Informational Social Influence And Product Evaluation Essay
Price is a major influence on travel purchases; however, traveler reviews have also become a prevalent source of influence. Theories of social influence and cognitive dissonance provide insight into consumer decisions. This research investigated the effect of social influence in the form of traveler reviews and price on consumer decisions and postdecision dissonance. Student subjects evaluated two resorts for a Spring Break vacation in Cancun using a 2 (valence: positive or negative) × 2 (unanimity: unanimous or nonunanimous) × 3 (price: same, slightly lower, much lower) experimental design. The results reveal that social influence had a strong effect on both resort evaluations and postdecision dissonance. Nonunanimous reviews reduced the prevailing valence of reviews, but increased dissonance. The lack of results for price suggests that price may not be the predominant influence on decisions, as previously thought. This research provides new insight into the effect of traveler reviews on decisions by evaluating the unanimity of social influence, the effect of price differences, and the extent to which consumers engage in postdecision dissonance reduction.
, Pages 137-143
, Australian Graduate School of Management
, Australian Graduate School of Management
, Australian Graduate School of Management
This paper reports an exploratory field study of the relationships between consumer types, normative social influence, information search using personal sources, and eventual choice. Using the fashion purchases of a sample of 324 female consumers it was found that some consumers perceived clear signals from their peers which obviated the need for external information search. In contrast, some required active reinforcement of their normatively derived preferences during product evaluation, and some made product choices without peer guidance. These complex interactions between normative and informational influence were related to the individual sociodemographic characteristics of the consumers.
This paper presents the results of an exploratory field study investigating the relationships between different types of consumer, the normative interpersonal influence of their peers, exposure to informational influence during the external search process, and eventual product choice. In particular we examine:
1. the likelihood that various types of consumer will perceive different peer group norms concerning the available assortment of products within a product category,
2. whether such normative influence is connected with patterns of external information search - especially the use or non-use of interpersonal sources during product evaluation, and
3. whether the patterns of both normative and informational influences are related to eventual purchase choice within the category.
We chose clothing styles as the stimuli for our research of the following reasons. First, the inherent social symbolism of clothing enhances the likely strength of the phenomenon under investigation (interpersonal influence). Second, the constant variety and change within this product category requires consumers to make relatively frequent and difficult choices. These decisions are therefore unlikely to become routinized or habitual - a circumstance which might make the effects of peer influence harder to isolate. Third, the nature and importance of these choices makes it feasible to rely on respondent recall as the basic source of data about the events that occurred prior to their purchase decisions.
The following sections briefly discuss the relevant literature before describing the methodology and results of this study.
Types of Consumer
It seems self-evident that individual consumers react differently to clothing fashions and indeed there is evidence for this in the literature. The early literature tends to-focus mainly on innovators and opinion leaders but in a wide-ranging review of fashion research Sproles (1981) suggests as many as nine market segments. In a more recent empirical study using classification variables form the fashion literature Midgley (1983) found only four groups, namely 'high status', 'profashion', 'singles' and 'uninvolved'. His study provides the impetus for the current paper as it also relates market segments to patterns of information search.
The idea that influence and choice processes might vary across market segments would not seem to be controversial, yet few of the influence or information search studies in the literature adopt such a stance. A segment-based perspective would also enhance the practical benefits of the research to the marketing decision-maker. The challenge, of course, is to build a useful theory of how different types of consumer make their purchases decisions. Hopefully the results reported here shed light on some aspects of such a theory.
Search, Influence and Choice
The study of personal influence in the context of clothing has a venerable tradition, going back beyond Katz and Lazarsfeld (1955) to Simmel (1904). The need of individuals to both conform to peer norms, and to distinguish themselves from other groups in society, is well documented in this literature. Personal influence processes have also been examined in other consumer settings. For example, work on reference groups by Venkatesan (1966) and Witt and Bruce (19 '2). in his work on social comparison Moschis (1976) distinguishes between active influence and passive (observational) influence, and more recently Price and Feick (1984) distinguish between normative social influence and informational influence. They see normative social influence as 'conforming with the expectation of others' (Cohen and Golden 1912) while informational influence occurs as the consumer attempts to resolve the product evaluation task by the collection of external information. Thus the study of personal influence forms one element of the broader investigation of consumers external information seeking behavior.
