An Investigation of the Differential Impact of Purchase Situation on Levels of Consumer Choice Behavior

John L. Stanton, Temple University
P. Greg Bonner, Rider College
ABSTRACT - The importance of situational variables in consumer research has been widely discussed in the recent marketing literature. Results indicate that situational variables can significantly add to the explanation of consumer choice. However, criticisms questioning the methodology employed and the lack of a theoretical perspective have appeared. This research employs a methodology which utilizes an unobtrusive measurement of purchase behavior. Furthermore, the authors discuss a theoretical perspective which clearly delineates between the purchase situation and consumption situation, hypothesize that situational variables differentially impact on various levels of consumer choice, and analyze the predictive ability of situational variables on three distinct levels of consumer choice. The findings confirm the hypothesis that situational variables do differentially influence the different levels of consumer choice.
[ to cite ]:
John L. Stanton and P. Greg Bonner (1980) ,"An Investigation of the Differential Impact of Purchase Situation on Levels of Consumer Choice Behavior", in NA - Advances in Consumer Research Volume 07, eds. Jerry C. Olson, Ann Abor, MI : Association for Consumer Research, Pages: 639-643.

Advances in Consumer Research Volume 7, 1980     Pages 639-643

AN INVESTIGATION OF THE DIFFERENTIAL IMPACT OF PURCHASE SITUATION ON LEVELS OF CONSUMER CHOICE BEHAVIOR

John L. Stanton, Temple University

P. Greg Bonner, Rider College

ABSTRACT -

The importance of situational variables in consumer research has been widely discussed in the recent marketing literature. Results indicate that situational variables can significantly add to the explanation of consumer choice. However, criticisms questioning the methodology employed and the lack of a theoretical perspective have appeared. This research employs a methodology which utilizes an unobtrusive measurement of purchase behavior. Furthermore, the authors discuss a theoretical perspective which clearly delineates between the purchase situation and consumption situation, hypothesize that situational variables differentially impact on various levels of consumer choice, and analyze the predictive ability of situational variables on three distinct levels of consumer choice. The findings confirm the hypothesis that situational variables do differentially influence the different levels of consumer choice.

INTRODUCTION

A strong interest in situational effects on consumer behavior is evident in the recent marketing literature. To date, the research has shown rather conclusive evidence that situational variables add significantly to the explanation of consumer choice behavior (Sandell 1968; Belk 1974a, 1974b, 1975a, 1975b; Lutz and Kakkar 1975; Miller 1975; Berkowitz, Ginter and Talarzyk 1977). However, two shortcomings exist in this research. First, the conceptualizations of situation have not adequately differentiated the purchase situation and consumption situation. Second, the methodology has been primarily restricted to laboratory experiments. Most research has utilized the laboratory setting where subjects were asked to self-report purchase intention in relation to hypothetical situations, which may or may not have been relevant to the subject.

This paper will build on the existing literature on sit-nation and behavior in three significant ways. First, a discussion of situational behavior theory is presented and a rudimentary model is posited. Second, the distinction between the purchase situation and the consumption situation is made. This distinction in the definition of the situation allows for a hybrid measurement of the situation. Both an objective component (Belk 1975) and psychological component (Lutz and Kakkar 1975) are included. Third, a methodological enhancement was made in that the sample of 2915 purchase occasions in this experiment were actual shoppers involved in their usual shopping behavior on the day and in the place of their choice and were totally unaware of their participation in an experiment when they began their regular shopping.

The effect of situational variables on choice will be assessed by comparing the predictive ability of situational variables to the predictive ability of the more traditional demographic and attitudinal variables. Purchaser choice, with respect to the specific product selection during the purchase situations, involves at least four levels. These four levels are not all explicitly discussed in the recent literature. The first level is the product class decision. For example, purchasers must at some point decide that they want dog food, snack foods, etc. Once this decision is made, the form, brand, and variety of the product must be considered. That is, assuming dog food is selected, the purchaser must select a form (dry versus canned), a brand (Purina versus Alpo), and a variety (beef versus liver). The significance of this taxonomy is that situation may differentially influence any of the above "decisions." This research will assess the predictive ability of situational variables at the form, brand and variety levels of purchaser choice.

CONCEPT OF SITUATION

Various conceptualizations of situation have been discussed in detail (Belk 1974a, 1975a, 1975b, 1975c, 1976; Barker 1975; Wicker 1975; Russell and Mehrabian 1976; Lutz and Kakkar 1975). The authors do not feel the need to reiterate these previously explained concepts. However, there exists some important issues which have not been specifically addressed in the existing literature. A clear delineation between the purchase situation and the consumption situation is requisite before psychological and objective measurements of situation can be merged.

