Consumer Search: An Extended Framework PETER H. BLOCH DANIEL L. SHERRELL NANCY M. RIDGWAY* While consumer search behavior has been studied for many years, its treatment has been limited to purchase con texts. This arti cle defines ongoing search as sea rch occun ing outside of the purchase process, and places it within an overall frame work for consumer search. In addition, it presents results of an exploratory study of ongoing search indicating that recreational or hedonic motives for ongoing search are more significant than practical, informational motives. This study also shows that product involvement is strongly linked to ongoing search and that ongoing searchers appear to be important elements in the marketplace. A lthough consum er search has been investi gated for years, nearly all of the many studies have focused on prepurchase events—that is, information gathering relevant to a specific consumption problem (see Clax- ton, Fry, and Portis 1974; Furse, Punj, and Stewart 1984; Newman 1977; Punj and Staelin 1983). This conceptualization embraces the traditional decision- making perspective, where a buying problem is recog- nized and search activity follows to help solve that problem . It is proposed here that an orienta tion focusing solely on prepurchase search is deficient and unable to account for search activity that is recreational or that occurs without a recognized consumption need. Such activities as browsing in an antique shop or subscribing to an automotive magazine by persons not in the market for these products are not addressed by traditional search theories. In such situations as these, product in- formation is obtained, yet the plan to purchase within the product class may be indistinct, temporally re- moved, or in some cases, nonexistent. In addition to conceptual shortcomings, limiting the study of search to prepurchase settings can understate the amount of information consumers have at their dis- posal when making a purchase. Many studies have found that despite the obvious benefits derived from prepurchase search, consumers have surprisingly little enthusiasm for the pursuit, even when buying expensive • Peter H . Bloch is Assistant Professor. Dan iel L. Sherrell is As- sociate Professor, and Nan cy M . Ridg way is Assistant Professor, all in the Department of Marketing. Louisiana State University. Baton Rouge. LA 70803. or socially risky goods (see Newman 1977). Claxton, Fry, and Po rtis pr ovide one explanation for the ob ser\ed lack of prepurchase search reported by consumers (1974, p. 35): Infonnation gatheritig i s a continuous process, even when the purchase is not foreseen. As a result, when the de- cision is made to make a purchase, relatively little explicit search is required. Therefore, studies relying on prepurchase contexts may only assess a subset of consumers' total search activity. Although consumer search has long been concep- tualized as a prepurchase activity, a fe w researchers have recognized that consumers search at other times and for other reasons. Tauber (1972) delineated a list of shopping motives, most of which were separate from desires to make a good purchase. Bellenger and Kor- goankar (1980) found that many consumers enjoy the act of shopping itself, without respect to buying, while Bloch and Richins (1983) found that retail store brows- ing was positively related to product interest, product knowledge, and word-of-mouth activity concerning the product. While the studies just described have provided a broader perspective on search behavior, they are still limited to retail settings. Search outside a purchase context can involve a wide range of activities; for ex- ample, reading a fashion magazine to see the newest styles or discussing fashion trends with friends can serve the same purpose as browsing through a clothing bou- tique. The research described in this article addresses limitations of the existing literature noted above by ex- amining search activity that is not linked to purchase 119 JOURNAL OF CONSUMER RESEARCH • Vo l. 13 • Jun e 1986
While consumer search behavior has been studied for many years, its treatment
has been limited to purchase con texts. This article defines ongoing search as sea rch
occun ing outside of the purchase process, and places it within an overall frame work
for consumer search. In addition, it presents results of an exploratory study of
ongoing search indicating that recreational or hedonic motives for ongoing search
are more significant than practical, informational motives. This study also shows
that product involvement is strongly linked to ongoing search and that ongoing
searchers appear to be important elements in the marketplace.
A lthough consum er search has been investigated foryears, nearly all of the many studies have focused
on prepurchase events—that is, information gatheringrelevant to a specific consumption problem (see Clax-ton, Fry, and Portis 1974; Furse, Punj, and Stewart1984; Newman 1977; Punj and Staelin 1983). Thisconceptualization embraces the traditional decision-making perspective, where a buying problem is recog-nized and search activity follows to help solve thatproblem . It is proposed here that an orienta tion focusingsolely on prepurchase search is deficient and unable toaccount for search activity that is recreational or thatoccurs without a recognized consumption need. Suchactivities as browsing in an antique shop or subscribingto an automotive magazine by persons not in the marketfor these products are not addressed by traditionalsearch theories. In such situations as these, product in-formation is obtained, yet the plan to purchase withinthe product class may be indistinct, temporally re-moved, or in some cases, nonexistent.
