Considerations For Situational Research

ABSTRACT - The amount of situational research in the consumer behavior literature is most impressive. See Leigh and Martin (1981). This paper extracts a number of focal issues from this literature and discusses their implications for future situational research. It is felt that proper attention to the issues raised herein prior to one's research design should alert the researcher to both potential pitfalls in the planned research and potential contributions in the study of situation.

Citation:

P. Greg Bonner (1985) ,"Considerations For Situational Research", in NA - Advances in Consumer Research Volume 12, eds. Elizabeth C. Hirschman and Moris B. Holbrook, Provo, UT : Association for Consumer Research, Pages: 368-373.

Advances in Consumer Research Volume 12, 1985      Pages 368-373

CONSIDERATIONS FOR SITUATIONAL RESEARCH

P. Greg Bonner, Rider College

ABSTRACT -

The amount of situational research in the consumer behavior literature is most impressive. See Leigh and Martin (1981). This paper extracts a number of focal issues from this literature and discusses their implications for future situational research. It is felt that proper attention to the issues raised herein prior to one's research design should alert the researcher to both potential pitfalls in the planned research and potential contributions in the study of situation.

INTRODUCTION

The intent of this paper is to address certain problem areas which the aspiring researcher should review prior to undertaking situational research. Due to space limitations, this discussion is not meant to be all inclusive. Most notably, it will not address the definition of situation, the development of situational taxonomies or the delineation of situation types. See Leigh (1981) or Bonner (1983) for expanded discussions of these topics. This paper will consider (1) the study setting, (2) the mode of presentation, (3) the data collection methodology, (4) the objective criterion specification, (5) the situation specification/measurement, (6) control for demand characteristics and (7) interaction in situational research. The importance of each of these considerations to the current review will be developed within the discussion of each topic.

THE STUDY SETTING AND MODE OF PRESENTATION

The study setting, which for purposes of this review will be dichotomized as laboratory [Laboratory is broadly defined to include the classroom; that is, a study setting which utilizes students in a classroom setting will be considered a Laboratory setting.] or field, is important in that it may affect what we measure. Brunswik (1956) strongly suggests that we study variables (situations) as closely as possible to the actual conditions in which the individual will come into contact with them. Lowenthal and Riel (1972) present findings which support Brunswik's contention. They compared responses of walkers through an environment (apprehended) with responses to a verbally described (semantic) environment (in which the respondent had knowledge of the environment but was asked to imagine it, not walk through it), and found the perceived linkages to be confined within the mode of perception; that is, perceived or imagined. Lowenthal and Riel (1972, p. 205) conclude, "What we think we like or should like (or dislike) about certain kinds of environments is often not what we do like (or dislike) when we actually experience them."

The mode of presentation refers to the nature of the engagement between the subject and the variable of interest. It will be reviewed in tandem with the study setting due to the influence and restrictions the study setting places on the mode of presentation. Pervin (1978) views various modes of presentation as falling on a continuum from direct engagement to imagined engagement. Note that direct engagement is not synonymous with observation in the natural setting. For example, a stimulus object can be presented in the laboratory setting. The relationship between the study setting and the mode of presentation is important in that the nature of situation precludes a direct contact presentation in the laboratory except, of course, when the situation under study is a "laboratory" situation. The following matrix illustrates the possibilities of a mode of presentation/study setting classification. Pervin identifies four stages along the mode of presentation continuum but all are not necessary to the current topic development and the two end points, direct contact and written, imagined phenomena, are employed here.

Cell 1 has not been utilized in situational research because the very construction of a direct contact situation in the laboratory imputes an artificiality to the situation which immediately renders its study inaccurate.

