Four Situations and Their Perceived Effects on Husband and Life Purchase Decision Making

Alvin C. Burns, Louisiana State University
Stephen P. DeVere, Louisiana State University
ABSTRACT - This study investigates the effects of four purchase decision making settings - at home alone, at the store alone, at home with a salesman present, and at the store with a salesman present - on husbands' and wives' perceptions of three dependent variables: relative influence, discussion times, and resolution modes given disagreement. An equivalent groups design was used with five subdecisions for each of four products. Physical surroundings and social circumstances were found to affect estimated discussion times and disagreement resolutions, but no effects were evident for perceived shared influence.
[ to cite ]:
Alvin C. Burns and Stephen P. DeVere (1981) ,"Four Situations and Their Perceived Effects on Husband and Life Purchase Decision Making", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 736-741.

Advances in Consumer Research Volume 8, 1981      Pages 736-741

FOUR SITUATIONS AND THEIR PERCEIVED EFFECTS ON HUSBAND AND LIFE PURCHASE DECISION MAKING

Alvin C. Burns, Louisiana State University

Stephen P. DeVere, Louisiana State University

ABSTRACT -

This study investigates the effects of four purchase decision making settings - at home alone, at the store alone, at home with a salesman present, and at the store with a salesman present - on husbands' and wives' perceptions of three dependent variables: relative influence, discussion times, and resolution modes given disagreement. An equivalent groups design was used with five subdecisions for each of four products. Physical surroundings and social circumstances were found to affect estimated discussion times and disagreement resolutions, but no effects were evident for perceived shared influence.

INTRODUCTION

Recent findings in the area of husband and wife purchase decision making influence have begun to reveal the complexity of the topic, It is common knowledge that influence distributions vary across products, across decisions within products, across phases in the decision making process, across families, and across spouses (Davis, 1976). Although context has been largely ignored in husband and wife purchase decision making studies, there is reason to believe that variance will also be found across situations in which the interaction takes place. In terms of individual purchase decision making, ample support for this contention comes from the arguments by Belk (1975) and Russell and Mehrabian (1976).

The husband and wife decision process would seem to be especially rich with regard to the opportunity for differences between situations inasmuch as the process necessarily occurs through interpersonal communication channels, most of which are overt. That is, spouses must communicate to each other their thoughts, opinions, and preferences. Furthermore, they interact to reach decisions. This requisite for joint decision making brings the process out into an observable realm. In fact, Burns and Granbois (1979) contend that analysis of tape recordings of husband and wives undertaking decisions is comparable to the protocol analysis approach applied to individuals and will greatly help in the understanding of collective purchase decision making.

One can envision situations in which the communication between spouses might be affected by the context. Obviously, individuals will differ in their sensitivities for situational factors, but when there exists the possibility of scrutiny of the conversation by a third party or parties, and the prospect of social evaluation of roles, it is conceivable that situation will influence the purchase decision making process. Consequently, this study may be regarded as an initial attempt to ascertain the effects of four different purchase contexts on three dimensions of the husband and wife purchase decision making process.

CONCEPTUAL FRAMEWORK AND HYPOTHESES

Much of the background for this study is described in a working paper by the authors (1979). This particular paper represents an initial empirical study investigating selected aspects of the comprehensive model proposed in the first paper. The model delineates attributes comprising each of the five "blocks" represented in the general schema:

FIGURE 1

A GENERAL MODEL OF INTRAFAMILY INFLUENCE

The study focuses on two situational dimensions in which husband and wife purchase decisions can occur: the physical surroundings and the social circumstances. Taken as a 2x2 paradigm, the four situations of interest are: (1) at home alone; (2) in the store alone; (3) at home with a salesperson present; or (4) in the store with a salesperson present. Each of these circumstances implies different degrees of potential scrutiny by commercial agents. Similarly, the store atmosphere implies a less familiar context than does the home situation. Couched somewhat differently, one could contend that the greater the potential for social and commercial scrutiny and pressure, the more pressure for normative compliance, while the opposite would be the case in situations where influence interaction takes place in the privacy of one's home. Consequently, it seems appropriate to hypothesize differences in interaction results from situation to situation.