In addition to conceptual shortcomings, limiting thestudy of search to prepurchase settings can understatethe am ount of information consum ers have at their dis-
posal when making a purchase. Many studies havefound that despite the obvious benefits derived fromprepurchase search, consum ers have surprisingly littleenthusiasm for the pursuit, even when buying expensive
• Peter H . Bloch is Assistant Professor. Dan iel L. Sherre ll is As-
sociate Professor, and Nan cy M . Ridg way is Assistant Professor, all
in the Department of Market ing. Louisiana State University. Baton
Rouge. LA 70803.
or socially risky goods (see Newm an 1977). Clax ton,Fry, and Po rtis provide one explanation for the ob ser\edlack of prepurchase search reported by consumers(1974, p. 35):
Infonnation gatheritig is a continuous process, even whenthe purchase is not foreseen. As a result, when the de-cision is made to make a purchase, relatively little explicitsearch is required.
Therefore, studies relying on prepurchase contexts mayonly assess a subset of consumers' total search activity.
Although consumer search has long been concep-tualized as a prepurchase activity, a few researchers haverecognized that consumers search at other times andfor other reasons. Tauber (1972) delineated a list ofshopping motives, most of which were separate fromdesires to make a good purchase. Bellenger and Kor-goankar (1980) found that many consumers enjoy theact of shopping itself, without respect to buying, whileBloch and Richins (1983) found that retail store brows-ing was positively related to product interest, productknowledge, and word-of-mouth activity concerning the
product. While the studies just described have provideda broader perspective on search behavior, they are stilllimited to retail settings. Search outside a purchasecontext can involve a wide range of activities; for ex-ample, reading a fashion magazine to see the neweststyles or discussing fashion trends with friends can servethe same purpose as browsing through a clothing bou-tique. The research described in this article addresseslimitations of the existing literature noted above by ex-amining search activity that is not linked to purchase
119
JOURNAL OF CONSUMER RESEARCH • Vo l. 13 • Jun e 1986
•Increased sat isfact ionwith the purchase outcome
ONGOING SEARCH
• Involvement with the
product• Market environment• Situational factors
Buiid a bankof informa.tion forfuture use
Experiencefun andpleasure
• Increased product andmarket knowledgeleading to :
- future buyingeff iciences
-personal inf luence
•Increased impulsebuying
• Increased satisfaction
from search, and otheroutcomes.
intent and by incorporating search elements outside theretail setting.
CONCEPTUAL FRAMEWORK
The framework presented here delineates the deter-minants, motives, and outcomes of prepurchase andongoing search, which are summarized in the Figure.Before discussing these framework components, how-ever, it is important to distinguish between prepurchase
search and ongoing search. Prepurchase search has beendefined as (Kelly 1968, p. 273):
Information seeking and processing activities which oneengages in to facilitate decision making regarding somegoal object in the marketplace.
Search occ unin g ou tside of purchase contexts, as notedearlier, is a considerably more obscure concept and isthe primary focus of the present investigation. Ongoingsearch is conceptualized here as search activities thatare independent of specific purchase needs or decisions.Th at is, ongoing search does not occu r in order to solvea recognized and immediate purchase problem.
While itis
conceptually useful to differentiate ongoingsearch from prepurchase search, the two concepts aredifficult to separate in practice. The problem lies withprecisely specifying when a purchase problem has beenrecognized and the decision process started. The will-ingness and ability of the consumer to m ake a purchaseat the time the search occurs may be the best distin-guishing characteristic of these two search domains.When a person is searching with an interest in a productbut without a demand (i.e., financial resources or in-tent ion to buy are lacking), s/he is engaging in ongoingrather than prepurchase search. The border between
the two processes is further obscured by the possibilityof impulse purchasing.
The term "ongoing" indicates that the behavior oc-curs on a relatively regular basis, independent of spo-radic purchase needs. Ongoing search for autom obiles,for example, may include subscriptions to automotive
magazines. In contrast, prepurchase search may involvereading the same magazines, but only once every threeor four years, when a new car purchase is at hand . Wh ileprepurchase search and ongoing search can be differ-entiated by their purposes, the activities involved wouldappear identical to an outside observer (Furse, Punj,and Stewart 1984).
