Shifting Roles in Family Decision Making

Julie Ruth, University of Washington
Suraj R. Commuri, University of Nebraska-Lincoln
ABSTRACT - While the roles played by husbands and wives have been shown to change over stages of the family decision-making process, a gap exists in the literature with respect to whether and how decision role structure may change over time. To address this gap, a qualitative research study was conducted with urban, middle class couples in India regarding their decision-making processes for a diverse set of product categories. The results indicate that macro-level environmental trends toward urbanization, westernization, and the desire to "work hard, grow rich" were associated with shifts in the decision role structure over the eight-year period.
[ to cite ]:
Julie Ruth and Suraj R. Commuri (1998) ,"Shifting Roles in Family Decision Making", in NA - Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT : Association for Consumer Research, Pages: 400-406.

Advances in Consumer Research Volume 25, 1998      Pages 400-406

SHIFTING ROLES IN FAMILY DECISION MAKING

Julie Ruth, University of Washington

Suraj R. Commuri, University of Nebraska-Lincoln

ABSTRACT -

While the roles played by husbands and wives have been shown to change over stages of the family decision-making process, a gap exists in the literature with respect to whether and how decision role structure may change over time. To address this gap, a qualitative research study was conducted with urban, middle class couples in India regarding their decision-making processes for a diverse set of product categories. The results indicate that macro-level environmental trends toward urbanization, westernization, and the desire to "work hard, grow rich" were associated with shifts in the decision role structure over the eight-year period.

The family and family decision-making (FDM) processes have been a focus of consumer research, as families provide a contextual framework for understanding consumers’ economic and social well-being. Much of this research has focused on issues such as decision role structure - whether the decision of what to buy is joint, dominated by one member, or split among the members and decided autonomously (Corfman 1987); relative influence of family members across various stages of the decision-making process (Belch, Belch, and Ceresino 1985; Davis and Rigaux 1974); and the impact of sex-role orientation (Qualls 1987) and ethnic group identification (Webster 1994) on purchase decision outcomes (for reviews, see Davis 1976; Foxman, Tansuhaj, and Ekstrom 1989).

While many important issues have been addressed effectively, the questin of how and why decision role structure changes over time in family decision making has received relatively little theoretical or empirical attention. In fact, there has been very little attention to longitudinal studies of family decision making in general (Roberts and Wortzel 1984), even though, for example, the conceptual model of family life cycle would imply that the decision roles of husbands and wives may change over time due to stage-related changes in the family life cycle.

While decision role structure may change over the course of the family life cycle, another important reason for role changesCand a focus of this paperCis that the macro-level environment in which consumers operate may change (e.g., economic, social, or cultural forces), which in turn may provide impetus for decision role structure to change. Over a period of time such macro-level changes tend to influence the behavior of people in a society (Sheth 1974), and the family as a decision unit is not likely to be immune to these changes. Thus, this research examines the relationship between such macro-level changes over time and the shifts in family decision role structure, using retrospective reports to do so.

FAMILY DECISION MAKING OVER TIME

The most comprehensive framework for examining FDM is the Davis and Rigaux (1974) adoption of the feasibility triangle suggested by Wolfe (1959). The framework describes decision role structure as: Syncratic (joint decisions), Wife-dominated, Husband-dominated and Autonomous (either husband or wife but independently). Davis and Rigaux (1974) examined shifts in decision role structure across stages of decision-making (problem recognition, search and final decision). While decision role structure has been found to change with stages of one decision episode, it can be argued that the framework has never been examined for changes over time.

While the feasibility triangle provides a framework in which changes in decision role structure over time can be assessed, it has a limitation for our study. It classifies autonomous decisions as one group and does not look at autonomic-wife and autonomic-husband decisions separately. By virtue of classifying these two decisions in the same category, it is possible that smaller yet significant changes such as shifts in the influence of one of the partners on the overall decision may be ignored. Since the objective of the present research is to understand changes, both subtle and significant, we felt it important to make a distinction between the two types of autonomous decisions.