A number of studies have identified various patterns of information search including the active use of information from peers, the use of information from media and salespersons, and also whose consumers who appear not to search to any great degree (eg. Claxton, Fry and Portis 1974). Olshavsky and Granbois (1979) thought that active search-based (informational) influence might be more relevant for functional products, while the more passive conformity to group norms might be more relevant to symbolic products such as clothing. Midgley (1983) essentially argues that nature of influence and information search might well depend on the type of individual. He presents evidence to demonstrate the co-existence of both active information search and other more inferential processes for different types of consumers in the purchase of men's suits, and evidence that different processes were associated with different market segments. Also, Price and Feick (1984) argue what normative influence could set a frame of reference within which informational processes might be used to resolve some facets of the decision. In a study of consumer durable purchasing they concluded that informational influence is more prevalent than shown by prior research. Their study, however, did not allow the formal testing of informational and normative influence as rival motivational hypotheses'. Nonetheless, it is arguable that individuals can be exposed to either normative or informational influence, or exposed to both in some complex manner, and that the precise nature of this exposure depends on both their characteristics and the decision they have to make.
Relationship of Influence to Types of Consumer And Choice
Midgley (1983) found some support for the idea that individuals with status or courtship motivations will display different search and influence processes to those lacking such motivations. In particular he found that for fashion items those higher status individuals (concerned to make socially correct decisions) appeared not to actively seek information from peers, but probably inferred the decision either form group norms or from their spouses. In contrast his 'singles' (who displayed some fashion leadership characteristics) appeared to actively consult their peers, possibly in order to evaluate the wide range of new styles. Finally, those consumer types without these basic and strong motivations either displayed extensive search patterns (his attitudinally based 'profashion' segment), or did little apparent search at all (the 'uninvolved').
Unfortunately, while Midgley's (1983) study included both normative and informational influence measures heir relative role was not separated in the analysis. Furthermore, and in common with much of the literature, the choice of style was not examined in depth. This is an important limitation of prior research in that consumers generally have to choose between product forms and/or brands - a choice which presumably may be influenced normatively or informationally. Yet many studies simply examine the act of category purchase - thereby ignoring some possibly interesting comparative aspects of choice behavior.
In this study we find that different market segments do perceive quite different peer norms, that some peer norms are strong enough to allow purchase choice to be made without external search of personal sources, and that some norms require the active reinforcement of peers during the product evaluation phase. We also find that the interaction of consumer types, peer norms and information search has a markedly varied impact on the likelihood that the consumer will make a socially 'correct' choice form the product assortment.
The research instrument was a pre-tested, self-completed questionnaire which was mailed to a s-ratified random sample of women aged 18 years and older in Sydney, Australia. After editing, 324 completed questionnaires were obtained - which represented a response rate of 31 70. The questionnaire included a wide variety of items concerning clothing purchase, only some of which were relevant to this paper. Also included were high-quality reproductions of nine fashion styles. These styles had been pre-selected by clothing designers as being prototypical of those available at the time, and as appropriate to a specific situation. The questionnaire contained a description of this situation (going to a dinner party) and respondents were asked to relate their answers to this context. By this means the situation - specific nature of clothing was controlled for.
The variables of interest to us here included a set of seven measures of an individual's sociodemographic characteristics, and nine measures of the individual's self-perception of how their friends might react to them wearing each of the nine styles. Friends' reactions were measured on a set of seven point favorable to unfavorable scales. The latter were defined as measures of perceived normative social influence. Both the individual characteristics and the normative influence measures were separately cluster analyzed by K-means techniques (MacQueen 1967) to produce parsimonious and easily understandable taxonomies for further analysis. For the individual characteristics a three cluster solution was an adequate representation of the data, while for the friends' reaction measures four clusters were needed. (The criteria used to select these cluster solutions were: face validity based on the prior research referenced above, adequate cell sizes for further analysis, and indicative F statistics).
Respondents were also asked a variety of questions concerning recent purchases of clothing similar to any of the nine styles shown in the questionnaire. ('Recent' was defined as within the last three months). Related to this purchases behavior was a set of twelve measures of information search and shopping patterns. We used these to create an index of whether the individual made their most recent purchase with the active involvement of their peers, or whether they made this decision without such assistance. This index was defined as the measure of informational influence. Finally, respondents were asked which of the nine styles they had bought recently. These purchases were then contrasted with the informational influence index and a measure of whether or not the style purchased conformed with the perceived expectations of peers to create a three way classification. See Tables 4 and 5 below.
TYPES OF CONSUMERS
Cross-tabulation was used for most of the analyses presented here - with the adjusted standardized deviation of each cell being the primary statistic of interest (Haberman 1973). Note that we formed taxonomies or indices on separate sets of variables which were then cross-tabulated. An alternative approach would be to cluster analyze all the measures simultaneously. However, we have reservations about simultaneously clustering measures from quite different conceptual domains. Also, simultaneously clustering all measures may run greater risks of capitalizing on chance than the procedure adopted here.