Search Determinants
In a prepurchase context, the extent of search is de-termined, in part, by the buyer's short-term involve-ment with the consumption problem resulting from risk
perceptions (Clarke and Belk 1979; Moore and Leh-man n 1980; Newm an 1977). Other factors influencinglevels of prepurchase search include the market envi-ronment, situational factors, and product familiarity(Moore and Lehm ann 1980). Levels of ongoing searchare also a function of involvement (Bloch 1981; Blochand Richins 1983; Tigert, Ring, and King 1976). In thiscase, however, the involvement is enduring in character,reflecting a continuing interest or enthusiasm ratherthan the temporary product interest resulting frompurchase requirements. As with prepurchase search,levels of ongoing search are also influenced by marketfactors such as the availability of product information
and time or other situational constraints.
Search Motives
The consumer's primary motive for prepurchasesearch is to enhance the quality of the purchase o utcom e(Punj and Staelin 1983). Ongoing search, on the otherhand, may involve two basic motives. The first is toacquire a bank of product information potentially usefulin the future (either for personal use or for dissemina-tion to others; see Hirschman and Wallendorf 1982).The second motive is pleasure or recreation. In this case,consumers engage in ongoing search for its intrinsic
satisfactions. In practice, it may prove difficult to sep-arate these two motives for ongoing search. There isempirical and conceptual support, however, for makingsuch a dichotomy. Hirschman (1980), for example, in-troduced the notion that some consumers seek cognitiveor informational stim ulation , while others seek sensorystimulation in the consumption experience. Venkatra-man and Maclnnis (1985) extended the cognitive/he-donic dichotomy to search activities, albeit only in aprepurchase context. It is proposed here that the infor-mational/hedonic dichotomy can be extended to on-going search.
Search to Build Information Banks. The first on-going search motive, information banking, involves in-creasing product expertise, but for reasons other thanto optimize the outcome of a planned purchase. Ex-pertise can also make a consumer feel well informed,enhance product care, add to feelings of self-actualiza-
tion, and improve the quality of future product selec-tions (Fleischmann 1981). Ongoing search to buildproduct knowledge and prepurchase search are similarin their functional orientations. The central differenceis the consumer's ability and/or readiness to buy at thetime of the search activity. For ongoing search, a pur-chase problem is not recognized and the buying decisionmay be temporally quite removed. Because it is difficultto specify precisely when the purchase process begins,it may be profitable to consider the information-bankingaspect of ongoing search as an extension of prepurchasesearch contexts.
Searchas
Recreation. The second proposed moti-vation for ongoing search activity is to have fun or toexperience po sitive affect. Punj and Staelin (1983) havenoted that some consumers get pleasure from seekinginformation about products, and many authors havestudied inherent pleasure as the motivation for a varietyof behaviors (see Csiksze ntmihaiyi 1983; Miller 1973).In these situations, activities are not guided bygoals oroutcomes, but by the process itself. This process ori-entation to behavior has been given a variety of names,including Funktionslust (Buhler 1930), ludic behavior(Berlyne 1960), and intrinsic m otivation (Deci 1975). Itis proposed here that for some individuals, ongoingsearch behavior is such an activity—that is, it representsa leisure pursuit performed as an end in itself.
Outcomes of Search
Search activities lead to a variety of outcomes. Forprepurchase search, outcomes noted in the literatureinclude better choice decisions, increased product andmarket expertise, and heightened satisfaction with apurchasing job well done (Punj and Staelin 1983).Whether motivated by desires for product knowledgeor hedon ic response, ongoing search also culminates inseveral outcom es described below.
Future Purchasing Efficiencies. The ongoing searcheracquires product and marketplace information on aregular basis, even if the primary motivator of the ac-tivity is pleasure seeking. Therefore, product expertisedeveloped through ongoing search makes a consumermore efficient—able to rely more heavily on less costlyinternal search when a relevant consumption problemarises in the future (Bettman 1979). When consideredin an ongoing search framework, even consumers re-porting little prepurchase search may be well preparedto make a satisfying purchase decision.
Personal Influence. Product information obtainedthrough ongoing search is likely to be disseminated tofriends and family, as product expertise is related toopinion leadership (Leonard-Barton 1985; Price andFeick 1984). A recent study by an automotive marketresearch firm found that two-thirds of new car buyers
relied most heavily on information provided by theirsocial co ntacts indeciding on a make of car (J. D. Powerand Associates 1984). In addition, a survey by Car andDriver (1978) found that each of its readers makes anaverage of 19 new car recommendations a year. Thesefindings indicate that ongoing searchers may have con-siderable influence in the marketplace.