Perhaps the only study that examined whether FDM changes over time is a retest of the original Sharp and Mott study (1956), where Cunningham and Green (1974) observed that in various product categories there was a shift between 1955 and 1974 in whether decisions were made autonomously or jointly between husbands and wives. Their research, however, did not investigate specifically why such shifts occurred. In addition, like most research in the area, their research classified decision making into simply an issue of who makes the purchase decision of what to buy. It can be argued that other aspects of the purchase decision are important including, for example, when to buy and how much to buy. Also, neither of these studies addressed the influence of macro-level economic, social and cultural factors on decision role structure.

The purpose of this research, then, is to address some of these inadequacies in the literature by investigating the relationship of macro-level socio-economic and cultural factors to changes in family decision making. Specifically:

1. In what way do macro-level social and economc changes affect FDM and, more specifically, decision role structure?

2. What do such changes mean to the relative roles of husband and wife vis-a-vis purchase decisions in various product categories?

This paper reports the findings of qualitative research conducted with couples in two large metropolitan cities of India. India was considered an apt environment because of the rapid socio-economic changes that have taken place in the last ten years, offering an accelerated view of the possible impact of long-term, macro-level socio-economic and cultural changes. Using critical incident methodology, the couples were asked to recall decision-making processes from the past (eight years prior) as well as current decision-making processes for the same product categories. The resulting "shifts" in decision-making patterns by the couples were the subject of verbatim responses, which were analyzed to understand their relationship to societal-level shifts in socioeconomic and cultural norms and conditions. Issues related to shifts in the decision role structure and the reasons for such shifts are addressed.

RESEARCH DESIGN AND METHODOLOGY

The research design and methodology were guided by many factors that emerged from prior research on family decision making and an understanding of the societal-level characteristics and norms associated with the subject respondents. First, past research has generally observed that husbands and wives, compared to children, exert significant influence on various stages of decision making (Belch, et al. 1985; Foxman et al. 1989). Since the purpose of this research was to focus on the roles of husband and wife in FDM and because husband and wife contribute to the bulk of decision making, children as explicit influencers in the decision-making process were deliberately kept out of the scope of the study. Instead, we elected to control for the presence of children by selecting only respondents who have at least one child living at home. This also controlled for the stage in the family life cycle.

Second, many studies have explored FDM within the context of house or car decisions. The characteristics associated with these products (i.e., potentially high risk, non-recurrent decision situations) may limit generalizability of findings. Accordingly, a wide spectrum of frequently purchased products was adopted for study here. Third, Davis (1976) discussed that conclusions about who (husband vs. wife) was involved were very sensitive to which respondent was interviewed. In order to address that issue, importance was placed on designing the study so that husbands and wives would respond jointly.

Keeping these issues in mind, joint depth interviews were conducted using critical incident methodology (for similar use in another consumer behavior context, see Mick and DeMoss 1990). Qualitative research was used because of two considerations. First, as the respondents would be required to discuss the past (decisions made up to eight years ago), it was felt that structured facilitation by the interviewer would be more effective than a survey instrument. Second, as the focus of the study was on uncovering the reasons for the shifts, it was felt that guided probes during free-flowing discussions with the respondents would yield better results.

The research design involved several stages: (1) screening and recruitment of respondents and identification of product categories for study; (2) an in-depth interview where respondents recalled from eight years ago their decision processes for 8-10 products; and (3) a second, follow-up interview, where respondents were asked to describe their present-day decision-making processes for those same product categories. The first interview was conducted within 24 hours of recruitment and typically lasted about one hour. The second interview was conducted within 48 hous of the first and lasted about 20 minutes. Both interviews were conducted in the homes of the respondents. Privacy during the interviews was ensured to the maximum extent possible. All interviews were conducted by one of the authors.

