A Model of the Determinants of Retail Search

ABSTRACT - The purpose of this study was to develop a model of the information search process that better reflects consumers' search strategies. In this model, the amount of retail search undertaken, and the extent to which a consumer relies on interpersonal and neutral sources, are hypothesized to be dependent upon the importance of the product, time availability, perceived risk, subjective knowledge, and enjoyment. Building upon previous research, the authors hypothesize that consumers who rely upon interpersonal advice or neutral sources will, in turn, undertake less retail search. The structural model was tested using LISREL VII, with a sample of 172 consumers who had recently made a major household purchase. The authors suggest that the low explanatory power of the model may be due to the changing retail environment (e.g., new types of stores, changing socio-demographics, technological advancements, etc.), and provide directions for future research.

Citation:

Jeffrey G. Blodgett, Donna J. Hill, and George Stone (1995) ,"A Model of the Determinants of Retail Search", in NA - Advances in Consumer Research Volume 22, eds. Frank R. Kardes and Mita Sujan, Provo, UT : Association for Consumer Research, Pages: 518-523.

Advances in Consumer Research Volume 22, 1995      Pages 518-523

A MODEL OF THE DETERMINANTS OF RETAIL SEARCH

Jeffrey G. Blodgett, University of Mississippi

Donna J. Hill, Bradley University

George Stone, University of Mississippi

ABSTRACT -

The purpose of this study was to develop a model of the information search process that better reflects consumers' search strategies. In this model, the amount of retail search undertaken, and the extent to which a consumer relies on interpersonal and neutral sources, are hypothesized to be dependent upon the importance of the product, time availability, perceived risk, subjective knowledge, and enjoyment. Building upon previous research, the authors hypothesize that consumers who rely upon interpersonal advice or neutral sources will, in turn, undertake less retail search. The structural model was tested using LISREL VII, with a sample of 172 consumers who had recently made a major household purchase. The authors suggest that the low explanatory power of the model may be due to the changing retail environment (e.g., new types of stores, changing socio-demographics, technological advancements, etc.), and provide directions for future research.

INTRODUCTION

A considerable amount of research in consumer behavior has attempted to explain why some consumers search for a great deal of information prior to making a purchase while others search very little, if at all (Beatty and Smith 1987; Bloch, Sherrel, and Ridgway 1986; Furse, Punj, and Stewart 1984; Punj and Staelin 1983; Duncan and Olshavsky 1982; Kiel and Layton 1981; Moore and Lehmann 1980; Westbrook and Fornell 1979; Jacoby, Chestnut, and Fisher 1978; Newman 1977; Claxton, Fry, and Portis 1974; Newman and Staelin 1972). One of the underlying goals of this research has been to explain why different individuals rely on different search strategies; that is, why some consumers reference several consumer rating guides (such as Consumer Reports), or spend a considerable amount of time visiting several retail stores and evaluating several brands, while other consumers visit only one retail store and rely mainly on the advice of a friend or a salesperson when deciding which brand to buy. Although these studies have increased our general understanding of the information search process, researchers have yet to explain much of its variance.

Blodgett and Hill (1991) suggested that one reason for the low explanatory power of current models of information search has to do with the conceptualization and measurement of the various dimensions of search (in particular, interpersonal search and neutral sources search). Traditionally, researchers have measured these two dimensions of information search by asking consumers how many times each particular search behavior was undertaken; for example, how many people they talked to, how many magazines or consumer guides they read, etc. (Kiel and Layton 1981; Duncan and Olshavsky 1982; Furse et al. 1984). A serious limitation of these measures is that they do not indicate how much the consumer relied on each particular type of information source when making the purchase decision. To overcome the deficiencies of these amount-of-search measures Blodgett and Hill (1991) suggested the use of "instrumentality" measures. These measures focus on the extent to which a consumer relied on a particular information source when making a purchase decision. As such, these instrumentality (or reliance) measures have the potential to provide greater insight into consumers' search strategies. Using these reliance measures, Blodgett and Hill (1991) provided evidence that interpersonal and neutral sources can serve as substitutes for retail search. These findings are consistent with those of previous researchers (e.g., Furse et al. 1984; Kiel and Layton 1981) who have created profiles of consumers' search strategies. The purpose of this paper is to present a causal model of the information search process, one that better reflects consumer's search strategies. In doing so, this research builds upon, and complements the work of Beatty and Smith (1987) and other researchers (Bloch et al. 1986; Furse et al. 1984; Punj and Staelin 1983; Duncan and Olshavsky 1982; Kiel and Layton 1981; Moore and Lehmann 1980; Westbrook and Fornell 1979; Jacoby et al. 1978; Claxton et al. 1974), and takes into account the suggestions of Blodgett and Hill (1991). In this model, the amount of retail search undertaken, and the extent to which a consumer relies on interpersonal and neutral sources, are hypothesized to be dependent upon the importance of the product, time availability, perceived risk, subjective knowledge, and enjoyment. Building upon previous research, we hypothesize that consumers who rely upon interpersonal advice or neutral sources will, in turn, undertake less retail search. In doing so, we model information search as a dynamic process, thus (potentially) providing greater insight into consumers' search strategies. (See Figure 1 for the hypothesized causal model.)