FIGURE

However, as noted above, cell 1 is widely used when the variable of interest is a simple stimulus object. Similarly, when one or two situational attributes are the study focus; for example, shelf facings and aisle layout, cell 1 may be a viable approach to uncovering causal effects since it allows for experimental manipulation. Cell 2, which contains Sandell (1968), Bishop and Witt (1970), Belk (1974b), Kakkar and Lutz (1975), Lutz and Kakkar (1975), Reingen (1976), Srivastava, Shocker and Day (1978), Becherer, Morgan and Richard (1979), and Dicksen (1982), and cell 4, which contains Belk (1974a, 1975a), Bearden and Woodside (1976, 1977), Miller and Ginter (1979), Srivastava (1980), and Srivastava, Leone and Shocker (1981), contain much of the situational research and have advantages and disadvantages in that was one moves from direct contact to written description one gains in ability to sample an array of phenomena but loses in certainty of relationship of responses to the actual variables of interest." (Pervin, 1978 p. 88). Written descriptions ( imaginary) of situations allow one to measure reports of behavior across numerous situations and; therefore, an assessment of person-situation interaction is possible. Cell 3, of which Barker's (1968) work on behavioral settings may be the best known, has the advantage of studying the variables of interest more directly. Belk (1979) is the only consumer researcher prior to Stanton and Bonner (1980) to utilize direct contact. Brunswik's (1956) work and the findings of Lowenthal and Riel (1972) discussed above highlight the limitations of situational research carried out in cells 2 and 4. Implications of these limitations will be discussed below in the situation specification review.

DATA COLLECTION METHODOLOGY

Sandell (1968), Bishop and Witt (1970), Belk (1974b, 1979), Kakkar and Lutz (1975), Lutz and Kakkar (1975), Reingen (1976), Srivastava, Shocker and Day (1978), Becherer, Morgan and Richard (1979), and Dicksen (1982) utilized undergraduate or graduate students in their research. In addition to the normal questions of generalizability this respondent group engenders, the particular focus of much of this research on the possible person-situation interaction effect as well as person and situation main effects may well be overly optimistic since the homogeneity of the respondent group may lead to understatement of the person main effect and consequently, inflate the interaction effect. Belk (1974a, 1974b, 1975a) also utilized adult respondents drawn from business and social groups. Again, the respondent homogeneity may be expected to be greater than that existing in the population. Bearden and Woodside (1976, 1977), Miller (1975), Miller and Ginter (1979), Srivastava (1980), and Srivastava, Leone and Shocker (1981) draw their samples from existing mail panels or from mail panels generated from telephone books. One would expect their results to be more universally applicable.

All but one of the consumer behavior researchers noted above employed a questionnaire which required the respondent to complete the questionnaire from recall or hypothetical encounter with the situation. Belk (1979) used a diary method in which respondents used a free-response approach to situational description. Information is not provided on the time lapse between the situational experience and the recording 80 no real estimate as to the extent recall was utilized by the respondents can be made. In all cases, self-reports of behavior (or hypothesized behavior) were collected. Since behavior of the responding organism is one of the variables of interest, response contingent (Pervin, 1978) measures must be employed. However, researchers still have a choice to employ either reactive or nonreactive (i.e. unobtrusive) measures of behavior. Self-reports are reactive and suffer the associated disadvantages of this type of measurement. Since self-reports involve subject cooperation, error and bias in measurement may result. This point is further explored in the consideration of demand characteristics below. Measurement of the non-objective aspects of the accompanying situation must be reactive due to the necessity of respondent knowledge of participation in the research.

OBJECTIVE CRITERION SPECIFICATION

In its assessment of situational effect, the consumer research literature looks at the relationship between situation or situation attributes and another variable. Since much of the literature at least implicitly posits a causal relationship or explanatory significance to situation, the "other" variable will be referred to as the objective criterion. The level [The notion of "level" comes from the concept of a product hierarchy. See Lunn (1972).] of specification of this criterion measure is important since its specification constrains the generalizability and theoretical implications of the results. The concept of level of criterion specification will be illustrated by use of the following contrived example. Suppose one studies two choice situations in which mode of transportation is the criterion. The two situations involve (1) travel from Philadelphia to Boston and (2) travel from 6413 second street to 6515 second street (about 500 feet). Further the criterion specifications are air travel and walking. Prediction is at the product class level but are the results really meaningful? The Philadelphia to Boston decision could be more profitably studied at the form level (air, automobile, bus or train) or the brand level (US Air or Delta) or some combination thereof. The second situation may be studied at the form level (automobile, walk, bicycle). While this example is trivial, the implications for situational research are important.

