

B1

ALL

Does your establishment use e-mail? (1) yes

(2) no

(3) DK

B2

ALL

Does your establishment have access to the World Wide Web, i.e. the Internet?

(1) yes

(2) no

(3) DK

B3

ALL

Does your establishment have an Intranet, i.e. an internal computer network that uses the Internet protocol?

(1) yes

(2) no

(3) DK

B5

ALL

Does your establishment use EDI, i.e. electronic data interchange using the EDI standard?

(1) yes

(2) no

(3) do not know what this is [IF SPONTANEOUSLY SAID]

(4) DK

B6

IF B5=1

Is your EDI Internet based? (1) yes

(2) no

(3) do not know what this is [IF SPONTANEOUSLY SAID]

(4) DK

B7

ALL

Does your establishment use video-conferencing in your own facilities?

(1) yes

(2) no

(3) DK


No

Branching


98

B8

ALL

Does your establishment use a call center for communication with customers or other external contacts?

(1) yes

(2) no

(3) DK

B11

IF B1=1

Which applications can be accessed by the majority of your office workers?

Can the MAJORITY OF YOUR OFFICE WORKERS

... send e-mails to external addresses?

(1) yes

(2) no

(3) DK

B12

IF B2=1

(What applications can be accessed by the majority of your office workers?)


... browse Internet sites?

(1) yes

(2) no

(3) DK

B13

IF B3=1

(What applications can be accessed by the majority of your office workers?)


... browse INTRANET sites?

(1) yes

(2) no

(3) DK

Module C: E-commerce DMS Transition C

ALL

Now we would like to ask you some questions about E-commerce. Please refer to your establishment when answering.

C1

ALL

Does your establishment put information on the Internet, for example by means of a website?

(1) yes

(2) no

(3) DK

C2

IF C1=1 or 3

Do you sell goods or services via the Internet? (1) yes

(2) no

(3) DK

C3a

IF C1=1 or 3

Do you offer online reservation? By this we mean that your customers can make a reservation for a product or service through the Internet.

(1) yes

(2) no

(3) DK

C3b

IF C2=1

Do you distribute digital products or services online? By this we mean that the product is transferred to the customer online, or the service is provided online.

(1) yes

(2) no

(3) DK

C4a

IF C2=1

Are some of your online sales to businesses? (1) yes

(2) no

(3) DK

C5a

IF C4a=1

How large a share of your total sales to businesses are conducted online?

Would you say ...

INT.: READ OUT. SINGLE ANSWER

(1) less than 5%

(2) 5 up to 25%

(3) 26 up to 50%

(4) 51 up to 75%

(5) more than 75%

(6) DK

C4b

IF C2=1

Are some of your online sales to consumers? (1) yes

(2) no

(3) DK


No

Branching


99

C5b

IF C4b=1

How large a share of your total consumer sales are conducted online?

Would you say ...


(1) less than 5%

(2) 5 up to 25%

(3) 26 up to 50%

(4) 51 up to 75%

(5) more than 75%

(6) DK

C4c

IF C2=1

Are some of your online sales to the public sector? (1) yes

(2) no

(3) DK

C5c

IF C4c=1

How large a share of your total sales to the public sector are conducted online?

Would you say ...


(1) less than 5%

(2) 5 up to 25%

(3) 26 up to 50%

(4) 51 up to 75%

(5) more than 75%

(6) DK

C6

IF C2=1

Are your online sales MAINLY to a local, national or global market?

INT.: SINGLE ANSWER.

(1) local market

(2) national market

(3) global market

(4) DK

C7

IF C1=2

OR

IF C2=2 or 3

I am now going to read you a list of statements about selling online. For each statement, please tell me whether you agree completely, agree somewhat or do not agree from the point of view of your establishment.

How about the statement ... [item].

Do you ...

INT.: READ OUT ANSWER CATEGORIES. ONE ANSWER PER ITEM.

(a) Selling our products and services requires face-to- face interaction with customers

(b) The necessary technology is expensive

(c) The costs for the promotion of the online offer are high

(d) The revenue potential of online sales is low

(e) Customers might be concerned about data protection or security issues

(f) Adapting corporate culture to e-commerce is difficult

(g) The necessary skills are not readily available

(h) Handling the delivery process causes problems

FOR EACH:


(2) agree somewhat


(4) DK


No

Branching


100

C8

IF C2=1

You said earlier that you make sales online.

According to your experience, what effect has sel ling online on ... [item]?

Would you say the effect is ...


(a) your sales

(b) your costs

(c) your sales area

(d) the quality of your customer service

(e) the efficiency of your internal business processes

FOR EACH:

(1) very positive

(2) rather positive

(3) neither positive nor negative

(4) rather negative

(5) very negative

(6) DK

C9

IF B2=1 or 3

Do you use the Internet or other online services to purchase goods or services?