Three dependent variables are addressed in this study, all drawn from the "Results" block. Perceived distribution of purchase influence has been investigated as the primary dependent variable in most family purchase decision making studies (Davis 1976, Burns and Granbois 1979). The second dependent variable is the estimated tine of discussion for various purchase decisions. This variable has not received attention, except in observation studies or analysis of spouses' discussions such as those reported by Kenkel (1957). The temporal dimension appears to be a logical impact point for situational influences. The third dependent variable, "disagreement outcomes," is analogous to the "recognized authority" variable studied and reported by Burns and Granbois (1977) wherein spouses indicated whose choice would be the final decision in the event of a disagreement on the resolution of a purchase decision. Again, the nature of disagreement outcomes is logically linked to discussion context, While other dependent variables are described in the model, these three were selected as representative and as ones most likely to be affected by the independent variables.

The study's hypotheses are stipulated below.

Hypothesis 1

Differences will be evident in all three dependent variables across the two situational dimensions.

Hypothesis 2

The "At Home Alone" situation will be associated with: (a) more equal sharing of influence; (b) longer purchase decision discussions; and (c) more joint resolution of disagreements than will the "At the Store with Salesman Present" situation.

The expectations of the behaviors of the dependent variables under the other two situations are not clear at this point. Consequently, only the two extremes are addressed in this study.

Other hypotheses were also generated for the study as a result of knowledge about the husband and wife purchase decision making process. In general, the following hypotheses require support as corroboratory evidence that the current study is consistent with other studies.

Hypothesis 3

Husbands' and wives' perceptions of purchase decision influence will differ.

Hypothesis 4

Differences will be evident in purchase decision influence across products.

Hypothesis 5

Differences will be evident in purchase decision influence across product subdecisions.

METHOD

The Four Situations

Brief scenarios were constructed for each of the four situations, and all four descriptions are included in the Appendix. In each, the spouses were instructed to maintain a mental image of the context while responding to the questionnaire. In addition, the beginning of each dependent variable section in the questionnaire reminded respondents of the purchase situation they were to retain. Admittedly, these scenarios are weak stimuli, but the authors believe that significant findings in such a case are compelling.

Products and Decisions Studied

Four products were chosen which could "fit" into any and all of the situations. Additionally, certain constraints on the product choice were imposed by the nature of the sample of married couples (see below). The four products included were: (1) vacuum cleaner; (2) life insurance; (3) encyclopedia set; and (4) pots and pans. For each product, five subdecisions were distinguished: (1) verify the need to buy one; (2) what price to pay; (3) how to pay for it; (4) what brand to buy; and (5) what model or type to buy. These five subdecisions or close variations have been studied by others (Burns and Granbois 1979).

Dependent Variables

Relative purchase influence was operationalized as a specification out of 100 percentage points (constant sum scale) as to the amount of influence the husband and wife, respectively, would have for each of the five subdecisions for each of the four products. Jenkins (1978) has shown this approach to be somewhat more reliable than others, while Szybillo, Sosanie, and Tenebein (1979) have shown the equivalence of a constant sum scale to 5-point scales. Discussion time was operationalized as an estimate of how long the couple would talk about each of the subdecisions for each product. Finally, disagreement outcome were indicated by a specification of how the decision would be resolved assuming some disagreement did arise. Spouses could indicate "husband" or "wife," signifying that that person's choice would be selected, or "together," indicating that the disagreement would require a joint discussion and determination of the outcome. Test-retest correlations were determined to assess reliability. Forty-two respondents (21 husbands and 21 wives) were telephoned within weeks of the survey and requestioned on representative items from each variable. The correlations were .88 and .77 for the influence and time estimates, respectively. Also, 79% gave perfectly consistent responses to the disagreement outcomes item retested. Two respondents could not recall the situation presented; all others recalled it correctly (95%).

Research Design and Sample

Questionnaire length and contamination effects precluded asking each respondent to go through all four situations; consequently, a split-plot design with repeated measures on products and decisions was adopted. The sample was drawn from married students living in married student housing on the LSU campus. Only American students were used in the sample, and spouses received identical questionnaires. A field worker introduced the study to respondents and left the questionnaires with the couple. Questionnaires were color-coded; moreover, spouses were requested to sign a statement certifying that they had not compared answers and made any changes as a result. Sample selection was based on the desire to minimize variability between groups and convenience. The final sample size for the study was 148 couples with the following group sizes: At Home Alone, 38 couples; Alone in the Store, 37 couples; At Home with a Salesman, 39 couples; and In the Store with a Salesman Present, 34 couples.