Bearden and Woodside (1976) specify the criterion measure at the brand level and employ Coke, Pepsi, Seven-Up, Diet Pepsi, and Tab as the criterion products. In discussing their results, Bearden and Woodside (1976, p. 768) theorize, "...Intentions for Coke may be formed more on the basis of produce attributes or attitudinal measures as opposed to situational influences. However, for Diet Pepsi, the situational component was more influential in forming individual intentions....The formation of behavioral intentions for the low-calorie soft drinks were (sic) more influenced by situational factors than by the attitudes held toward the choice objects." I have no disagreement with these findings; in fact, they are beneficial in stressing the importance of the criterion measure specification. Bearden and Woodside are left to speculation in examining the link between situation and soft drink choice and offer some potentially important insights. However, a weakness in their criterion measure specification leaves them powerless to glean much more from their analysis. Their five brands do not represent strictly a brand choice, but, as their own quote above suggests, variety differences (low calorie) play an important role in situational effect. Furthermore, all brands are not even colas 80 that different types (cola and non-cola) exist in the study. In addition, no consideration to form (canned or bottled) is given. The functioning of situational effect may well operate differentially in the variety choice as opposed to the brand choice. Again, Bearden and Woodside suggest this possibility in their discussion. However, without multiple categories per specification level these differential effects can not be studied and theory development is limited.

Product class and brand have, by far, been the dominant criterion specifications utilized. Becherer, Morgan and Richard (1979), Belk (1974a, 1974b), Kakkar and Lutz (1975), Lutz and Kakkar (1975, 1976), Reingen (1976), and Sandell (1968) all used product class as the criterion specification. Bearden and Woodside (1976, 1977), Belk (1975a), Berkowitz, Ginter and Talarzyk (1977), Miller (1975) and Miller and (1979) used brand as the criterion specification. Bishop and Witt (1970) utilized leisure activities as the criterion and, as such, their research was even broader than product class. Belk (1979) was more concerned with developing situational taxonomies and did not utilize his product class (clothes) as a criterion variable. However, he did collect brand, variety, and type information and used the resulting clothes ensembles as observations in a factor analysis. Srivastava (1980), Srivastava, Leone and Shocker (1981), and Dickson (1982) employ variety as the criterion measure. Srivastava, Shocker and Day (1978) utilize an iterative procedure to develop appropriate criterion specification. They begin with the "bresh freshener" market as an end-use market and work backwards through situations and products, Consequently, the authors product list included many brands, products (generic), forms and varieties. However, not enough detail is presented to explain situational effect at various criterion level specifications and; in fact, their factor loading map (p. 36) appears to cluster primarily on product class as one would expect based on the situation-l descriptions employed.

The impact of specifying the criterion variable at only one level; that is, looking at only one dimension of the choice outcome, is twofold. Consumer behavior researchers have not studied situational influence on various components of choice within the same study; therefore, no assessment of differential influence relative to the various possible criterion measures can be made. Seconds no development of theory in the area of situational influence on the choice process has been possible since a unidimensional measurement of choice allows little in the way of process variance (relative to different aspects of choice).

SITUATION SPECIFICATION

Specification of the situation is extremely difficult. Wicker (1975) comments on the almost impossible task faced by the researcher who attempts to symbolically communicate the situation. Pervin (1978) states most situational research has employed imagined situations as a mode of presentation. Endler and hunt (1968) utilize one-sentence descriptions of situations. Sandell (1968) employs short sentences or phrases. This is significant since Endler and Hunt (1966, 1968, 1969) and Sandell (1968) appear to be the "seeds" from which most situational research in consumer behavior has grown. Pervin (1978, p. 88) sums up his feelings on this issue when he states, " . . . imagined modes of presentation allow for responses to a wide variety of stimuli, situations, and environments but one is left wondering about relationships to the actual variables of interest." Returning to our trivial example of the previous section, how does one state the various situations arising in one's travel from Philadelphia to Boston? Obviously, one might include time pressure, vacation/business, companions, financial considerations and numerous other dimensions of importance. However, and this is the crus of the issue, can one ever be sure that he has included all salient dimensions for the respondent?