(1) yes

(2) no

(3) DK

C10

IF C9=1

Roughly what proportion of the maintenance, repair and organisation goods your establishment buys are purchased online, measured in amount spent?

Would you say ...


(1) less than 5%

(2) 5 up to 25%

(3) 26 up to 50%

(4) 51 up to 75%

(5) more than 75%

(6) DK

C11

IF B2=2

OR

IF C9=2 or 3

I am now going to read you a list of statements about purchasing online. For each statement, please tell me whether you agree completely, agree somewhat or do not agree from the point of view of your establishment.

How about the statement ... [item].

Do you ...


(a) Purchasing procurement products or services requires face-to- face interaction with suppliers

(b) Our suppliers do not sell online

(c) The necessary technology is expensive

(d) The cost advantage is negligible

(e) We are concerned about data protection or security issues

(f) The legal protection of online contracts is not sufficient

(g) The necessary skills are not readily available

(h) Suppliers´ technical systems are not compatible with ours

FOR EACH:


(2) agree somewhat


(4) DK


No

Branching


101

C12

IF C9=1

You said earlier that you purchase goods or services online. According to your experience, what effect has online procurement on ... [item]?

Would you say the effect is ...


(a) your procurement costs

(b) stock-keeping of maintenance, repair and organisation goods

(c) the number of suppliers

(d) your relations to suppliers

(e) the efficiency of your internal business processes

FOR EACH:

(1) very positive

(2) rather positive

(3) neither positive nor negative

(4) rather negative

(5) very negative

(6) DK

C13

IF C1=1

Does your establishment have an EXTRANET, i.e. a private, secure network running on the Internet protocol and accessible for selected external users?

(1) yes

(2) no

(3) DK

C14

IF C13=1

For which of the following purposes do you use your Extranet? Do you use it for ... [item]

INT.: ONE ANSWER PER ITEM.

(a) communication with customers or clients?

(b) communication with suppliers?

FOR EACH:

(1) yes

(2) no

(3) DK

C15

IF B2=1

Do you have access to the Extranet of one of your supplier, partner or customer organisations?

PROGR.: IF C1=2 or 3, add:

By Extranet I mean a private, secure network running on the Internet protocol and accessible for selected external users.

(1) yes

(2) no

(3) DK

C19

IF B2=1

Does your establishment trade goods or services through an e-marketplace? By e-marketplace I mean a business-to-business Internet trading forum in which multiple buyers and sellers exchange goods and services within an industry group or geographic region.

(1) yes

(2) no

(3) DK

C20

IF C19=1

On e-marketplaces, different types of business transactions can be accomplished. In which of the following types is your establishment actively involved?

INT.: READ OUT AND CODE ALL THAT APPLY

(1) catalogue-based offering of products or services

(2) catalogue-based purchasing of products or services

(3) auctions -- as a seller

(4) auctions -- as a bidder

(5) launching calls for tenders

(6) answering calls for tenders

(7) powerbuying, i.e. joint purchases together with other organisations to save costs

(8) none of these

(9) DK

Module D: e-Business security DMS Transition D

IF C1=1

Let us now turn to the topic of information security. Again, please refer to your establishment when answering.


No

Branching


102

D1

IF C1=1

Many establishments are affected by security breaches such as identity theft, online fraud, manipulation of software applications, computer viruses or unauthorised entry to internal networks.

Have any breaches of your information security occurred in your establishment in the last 12 months?

(1) yes

(2) no

(3) DK

D2a

IF D1=1

Progr.: Note for D2a to D2b:

For each item in D2a=1, ask directly D2b; then go to next item in D2a!!

Which of the following types of information security breaches have occurred in your establishment in the last 12 months? Did you experience cases of ... [item]?

INT.: READ OUT. ONE ANSWER PER ITEM.

(a) Identity theft

(b) Online fraud

(c) Manipulation of software applications

(d) Computer virus infections

(e) Unauthorised entry to internal networks

FOR EACH:

(1) yes

(2) no

(3) DK

D2b

(For Each Item) IF D2a=1

And how substantial were the consequences of this security breach for your establishment? Would you say they were ...

INT.: READ OUT ANSWER CATEGORIES. SINGLE ANSWER (PER ITEM ASKED)

FOR EACH ITEM IF D2a=1

(1) very substantial

(2) rather substantial

(3) not substantial

(4) DK

D3

IF D1=1

Where do you believe these breaches mainly came from? Do you think the largest threat to online security came from ...