HYPOTHESES

Since previous studies (e.g., Bloch et al. 1986; Punj and Stewart 1983), and consumer behavior textbooks (e.g., Engel and Blackwell 1982) have more than adequately discussed the underlying theory of information search we will focus directly on the hypothesized model. As previously mentioned, the explanatory variables included in the model are product importance, subjective knowledge, perceived risk, time availability, and pleasure. We do not claim to have included all possible explanatory variables in this study. We recognize that other authors have investigated other possible determinants (e.g., perceived variance among alternatives, satisfaction with prior purchase, ego involvement, etc.); however, based on a search of the literature this subset appears to represent most of the major explanatory variables. Another reason for including this particular set of variables is that we have tried to closely mirror the Beatty and Smith (1987) study so that we can assess the explanatory power of the proposed model in relation to their findings. In the following section we will elaborate on these variables, discuss previous empirical findings, and present our hypotheses.

Time Availability. Time availability refers to the amount of time consumers have to educate themselves about the product category and the alternatives that are available, prior to making a purchase decision. As such, time availability encompasses both the time it takes to physically search for information, and the time it takes to mentally process that information. All things being equal, increased amounts of time allows one to conduct greater amounts of external search, and to adequately process that information prior to making the purchase decision.

The empirical evidence regarding the effect of time availability on retail search is mixed. Although Moore and Lehmann (1980) and Beatty and Smith (1987) found that time availability leads to increased levels of retail search, Kiel and Layton (1981) and Newman and Staelin (1972) both found no relationship between perceived urgency of purchase and retail search. Other researchers have found that consumers who have sufficient amounts of time are more likely to consult neutral sources (Beatty and Smith 1987), while consumers who are under time constraints are more apt to rely on a simple heuristic, such as asking for advice from a more knowledgeable friend (i.e., a purchase pal) when making the purchase decision (Furse et al. 1984). Based on this research, we hypothesize that consumers who have sufficient amounts of time to search for information will undertake greater amounts of retail search, and are more likely to rely on neutral sources when making their purchase decisions, while consumers who have little time in which to search will rely more heavily on interpersonal sources.

FIGURE 1

A MODEL OF THE DETERMINANTS OF INFORMATION SEARCH

Subjective Knowledge. Subjective knowledge is a global self-assessment of how much one knows about a product class, vis-a-vis other consumers, and is an indication of one's self-confidence regarding a particular product category (Brucks 1985; Park and Lessig 1981).

Park and Lessig (1981) and Selnes and Gronhaug (1986) state that subjective knowledge is strongly related to the use of different types of information. Consumers who perceive themselves as being highly knowledgeable about a product category have been found to undertake less external search (Beatty and Smith 1987), to seek less advice from friends (Beatty and Smith 1987), to seek more information from neutral sources (Selnes and Gronhaug 1986; Kiel and Layton 1981), and to rely less on dealer evaluations (Brucks 1985). On the other hand, those consumers who perceive themselves as having little knowledge about a product class have been found to undertake the greatest amount of search activity (Kiel and Layton 1981; Furse et al. 1984), and to be more apt to ask for advice from a friend (i.e., a purchase pal) when making the purchase decision (Beatty and Smith 1987; Furse et al. 1984; Duncan and Olshavsky 1982). Based on this research, we hypothesize that consumers who perceive themselves as being highly knowledgeable about a particular product category will undertake lessor amounts of retail search, and will rely more heavily on neutral sources, while consumers who perceive themselves as less knowledgeable about a product category will undertake greater amounts of retail search and will rely more heavily on interpersonal information.

