Using Household Surveys to Study the Economic Using Household Surveys to Study the Economic and Social Implications of Migration: and Social Implications of Migration:

A Methodological Evaluation*A Methodological Evaluation*

Regional Training Workshop on International Migration StatisticsCairo-Egypt

30 June – July 3, 2009

Jeronimo Cortinawww.jeronimocortina.com

*The views expressed here are those of their respective authors and do not necessarily represent the official views of the University of Houston or of the United Nations Children's Fund (UNICEF).

Global project on migrationGlobal project on migration• What is the Issue?

– Little attention, has been devoted to weigh the impact of international migration on sending countries and specifically on the left-behind

• Why Did the Issue Arise?– Reliable national-level data about the incidence and magnitude of international

migration (of adults or children) and on those left behind are rare because estimating these numbers is extremely difficult due to a range of methodological problems

• What are the Implications of not Having Reliable and Valid Data?– Lack any comparability among other sources due to different uses of categories and

formats – Limit the generalizability of these data – Public policy formulation may rest on analyses that may lack any reliability or validity

and thus have potential implications regarding implementation and outcomes

Household surveys to “measure”/”study” migration?Household surveys to “measure”/”study” migration?

Why?1. Provides a more dynamic

picture of migration 2. Cheaper than a national

census3. Research specific (what

kind of indicators are expected?)

How?1. Stand alone survey2. Migration module can be

introduced in national labor surveys, income and expenditure, LSMS, DHS, etc. with a specific longitudinal framework

Albania, Ecuador and MoroccoAlbania, Ecuador and Morocco

• Objective: Impact of migration on children-left behind in two dimensions 1. Economic: remittances uses and purposes, HH SES

characteristics (pre/post migration)2. Subjective:

• Health (mobility, self-care, usual activities, pain/discomfort, anxiety/depression - 243 health state indicators)

• Life-Satisfaction (family, friends, school, self, living environment - life satisfaction scale)

• Communication patterns

DefinitionsDefinitions• Migrant Household: A household with children (0-17 yrs. of age) in which at

least one member (who ate out of the same pot and slept in the same dwelling) who is still considered to be a part of that household has moved to live in a foreign country.

• Non-migrant Household: A household with children (0-17 yrs. of age) in which no members (who eats out of the same pot and sleep in the same dwelling) have ever moved away from the household.

• Region: An artificial construct, created by selecting geographical or administrative areas for which it is known or expected that they contain relatively high proportions of migrant households. A region may contain areas that are not necessarily contiguous and may be made up of one or more such areas.

Sampling strategySampling strategy• Two-phase sampling strategy devised by EuroStat and NiDi

1. Chose study regions: high economic development, low economic development and high vs. low migration

2. Classify geographical areas within each region according to the estimated prevalence of households with international migration

3. Create strata to classify areas according to the (relative) prevalence of households with international migration experience.

4. Sample areas from each “prevalence rate” stratum, whereby areas with higher expected prevalence of migrant households are over-sampled

5. Screen households in areas selected and prepare a list identifying all households as migrant or non-migrant households.

6. Create strata of migrant and non-migrant households for each sample area.7. Allocate a disproportionate share of the sample for each area to the stratum of migrant

households (i.e., over-sample migrant households).8. Interview clusters of non-migrant households in each area as needed to economize on

fieldwork time and costs.

Sampling strategySampling strategy• Migrant and Non-Migrants households are included in the sample and

prescreened. The former are oversampled

• Advantages– Designed to take into account the inherit characteristics of migration– “Relatively” easy to adapt to different available information and particular

country needs– Sample design takes into account population of interest and “control” group

• Disadvantages– National representativeness might be an issue– More field work– Without knowledge/information may be difficult to implement

Listing formListing form

IndicatorsIndicators

SurveySurveyEcuador

1,401 respondents595 Children

806 Adults

Albania1,482 respondents

611 Children 871 Adults

ResultsResults• No significant difficulties in filling the health and life-satisfaction modules• Migration is associated with

– Sadness and solitude– An opportunity to have access to a better quality of life.

• Migration has both positive and negative impacts• Family as a whole suffer emotionally• Marriages tend to be affected by migration• Caretakers experience high levels of stress and responsibility• Remittances are spent on health, education, basic consumption, clothing

and daily needs• Migrants and their family members left-behind communicate frequently

Non-migrant households

Migrant households

.6.7

.8.9

Pr(

Sat

isfie

d)

8 10 12 14 16 18Age

Predicted probabilities Albania

Predicted probabilities Ecuador

Migrant households

Non-migrant households0.0

02.0

04.0

06.0

08.0

1P

r(U

nsat

isfie

d)

8 10 12 14 16 18Age

Migrant households

Non-migrant households

.5.6

.7.8

Pr(

Sat

isfie

d)

8 10 12 14 16 18Age

Migrant households

Non-migrant households0.0

02.0

04.0

06.0

08.0

1P

r(U

nsat

isfie

d)

8 10 12 14 16 18Age

Morocco & 2009 National Morocco & 2009 National Demographic SurveyDemographic Survey• Government, Civil Society, International

Organizations• Negotiations to implement an oversample of

migrants• Will work together to design sampling frame

and training

Conclusions & next stepsConclusions & next steps

• Migration’s material and immaterial impacts on those left-behind

• Capacity building• Collaborative work between Government,

International Organization and Civil Society