INT.: READ OUT ANSWER CATEGORIES. CODE ALL THAT APPLY

MULTIPLE ANSWERS

(1) Customers

(2) Suppliers/competitors

(3) Former employees

(4) Computer hackers

(5) Internal users

(6) Others, not mentioned yet

(7) DK

D4

IF D1=1

How have you learned about these breaches, in most cases? Were you ... [item]

INT.: READ OUT, CODE ALL THAT APPLY

MULTIPLE ANSWERS

(1) alerted by a customer/supplier

(2) alerted by employees or did you notice yourself

(3) notified by your own information security system

(4) made aware by damage or loss of data

(5) alerted by the providers of outsourced security services

(6) in another way (DO NOT READ)

(7) DK


No

Branching


103

D5

IF C1=1

Does your establishment or your organisation have an information security policy?

(1) yes

(2) no

(3) DK

D6

IF D5=1

How would you describe it? As formal or informal? (1) formal

(2) informal

(3) DK

D7

IF D5=1

Which are your information security priorities?

How much priority is given to ... [item]


(a) Blocking of unauthorised access

(b) Expanding budget for security measures

(c) Defining the security architecture

(d) Outsourcing security management

FOR EACH

(1) high priority

(2) medium priority

(3) low priority

(4) DK

D8

IF C1=1

How important are the following factors as barriers to effective information security inside your establishment?

How about ...[item]:

Is this factor as a barrier to effective information security inside your establishment...


(a) High costs for security measures

(b) Lack of staff training

(c) Lack of staff time

(d) Complexity of the technology

(e) Lack of employee co-operation

FOR EACH:

(1) very important

(2) fairly important

(3) not important

(4) DK

D9

IF C1=1

Which of the following tools do you use for information security in your establishment? Do you make use of ... [item]


(a) Control of access to the computer system

(b) Cryptography/ data encryption

(c) Vulnerability Assessment Tools

(d) Firewalls

(e) Security Training and Awareness Rising Activities

(f) Intrusion Detection Systems

(g) End-user Security Training Classes

FOR EACH:

(1) yes

(2) no

(3) DK

Module F: e-Government DMS Transition F

IF B2=1

Now let´s turn to the topic of using online services for interacting with public administration.


No

Branching


104

F1

IF B2=1 AND A11 (NACE-Code) NOT =75 (Public Admin)

Progr.: Note for F1 to F2:

For each item in F1=2, ask directly F2; then go to next item in F1!!

I am going to read you a list of activities for which establishments have to get in touch with public administration.

For which of these activities do you already use online media such as EDI or the Internet?

What about ...[item]? Do you use online media such as EDI or the Internet for this?


(a) Payment of social contribution for employees

(b) Corporation tax declaration

(c) VAT declaration

(d) Submission of data to statistical offices

(e) Obtaining environment-related permits

(f) Participation in public invitation to tender

FOR EACH

(1) yes

(2) no

(3) DK

F2

(For Each Item)

IF F1=2

Would your establishment prefer to use online media such as EDI or the Internet for this purpose?

FOR EACH ITEM IF F1=2

(1) yes

(2) no

(3) DK

Transition F3

IF B2=2 or 3

Now let´s turn to the topic of using online services for interacting with public administration.

It is now possible to conduct at lest some of the interaction with public administration online, i.e. by using EDI or the Internet.

F3

ALL

Now I will read you a list of statements about using online media for interacting with public administration. Please tell me for each statement whether you agree completely, agree somewhat or do not agree.

Public services on the Internet ... [item].

Do you ...


(a) are not useful enough

(b) are faster than the traditional way

(c) require that you install special equipment or software

(d) reduce the number of mistakes public authorities make

(e) do not seem as safe as using the traditional way

(f) make it possible to deal with the authorities at more convenient times

(g) make it possible to deal with the authorities at more convenient locations, e.g. from the workplace

(h) are difficult to use

FOR EACH


(2) agree somewhat


(4) DK


No

Branching


105

Module G: Website accessibility DMS Transition G

IF C1=1

Now a few questions about the accessibility of your website for people with special needs.

G1a

IF C1=1

What priority has making your website user friendly for ... [item] in your establishment?


(a) People with visual disabilities or sight difficulties

(b) People with reduced or limited dexterities

(c) People with limited literacy

FOR EACH:

(1) high priority

(2) medium priority

(3) low priority

(4) DK

G1b

IF G1a (a) =2,3

or IF G1a (b) =2,3

or IF G1a (c) =2,3

Bearing the these groups in mind: Would you say that your website could be adapted rather easily, would prove difficult to adapt, or could not at all be adapted to these people's needs?


(1) could be adapted rather easily

(2) would prove difficult to adapt

(3) could not at all be adapted

(4) DK

G2

IF G1a (a) =1,2

or IF G1a (b) =1,2

or IF G1a (c) =1,2

Does your establishment or your organisation have formal Guidelines for making your website accessible to people with such special needs? By guidelines I mean rules which have to be followed by your website developers?