Product Importance. The concept of product importance recognizes that consumers attach more "worth" to some products than to others (Bloch and Richins 1983). When the product is perceived as being important consumers will conduct a more extensive search process in order to make a good decision (Beatty and Smith 1987; Jacoby, Chestnut, and Fisher 1978; Newman 1977). When a product is perceived as being important consumers are also more likely to seek out information from friends and relatives (Beatty and Smith 1987), and should be more willing to consult neutral sources. Based on this research, we hypothesize that when the product is perceived as being important consumers will undertake greater amounts of retail search, and will also rely more heavily on both interpersonal and neutral sources.

Pleasure. Pleasure refers to the level of enjoyment that consumers experience with a certain product category (Bellenger and Korgoankar 1980; McQuarrie and Munson 1987). Some people (e.g., car enthusiasts, camera buffs) inherently derive pleasure out of certain product categories while other people view the same product category strictly in functional terms.

Consumers who inherently derive pleasure from a certain product category have been found to conduct greater amounts of external search (Bloch et al. 1986). People who experience greater amounts of pleasure with a product category should also be more apt to read hobbyist magazines and articles in consumer rating guides regarding that product category. Indeed, Beatty and Smith (1987) found that a similar variable, ego involvement, led to increased neutral source search. Based on this research, we hypothesize that consumers who inherently derive pleasure from a certain product category will undertake greater amounts of retail search, and will rely more heavily on neutral sources of information.

Perceived Risk. Perceived risk refers to the perceived costs of making a poor purchase decision. These costs (or on the other hand, perceived benefits) can be both economic and psychological (Newman 1977). Several researchers have found that perceived risk leads to greater depth of external search (Swan 1972; Capon and Burke 1980; Duncan and Olshavsky 1982). Consumers who perceive a high degree of risk are also more likely to seek advice from friends and relatives (Cox 1967; Lutz and Reilly 1974) and to consult neutral sources (Clarke and Belk 1979; Moore and Lehmann 1980; Newman 1977). Therefore, we hypothesize that when consumers perceive a high degree of risk they will conduct greater amounts of retail search, and will rely to a greater extent on the advice of friends and relatives, and on neutral sources.

Reliance on Interpersonal Sources. Rather than undertake an extensive search process, some consumers rely more heavily on advice from friends and relatives when making the purchase decision (see Kiel and Layton 1981; Westbrook and Fornell 1979). We explicitly recognize the dynamic nature of the information search process by hypothesizing that consumers who rely more heavily on interpersonal sources will, in turn, undertake less retail search. Support for this hypothesis also comes from Bloch et al. (1986) who pointed out that discussions with friends can serve the same purpose as browsing through a store, and by Blodgett and Hill (1991), who found a negative relationship between reliance on interpersonal sources and reliance on retail search.

Reliance on Neutral Sources. The extent to which consumers rely on neutral sources can also affect the amount of retail search undertaken. Rather than personally inspect and compare brands, or rely on a salesperson to explain the features and benefits of a product, oftentimes consumers instead will turn to hobbyist magazines or consumer rating guides. This type of information search, in effect, helps consumers to narrow their "evoked set" so that less retail search is needed. Therefore, we hypothesize that consumers who rely more heavily on neutral sources will undertake less retail search.

METHOD

The data were collected via a self-report questionnaire administered to staff members at a large midwestern public university, and from evening MBA students at a private midwestern university. A total of 172 useable questionnaires were collected. Respondents were asked to report on their last major household purchase (excluding an automobile) within the last twelve months, and were given $3 in exchange for their participation. Focal products included televisions, microwave ovens, VCR's, cameras, camcorders, stereos, refrigerators, and other major appliances (this group of products is similar to that used by Beatty and Smith 1987). The average cost of the focal product was $684 (s.d.=$809). Respondents, on average, visited 3.11 stores and considered 2.88 brands. They asked for advice from 2.19 friends, and obtained information from .62 neutral sources.

Approximately fifty-seven percent of the respondents were female, while 43% were male. Fifty percent were married or divorced, while 50% were single. Sixty-eight percent of the respondents were university staff members (mainly secretaries and clerks), while 32% were professional employees enrolled in an evening MBA program. Forty-nine percent were college graduates, 36% had attended some college, while 15% reported that their highest educational level was high school. Approximately 51% of respondents stated that they were making a "first time" purchase, 25% had purchased a similar product once previously, while 21% had purchased a similar product more than once previously.