Impulse Buying. An ongoing searcher browsing ina retail establishment without an intent to buy may,nevertheless, leave the store with a purchase. It is rea-sonable to assume that the more frequently one is in astore, for whatever the reason, the greater the likelihoodthat one will buy (Bellenger, R obertson, and Hirschman
1978). Given the growing significance of catalog mar-keting and toll-free order lines, such impulse buyingalso may be found among ongoing searchers in the printmedium.
The outcomes of ongoing search described aboveconstitute a representative, but not exhaustive list. Anumber of other outcomes are possible, includ ing im-provements in problem solving skills, increased leisuresatisfaction, and regulation of variety drives.
EMPIRICAL STUDY
The quality of available evidence for the above
framework is uneven. While hypothesized relationshipsfor prepurchase search have been extensively investi-gated, those for ongoing search are based on more ten-tative evidence. To bolster knowledge ofongoing searchand to provide empirical support for portions of theframework above, an empirical study was undertaken.In conducting this study, there were three specific ob-jectives:
1. To determine whether informational and recreationalmotives for ongoing search have significant andequalimpact on level of ongoing search, or whether onemotive is a stronger determinant.
2. To examine the relationship between ongoing searchand product involvement.
3. To assess the impact ongoing searchers have in themarketplace.
Product Classes Studied
In choosing products for this study, the primary cri-terion was the potential for ongoing search amo ng con-sumers. In particular, products were sought for whichavailable information is high and for which a relativelywide range of ongoing search levels across subjects can
O P E R A T I O N A L I Z A T I O N O F V A R I A B L E S
Ongoing search measures*Perceived informat iveness and enjoyment
of ongoing search activit ies"
Clothing
• How often do you visit clothing stores or departments, just to look around or get information,
rather than to make a specific purchase?
• In a typical month, how many clothing catalogs do you look through?
• How many dothing or fashion-related magazines do you subscribe to/read regularly, but do not
subscribe to?
• How often do you talk to friends and acquaintances to get information or advice conceming
dothing and clothing styles?
Persona l computers
• How often do you visit computer stores or departments, just to look around or get information,
rather than to make a specific purchase?
• How of ten do you browse through computer book sect ions at bookstores?
• How many personal computer magazines do you subscribe to/read regulariy, but do not
subscribe to?
• How often do you talk to friends and acquantainces to get information or advice concemingpersonal computers?
Clothing• Visits to clothing stores
• Discussions with retail salespersons
• Looking at dothing catalogs
• Reading ads in fashion m agazines
• Reading articles in fashion magazines
• Discu ssions vkrith friends
P e r s o n a l c o m p u t e r s
• Visits to computer stores
• Discussions with retail salespersons
• Browsing through computer book
sections
• R eading ads in computer magazines
• R eading art ides in computer magazines
• Discussions with friends• Discussions with other computer users
• Because ranges ol individual items vaned. me items were standardaed prior to summation or rekabisty testing.
• Intormatiweness was measured on 5-point scales that ranged from -N ot at a ll informative to "Extreme ly informative.' Enjoyment was m easured on companion scales that ranged from"Not at all enioyable' to Extremely enioyabie."
be observed. Based on these considerations and the de-sire to add generalizability to study findings, two prod-uct classes were chosen: clothing and personal com-pu ters. While these produ ct classes differ in many ways,they both offer am ple op portunity to engage in ongoingsearch. There are numerous retail stores and special
interest magazines focusing on these classes, and prod-uct differentiation and rates of prod uct change are highenough to stimulate search behavior.
Method and Sample
Questionnaires were mailed to a probability sampleof 1,500 respondents in a Sunbelt SMSA. Since it isunlikely that any general SMSA frame would containenough subjects who engage in high levels of ongoingsearch for a given product to provide adequate constructrange and response variance, the probability sample wassupplemented with two additional samples of subjectspresumed to be mo re involved with the product classesunder study. Thus, surveys were also sent to 383 indi-viduals on custom er m ailing lists provided by wom en'sand men's clothing specialty stores and to 460 personson local computer-store mailing lists. Since relation-ships between constructs rather than generalization tospecific po pulations were of interest, this procedure wasdeemed appropriate (Calder, Phillips, and Tybout1982). A second questionnaire wave was sent approx-imately 10 days after the first to stimulate response.
Of the approximately 2,300 surveys mailed. 712 werereturned, providing a 29 percent overall response rate.
The response rates differed, however, across the threesubsamples employed in the study. For the personalcomputer, clothing, and general random samples, theresponse rates were 54 percent, 40 percent, an d 19 per-cent, respectively. The relatively low response of theprobability sample can be attributed to lower interest
levels and to the relatively high number of rural andlow-income addresses contained in that sample. Editingproduced a total usable sample of 679.