Screening and Recruitment

The respondents were screened for certain demographic characteristics including at least eight years of marriage, at least one child living at home, age of the spouses (approximately 35-40 years old), and socio-economic classification (SEC B1/B2; SEC’s B1 and B2 constitute the bulk of what can be termed the Indian urban middle class). Though the role of child was outside the scope of the study, the criterion of a child was included to control for the effect of children. In those cases where respondents fit the demographic profile and were willing to participate, the researcher scheduled a visit for a time when both husband and wife would be available.

Interview #1

Although the main purpose of Interview #1 was to obtain responses regarding past product decisions, an important first step was to identify product categories that husbands and wives were involved in and affected by the purchase decisions. In order to do so and to facilitate accurate expression, an innovative visual-based method was developed to capture the relative involvement of household members.

Identifying Products for Study. Respondents (husband and wife) were first asked to identify products/services purchased with some regularity and, then, any other products purchased in the last four months. Each of these products was recorded on a separate card, which was then placed on a template with (unlabeled) concentric triangles (see Figure 1; the actual template contained eight concentric triangles). Then, each respondent was given three 12 inch sticks with the word "husband" or "wife" or "others" pasted on them. At the same time, each respondent was asked to place the sticks around the first product name card in an effort to form a fence around it. Respondents were asked to bring the sticks only as close to the product card as they perceived the involvement of that person (husband/wife/others) to be in the purchase decision of that product. Respondents were instructed to freely choose any range for each stick. The ranges were not identified in any semantic or numeric fashion. The respondents were given two examples (with a different context) to ensure that they understood the task completely. The exercise was then undertaken for each of the product categories that respondents had mentioned.

The complete exercise was then repeated to assess the extent to which each respondent was affected by the decisions in each of the product categories. This time, the respondents were asked to place a stick close to the product card if a wrong decision with that product affected that person the most and vice versa.

The products to be discussed in the interviews were selected when they met both of the following criteria: (1) Only the husband and wife sticks featured in first and second ranges (from the center) for the first task; and (2) At least the husband and wife sticks featured in first and second ranges (from the center) for the second task. These criteria ensured that the product categories discussed in the interviews were ones in which respondents were actively involved and ones that were relevant to the respondents. This technique was developed exclusively for this research and pre-tested before being used here.

The Interview Procedure. Depth interviews were conducted using critical incident methodology. The respondents were guided to go back in time to eight years ago. Respondents were asked to recall where they lived/worked eight years ago ("Let us now go back to 1985. Where were you living then? Were you here in this city?") and similarly prompted until an anchor was detected around which relevant critica incidents could be constructed. The interviews were then guided with prompts about the respondents’ general life situations eight years ago, until the point when the researcher was convinced that the respondents were sufficiently anchored in the past.

Then, the interviews were guided towards discussing decision role structure for the product categories. For each product, respondents were asked to focus specifically on three aspects of decision making: which brand to buy, what quantity to buy and when to buy. The respondents were asked to recall how they went about making these three decisions. Forty seven joint interviews were conducted at this stage.

Interview # 2

The objective of the second round was to explore current decision making vis-a-vis the same product categories explored in the first depth interview. In addition, the respondents were also asked to discuss the reasons for any changes in decision making between the two time periods. Respondents were not, however, cued to think of any macro or micro-level variables. The interviews were merely built on whatever emerged. At the end of the second interview, respondents were given an option to change anything they said in the first interview. The purpose of giving respondents this option was to allow them to discuss any statements from the first interview that might have caused post-interview dissonance. None of the respondents exercised this option.

FIGURE 1

TEMPLATE INDICATING RANGES FOR THE STICK TASK

The rationale for splitting the interviews into two stages was to limit the possibility that the responses about one time period might bias those about the other time period, where respondents might be inclined to structure their responses to the second time period in light of or in comparison to the earlier time period. Thus, respondents were not informed of a second visit until the stage of seeking an appointment 48 hours after the first visit. Three households could not be revisited for this round due to non-availability of respondents. Also, the researchers specifically chose to collect the data regarding the past in the first interview because of the degree of effort required to recall events in the past. A second round interview to describe current decision-making behavior placed less strain on the respondents and enhanced the likelihood of agreeing to a second interview.