Measures. After first specifying the domain of each construct multiple item scales were developed, as suggested by Churchill (1979). All of the items were based on previous research, and were measured on seven-point scales. In order to minimize response biases several of the items were negatively worded (Feldman and Lynch 1988). Product importance was based on scales by Zaichkowsky (1985) and Laurent and Kapferer (1985). Subjective knowledge and time availability were based on items developed by Beatty and Smith (1987). Pleasure and perceived risk were taken from scales developed by McQuarrie and Munson (1987). Reliance on interpersonal sources and reliance on neutral sources were measured with scales developed by Blodgett and Hill (1991). Retail search was measured with a two-item scale developed by Duncan and Olshavsky (1982). One item measured the number of stores that the consumer visited, while the second one measured the number of brands considered. Based on statistical analyses (Cronbach's alpha, confirmatory factor analysis) a few of the initial items were dropped from subsequent analyses.

RESULTS

The structural model was analyzed using LISREL VII (Joreskog and Sorbom 1990). The original model provided a poor fit to the data, with many of the hypothesized relationships being non-significant. Therefore, all of the hypothesized paths that were found to be non-significant (at the .05 level) were fixed to zero, one at a time. This model trimming process continued until only those paths that were significant at the p=.05 level remained. The final model fit the data quite well, producing a non-significant chi-square (X²156=174.59, p=.147), a goodness-of-fit (GFI) index of .906, and a Tucker-Lewis index of .987. Although the final model fit the data quite well, the overall explanatory power was quite low. (See Table 1 for a summary of the structural coefficients and the fit indices.)

Surprisingly, very few of the hypothesized paths were found to be significant. As hypothesized, time availability had a positive effect (.240) on retail search; however, time availability had no effect on consumers' reliance on interpersonal sources or on neutral sources. Subjective knowledge had a significant effect (.140) on consumers' reliance on neutral sources, but it had no effect on retail search or on consumers' reliance on interpersonal sources. Although pleasure had no significant effect on retail search, it did have a positive effect (.216) on consumers' reliance on neutral sources, as was hypothesized. Contrary to expectations, product importance and perceived risk had no effects on search whatsoever (i.e., retail, interpersonal, or neutral). In addition, we had hypothesized that consumers who relied more heavily on interpersonal or neutral sources would undertake lessor amounts of retail search. However, reliance on interpersonal sources had no effect on retail search, while the effect of reliance on neutral sources on retail search was in the opposite direction of what we had anticipated. Rather than having a negative effect, reliance on neutral sources had a significant, positive effect (.192) on retail search.

TABLE 1

STRUCTURAL MODEL

DISCUSSION

The purpose of this research was to build a model of the information search process that more accurately depicted consumers' search strategies. In doing so, we had hoped to explain a greater percentage of the variation of information search than had previous studies. However, whereas Beatty and Smith (1987) were able to explain 29% of the variance of retail search, the current model explained only 9.9% of the variance of retail search. This result is very surprising, especially since we used essentially the same set of explanatory variables as did Beatty and Smith (1987). Given the seemingly logical premise that interpersonal search and neutral sources can serve as substitutes for retail search, we had expected that the addition of the reliance measures would greatly enhance the explanatory power of the model. Based on previous research (Bloch et al. 1986; Furse et al. 1984; Kiel and Layton 1981; Blodgett and Hill 1991) we had expected that consumers who relied on interpersonal sources would undertake less retail search; however, we found reliance on interpersonal sources to have no effect on retail search. We also expected that those consumers who relied on neutral sources would undertake less retail search; however, reliance on neutral sources actually had a positive effect on retail search. In retrospect, this finding is actually consistent with Westbrook and Fornell (1979), who found that one group of shoppers who relied heavily on neutral information sources visited several retail stores. Apparently, at least for some people, neutral sources supplement, facilitate, and enhance retail search. This latter finding implies that retailers could use neutral sources of information to their advantage. Indeed, many retailers have installed electronic displays that provide neutral information in a Consumer Reports type of format (King and Hill 1994). For example, in many of its stores, K-Mart has installed "walk-up information centers" (Fox 1992) which provide comparative information on K-Mart's selection of large-screen TV's, camcorders, and cameras, etc. Neutral information presented in this type of format might cause consumers to consider certain retail stores and compare certain brands that they might not otherwise have included in their evoked set.