Measures
Ongoing Search. Product-specific multi-item indiceswere used to assess levels of ongoing search in severaldomains: retail settings, the mass media, and interper-sonal sources (see the Exhibit). Each item was wordedto emphasize activity tha t occurs on a regular basis andthat can thus be presumed to be independent of pur-chasing. Individual items were standardized, then
summed. Item analyses of the two ongoing search in-dices showed satisfactory internal consistency with al-pha coefficients of 0.70 for clothing and 0.80 for per-sonal computers.
Scores among the three subsamples were comparedto partially validate the search indices. For clothing,the mean search index score for the clothing boutiquecustomers was 1.63, while the general sample and thepersonal compu ter sample produced scores of- 0.2 6 and-1.34, respectively. Results were also as expected forthe personal computer index. The computer-store sam-ple mean was 2.20, compared to -1.19 for the general
sample and -1.46 for the clothing-store sample. Thedifferences in group me ans were significant (p < 0.001)for both product classes (F = 50.9, df = 2,582; F= 138.6, df= 2,651).' These results not only provideevidence of measure validity, but also suggest that on-going search is a product-specific phenomenon for a
particular consumer segment. There was also a widerange of scores on the two indices, generally. Over theentire sample, clothing-search scores ranged from -3.58to 8.97, and the computer-search scores ranged from-2.7 3 to 10.22. These findings support the use of cloth-ing and personal com puters as products w ith wide vari-ance in ongoing search.
Search Benefits. To investigate the two proposedongoing search motivations and address the first re-search objective, two sets of measures were developed.Respond ents were asked to rate on five-point scales theinformativeness of various ongoing search activities andthe level of enjoyment derived from these activities (see
the Exhibit).Consumer Difference Variables. Enduring product
involvement was measured by a pair of multi-item in-strum ents that draw from the previous work of severalresearchers (Bloch 1982; Tigert, Ring, and King 1976).The product-specific involvement measures employedstandardized items and summated scoring.^ Impact onthe marketplace was assessed by three items: (1) levelof expenditures in the product category, (2) the impor-tance to the respondent of new product developmentsas an indicator of innovativeness, and (3) opinion lead-ership concerning the product.
ANALYSIS AND RESULTSMotives for Ongoing Search
It was posited earlier that consumers engage in on-going search in order to build a useful bank of productinformation and/or to experience pleasure in a product-oriented activity. This study related respondents' in-formativeness and enjoyment ratings to levels of on-going search using both bivariate and multivariateanalyses. Table 1 presents the results of these analyse s.'
' In validating the ongoing search indices, sex differences were alsoexamined. For both product classes, significant differences were ob-
served. In the case or clothing, females engaged in more ongoingsearch than did males {x = 1.56 for females, x = -1.24 for males: (= 190.50; df= 1,677; p < 0.001), while the opposite was true forcomputers (x = 0.92 for males, x = -1.09 for females; / = 75.21; rf/"= 1,652; p < 0.001).
' The specific items used for each involvement index were productinterest, time spent thinking about the product, and average impor-tance of the product to the performance of several social and careerroles. The first item employed a four-point scale, while the other twoitems used five-point scales. The two three-item indices showed sat-isfactor>- reliability, with alphas o f 0.83 and 0.77 for the clothing andcomputer measures, respectively.
' To ensure that reports of ongoing search were not contaminatedby prepurchase activities, analyses were also performed that excluded
Informativeness ratings were used first as predictors ofongoing search, with enjoyment ratings added in a sec-ond stage of analysis. The practical or informationalbenefits of ongoing search were considered fir st becau sethey represent an extension of traditional conceptual-izations of search in prepurchase settings. The less well-
known hedonic benefits of search were added to themodel to determine how much additional explanatorypotential they provide.
In the first multiple regression analysis, the perceivedinformativeness ratings of various ongoing search be-haviors were used as predictors of companion itemsfrom the search indices. For exam ple, the perceived in-formativeness of reading ads and articles in magazineswas used to predict frequency of magazine readership.Separate analyses were run for each product class andfor each of the ongoing search dom ains: retail browsing,media search, and interpersonal discussions. For bothproduct classes, results showed weak predictability forthe informativeness items alone. In the clothing classthe informativeness items accounted for very little ofthe variance in the ongoing search measures, regardlessof the information source. For computers the infor-mativeness items were also poor predictors of ongoingsearch activities, explaining about one percent of thevariance in the dependent variables.