ANALYSIS

Overall, 47 interviews with husbands and wives were conducted. Three couples could not be revisited for the second round. Eight interviews were not used in the analysis due to the presence of obvious and numerous inconsistencies in the verbatims, indicating difficulty in recollection. Thirty-six interviews were complete and suitable for final analysis. All the interviews were audio recorded and verbatim transcripts were content analyzed by the researchers. The following product categories were used in the analysis because of their mention in most of the interviews. However, shifts in decision role structure for eight products are reported here.

Household        Personal                   Others

Coffee               Bathing soap           Entertainment

Cooking oil       Men’s clothing         Restaurants

Durables            Women’s clothing    School for the child

Snacks               Toothpaste             Visiting family/friends via trips

RESULTS

Before evaluating whether FDM has changed over the years, it is important to examine decision making in a framework that would allow us to critique these changes appropriately. Given our research objectives and methodology, it became apparent that autonomous decisions should be examined separately for wife and husband. In addition, the three purchase related decisions-what brand, what quantity and when to buy, were used as inputs. After formal content analysis of who made each of these three decisions, the following titles were attached:

                                      Out of the three decisions: what brand, what quantity, when to buy.

Wife dominated:             Two decisions by wife and the third by husband.

Husband dominated:      Two decisions by husband and the third by wife.

Solo-Husband:               All three by husband.

Solo-Wife:                     All three by wife.

Joint:                              At least two decisions together.

These decision classifications appeared to fall into the pattern described in Figure 2. This classification makes an implicit assumption that none of these three decisionsCwhat brand, what quantity, when to buyCis more important than the others. Nevertheless, in situations where one spouse makes two of the three decisions, by inference, the other spouse does exert some influence on the final decision.

In order to examine the shifts in decision making, each product category was plotted on the grid and its movement traced between the two time periods. The shifts between the two time periods are expressed in Figure 3. The arrows indicate the direction of movement between the two time periods. So, for example, the results show that the decision about a school for a child moved from being a husband-dominated decision to a joint decision. Similarly, entertainment used to be a joint decision but later shifted to a wife-dominated decision; i.e., at least two of the three purchase decisions (what, when, how much) used to be joint whereas now two of the decisions are by the wife and the third by the husband. Decisions regarding toothpaste did not show any difference in its decision role structure over the eight-year time period.

Shifts in Decision Making

It is clear that there is a trend towards decision making with both the partners involved. More product categories are now featured in a diagonal band extending from bottom left to top right, representing some degree of collaboration between husbands and wives, rather than the single spouse dominated areas at the top left and bottom right. The specific shifts of some of the categories are described here. While some of the reasons are briefly discussed here, the next section deals exclusively with reasons for the shifts.

FIGURE 2

FAMILY DECISION MAKING GRID

FIGURE 3

SHIFTS IN FAMILY DECISION MAKING

Furniture. Decisions regarding the purchase of furniture moved from being solo-husband to joint. Content analysis clearly revealed that this shift is largely because of macro level variables. The heightened awareness of women regarding fashion and trends and what was in style and what was out of style as well as greater incidence of the wife contributing to the budget appeared to drive this shift.

I am not like some women who do not know about styles. I know what looks good in this house. Maybe I did not know this all the while. But when we see good houses on TV, I get a feeling that I should also buy like that. So I tell my husband. (Wife)

Initially it was different. Now she is also working. So we are more like partners. We decide everything together. (Husband)

Snacks. While we have classified snacks as one category, respondents talked about two sub categories; 1) the regularly purchased low priced ones and 2) the infrequently purchased high priced ones. The first category moved from solo-wife to wife dominated while the second category moved from solo-husband to husband dominated. In both the cases, each partner has given up a little bit of the autonomy in favor of the other.

Men’s Clothing. Husbands drew a distinction between purchases initiated by them and purchases made at others’ initiative (usually the wife). The latter type was perceived to be a giftCeven in cases where the spouse was not employed and therefore had no individual income. The shift reported in Figure 3 is for the first category where men appear to be taking more control of their wardrobe decisions.