The three types of consumers which we identified form the cluster analysis of individual characteristics were labeled 'Young Singles', 'Active Parents' and 'Higher Status'. The profiles of these clusters are shown in Table 1 and bear a reasonable, though not perfect, resemblance to prior work in this area (eg. Sproles 1981, and Midgley 1983).
The four clusters reflecting the perceived reaction of friends to the individual wearing each of the nine styles can be described as follows. The first was a cluster who though their friends would react positively to them wearing six of the styles, and would be indifferent to the remaining three. We labeled this cluster 'Peers Favor Most' (26% of the sample). The second was a group of individuals who considered that their friends would react quite negatively to three of the styles and would be indifferent to the rest. We labeled this cluster 'Negative Peers' (39% of the sample). The final two clusters were comprised of individuals who thought their friends would react quite positively to some styles and quite negatively to others. Since the focus of our paper is not on which particular styles received negative or positive evaluations, and since one of these clusters was small, we combined them into one group which we then labeled 'Peers Discriminate' (36% of the sample).
Are Different Types of Consumers More or Less Likely to Perceive Different Peer Group Norms?
Table 2 displays the highly significant relationship (pc 0.0001) between the Types of Consumer' taxonomy and the "Perceived Normative Influence" taxonomy. As shown by the adjusted standardized residuals the 'Young Singles' were most likely to perceive their peers as discriminating between styles, and less likely to perceive their friends as being either mainly positive or mainly negative. Hence the 'Young Singles' expected their peers to provide clear market signals as to which styles were 'in' and which were 'out'. At the other end of the spectrum both the 'Active Parents' and the 'Higher Status' did not expect to receive discriminating reactions from their friends. Indeed, both groups were strongly associated with the 'Negative Peers' pattern. However, the 'Active Parents' cluster also demonstrated some association with the 'Peers Favor Most' classification. Midgley argued that fashion consumers without the strong motivations of a single lifestyle, or of a higher status position, might well have no one dominant pattern of behavior.
TYPE OF CONSUMER VERSUS PERCEIVED NORMATIVE SOCIAL INFLUENCE
PERCEIVED NORMATIVE SOCIAL INFLUENCE VERSUS REPORTED INFORMATIONAL INFLUENCE
PERCEIVED NORRNAIIVE SOCIAL INFLUENCE VERSUS CONFORMITY OF STYLE CHOICE
Is Normative Influence Related to External Search?
Table 3 displays the significant relationship between perceived normative social influence and reported informational influence. The first notable result in Table 3 is that those respondents who perceived their peers as viewing some or all of the styles positively were much more likely to make a category purchase than those who thought their peers would be negative. The second result is that the 'Peers Favor Most' perception was associated with external information search and consequent exposure to informational influence. In contrast, and for those who made a purchase, the 'Peers Discriminate' perception was most strongly associated with the absence of external search for peer information. It could be that the perception of clear normative signals form their peers obviated the need for search when making an actual purchase - at least for some individuals.
Are Patterns of Normative and Informational Influence Related to Eventual Choice?
Table 4 presents the significant relationship between perceived normative social influence and conformity of style choice, while Table 5 displays the significant relationship between reported informational influence and conformity of style choice. 'Conformity of Style Choice' was defined within the context of the normative social influence groups. That is, did the individual purchase a style that was highly favored by their peer group or did they purchase another style less favored? Hence there is a structural zero in Table 4 because those respondents who perceived their peers as negative towards the styles could not, by definition, make a 'favored' purchase. Tables 4 and 5 were also difficult to compare in that while normative social influence can legitimately lead an individual not to make a purchase this is not so for informational influence. As measured here informational influence could only occur within the context of an actual purchase. This is a significant methodological issue for both our study and the literature in general; and one we will return to in the subsequent discussion.
Table 4 contains some interesting results, especially when contrasted with Table 3. First, the very strong association between the 'Peers Favor Most' normative grouping and the purchase of a 'Highly Favored' style. This, of course, is because of the fact that these individuals' peers favored 6 out of 9 styles. It was therefore relatively easy for them to make a 'socially correct' purchase. The more striking result was that despite these lower risks of an incorrect style choice this group was still more likely than the others to conduct external information search (as shown in Table 3). Second, any purchase by those respondents who perceived their peers to be negative had, by definition, to be of a non-favored style. Perhaps the more interesting result here was that 47% of this 'Negative Peers' group actually make a purchase, and in doing so were less likely to conduct information search amongst their peers than the other normative groups (again from Table 3). Third, those respondents who perceived their peers to discriminate were also likely to purchase a 'favored' style. The risks of an incorrect social decision were greater here as the normative social influence for this grouping highly favored 3 out of 9 styles, strongly disliked 2 out of 9, and was indifferent to 4 out of 9 (the 'disliked' and 'indifferent' forming our 'other styles' category). While some of the 'Peers Discriminate' grouping did purchase socially incorrect styles (perhaps because they did not involve their peers ln the actual purchase - Table 3) the stronger association was for them to purchase the 'Highly Favored' styles.