The prior discussion leads logically to the question of who should specify the situation or situational attributes and how should this specification be made. Except for Belk (1979) and Stanton and Bonner (1980), all other consumer behavior researchers have utilized researcher initiated [Srivastava, Shocker and Day (1978) employed researcher initiated situational descriptions but these were developed with subject input.] situational descriptions and an imagined mode of presentation. For example, Belk (1975a) presented subjects with the following situation: You are planning a picnic with your friends. The choice behavior concerned patronage of fast foot and take-out restaurants. The lack of specificity in the situation allows subjects to respond to different imagined "picnic" situations. Variable unspecified attributes might be number of friends, back yard or park, weather forecast, cooking facilities, and 80 forth.

The problem of properly addressing the salient situational characteristics is explored in depth in Belk (1979) and Bonner (1983). One approach is to employ respondent generated situational descriptions through the use of a free-response approach. Frederiksen (1972) addressed the primary difficulty in developing situational taxonomies in stating, "There is no prescription that can be given to the would-be developer of a taxonomy of attributes of situations with regard to how to proceed." A researcher can never be sure if a researcher-generated list of situational attributes is inclusive of the full range of attributes necessary for each subject individual to adequately define the differentiating characteristics that make two distinct situations operationally similar or disparate Obvious as this problem is to the laboratory researcher who must communicate situations symbolically to his subjects, the field researcher must address the same issue in measuring the focal situation.

CONTROL FOR DEMAND CHARACTERISTICS

Lutz and Kakkar (1975, p. 444), in a partial replication of Belk's (1974a) work, report that considerably less variance is explained by the interaction of situation and products than Belk reported. The authors suggest that "the most plausible interpretation of the discrepancy in results may be that demand characteristics were operating in Belk's (1974a) experiment, causing subjects to exaggerate supposed shifts in consumption behavior across situations." Lutz and Kakkar did not present multiple situations to the subjects while Belk did. Since products were being held constant and situations were changed, it is not difficult to imagine that many subjects could figure out that hypothesized behaviors in varying situations were being measured and respond accordingly. Reingen (1976. p. 130) states, "When a within-subjects design is employed, this may provide the subjects with the cue that they are "supposed to" shift in their buying choices across the situations." Reingen (1976) went on to test for the existence of demand bias by manipulating both demand awareness and subject roles. Results indicated that the within-subjects design often employed in situational research is particularly demand prone. Reingen (1976, p. 132) calls for wa greater reliance on between-subjects design in future research. Lutz and Kakkar (1975, p. 444) provided approximately the same advice is stating, "Careful attention should be focused on this point (demand characteristics) [Parenthesis are mine.] in future development of methods for studying situational influence.

The implications of this prescription to control for demand characteristics are far ranging. They impact, at least, on the mode of presentation utilized, the data collection methodology employed, and the situation specification used. For example, to study a person in multiple situation scenarios necessitates respondent cooperation. The task of hiding the intent of the research from the respondent to preclude either compliant or non-compliant performance seems impossible.

It is important to note that it is the demand characteristics of the study design that cause concern and generally not the demand characteristics of the situations being studied. However, if one were to describe the situation in such detail that the situational specificity necessitated all rational beings behaving in the same way, then the "demand" characteristics of the situation being studied may be so strong as to preclude any generalization beyond the study. As stated, this is not our concern here.

INTERACTION IN SITUATIONAL RESEARCH

Belk (1974a, p. 158) takes a stong stand in favor of the explanatory power of interactions when he reports:

"the results in Table 2 confirm the dominance of interactions over primary source effects in contributions to variance. For snack products, the persons by product interaction is the most important component....The effect commonly implied when referring to situational influence is reflected in the products by situations interaction term. This is the second most important effect in the snack products results, and demonstrates that choice among snack products is dependent upon the consumption and purchase situations examined."

Bearden and Woodside (1976, p. 764-768) similarly attach importance to interaction in hypothesizing and reporting:

Person x Situation interaction may offer greater explanation of consumer behavior .... use of a multiplicative interaction term to represent the Situation X Attitude Toward the Object interaction will further improve prediction is partially supported.