Of course she comes along (when shopping for clothes) and I ask her, but you see it depends on whether I like it too. I know what goes in the office. So I will have to like it finally. So it is my decision. (Husband)

Entertainment. Decisions regarding movies were classified under entertainment. This decision shifted from joint to wife-dominated. In fact, in many cases, verbatim comments hinted even traces of wife-soloness among these decisions. The rising awareness and assertiveness among women did appear to be a probable cause. However, the dominant reason appeared to be the guilt in the husband that he was no longer spending adequate time with the family (due to rising pressures at work); husbands reported that therefore there was a tendency to spoil the family whenever they could.

I do not like Govinda (a popular Indian film actor), but I missed my promise to take them out. So last Thursday when we went to a movie we did go to a Govinda one. I just felt I had to do that. (Husband)

School for the Child. This factor came up in interviews where there were at least two children in the family. The contrast was between how the decision with regard to the first child differed from that about the second. In this case, the three decisions of what kind of school, at what age should the child go to school, and which parent would go and admit the child in the school were considered. Increasing demands on the husband’s time, learning curve and the growing literacy among wives appeared to contribute to this shift from husband-dominated to joint decision making.

Restaurants. These decisions appeared to move along the lines of entertainment and in some cases the reason also appeared to be a trend toward later marriages. The delayed marriages appeared to allow the women to develop their own preferences. This also appeared to make them more assertive and a credible source of information and experience when participating in these decisions.

Ours was a love marriage. When we went out initially, she chose a restaurant. We liked it there. We go there very often. . . . I would not have thought of that because I am not much of a regular at restaurants. (Husband)

Cooking Oil. The shift of cooking oil from solo-wife to wife-dominated appears to be largely driven by the extenive branding and marketing communication taking place in the product category. Though husbands mentioned incidents of experimenting with cooking and therefore being more involved with the product category, most of them did report assisting with cooking in the past too. In addition, the repeated mention of specific ads suggests a strong effect of marketing activity on husbands’ heightened involvement with the category.

It is like they show in the adI am not saying that something like that will happen, but you see, it is an issue of health. I know what is good in oil and what is not. (Husband)

Everyone talks about less fat and all that. It is important. All of us are prone to heart attack. So when I know what is the content of fat, I suggest to her. (Husband)

Toothpaste. This was a category that did not exhibit any shift, remaining the sole domain of wives as decision-makers with respect to brand, quantity, and timing. Verbatims also suggested very low involvement with the product category.

Other products that did not appear to show any shifts were coffee (wife-dominated) and women’s clothing (solo-wife). Bathing soap (solo-wife), while not exhibiting any shift in the grid, did change in the sense that some families mentioned more than one soap in the family now rather than eight years ago; respondents attributed additional income and the presence of children as the reasons. Decisions regarding visiting family/friends via trips (solo-husband) did not move on the grid but verbatims indicated a growing role of the wife in these decisions.

Overall, the more sharing of authority and less division of labor predicted for the American families (Foote 1954) appears to be coming true in this context too. While the data does not specifically indicate whether there is any specialization of task performance taking place, it can be argued that on the face of it, it does not appear to be so.

Reasons for Shifts in Decision Making

In order to identify the reasons for the shifts, respondents were asked to recall the factors that, according to them, could have caused them to operate differently over the two time periods. These explanations appeared to fall into six broad groups.

Urbanization. The recent past has witnessed a marked movement of the population into urban India. Some of the families that were interviewed had in fact relocated to an urban center within the last 8 years. This movement from semi-urban centers to urban centers appeared to have an influence on FDM behavior, in general, and decision role structure in particular.

The biggest impact appeared to be on the involvement of women in a way that was not common before. Women were more likely to be interested in learning English and being exposed to English media. This appeared to open up a new scheme of aspirations and knowledge to them. In addition, living in an urban setting also created time pressures on the spouse that was employed (husband in all the cases and wife too in some of the cases). It appeared that longer travel times and more work hours for the working spouse tended to place demands on and thereby expand the role of the other spouse.