Belk (1974a) uses the following situational specifications for meat products:

(1) You thought you would stop by a fast food take-out restaurant to have a quick meal.

(2) No one around the house has been very pleased with dinners lately and you are discussing what you might all try for dinner this weekend.

Both of these situations load heavily on "impromptu situations.'' However, the latter situation (2) seems to describe a situation dominated by its consumption aspect. The former situation (1) seems to define a situation where the purchase aspect clearly interacts with (and to a great extent defines) the consumption aspect of the behavior. There is a need to clearly distinguish between both the purchase and consumption situations and the purchase and consumption behaviors in the theory of situational effects on consumer behavior.

The authors realize that Belk's (1975b) model (Figure A) was intended for expository purposes only. However, Wicker(1975) observes that progress in the situational area will depend upon the development of theories which relate situation and behavior. The authors posit Figure B as representing a rudimentary model of the situational effect on consumer behavior.

FIGURE A

The authors believe that there is a very different set of relevant variables (including situational variables) operative at these two distinct behavior occurrences.

These could be called the two "moments of truth." The first moment is the activity and/or situation in which an object is acquired or purchased for future use. This future use may be distant in the case of "inventory purchases'' of food, additional clothes, etc. or may be almost immediate in the case of fast foods, prescription drugs, etc.

FIGURE B

The second moment of truth involves the consumption of a previously acquired or purchased good. Although the time period may be quite short between the two moments, they are normally not simultaneous. In fact, observation would suggest that for most products the purchase situation is different from the consumption situation. For example, the purchaser may be influenced by situational influences like point of purchase advertising whereas the consumer of that product may be far removed from any advertisement. Figure B suggests that the intended consumption situation influences the psychological component of the purchase situation. This dichotomy of behaviors and situations is consistent with a theoretical perspective that the roles of acquisition and consumption are different.

Clearly two influences can be identified during the purchase event. One influence is the effect of the physical surrounding at the time of purchase. Another is the buyer's psychological interpretation of both the physical surroundings and the anticipated end use of the product. In essence, the intended consumption situation may influence the purchase situation which, in turn, influences the purchase behavior.

METHODOLOGY

The subjects for this study were real shoppers who entered their usual supermarket without previous knowledge of their participation in an experiment. They were intercepted before any food item was purchased and were asked if they would begin their usual shopping by first shopping a particular aisle. [The requirement that the shopper begin in a particular aisle is not necessary to the methodology or research reported here. However, concurrent research, to be reported later, required the shopper begin in the experimental aisle.] No other indication was given to the shoppers at to their participation in an experiment. This aisle, unknown to the shopper, contained the experimental product. The product was a food item which appears in many forms, brands and varieties and is purchased by almost every household during the year. [The specific product cannot be revealed by request of the sponsor.] The experimental product was stocked on the shelf in the usual location. Once the shopper completed a trip down the aisle, she was intercepted again and, if she purchased the experimental product, she was asked questions about her purchases. From this questioning the variables of interest were measured (see Table 1). Since the interviews took place immediately after selection of the experimental product and prior to the financial transaction completing the purchase, it is possible some purchasers would have reconsidered and returned the experimental product to the shelf. However, this moment was deemed superior to an interview after payment since the intervening time lapse might obscure the psychological interpretation of the intended consumption situation at the moment of selection.

TABLE 1

TYPES OF VARIABLES MEASURED IN EXPERIMENT

Examples:

Socio-demographic - Number of children is coded 0,1,2...n

Attitudinal                - Ingredient content viewed as benefit is coded 0,1.

Situational                - End use will be as main dish is coded 0,1.

The methods used in this study overcome many of the critical shortcomings of the previous situation research. Wicker (1975) points out a number of problems in the laboratory approach to situational research. First, Wicker (1975, p. 166) addresses the limitation which results from the fact that "Respondents are merely asked to imagine different situations, and their responses are generally self-reported probabilities or intentions of behaving in specified ways." The precarious nature of utilizing a linkage between expressed intentions and behavior is well known. Second, Wicker (1975, p. 167) notes the impossibility of creating in the laboratory the complex conditions necessary to accurately structure the situation, and he suggests "Unobtrusive observations in natural settings are needed for the initial development of descriptive dimensions of consumer situations..."