Following these analyses, the enjoyment items wereadded to the informativeness items in the individualanalyses to determine whether the recreational aspectsof search added to the explained variance in ongoingsearch behaviors. Differences in obtained R^ values be-tween the informativeness items alone and the infor-mativeness + enjoyment items were examined andtested for significance. For both product classes, the en-joyment items added significantly (p < 0.001) to theexplained variance for all categories of ongoing searchbehavior. When enjoyment ratings are used in con-junction with informativeness ratings, approximately25-30 percent of the variance in search behavior is ex-plained. These results, combined with the simple cor-relations, indicate that the perceived enjoym ent of on-going search ac tivities is a relatively effective pred ictorof the frequency of such activities. Thus, the traditionalorientation that considers search to be determined solelyby the practical information it provides is deficient.
Ongoing Search and Product InvolvementAs noted above, product involvement appears to be
a basic determinant of ongoing search. To exam ine therelationship between product involvement and ongoing
those respondents expecting to make a large purchase in one month
for clothing and in three months for computers. Such screening did
not produce significant changes in any of the results. The remaining
sections of the article report results pertaining to the unscreened
IMPACT OF PERCEIVED INFORMATIVENESS AND ENJOYMENT OF ONGOING SEARCH BEHAVIORS ON LEVEL OF SEARC H*
Ongoing search behavior
Frequency of store browsing
Visits to retail storesDiscussions w ith salespersons
R' (informativeness items)R' (informativeness and enjoyment
items)F for change in R*
Amount of catalog/book browsing
Reading clothing catalogs/Browsingin PC txx>k sections
R' (informativeness items)R' (informativeness and enjoyment
items)
F for change in R'
Amount of magazine readership
Reading magazine ad vertisingReading magazine artides
R' (informativeness items)R' (informativeness and enjoyment
items)F for change in R'
Frequency of product discussions
Discussing product with friendsDiscussing PCs with other users
R' (informativeness items)R' (informativeness and enjoyment
items)
F for change in R'
* All F ratios for ctiange in R' are significant al p < 0.001.
• « " A.STA.
•«(f =2.676.
Informativeness
rating r
.24
.08
Clothing
Enjoyment
rating r
.55
.23
Multipleregression
results
.00
.3175.35"
Informativeness
rating r
.08
.05
Computers
Enjoyment
rating r
.50
.24
Multipleregression
resuits
.01
.2657.13"
.38 .48.01
.23 .48.00
3538
40
.49
.48
.51
.24
105.08 '
.01
.2659.20"
.03
.29135.60 '
.20
.19
.44
.30
.23
.29
.53
.37
.2398.94 '
.00
.0916.16"
.02
.3175.53"
search, the summed search and involvement indiceswere correlated. These correlations (r = 0.70 for cloth-ing; r = 0.67 for computers, p < 0.001 for both) supportprevious research (Bloch 1981; Bloch and Richins 1983)showing a strong connection between a consum er's en-during involvement in a product class and the propen-sity to engage in ongoing search.
Marketplace Impact of Ongoing Searchers
To address the third research objective, respondentswere divided into groups of heavy and light searchersbased on a median split of ongoing search-index scores,and group differences on the three marketplace-impactvariables were tested for significance (see Table 2). Inaddition, ongoing search-index scores were correlatedwith each marketplace item. Due to the strong rela-tionship between ongoing search and involvement re-
ported earlier, partial correlations controlling for in-volvement levels were also obtained. These additionalanalyses would indicate the extent to which ongoingsearch and the marketplace variables were related in-dependent of product involvement levels.
Ongoing searchers appear to be significant forces inthe marketplace. Among the clothing sample, heavy
ongoing searchers spend over twice as much in the sametime period as do light searchers, and am ong com puterowners, heavy ongoing searchers have over three timesas much invested in their computer systems as do lightsearchers. The partial correlations indicate that evenafter respondents' level of product involvement is ac-counted for, ongoing search is significantly related toproduct expenditures.
Levels of ongoing search are also related to the per-ceived importance of keeping up with new product de-velopments, indicating a potential for innovativeness.
Dollars spent: on clothes in last sixmonths/on PC system among PCowners
Importance of keeping up withnew product developments(5-point scale)
Amount of product informationgiven to o thers (5-point scale)
Lowongoing
searchers
$282
2.42
1.35
Clothing
Highongoing
searchers
$631
3.65
2.80
Simpler"
.35
.56
.70
Partialr'
.19
.05'
.46
Lo wongoing
searchers
$1,255
2.02
1.36
Computers
Highongoing
searchers
$4,613
3.87
2.89
Simpler
.46
.57
.68
Partialr
.29
.17
.51
* AI tests of mean differences, correlation coelTicients. and partial correlations are significant at p < 0.01 except where noted with *.