Finally, Lutz (1980, p. 660) drives this point home when, in an otherwise favorable review of Stanton and bonner (1980), he bemoans the lack of an assessment of person-situation interaction in view of the fact that:

"The literature on situational effects has consistently shown that the interaction of situational and personal factors accounts for the bulk of explained variance, as compared to 'main effects' for either factor alone."

This apparently agreed upon consistency is somewhat a myth, which stems at least in part from varying contextual uses of the term "interaction." Lutz (1980) is obviously referring to the Sandell-Belk findings in his statement above. However, a closer examination of Belk (1974a) and Sandell (1968) show that PXS (Person X Situation) is relatively low in both studies, and it is SXP (Situation X Product) interaction which is dominant. It is important, therefore, to see just what this interaction means.

The variance explained in the analysis of variance models used in situational research is that which varies from "not at all likely" to "extremely likely" on a five point scale. (Sandell used a seven point willingness scale.) Subjects respond as to their hypothesized likelihood of choosing each of a battery of pre-grouped products within each of a number of described situations. This approach to explaining behavioral variance is taken from Endler and Hunt (1966, 1968, 1969), and it is instructive to review this work.

Endler and Hunt (1966) attempt to measure the contribution to the total variance in reports of behavior from persons and situations for various indicator responses of the trait of anxiousness. The indicator responses, logically, are all designed to be indicative of "anxiety." For example, Endler and Hunt use indicator responses such as "Heart beats faster," "Get an uneasy feeling," "Perspire" and 80 forth. This information is collected from different subjects for each of many differently described situations. Scales are directionalized so that (5) indicates a high level of anxiety and (1) indicates a low level of anxiety. The total variance in these anxiety scales is partitioned using 3-way analysis of variance in which the main effects are person, situations and response indicators. The authors report the main effects of persons and situations to be small (around 6%), while that of response indicators to be quite large (about 24%). This is to be expected since not all of the response indicators of anxiety can be expected to be equally common throughout the population. (For example, people may often "get an uneasy feeling" but very infrequently experience "having loose bowels.") The three simple interactions account for about another third of the variance. This, then, is the methodology of Sandell and Belk.

Returning to the question of the variance explained in the Belk studies, it is inconsistent with the methodology to expect situation main effects" to be important. For Belk's snack product inventory, and following Endler and Hunt (1966), the indicator responses (snack products) are designed to be indicators of choosing snack products. The situations are designed to be "snack" situations. It would seem a high "main effect" for situation, which means it contributes significantly to the generalized "not at all likely" - "extremely likely" scale variance, would indicate only that it did not belong in the "situation set" employed. Since one's intention is not to look at an underlying trait (anxiety) in consumer behavior situation research but to look at choice among snack products (and not a measure of snack-product proneness), the situation-response indicator interaction (that is, situation by product) is an "interaction" in analysis of variance terminology only. It is really a main effect of situation if the object of the research is to determine what drives the specific product choice and Belk notes it is this "interaction" term which is the effect which is commonly implied by situational influence. A different methodology, for example discriminant analysis, may find a strong "main" effect for situation which corresponds to the "interaction" effect in the analysis of variance methodology. Consumer behavior research should concentrate on approaches where the choice object is the dependent variable when the portent of the research is to explain choice variance. Bearden and Woodside (1976), quoted above, follow this prescription in defining their dependent variable. However, the question of interaction remains.

The interactionist approach in psychology views interaction as that occurring between the person and the situation. Endler (1975, p. 17) summarizes the interactionist position as "...that examines how situations and persons interact in evoking behavior....we are referring to the interaction of two independent variables (person and situation) in effecting behavior (the dependent variable) and not to the interaction between independent and dependent variables." However, Overton and Reese (1973) distinguish between reactive (mechanistic) and active (organismic) models of man. Here, two types of interaction are operative. The first is between independent variables [For example, person and situation.] (as above) that determine behavior and is termed mechanistic interaction. In the organismic model, interaction refers to reciprocal causation between environment and behavior; that is, causation is bidirectional in that behavior also influences persons and situations, and it is termed dynamic interaction.