We are now in a city. We have to be like these people here. Old ways do not work. My mother had a difficult time adjusting to this. But we have to change. I cannot sit at home with my wife like my father used to do with my mother in the village. (Husband)

Everybody is educated here (city). I did not want any help. So I attended these classes for English; sometimes I read the ads in the papers. (Wife)

We were transferred here four years back. Everything is fast here. So we have to run too . . . many times I do the shopping myself. (Wife)

Westernization. This appeared to be one of the stronger factors influencing the change. Exposure to the so-called "western" lifestyles through media and peers appeared to influence the way a family functioned. There appeared to be a strong urge to emulate western lifestyles.

Today many women go for shopping themselves. I mean even into big shops. It is like in foreign countries now. I feel I should not be left behind. (Wife)

I saw on the TV the other day that the husbands now listen to the wives and the wife runs the show. If you are not like that it means that you are old fashioned. . . . I am not saying we are like that (what is seen on TV) but who wants to be old fashioned these days? (Wife)

The concept of a western liberated and independent woman appeared to be very aspirational and intriguing to the wives interviewed. There appeared to be a lot of curiosity and therefore a desire to learn/emulate such lifestyles.

"Work hard, grow rich" Attitude. This appeared to be a critically important factor that shaped the way families evolved. This meant that in most cases the husband now spent very little time at home and in some cases neither of the spouses had enough time for each other. This appeared to influence the sharing of decision-making activities. It appeared to lead to more collaborative behavior than role specialization. The immediate fallout appeared to be the relinquishing of decision making power by men.

In many ways it is not like what it used to be. We have always been going out on Saturdays. But now I am not able to do that because there is so much work. I am also doing part-time (job). So when we skip because of me, I feel bad for her and I let her decide where to go and what to eat. It is her choice. (Husband)

For the (School for) first son it was OK. But for this boy (the second son) I could not go to school for his admission. She went alone. She is more confident now. (Husband)

Literacy of Women. Closely related to the other factors was that more of these women are now literate. This not only appeared to open up more information to them but also made them more assertive. Some of the issues mentioned by the respondents were exposure to media, independence while shopping (reading package labels), and exposure to new products.

That is the difference now. We are also educated and we know what is good and what is bad. Sometimes I decide better than he does (laughter). (Wife)

Women entering the Work Force. Closely related to urbanization, work hard grow rich attitude and literacy of women was the fact that more women entered the workforce. This appeared to have an impact in many ways on how a family now managed its decisions. Contribution to the family income, independent schedules, and separate set of friends, all seem to offer the wife a greater degree of autonomy. In addition, the job-related demands on her time seem to have brought the husband closer to several household chores and, as an extension, closer to several household decisions (e.g., produce and other categories not reported here).

I am now in the job. Since he (son) now goes to school, I have more time. Also in city one income is not sufficient. . . . I know more now. We (colleagues) discuss during lunch so we share information. In fact husbands do not discuss like that about shopping. So we women often know more and we suggest. (Wife)

If both are working, only Sundays are free and shops are closed on Sundays. So I usually shop on my way back (from work) and he picks me up if the shopping is heavy. (Wife)

He can saywhat he wants to say. But since I also have my salary, I can still buy what I want and he will only know after I come home (laughter). (Wife)

Later Marriages. This feature emerged among younger respondents. The fact that the marriages occurred at a later age (particularly for women) appeared to vest women with more time to explore the world on their own and therefore develop independent preferences. Women’s experience in the product category, then, resulted in a certain confidence in influencing the decision to some degree.