FINDINGS

Equivalence Tests

Equivalence tests verified the equality of the four groups. Several demographic variables were subjected to analysis of variance, and no statistically significant differences were determined. The demographic variables and their means were: Husband's Age (24.7 years); Wife's Age (23.2 years); Years Married (2.7 years); Number of Children (.6 children); Husband's College Education (4.6 years), Wife's College Education (3.4 years); husband's Workweek (22.2 hours); and Wife's Workweek (27.8 hours).

Purchase Influence

The investigation of the distribution of purchase influence began with the mean percentage allocations for husbands and wives assigned by husbands and assigned by wives which are presented for inspection in Table 1. This information served as the data for an analysis of variance determination of the hypothesized differences. For this analysis, only the estimate of the husband's influence was used due to the constant sum interdependent nature of the scale. Because of the confounding efforts of the repeated measures, multiple error terms are required. The couple was used as the basic unit of analysis. The results of this analysis, and salient means are reported in Tables 2 and 3, respectively. The corroboratory hypotheses for products and decisions were supported; however, the hypothesis of differences in spouses' perceptions was not. Main effects for situations are not evident, but significant interaction effects were found, however, between physical situation and product as well as decision. At best, therefore, situation can be viewed as a moderator variable for the main effects of these two variables already known to influence perceived purchase decision making influence distributions.

Estimated Discussion Times

The analysis of the estimated time of discussion for each subdecision assumed a similar tact. The means for the husbands and wives groups are reported by situation in Table 4. Tables 5 and 6 contain analysis of variance results and relevant means, respectively. In this instance significant differences were found for the main effects of all independent variables (physical situation, social situation, spouse, product and decision); while interaction effects were determined for isolated cases.

TABLE 1

AVERAGE INFLUENCE ASSESSMENTS BY SPOUSES ACROSS SITUATIONS

TABLE 2

ANALYSIS OF VARIANCE RESULTS FOR INFLUENCE ASSESSMENTS

TABLE 3

MEANS FOR INFLUENCE ASSESSMENTS

Further presentation of the means is in Table 7 which compares the husbands group to the wives group. Duncan multiple range tests revealed that in relative terms, the estimates are consistent across spouses, yet differences persist in absolute levels with wives always specifying more estimated time than did husbands. Product and sub-decision effects differences are in evidence with the Pots and Pans decisions expected by both spouses to take the least time and life insurance decisions expected to take the most time. Need verification is expected to take the longest time by both spouses, while the resolution of how to pay is anticipated to be shortest in duration. The "At Home Alone" situation is perceived to be the one in which the discussion times would be the longest while the "At the Store With a Salesman Present" discussions are estimated to take about half as long, on the average. The second hypothesis is therefore supported.

TABLE 4

AVERAGE ESTIMATED DISCUSSION TIMES IN MINUTES

TABLE 5

ANALYSIS OF VARIANCE RESULTS FOR DISCUSSION TIME ESTIMATES

TABLE 6

MEANS FOR DISCUSSION TIME ESTIMATES

TABLE 7

MEANS FOR DISCUSSION TIME ESTIMATES: HUSBANDS GROUP COMPARED TO WIVES GROUP

Disagreement Outcomes

The final set of analyses concerned the recognized authority variable. Preliminary investigation began with husbands and wives groups distributions for each situation treatment. The distributions are too cumbersome for display here. Analysis of differences was performed with the use of a generalized least squares linear model approach developed by Grizzle, Starmer, and Koch (1969). The method partitions multinomial responses to test for main and interaction effects. The three corroboratory hypotheses were supported, and social situation main effects were evident but no support was found for physical situation effects. Significant interaction also emerged. Table 9 compares the two situations of interest for the second hypothesis. Only the percent of husbands and wives stipulating "joint" resolution is reported. The hypothesis is not supported. In fact, the reverse of the predicted result seems evident: more joint resolution is believed to occur in the store with salesman present than at home alone.