Another surprising finding is that none of the explanatory variables included in the model (time availability, product importance, subjective knowledge, perceived risk) had a significant effect on consumers' reliance on interpersonal sources. We had expected that consumers who perceived themselves as having little knowledge about the product class, who were purchasing a product that they considered to be important, who perceived the financial or psychological risk of making a poor decision to be high, or who had little time in which to visit several retailers and compare different brands would instead rely on the advice of a friend (i.e., a purchase pal). Previous research (Beatty and Smith 1987; Furse et al. 1984; Duncan and Olshavsky 1982; Kiel and Layton 1981; Moore and Lehmann 1980; Jacoby et al. 1978) had found these four variables to affect the amount of interpersonal search undertaken by consumers. Given the limitations of traditional, amount-of-search measures, we had anticipated that these four explanatory variables would be even more strongly related to the reliance measure.

The low explanatory power of the model suggests that there are a number of unmodeled antecedents of retail search. Previous research has indicated that retail search is influenced by a number of factors, not all of which were included in our study. Specifically, such factors as the perceived variance among brands (Duncan and Olshavsky 1982), the stability of product categories (Urbany and Dickson 1987), and the retail environment (Urbany 1986) were not included in our model. Since most of these variables are market related, failure to include these variables would be especially problematic if the retail marketplace had changed significantly over the last several years. Indeed, the retail industry has undergone dramatic structural changes since the 1980's, as evidenced by the emergence of new and different types of retailers. Category killers (e.g., Toys 'R' Us, Circuit City, Best Buy, Home Depot), for example, now dominate specific product categories with a wide variety of brand names at guaranteed low prices. Warehouse clubs (e.g., Sam's Wholesale Club, Price Club) and superstores, which were in their infancy several years ago, have emerged as major players, while discount stores (e.g., Wal-Mart) have continued to thrive and expand, thus creating fierce price competition among the various retailers (Mason, Mayer, and Wilkinson 1993). In order to compete more effectively against these new rivals the large mass retailers (e.g., Sears and Montgomery Ward) have expanded their product selection to include a variety of name brands, and have instituted "everyday low prices." With most of these retailers now guaranteeing "Satisfaction or your money back!", consumers know they can return any item that does not live up to their expectations. With so many retailers carrying similar brands, guaranteeing low prices and "Satisfaction or your money back!", the risk of making a poor purchase decision is less today than what it was just five years ago. Hence, consumers today may not feel the need to visit several different retailers to compare prices and product selection. Likewise, consumers may also feel less need to seek the advice of friends or to consult neutral sources. Because of the many changes in the retail environment, it may be that current models of information search are outdated, and that a new paradigm is needed. In order to better reflect today's retail environment, we suggest that researchers incorporate market structure variables (e.g., product differentiation, stability of product category, retail environment) into future models of information search.

Sociodemographic changes have also affected the retailing industry. With the increase in two-wage earner families since the late 1970's (Mason et al. 1993) consumers have come to place increased importance on time. With all of the demands placed on their time consumers are less willing to spend their free time shopping around. As a result, consumers today may actually do very little comparison shopping. With many stores offering wide product selections at everyday low prices and "Satisfaction or your money back" time-pressed consumers may feel comfortable visiting only a single store, especially if they perceive little variation among the brands (Duncan and Olshavsky 1982) carried by the various retailers, and are satisfied with a previous purchase (Kiel and Layton 1981). Olshavsky and Granbois' (1979) contention, that very little search occurs prior to most purchases, may ring even more true today. Again, it appears that the changing retail environment may play an increasingly important role in the information search process.

The concept of ongoing search (Bloch et al. 1986) also deserves greater attention. Bloch et al. (1986) stated that one of the primary motivations of ongoing search is pleasure. Some consumers actually enjoy casually browsing among various stores to learn more about certain products C despite having no immediate purchase intentions. Although they might not currently be in the market for such a product, these consumers may be preparing themselves for a future purchase. The cumulative effect of this ongoing search is that the final purchase decision requires little additional retail search. Hence, when researchers ask consumers how many stores they visited and how many brands they considered the amount of retail search is underreported because ongoing retail search is not adequately taken into consideration; the same may be true for interpersonal and neutral search also. Although ongoing search, and its effects, are somewhat reflected (albeit poorly) in the present model (through pleasure and subjective knowledge), we suggest that this concept needs to be more fully developed and integrated into the information search model.