REPORTED INFORMATIONAL INFLUENCE VERSUS CONFORMITY OF STYLE CHOICE
Table 5 was only marginally significant but does provide some evidence that (all other things being equal) informational influence during purchases was more likely to reinforce social norms, while its absence increased the likelihood that a non-favored style was purchased. The "Did Not Buy" category was dropped form this Table because informational influence was defined only to occur in the context of an actual purchase.
There appear to be three main conclusions which can be drawn from this study. First, only those fashion involved individuals with a single lifestyle perceived that their peers would give them clear signals about which items of the available product assortment would be highly favored. Signals that were strong enough to obviate the need for external information search for many of the purchases made by this segment, and signals which despite the absence of search were quite likely to result in a socially correct decision in a risky choice situation (2 out of 3 chance of being wrong) The rest of the market either saw general support for most styles or a degree of negativism/indifference toward the available styles. Second, where individuals saw general support for most of the product assortment they were more likely to actively seek reinforcement for their decision by involving their peers in the actual product evaluation process. That is, these individuals probably used informational influence within the context of normative social influence. Third, where individuals saw their peers as being negative or indifferent to the available styles they were more likely not to buy, but if they had to make a purchase (for reasons not measured here) they possibly may have avoided collecting more (confusing?) peer information.
Thus our study provides some evidence that individual characteristics, normative social influence and external information search interact in a complex manner. There are, however, a number of problems in generalizing these results. First, the problem of 'not buying' alluded to earlier. We did not measure any information search which was conducted but where the individual failed to make a purchase. As far as we know neither do most of the studies in the literature. Hence we cannot be completely confident that 'informational influence' is primarily associated with the 'Peers Favor Most' perception. It could be that a different type of information search resulted in the 'Did Not Buy' category. Second, we only examined information search using one type of personal source. Information can be obtained form other sources such as media or salespersons. Third, our study is subject to the usual limitations of post hoc surveys which rely on respondent recall.
However, despite these limitations we hope that this study has been useful in demonstrating both that the phenomena studied may be more subtle than previously conceptualized, and that there is good potential for constructing more predictive theories in the area.
Claxton, John D., Joseph N. Fry and Bernard Portis (1974), "A Taxonomy of Prepurchase Information Gathering Patterns", Journal of consumer Research, 1 (December), 3543.
Cohen, Joel and Ellen Golden (1972), "Informational Social Influence and Product Evaluation", Journal of Applied Psychology, 56, 1, 54-99.
Haberman, Shelley J. (1973), "The Analysis of Residuals in Cross-classified Tables", Biometrics, 29, 2@5-220.
Katz, Elihu and Paul F. Lazarsfeld (1955), Personal Influence, New York: Free Press.
MacQueen, James B. (1967), "Some Methods for Classification and Analysis of Multivariate Observations," in Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, Volume 1, Lucien M. LeCam and Jezzy Newman (Editors), Berkeley: University of California Press.
Midgley, David F. (19 ,3),"Patterns of Interpersonal Information Seeking for the Purchase of a Symbolic Product", Journal of Marketing Research, 20(February), 74-83.
Moschis, George P. (1976), "Social Comparison an Informal Group Influence", Journal of Marketing Research, 13 (August), 237-244.
Olshavsky, Richard W. and Donald H. Granbois (1979), "Consumer Decision Making - Fact or Fiction", Journal of Consumer Research, 6 (September), 93- 100.
Price, Linda L. and Lawrence F. Feick (1984), 'The role of Interpersonal Sources in External Search: an Informational Perspective", in Advances in Consumer Research, Vol. 11, Thomas C. Kinnear (Editor), Ann Arbor: Association for Consumer Research, 250-3.
Simmel, Georg (1904), "Fashion", International Quarterly, 10 (October), 130- 155.
Sproles, George B. (1981), "Analyzing Fashion Life Cycles: Principles and Perspectives", Journal of Marketing, 45, (Fall), 116-124.
Venkatesan, M. (1966), "Experimental Study of Consumer Conformity and Independence", Journal of Marketing Research, 3 (November), 384-387.
Witt, Robert E. and Grady D. Bruce (1972), "Group Influence and Brand Choice Congruence", Journal of Marketing Research, 9 (November), 440-443.