Since this research methodology involves the interception of purchasers who have already made a real purchase, no "correlation assumption" using intention as a surrogate for behavior is required. Second, this type of experiment ia as close to unobtrusive as possible since an actual behavioral setting is employed and the only intervention prior to the behavior was a request to begin shopping in a particular aisle. The physical surroundings were controlled since the shoppers were in the same stores, facing similar shelf facings, temperature, and aisle layouts. Third, subjects were not given hypothetical situations which may or may not be relevant to the subject. The laboratory method requires that the "domain" or "conditions" of the situation be prespecified. Even the most arduous experiment could not define all relevant characteristics of a purchase. In this study the situations were not pre-specified but were subject initiated as opposed to experimenter initiated. This insures the relevance of the situation and a situational definition consistent with the salient aspects of the purchase decision at the time the decision was made.

Finally, Belk (1975, p. 161) claims that "Without some sort of hybrid measurement which merges these perspectives (psychological and objective), it appears that situational research must utilize both types of measurement.'' In the above methodology the purchaser's psychological interpretation of the relevant end use is integrated into a hybrid measurement of the situation. The situational variables in Table 1 are broken down into two separate classes of variables. Type I are those which reflect the "first moment of truth" or the purchase situation. Type II variables are those which reflect the perception of the intended consumption situation, the "second moment of truth." Since the experiment took place in the supermarket the situation was in part defined by the content, layout, lighting, etc. of the aisle containing the experimental product. Since these variables were stable across the experiment, they were not specifically measured. Note that the Type I variables address the familiarity of the purchaser with the store and product as well as other specific details of the purchase situation. Therefore, Type I variables are a measure of the objective component of the purchase situation. Type II variables address elements of the intended consumption situation as perceived by the purchaser. As such, these variables are in part a measure of the psychological component of the purchase situation, where the psychological component is defined as "An individuals internal responses to, or interpretations of, all factors particular to a time and place of observation which ... have a demonstrable and systematic effect on the individual's psychological processes and/or his overt behavior." (Lutz and Kakkar, 1975).

ANALYSIS AND RESULTS

As previously mentioned, the purpose of this analysis is to compare the predictive ability of the three classes of variables on the three levels of the purchase decision. The variable classes are situational, attitudinal, and socio-demographic. The three levels of the purchase decision are brand, form, and variety. The method of analysis will be discriminant analysis. The criterion variables represent various groups within the form, brand, and variety variables. Therefore, group membership was determined by purchase choice. Each level is analyzed separately. The independent variables are the specific socio-demographic, situational, and attitudinal variables measured. Table 2 illustrates two examples of product class.

TABLE 2

EXAMPLES OF BRAND, FORM AND VARIETY CHOICE

Prediction of Form

The experimental product can be obtained in five distinct forms, but the forms cannot be identified since they will clearly suggest the product. The authors shall refer to these forms as A through E. See Table 2 for examples of differences in form. Form A through E are all sold under the same generic product class name and appear together in the supermarket aisle. Table 3 presents the results showing the significant discriminant functions for the three sets of variables. Table 4 reports the prediction matrices for the three sets of variables.

TABLE 3

SIGNIFICANT DISCRIMINANT FUNCTIONS - FORM CHOICE: THREE SETS OF VARIABLES

Prediction of Brand

Many brands of the experimental product exist in the marketplace; however, not all brands are represented in all forms. Therefore, to avoid overstating the prediction of brand choice, brand choice was analyzed in two subsets. The first subset of choice considered 3 brand choices, the dominant national brand, a private label brand, and other brands, within forms A and B. The second subset of choice considered two brand choices, the dominant national brand and other brands, within forms C, D and E. Table 5 presents the results showing the significant discriminant functions for subsets I and II using the three sets of variables. Table 6 presents the significance tests for the prediction matrices and the percents correctly classified.

Prediction of Variety

The experimental product was grouped into ten basic varieties; however, the varieties cannot be identified since they would certainly disclose the product class. See Table 2 for examples of differences in variety. Table 7 presents the results showing the significant discriminant functions for the three sets of variables. Table 8 reports the significance tests for the prediction matrices and the percents classified correctly.