TABLE 8

ANALYSIS OF CATEGORICAL RESPONSES FOR DISAGREEMENT OUTCOMES

DISCUSSION

The findings of this exploratory investigation of situational influences on husband and wife purchase decision making have revealed that spouses are sensitive to context. In particular, both husbands and wives believe that discussions held at home alone will last longer than when they were held in the store with a salesman present. Also, the character of resolution of disagreements is affected by social circumstances. More joint resolution of disagreements is believed to occur in the store and salesman situation. Specification of the amount of shared influence purchase, however, appears independent of the physical or social surroundings in which it is envisioned to occur.

Precisely why this pattern has been found is a matter of speculation, and the external validity of the experiment may be criticized. Conceivably, these couples require a stronger stimulus than the one presented. Counterarguments to this contention are elicited in the significant findings for estimated times and disagreement outcomes, both of which followed the influence assessments section of the questionnaire. A more logical explanation, it seems, emerges from the sample's characteristics. Well-educated, young, newly married couples with few children and in an obviously income-constrained period of their marriages probably engage in high degrees of perceived equal decision making regardless of physical or social circumstances.

TABLE 9

COMPARISON OF DISAGREEMENT OUTCOMES

This study of situational influences has served to demonstrate the multidimensionality of the husband and wife purchase decision making process. Clearly these and other dimensions deserve more research consideration with sensitivity for context.

APPENDIX

DISCRIMINATIONS OF THE FOUR SITUATIONS

At Home Alone

For each of the following purchase decisions, please assume that you and your spouse are discussing them at home alone. That is, there are no other people in the room during your discussion. By "alone" we mean there would be no children, friends, neighbors or relatives present. It is a discussion between you and your spouse concerning various aspects of the purchase listed below.

Alone at the Store

For each of the following purchase decisions, please assume that you and your spouse are discussing them while standing in a store. Other customers are in the store, some of whom are walking by and some of whom are shopping for items at various places in the store. You are not being helped by a salesperson. Even though sales people are within view, no salesperson is helping you at present, and no salesperson is about to approach you. If you have children, they are not with you in the store, and there are no friends, neighbors, relatives with you either.

At Home with Salesman Present

For each of the following purchase decisions, please assume that you and your spouse are discussing them at home with a salesman present. That is, the salesperson for that product has come to your home and made a sales presentation to you. You are now discussing the various aspects of the purchase of that particular product. Assume that no children, if you have children, are present. Also no friends, neighbors, or relatives are with you in this room -- only you, your spouse, and the salesperson are present.

In the Store with Salesman Present

For each of the following purchase decisions, please assume that you and your spouse are in a store discussing them after a sales presentation has been made to you. That is. there are people walking through the store and shopping for various item at various places in the store, and you are discussing the various aspects of this purchase with the salesman standing beside you and listening to your conversation. If you have any children, they are not present with you at this time, and there are no friends. neighbors, or relatives with you at the time. The salesperson will be present throughout the conversation.

REFERENCES

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

Burns, A .C. and Granbois, D. (1977), "Factors Moderating the Resolution of Preference Conflict in Family Automobile Purchasing," Journal of Marketing Research, (February), 77-86.

Burns, A .C. and Granbois, D. (forthcoming), "Advancing the Study of Family Purchase Decision Making," Advances in Consumer Research, Vol. VII.

Davis, H. L. (1976), "Decision Making Within the Household," Journal of Consumer Research, (March), 2, 241-60.

DeVere, Stephen P., and Burns, Alvin C. "An Expanded Model of Family Purchase Decision Making," LSU working paper.

Grizzle, J. E., Starmer, C. F. and Koch, G. G. (1969) "Analysis of Categorical Data by Linear Models," Biometrics, 25.

Jenkins, R. "Minimizing Bias in the Measurement of Spousal Influence in Family Decision Making by Self-Report Methods," Research Frontiers in Marketing: Dialogues and Directions, 1978 Educators' Proceedings, Sabhash, C. Jan (ed.), 238-243.

Kenkel, W. F. (1957), "Influence Differentiation in Family Decision Making," Sociology and Social Research, (September-October), 42, 18-25.

Russell, J. A. and Mehrabian, A. (1976), "Environmental Variables in Consumer Research," Journal of Consumer Research, 3, 62-63.

Szybillo, G. J., Sosanie, A .K. and Tenebein, A. (1979), "Family Member Influence in Household Decision Making," Journal of Consumer Research, 6, 312-16.

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