I have been to all these places (restaurants) before marriage too. So I suggest because I know. It will not be the same if got married soon after college. I will have to depend only on him. (Wife)

DISCUSSION

The results of this research provide important insights into the dynamic nature of family decision making over time. Whereas past research has investigated shifts in decision roles over one decision-making episode (Davis and Rigaux 1974), this research has uncovered evidence that decision role structure appears to change over many episodes, that is, over time. Across a diverse set of product categoriesCranging from snacks to furniture, to restaurants to decisions about the child’s schoolChusbands and wives perceived that their decision roles had shifted over an eight-year period. Moreover, husbands and wives felt that these shifts were related to social and cultural dynamics in India over this time period. Indian census figures are consistent with these perceptions (comparing 1991 to 1981): a greater percentage of Indian women participate in the work force (22.3 Vs. 19.7%); urban population accounts for a greater percentage of the population (25.7 Vs. 23.3%); and literacy among women is much greater (showing a 66% increase). The national Readership Survey (1995) also shows enormous growth in homes with TV’s (up 64% from 1990 to 1994) and cable access, particularly, homes in urban areas with cable access (approximately 60% of the urban population).

These shifts in decision roles need not be dramatic (e.g., solo-husband to solo-wife) to be of theoretical and practical importance. For example, though we have classified only the center of the grid in Figure 3 as joint decisions, the diagonal band that runs from the bottom left to the top right of the grid is made up of decisions more collaborative in nature than the decisions in the top left or bottom right. The movement of products into this band indicates a shift toward both partners participating in some aspects of decision making. The nature of this "participation" is of theoretical interest (e.g., under what conditions is there a division of labor, with an individual spouse specializing in some aspects of the decision, versus more collaborative participation) and practical importance (e.g., marketers directing advertising toward the appropriate decision maker). Further, although many of the shifts point to wives’ greater exposure to and involvement in arenas outside of the home (e.g., TV portrayal of "good houses" and western styles), we also observed an expanding role for husbands in some aspects of the household that had previously been the domain of the wives, with similar cultural forces playing a prominent role for both parties. For example, for men, greater exposure to advertising and media coverage about cooking oil enabled husbands to feel knowledgeable about decisions regarding this product category, reflected in a shift from solo-wife to wife-dominated. Interestingly, cerealCalso a basic food commodityCwas and continues to be perceived as a solo-wife decision perhaps due to the lack of brand-based advertising and media coverage of this product category.

Moreover, by uncovering some of the perceived socioeconomic factos affecting FDM in this context, we are better able to see the important relationship between micro-level psychological processes and macro-level socio-economic forces. These relationships are, of course, important to many consumer behavior contexts, although the impact of macro-level variables such as "culture" or "exposure to western media" on psychological processes have not been the subject of a great deal of theoretical consumer behavior research, perhaps because of the difficulty of observing or measuring macro-environmental forces. The impact of social and cultural changes were "observable" here because of the speed of change affecting India’s urban middle class in the recent past.

Our reconceptualization of "decision" to include not only what to buy but also the quantity and the timing of the purchase has provided additional insight into issues of influence in family decision making. Whereas past research has investigated husbands’ and wives’ influence in various stages of decision making, by investigating more than "what to buy" we have seen, by implication, that husbands and wives can be influential also in when and how much to buy. In sum, then, we have added to the literature by providing a more complete picture of how husbands’ and wives’ perceived roles and, by implication, their influence may change over time.

Directions for Future Research

While our research demonstrates that family decision role structure is dynamic and related to macro-environmental factors, there are some cautions appropriate to this study. It can not be said for certain that our respondents were able to recollect accurately and describe their decision-making practices eight years ago. Although we attempted to create a memory anchor for that period of their lives, which should facilitate recall, we can not rule out the possibility of somewhat inaccurate responses or an inability to verbalize those decision-making episodes fully. In addition, because we interviewed husbands and wives simultaneously, we can not rule out some adjustments toward social acceptability in the presence of the spouse and/or the interviewer.

It is also important to point out that macro-level factors are not the only viable "reasons why" decision role structure may change, in general, or may have changed for our respondents. For example, although decisions might have been made solely by one spouse at one time, the family’s brand preferences may have developed over time such that the other partner has learned and developed very effective heuristics for brand, quantity and timing decisions. Although this type of explanation could have been uncovered in our analysis, our primary interest was in understanding whether and in what way macro-environmental factors were related to role changes in FDM. Moreover, the macro-environmental variables are not necessarily unrelated to one another, where for example, the move to urban centers may lead to exposure to western media and ideas such as "work hard grow rich". We have isolated these factors for ease of discussion.