* C^orrelatons b etween impact items and ongoing search index.
' Partial correlatians twtween impact items and ongoing search index while contromg lor reported irwolvemant index scores. Ariatyses ol covariance using involvement soores as a oovariate
were also performed for both product categories. With the exception of keeping up witfi new product developments for dothlrtg. the means for the low and high ongoing searchers renained
significantly different at p < 0 01.
However, the partial correlations indicate that this re-lationship is largely due to product involvement, par-ticularly in the case of clothing. Heavy ongoing search-ers also reported significantly higher levels of word-of-mouth activity than did light searchers. This result isnot surprising, since the product expertise gainedthrough ongoing search makes the searcher sought afterby other consumers. The relationship between opinionleadership and ongoing search remained strong for bothproduct classes even when involvement levels were
taken into account.
DISCUSSION AND CONCLUSIONS
This research has elaborated the concept of consumersearch and provided a framework for more compre-hensive examinations of this topic in the future. In ad-dition, it has developed measures of ongoing search po-tentially useful in future research. It has also demon-strated some of the deficiencies of traditionalorientations toward search as part of a general broad-ening of consumer research to include all aspects of
consumption. This framework suggests that a numberof consumers gather product information on a regularbasis from a variety of sources with two general objec-tives; to augment stores of product knowledge and toexperience pleasure. Although it is likely that ongoingsearch meets both objectives simultaneously, data pre-sented here indicate that hedonic benefits are consid-erably more relevant.
Conclusions concerning the relative impact of infor-mational and hedonic motives must be tempered bylimitations in the present research, however. It is pos-
sible that informativeness as measured here representsa less personal concept than enjoyment and so is easierto rate in a hypothetical ma nner. But rating enjoymentshould depend more on first-hand experience, andthereby produce richer responses. Certainly, more so-phisticated measures of these motivatio ns would furtherunderstanding of ongoing search.
This study has also demonstrated a strong relation-ship between ongoing search and enduring involvement.The reader should note, however, that the direction of
this relationship remains unclear. Though it is tacitlyassumed here that involvement represents an internalmotivating state of the consumer that manifests itselfin a variety of outco mes, one of which is ongoing search,ongoing search activity m ay have a reciprocal influenceon involvement. When ongoing search is pleasurableand product expertise is increased, one might expectincreased attachment to the product class through atype of conditioning.
Ongoing searchers are also distinguished by a prom-inent position in the marketplace. Heavy searchers werefound to be heavy spenders within the product class.These spending levels are likely the result of high new
product awareness and frequent contact with retailers.In addition, the ongoing searcher is eager to learn ofnew product developm ents and is willing to talk aboutthem. Although product involvement plays a role inthe observed relationships, ongoing search was shownto have a distinct effect on the marketplace.
In considering future investigations of ongoingsearch, several promising avenues of inquiry exist. Oneconcerns the demarcation line between prepurchasesearch and ongoing search. To identify this boundary,research must focus on the exact beginning of the pur-
chase process. In other words, can the point at which aconsumption problem is recognized be identified?Greater knowledge of the starting phase of the purchase-decision process will greatly enhan ce understan ding ofconsumption activities thought to be outside the buyingdomain. A related question concerns how patterns of
information gathering may differ between ongoingsearchers and prepurchase searchers. It would be inter-esting to determine whether ongoing search in a givenprodu ct class tend s to entail consu lting sources differentfrom those consulted during prepurchase search. In ad-dition , certain sources of information may be preferredmore by ongoing searchers than by prepurchase search-ers, regardless of product category.
{Received June 1985. Revised December 1986.]
REFERENCESBellenger, Danny N. and Pradeep Korgoankar (1980), "Pro-
filing the Recreational Shopper," Journal of R etailing.
58 (Spring), 58-81., Dan H. Robertson, and Elizabeth C. Hirschman
Berlyne, Daniel E. (1960), Conflict. Arousa l and Curiosity.New York: M cGraw-Hill.
Bettman, James R. (1979), An Information Processing Theoryof Consumer Choice. Reading, MA: Addison-Wesley.
Bloch, Peter H. (1981 ), "An Exploration into the Scaling ofConsumers' Involvement with a Product Class," in Ad -vances in Consumer Research. Vol. 8, ed. Kent B. Mon-roe, Ann A rbor, MI: Association for Consum er R esearch,61-65.