In situational research, another type of interaction may be identified. When situations are not defined in total, but situational attributes are measured and analyzed, one may assess interaction between situational attributes and the effect of this interaction on behavior. Following Magnusson and Endler (1977), this is termed within-situation interaction. [Although Magnusson and Endler point to within-situation interaction as being dynamic in that the person, in responding to situational cues, is himself a situational cue for others, within-situation interaction may also be viewed as mechanistic in that behavior may be studied as the dependent variable and the independent variables are not "whole" situations but situational cues and their interactions.] The prescription for situational research is to clearly define the interaction under discussion, while paying close attention to the appropriate methodology used to assess interaction.

SUMMARY

The problem areas identified provide opportunities for further research. A comparison of a direct contact study which utilized respondent-generated (free response) situation descriptions with a written imagined study which utilized researcher-generated situational descriptions would be of particular interest in determining the effect of the methodology employed. Obviously, no clear-cut prescriptions exist to eliminate all the problems encountered. For example, it taxes one's imagination to devise a study that would utilize observation of behavior, sample across numerous situations for each respondent, and be free of demand characteristics. However, a full consideration of these potential problems prior to the research design should aid in furthering research and theory development in the study of situational effect.

REFERENCE

Barker, Roger G. (1968), Ecological Psychology, Stanford: Stanford University Press.

Bearden, W. O. and A. G. Woodside (1976), "Interactions of Consumption Situations and Brand Attitudes," Journal of Applied Psychology 61 (No. 6), 764-769.

Bearden, W. O. and A. G. Woodside (1977), "Situational Influences on Consumer Purchase Intentions," in Arch G. Woodside, Jagdish N. Sheth and Peter D. Bennett, (eds.), Consumer and Industrial Buying Behavior, New York: Elsevier North-Holland, Inc.

Becherer, Richard C., Fred W. Morgan, Jr. and Lawrence M. Richard (1979), Person-Situation Interaction Within a Consumer Behavior Contest," The Journal of Psychology, Vol. 102, (July), 235-242.

Belk, Russell W. (1974a), "An Exploratory Assessment of Situational Effects in Buyer Behavior," Journal of Marketing Research, Vol. XI, 156-163.

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. 370-380.

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, 427-437.

Belk, Russell W. (1975b), "Situational Variables and Consumer Behavior," Journal of Consumer Research, Vol. 2, 157-163.

Belk, Russell W. (1979), "A Free Response Approach to Developing Product-Specific Consumption Situation Taxonomies," in Analytic Approaches to Product and Marketing Planning, (ed.), Allan Shocker, Cambridge, Massachusetts: Marketing Sciences Institute, 1; 7-196.

Berkowitz, Eric N., James L. Ginter and Wayne W. Talarzyk (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, 90-94.

Bishop, Doyle W. and Peter A. Witt (1970), "Sources of Behavioral Variance During Leisure Time," Journal of Personality and Social Psychology Vol. 16, No. 2, (Oct.), 352-360.

Bonner, P. Greg (1983), An Analysis of the Theoretical and Empirical Relationships Between the AnticiPated Consumption Situation and Purchase Choice Behavior, unpublished doctoral dissertation, Temple University.

Brunswik, E. (1956), Perception and the Representative Design of Psychological Experiments, Berkeley: University of California Press.

Dickson, Peter R. (1982), "Person-Situation: Segmentation's Missing Link," Journal of Marketing, 46 (Fall), 56-64.

Endler, Norman S. (1975), "The Case for Person-Situation Interaction," Canadian Psychological Review, 16, 12-21.

Endler, Norman S. and J. McV. Hunt (1966), "Sources of Behavioral Variance as Measured by the S-R Inventory of Anxiousness, Psychological Bulletin, 65, 336-346.

Endler, Norman S. and J. McV. Hunt (1968), "S-R Inventories of Hostility and Comparison of the Proportions of Variance from Persons, Responses, and Situations for Hostility and Anxiousness," Journal of Personality and Social Psychology, Vol. 9, No. 4, 309-315.

Endler, Norman S. and J. McV. Hunt (1969), "Generalizability of Contributions from Sources of Variance in the S-R Inventories of Anxiousness," Journal of Personality, 37, 1-24.

Frederiksen, N. (1972), "Toward a Taxonomy of Situations," American Psychologist, (February), 4-123.