Also, the respondents may have moved across the stages of the family life cycle during these eight years. Such movements can have their own implications for decision role structure. However, the research design has attempted to control for that by focusing on those decisions in which the maximum involvement is that of husband and wife. The impact of the presence of children and their role in decision making has been reduced this way. Also, the second interviews attempted to detect possible reasons for changes in decision role structure. No major changes in decision role structure largely because of movement in family life cycle were detected but cannot be ruled out. Finally, while we asked respondents to recollect their past decision-making processes, an alternative approach would be to capture current decision making and, some years later, interview the same respondents about their decision processes. This complementary approach would provide additional insight nto the nature of shifts in family decision making over time.

Future research might also use our reconceptualization of decision role structure (particularly the redefined solo-wife and solo-husband categories) to more formally investigate the diversity of ways in which husbands and wives might influence FDM in earlier stages of the process. In addition, such research can also examine whether joint decision-making heuristics, reported by Park (1982), also change as decision role structures evolve.

REFERENCES

Belch, George E., Michael A. Belch, and Gayle Ceresino (1985), "Parental and Teenage Child. Influences in Family Decision Making," Journal of Business Research, 13 (April), 163-176.

Corfman, Kim P. (1987), "Group Decision Making and Relative Influence when Preferences Differ: A Conceptual Framework," in Jagdish N. Sheth ed., Research in Consumer Behavior, vol. 2, JAI Press.

Cunningham, Isabella C.M. and Robert T. Green (1974), "Purchasing Roles in the U.S. Family, 1955 and 1973," Journal of Marketing, 38 (October), 61-61.

Davis, Harry L. and Benny P. Rigaux (1974), "Perception of Marital Roles in Decision Process," Journal of Consumer Research, 1 (June), 51-62.

Davis, Harry L. (1976), "Decision Making within the Household," Journal of Consumer Research, 2 (March), 241-260.

Ferber, R. and Lee L.C. (1974), "Husband-Wife Influence in Family Purchasing Behavior," Journal of Consumer Research, 1 (June), 45-50.

Foxman, Ellen R., Patriya S. Tansuhaj, and Karin M. Ekstrom (1989), "Family Members’ Perceptions of Adolescents’ Influence in Family Decision Making," Journal of Consumer Research, 15 (March), 482-491.

Mick, David Glen and Michelle DeMoss (1990), "Self-Gifts: Phenomenological Insight from Four Contexts," Journal of Consumer Research, 17 (December), 322-332.

Park, C. Whan (1982), "Joint Decisions in Home Purchasing: A Muddling-Through Process," Journal of Consumer Research, 9 (September), 151-162.

Qualls, William J. (1987), "Household Decision Behavior: The Impact of Husbands’ and Wives’ Sex Role Orientation," Journal of Consumer Research, 14 (September), 264-279.

Roberts, Mary Lou and Lawrence H. Wortzel (1984), in Roberts, Mary Lou, and Lawrence H. Wortzel ed. Marketing to the Changing Household, Harper & Row.

Sharp, Harry and Paul Mott (1956), "Consumer Decisions in the Metropolitan Family," Journal of Marketing, 22 (October), 149-156.

Sheth, Jagdish N. (1974), "A Theory of Family Buying Decisions," in Sheth, Jagdish N., ed., Models of Buyer Behavior: Conceptual, Quantitative and Empirical. New York: Harper & Row.

Webster, Cynthia (1994), "Effects of Hispanic Ethnic Identification on Marital Roles in Purchase Decision Processes," Journal of Consumer Research, 21 (September), 319-331.

Wolfe, D. M. (1959), "Power and Authority in the Family," in D. Cartwright, Studies in Social Power. Ann Arbor: University of Michigan Press.

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