Finally, another reason for the low explanatory power of the model may have to do with our measure of retail search. Our measure of retail search consisted of two items, one measuring the number of stores visited and the other measuring the number of brands considered (per Duncan and Olshavsky 1982). Some researchers (Beatty and Smith 1987; Kiel and Layton 1981) have used a somewhat broader measure of retail search, one that reflects how much time was spent at the retailer, and how many phone calls were made to various retailers. It could be that this broader measure more accurately reflects retail search, and that our narrower measure limited the explanatory power of our model.

SUMMARY AND DIRECTIONS FOR FUTURE RESEARCH

This study suggests several fertile directions for future research. First, the concept of reliance on different information sources needs to be further investigated. Conceptually, the amount of a particular type of information search, and the level of reliance placed on that information source, are two distinct constructs (Blodgett and Hill 1991; Engel, Kollat, and Blackwell 1968). It is unclear, however, as to which specific factors cause consumers to rely on different sources of information, and how the level of reliance on each information source affects the retail search process.

Second, it would be of interest to both managers and researchers to explore the effects of relatively recent changes in the retail environment on search behavior. These changes include new types of retail institutions (e.g., category killers, superstores, etc.), new technological developments (e.g., electronic catalogs, computerized information services such as Prodigy, etc.), and changes in economic and family structure (e.g., two-wage earner families with less time to shop).

In summary, the investigation of the consumer search process is crucial to marketers since such knowledge can influence both manufacturers' and retailers' marketing strategies. As this research indicates, there is great potential for developing more sophisticated models of the information search process. By building on the substantial search literature already available and by incorporating such variables as market structure, etc., our understanding and ability to use the search process to provide valuable insights into consumer decision making will be greatly enhanced.

REFERENCE LIST

Beatty, S.E. & Smith, S.M. (1987). "External Search Effort: An Investigation Across Several Product Categories," Journal of Consumer Research, 14 (June), 83-95.

Bellenger, D. N. & Korgoankar, P. (1980). "Profiling the Recreational Shopper," Journal of Retailing, 58 (Spring), 58-81.

Bloch, P.H. & Richins, M.L. (1983). "Shopping Without Purchase: An Investigation of Consumer Browsing Behavior," in Advances in Consumer Research, vol. 10, eds. Richard P. Bagozzi and Alice M. Tybout, Ann Arbor, MI: Association for Consumer Research, 389-393.

Bloch, P.H., Sherrel, D.L. & Ridgway, N.M. (1986). "Consumer Search: An Extended Framework," Journal of Consumer Research, 13 (June), 119-126.

Blodgett, J. & Hill, D. (1991). "An Exploratory Study Comparing Amount of Search Measures to Consumers' Reliance on Each Source of Information," in Advances in Consumer Research, vol. 18, Provo, Utah: Association for Consumer Research. 773-779.

Brucks, M. (1985). "The Effects of Product Class Knowledge on Information Search Behavior", Journal of Consumer Research, 12 (June), 1-16.

Capon, N. & Burke, M. (1980). "Individual, Product Class, and Task Related Factors in Consumer Information Processing," Journal of Consumer Research, 7 (December), 314-326.

Churchill, G. (1979). "A Paradigm for Developing Better Measures of Marketing Constructs," Journal of Marketing Research, 16 (February), 64-73.

Clarke, K. & Belk, R.W. (1979). "The Effects of Product Involvement and Task Definition on Anticipated Consumer Effort," in Advances in Consumer Research, vol.6, ed. William L. Wilkie, Ann Arbor, MI: Association for Consumer Research, 313-318.

Claxton, J.D., Fry, J.N. & Portis, B.P. (1974). "A Taxanomy of Prepurchase Information Gathering Patterns," Journal of Consumer Research, 1 (December), 35-42.

Cox, D.F. (1967). Risk Taking and Information Handling in Consumer Behavior, Cambridge, MA: Harvard Business School.

Duncan, C.P. & Olshavsky, R.W. (1982). "External Search: The Role of Consumer Beliefs," Journal of Marketing Research, 19 (February), 32-43.

Engel. J.F., & Blackwell, R.D. (1982). Consumer Behavior, Chicago, Il.: The Dryden Press.

Engel. J.F., Kollat, & Blackwell, R.D. (1968). Consumer Behavior, Chicago, Il.: Holt, Rinehart, and Winston, Inc.