TABLE 4

CLASSIFICATION MATRICES FOR FORM CHOICE: THREE SETS OF VARIABLES

TABLE 5

SIGNIFICANT DISCRIMINANT FUNCTIONS - BRAND CHOICE THREE SETS OF VARIABLES

TABLE 6

SIGNIFICANCE OF CLASSIFICATION MATRICES - BRAND CHOICE THREE SETS OF VARIABLES

TABLE 7

SIGNIFICANT DISCRIMINANT FUNCTIONS - VARIETY CHOICE THREE SETS OF VARIABLES

TABLE 8

SIGNIFICANCE OF CLASSIFICATION MATRICES-VARIETY CHOICE: THREE SETS OF VARIABLES

DISCUSSION AND CONCLUSIONS

Two conclusions can be drawn from the previous analysis. First, this research confirms the findings of Belk and other situational researchers that situational variables are clearly significant in explaining purchase behavior. Second, this research indicates that situation has a differential impact on the various levels of purchase choice. Table 9 shows that situational variables clearly outperform attitudinal and demographic variables in the prediction of form and variety choice. Results with respect to brand choice are mixed. The conclusion is, that depending on the product class, the situational variables do not influence all levels of the purchase decision to the same degree. Furthermore, the analysis does not address potential interactions between the three classes of variables.

TABLE 9

SUMMARY TABLE - CORRECT CLASSIFICATIONS

This analysis was not meant to specifically test the model presented in figure B, since consumption situation data was not available. However, the results seam to indicate that the intended consumption situation can significantly affect the choice outcomes of the purchaser during the purchase behavior process. The methodology was designed to allow for unobtrusive measurement of purchase behavior and then to almost immediately assess the purchaser's psychological interpretation of the intended consumption situation. The authors feel this to be a significant methodology in addition to the laboratory experimental setting which has been primarily used in situational research. In addition, situational effects were assessed within a particular generic product class as Opposed to a more general definition of product class as snack foods. Choice is seen to include more than product class choice or brand choice. Form and variety are significant areas of inquiry into choice. Finally, an initial model of situational effect was presented which can serve as a jumping off point for further theoretical discussion in the situational area.

REFERENCES

Barker, Roger G. (1975), "Commentaries on Belk, 'Situational Variables and Consumer Behavior'," Journal of Consumer Research, Vol. 2, 165.

Belk, Russell W. (1974a), "An Exploratory Assessment of Situational Effects in Buyer Behavior," Journal of Marketing Research, Vol. XI, 156-63.

Belk, Russell W. (1974b), "Application and Analysis of the Behavioral Differential Inventory for Assessing Situational Effects in Buyer Behavior," in Advances in Consumer Research, Vol. 1, eds. Scott Ward and Peter Wright, Urbana: Association for Consumer Research.

Belk, Russell W. (1975a), "The Objective Situation as a Determinant of Consumer Behavior," in Advances in Consumer Research, Vol. 2, ed. Mary Jane Schlinger, Chicago: Association for Consumer Research.

Belk, Russell W. (1975b), "Situational Variables and Consumer Behavior," Journal of Consumer Research, Vol. 2, 157-64.

Belk, Russell W. (1975c), "Communication Notes: 'Situating the Situation: A Reply to Barker and Wicker'," Journal of Consumer Research, Vol. 2, 235-6.

Belk, Russell W. (1976), "Communication Note: 'Situational Mediation and Consumer Behavior: A Reply to Russell and Mehrabian'," Journal of Consumer Research, Vol. 3, 174-7.

Berkowitz, Eric N., Ginter, James L., and Talarzyk, W. Wayne (1977), "An Investigation of the Effects of Specific Usage Situations on the Prediction of Consumer Choice Behavior," in Proceedings, 1977 Educators' Conference, eds. Barnett A. Greenberg and Danny N. Bellenger, American Marketing Association.

Lutz, Richard J., and Kakkar, Pradeep (1975), "The Psychological Situation as a Determinant of Consumer Behavior,'' in Advances in Consumer Research, Vol. 2, ed. Mary Jane Schlinger, Chicago: Association for Consumer Research.

Miller, Kenneth E. (1975), "A Situational Multi-Attribute Model," in Advances in Consumer Research, Vol. 2, ed. Mary Jane Schlinger, Chicago: Association for Consumer Research.

Russell, James A., and Mehrabian, Albert (1976), "Environmental Variables in Consumer Research," Journal of Consumer Research, Vol. 3, 62-3.

Sandell, Rolf Gunnar (1968), "Effects of Attitudinal and Situational Factors on Reported Choice Behavior," Journal of Marketing Research, Vol. V, 405-8.

Wicker, Allan W. (1975), "Commentaries on Belk, 'Situational Variables and Consumer Behavior'," Journal of Consumer Research, Vol. 2, 165-7.

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