Kakkar, Pradeep K. and Richard J. Lutz (1975), "Toward a Taxonomy of Consumption Situations," in Edward M. Mazze, (ed.), 1975 Combined Proceedings, Chicago: American Marketing Association, 206-210.

Leigh, James (1981), Informational and Organizational Bases of Shoreline Protection Decision-Making, unpublished doctoral dissertation, University of Michigan.

Leigh, James and Claude R. Martin, Jr. (1981), "A Review of Situational Influence Paradigms and Research," in Ben Enis and Kenneth Roering, (eds.) Review of Marketing 1981, American Marketing Association, 57-74.

Lowenthal, D. T. and M. Riel (1972), "The Nature of Perceived and Imagined Environments," Environment and Behavior, 4, 189-207.

Lunn, Tony (1972), "Segmenting and Constructing Markets," in R. M. Worcester, ( ed. ), Consumer Market Research Handbook, Maidenhead, Berkshire: McGraw-Hill, 345-377.

Lutz, Richard J., (1980), "On Getting Situated: The Role of Situational Factors in Consumer Research," in Advances in Consumer Research, Vol. 7, ( ed. ), Jerry Olson, San Francisco: Association for Consumer Research, 659-663.

Lutz, Richard J. and Pradeep Kakkar (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, 439-453.

Lutz, Richard J. and Pradeep Kakkar (1976), "Situational Influence in Interpersonal Persuasion," in Advances in Consumer Research, Vol. 3, (ed.), Beverlee B. Anderson, Ann Arbor: Association for Consumer Research, 370-378.

Magnusson, David and Norman S. Endler (1977), "Interactional Psychology: Present Status and Future Prospects," in David Magnusson and Norman S. Endler, (eds.), Personality at the Crossroads: Current Issues in Interactional Psychology, Hillsdale, New Jersey: Lawrence Erlbaum Associates.

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. 455-463.

Miller, Kenneth E. and J. L. Ginter (1979), "An Investigation of Situational Variation in Brand Choice Behavior and Attitude," Journal of Marketing Research, 16. (February), 111-123.

Overton, W. F. and H. W. Reese (1973), Models of Development: Methodological Implications," in S. R. Nesselroade and H. W. Reese, (eds. ), Life-Span Developmental Psychology: Methodological Issues, New York: Academic Press.

Pervin, Lawrence A. (1978), "Definitions, Measurements, and Classifications of Stimuli, Situations, and Environments, Human Ecology, Vol. 6, No. 1, 71-105.

Reingen, Peter H. (1976), "Demand Bias in the Assessment of Situational Effects on Buyer Behavior," in Advances in Consumer Research, Vol. 3, (ed. ), Beverlee B. Anderson, Ann Arbor: Association for Consumer Research, 130-133.

Sandell, Rolf Gunnar (1968), "Effects of Attitudinal and Situational Factors on Reported Choice Behavior, Journal of Marketing Research, Vol. 5, 405-408.

Srivastava, Rajendra K., (1980), "Usage-Situational Influences on Perceptions of Product Markets: Response -Homogeneity and Its Implications for Consumer Research," in Advances in Consumer Research, Vol. 7, (ed.) 9 Jerry Olson, San Francisco: Association for Consumer Research, 644-649.

Srivastava, Rajendra K., Allan D. Shocker and George S. Day (1978), An Exploratory Study of the Influences of Usage Situation on Perceptions of Product-Markets," in Advances in Consumer Research, Vol. 5, (ed.), H. Keith Hunt, Ann Arbor: Association for Consumer Research. 32-38.

Srivastava, Rajendra K., Robert P. Leone and Allan D. Shocker (1981), Market Structure Analysis: Hierarchical Clustering of Products Based on Substitution" Journal of Marketing, Vol. 45, (Summer), 38-48.

Stanton, John L., and P. Greg Bonner (1980), "An Investigation of the Differential Impact of Purchase Situation on Levels of Consumer Choice Behavior," in Advances in Consumer Research, Vol. 7, (ed.), Jerry Olson, San Francisco: Association for Consumer Research, 639-643.

Wicker, Allan W. (1975), "Commentaries on Belk, 'Situational Variables and Consumer Behavior'," Journal of Consumer Research, Vol. 2, 165-167.

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