Feldman, J.M. & Lynch, J.G. (1988). "Self-Generated Validity and Other Effects of Measurement on Belief, Attitude, Intention, and Behavior," Journal of Applied Psychology, 73, no.3, 421-435.

Fox, B. (1992). "Retailers Turn to Kiosks," Chain Store Age Executive, March, 35-36.

Furse, D.H., Punj, G.N. & Stewart, D.W. (1984). "A Typology of Individual Search Strategies Among Purchasers of New Automobiles," Journal of Consumer Research, 10 (March), 417-31.

Jacoby, J., Chestnut, R.W. & Fisher, W.A. (1978). "A Behavioral Process Approach to Information Acquisition in Nondurable Purchasing," Journal of Marketing Research, 15 (November), 532-44.

Joreskog & Sorbom (1990). LISREL VII: Analysis of Linear Structural Relationships by the Method of Maximum Likelihood: User's Guide, Mooresville, IN: Scientific Software, Inc.

Kiel, G.C. & Layton, R.A. (1981). "Dimensions of Consumer Information Seeking Behavior," Journal of Marketing Research, 18 (May), 233-39.

King, M. & Hill, D.J. (1994). "Electronic Decision Aids: Integration of a Consumer Perspective," Journal of Public Policy, 17, 181-202.

Laurent, G. & Kapferer, J.N. (1985). "Measuring Consumer Involvement Profiles," Journal of Marketing Research, 22 (February), 41-53.

Lutz, R.J. & Reilly, P.J. (1973). "An Exploration of the Effects of Perceived Social and Performance Risk on Consumer Information Acquisition," in Advances in Consumer Research, vol. 1, eds. Scott Ward and Peter Wright, Urbana, IL: Association for Consumer Research, 393-405.

McQuarrie, E.F. & Munson, J.M. (1987). "The Zaichkowsky Personal Involvement Inventory: Modification and Extension," in Advances in Consumer Research, 14, M. Wallendorf and P. Anderson, eds., Association for Consumer Research, 36-40.

Mason, J.B. & Mayer, M.L. (1993). Modern Retailing and Practice, Homewood, IL: Irwin.

Moore, W.L. & Lehmann, D.R. (1980). "Individual Differences in Search Behavior for a Nondurable," Journal of Consumer Research, 7 (December), 296-307.

Newman, J.W. (1977). "Consumer External Search: Amount and Determinants," in Consumer and Industrial Buying Behavior, eds. Arch Woodside, Jagdish Sheth, and Peter Bennett, New York: North-Holland.

Newman, J.W. & Staelin, R. (1971). "Multivariate Analysis of Differences in Buyer Decision Time," Journal of Marketing Research, 8 (May), 192-198.

Olshavsky, R.W. & Granbois, D.H. (1979). "Consumer Decision Making C Fact or Fiction," Journal of Consumer Research, 6 (September), 93-100.

Park, C.W. & Lessig, V.P. (1981). "Familiarity and its Impact on Consumer Decision Biases and Heuristics," Journal of Consumer Research, 8 (September), 223-230.

Punj, G.N. & Staelin, R. (1983). "A Model of Consumer Information Search Behavior for New Automobiles," Journal of Consumer Research, 9 (March), 366-380.

Punj, G.N. & Stewart, D.W. (1983). "An Interaction Framework of Consumer Information Acquisition," Journal of Consumer Research, 10 (September), 208-216.

Selnes, F. & Gronhaug, K. (1986). "Subjective and Objective Measures of Product Knowledge Contrasted," in R. Lutz (ed.), Advances in Consumer Research, 13, Provo, Utah: ACR, 67-71.

Swan, J.E. (1972). "Search Behavior Related to Expectations Concerning Brand Performance," Journal of Applied Psychology, 56 (August), 332-335.

Urbany, J.E. (1986). "An Experimental Examination of the Economics of Information," Journal of Consumer Research, 13 (September), 257-271.

Urbany, J.E. & Dickson, P.R. (1987). "Information Search in the Retail Grocery Market," working paper, The Ohio State University.

Westbrook, R.A. & Fornell, C. (1979). "Patterns of Information Source Useage Among Durable Goods Buyers," Journal of Marketing Research, 16 (August), 303-312.

Zaichkowsky, J.L. (1985). "Measuring the Involvement Construct," Journal of Consumer Research, 12 (December), 341-352.

----------------------------------------