(1) yes

(2) no

(3) DK

G3

IF G1a (a) =1,2

or IF G1a (b) =1,2

or IF G1a (c) =1,2)

Was your website ever evaluated concerning its accessibility for people with such special needs?

(1) yes

(2) no

(3) DK

G4

IF G3=1

Was this evaluation done internally or using external evaluators?


(1) internal evaluation

(2) using external evaluators

(3) both

(4) DK

Module E: R&D DMS E1a

ALL

You said before that xyz [PROGR.: Insert answer to question A5] employees work for your organisation at this establishment.

From this, how many work in research & development, i.e. R&D? Please add up possible part time R&D personnel to full-time personnel.

INT.: IF "DK", PROMPT:


INT.: IF NECESSARY, EXPLAIN:

Among R&D we include all creative work undertaken on a systematic basis in order to increase the stock of knowledge and the use of this stock of knowledge to devise new applications.

[OPEN]


INT.: IF NONE, CODE "0".

[DK]

Progr.: Answer to E1a (Number employed in R&D) must be ≤ Answer to A5 (Total number employed in establishment)

If not, re-ask E1a

E1b

IF E1a > 0

and E1a is NOT DK

R&D can be centralised in R&D units, or it can be distributed over various units of an establishment.

Do you have at least one central R&D unit at your establishment?

(1) yes

(2) no

(3) DK


No

Branching


106

E2

IF E1b=1

What is the size of the computer staff in your central R&D unit(s)? Please add up part time computer staff to full-time staff.

INT.: IF NECESSARY, EXPLAIN:

By computer staff we mean all staff that

- manages the computers, networks and digital resources, or

- manages the Internet access and presentation, or

- carries out information searches and computations as their major work tasks, or

- provides user training.



[OPEN]



[DK]

Progr.: Answer to E2 (Computer staff in R&D) must be ≤ Answer to E1a (Number employed in R&D)

IF NOT, re-ask E2

E3

IF E1a > 0

and E1a is NOT DK

Do you get IT services for R&D from internal computer staff that are not members of your central R&D unit(s)?

(1) yes

(2) no

(3) DK

E4

IF E3=1

What is the size of the internal computer staff outside of your R&D unit(s) who provide IT services for R&D projects? Please add up part time computer staff to full-time staff again.



[OPEN]



[DK]

Progr.: Answer to E4 (Computer staff outside R&D) must be ≤ Answer to A5 (Total number employed in establishment)

IF NOT, re-ask E4

E5

IF E1a > 0

and E1a is NOT DK

Do you buy IT services for R&D from external service providers? (1) yes

(2) no

(3) DK

E6

IF E5=1

What is the number of additional computer staff in your establishment that would be necessary to substitute for the IT services for R&D projects which are currently obtained from external service providers?



[OPEN]



[DK]

E7

IF E1a > 0

and E1a is NOT DK

Do your R&D activities suffer from a low supply of qualified computer staff in your establishment?

(1) yes

(2) no

(3) DK

E8

IF E7=1

Please specify the number of open jobs for computer staff needed to provide IT services for R&D projects in your establishment?



[OPEN]


[INT.: IF NONE, CODE "0".

[DK]


No

Branching


107

X1

ALL

Finally I would like to ask you for a brief assessment:

In the course of the interview we talked, among others, about the areas e-Commerce, i.e. selling and buying online, and e-Government, i.e. interacting online with public administration. That is about areas, which might not necessarily fall into your direct responsibility.

Thinking back to the questions about ... [item]: What would you say: How familiar were you with the topics covered in those questions? Would you say...


(a) e-Commerce, i.e. selling and bying online

(b) e-Government, i.e. interacting online with public administration

FOR EACH (1) very familiar (2) fairly familiar (3) not very familiar (4) not at all familiar (5) DK/ no answer

X2

ALL

And all in all: How interesting did you find the questionnaire as a whole?

Would you say ...

INT.: READ OUT ANSWER CATEGORIES. SINGLE ANSWER.

End Text

ALL

These were all my questions. I would like to thank you very much for participating in the interview.

Have a nice day/evening!

Data to be provided by survey organisation Categories P0 Survey Number 1 0 1 4 3 9

P1 Country Code |__|__|

P2 Interview Number |__|__|__|__|

P3 Date of Interview: Day |__|__|,

Month |__|__|

P4 Time of the beginning of the interview (USE 24 HOUR CLOCK): Hour |__|__|,

Minute |__|__|

P5 Number of minutes the interview lasted |__|__|__|

P9 Interviewer Number |__|__|__|__|__|__|__|__|

Date post:	02-Oct-2021
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Benchmarking Telecommunication and Access in the ...

Documents