(1982), "Involvem ent Beyond the Purchase Process:
Conceptual Issues and Empirical Investigation," in Ad -vances in Consumer Research. Vol. 9. ed. AndrewMitchell, Ann A rbor, MI: Association for C onsumer Re-search, 413-417 .
and Marsha L. Richins (1983). "Shopping WithoutPurchase: An Investigation of Consumer Browsing Be-havior," in Advances in Consumer Research. Vol. 10.eds. Richard P. Bagozzi and Alice M. Tyb out, Ann Arbor,MI: Association for Consumer Research, 389-393.
Buhler, Karl (1930), The Mental Development of the Child.New York: Harcourt B race.
Calder, Bobby J., Lynn Phillips, and Alice M. Tybout (1982),"The Concept of External Validity," Journal of Con-sumer R esearch. 9 (December), 240-24 4.
"Car and Driver Reader Survey" (1978), Car an d Driver. July.
Q ark e, Keith and Russell Belk (1979), "Th e Effects of ProductInvolvement and Task Definition on Anticipated Con-sumer Effort," in Advances in Consumer Research. Vol.6, ed. William L. Wilkie, Ann Arbor, MI: Associationfor Consumer Research, 313-318.
Claxton, John D., Joseph N. Fry, and Bernard Po rtis (1974)."A Taxonomy of Prepurchase Information GatheringPatterns," Journal of Consumer Research. 1 (December),35-42.
Deci, Edward L. (1975), Intrinsic Motivation. New York:Plenum.
Fleischmann, G erd (1981), "Sources for Product Ideas: AProactive View on the Co nsum er," in Advances in Con
sumer Research. Vol. 8, ed. Kent B. M onroe, Ann A rboMI: Association for Consumer Research, 386-390.
Furse, David H., Girish N. Punj, and David W. Stewart(1984), "A Typology of Individual Search Strategiesamong Purchasers of New Automobiles," Journal oConsumer Research. 10 (March), 417-431.
Hirschman, Elizabeth C. (1980), "Innovativeness, NoveltySeeking, and Consumer Creativity," Journal of Con-sumer R esearch. 1 (December), 283-295 .
and Melanie R. Wallendorf (1982), "Motive Under-lying Marketing Information Acquisition and Transfer,"Journal of Advertising. 11 (2), 25-31.
Kelly, Robert F. (1968), "Th e Search Com ponen t of the Con-sumer Decision Process: A Theoretic E xam ination," in
Marketing and the New Science of Planning, ed. RobKing, Chicago, IL: American Marketing Association271-274.
Leonard-Barton, Dorothy (1985), "Experts .as NegativeOpinion Leaders in the Diffusion of a Technological \n -novauon." Journal of Consumer Research. 11 (M arch914-926.
Miller, Stephen (1973), "Ends, Means, and GalumphingSome Leitmotifs of Play," American Anthropologist. 7587-98.
Moore, William L. and Donald R. Lehmann (1980), "Indi-vidual Differences in Search Behavior for a N ond urab le,"Journal of Consumer Research. 1 (December), 296-30
Newman, Joseph W. (1977). "Consumer External SearchAmount and D eterminants." in Consumer an d IndustriaBuying Behavior, eds. Arch Woodside. Jagdish Shethand Peter Bennett. New York: North-Holland. 79-94.
Price. Linda L. and Lawrence F. Feick (1984), "The Role ofInterpersonal Sources in External Search: An Informational Perspective," in Advances in Consumer ResearVol. 11. ed. Thom as Kinnear. P rovo, U T: A ssociationforConsumer Research, 250-255.
Punj, Girish N. and Richard Staelin (1983), "A Model ofConsumer Information Search for New Automobiles,"Journal of Consumer Research. 9 (March), 366-380.
Tauber, Edward M. (1972), "Why Do People Shop?" Journaof Marketing. 36 (October), 46-59 .
Tigert, Douglas J.. Lawrence J. Ring, and Charles W. King(1976). "Fashion Involvement and Buying Behavior,"
in Advances in Consumer Research. Vol. 3, ed. BeverB. Anderson. Ann Arbor, MI: Association forCon sumerResearch, 46-52.
Venkatraman, Meera P. and Deborah J. Maclnnis (1985)"The Epistemic and Sensory Exploratory Behaviors oHedonic and Cognitive Shop pers," in Advances in Consumer Research. Vol. 12, eds. Elizabeth C. Hirschmanand Morris B. Holbrook. Provo, UT: Association foConsumer Research, 102-107.
