Gender Differences and Household Decision-Making: Needed Conceptual and Methodological Developments

Mary Lou Roberts, Boston University
ABSTRACT - The papers in this session all have a common thread gender differences. The Qualls paper represents an uncommon thread -- household decision-making. If I tried to force them all into a single mold, we would lose the opportunity to compare and contrast these overlapping, but in many ways quite different streams of research.
[ to cite ]:
Mary Lou Roberts (1984) ,"Gender Differences and Household Decision-Making: Needed Conceptual and Methodological Developments", in NA - Advances in Consumer Research Volume 11, eds. Thomas C. Kinnear, Provo, UT : Association for Consumer Research, Pages: 276-278.

Advances in Consumer Research Volume 11, 1984      Pages 276-278

GENDER DIFFERENCES AND HOUSEHOLD DECISION-MAKING: NEEDED CONCEPTUAL AND METHODOLOGICAL DEVELOPMENTS

Mary Lou Roberts, Boston University

ABSTRACT -

The papers in this session all have a common thread gender differences. The Qualls paper represents an uncommon thread -- household decision-making. If I tried to force them all into a single mold, we would lose the opportunity to compare and contrast these overlapping, but in many ways quite different streams of research.

Consequently, I will first look rather briefly at each of the three papers individually. Then I will step back and take a broader view of the two areas of research, presenting my own perspectives on what we should be doing in terms of both conceptual and methodological developments. in each field.

THE GENDER DIFFERENCE PAPERS

There are some general comments about the Gentry and Haley and the Zikmund et al. papers.

I do not quarrel with the student sample in either instance. If you agree with Calder, Phillips, and Tybout (1981) that rigorous theory falsification research requires a homogeneous sample, then a student sample is acceptable as long as the stimuli are appropriate. In both cases, the stimuli were chosen for and pre-tested on students. At the same time, we must remember that this seriously limits the generalizability of the results

The weakness of the results is disappointing, however, especially in vew of the sound theoretical bases and careful choice of methodology in both instances. I will return to that issue later because it opens up some interesting speculation.

Although the stimuli were carefully developed, I can offer alternate hypotheses for at least some of the results based on the nature of the stimuli. Gentry and Haley's findings of greater recall of the blusher ad by females and greater ease of recall of the football ad for masculine schematics could be a function of involvement with the product category. Likewise, I could explain away Zikmund et al.'s insignificant finding relative to tire tread by hypothesizing that no one could be very involved in a choice between tire treads'

There is one aspect of Zikmund et al. which does bother me. It is not their research per se, but a continuing difficulty with the whole issue of informal communications, at least in the context of consumer products marketing.

Even if a marketing manager has incontrovertible evidence that, say, females conform more to peer pressure than do males, what does the manager do about it? I see so few instances where the marketer can influence the nature or the amount of informal communications that I am not enthusiastic about research in this field, even though it has a sound basis in theory. There may, however, be other approaches which could provide actionable findings -- more on that later.

THE HOUSEHOLD DECISION-MAKING PAPER

Qualls' paper uses a sample of "real people" - in this case a necessity because the decision being studied is complex and infrequently made. It is a commendable attempt to look at process, not just outcome.

Happily, the results make sense. Sex-Role Orientation was found to be associated with perception of spousal influence, level of preference agreement and mode of conflict resolution but not preference pattern or decision outcome. It seems intuitively reasonable that SRO would be related to processual variables but not to matters of individual taste or to the jointly-made final decision

One empirical finding did surprise me, however. In both sex-role modern and sex-role traditional couples, the absolute number of husband concessions was greater than that of wife concessions. That was counter-intuitive for sex-role traditional couples. [Qualls later advanced as an explanation for this finding a real estage adage that "the husband sells the house; the wife buys a house." A price range is established by the couple before actual house-hunting, but on most other sub-decisions husbands have a tendency to give in to their wives' preferences.]

RESEARCH INTO PSYCHOLOGICAL GENDER DIFFERENCES

Conceptual Approaches

Let me approach this whole area of research by taking an extreme position: I recommend that we terminate this entire stream of research on the grounds that

1) it has never been very productive

2) it will be even less productive in the future.

I have not seen a major review of the social psychological research in this field since Maccoby and Jacklin's (1974) book was published. Parenthetically, there has never been a major review of this aspect of research into consumer behavior; I suspect that such a review would reach a conclusion similar to that of Maccoby and Jacklin. They state quite strongly that there are few psychological gender differences that are empirically supported. In fact, their analysis concludes that hypothesized differences are often much smaller than the researchers themselves have argued (Maccoby and Jacklin, 1974, pp.3-8).

Having accepted their argument, I would then go a step further. Even where psychological gender differences have been found to exist, they are likely to disappear or at least decrease in the future. Social phenomena including, but not limited to, the increasing education of women and their labor force participation mean that many women now have an experiential background which was previously unique to men. Likewise, the fact that men are increasingly taking on child care and other household tasks provides them with a set of experiences which heretofore have been the sole province of women.

To the extent that psychological gender differences are not based on physiological differences - and there is little evidence that they are (Bernstein, Bank and Jarvik, 1980) - the increasing commonality of social experience for men and women suggests a decrease in psychological gender differences.

I would therefore argue that the entire subject of gender differences is relatively unimportant. However, I doubt that researchers interested in this area will accept the finality of my statements. Consequently, I will suggest two approaches that I think might be productive.

I agree with Gentry and Haley that studying within- gender differences may prove more worthwhile than studying between-gender differences. "Segments" of females have been studied often; the s Tot consumer rarely been the focus of careful scrutiny.

Somewhere between a "macro" view - all females compared with all males -- and only looking at within-gender differences is a middle ground which seems worth considering. It should be worthwhile to compare subgroups of males and females that are "matched" on some managerially or theoretically-relevant variable.

For example, I am working with a data base which contains responses from both husband and wife in 639 dual-income families. We asked both spouses whether they were primarily oriented toward their careers, toward their homes and families, or toward their careers and homes and families equally. It was no surprise when we found that over 50 percent of the wives said they were primarily oriented toward their homes and families. It did surprise us, however, that 30 percent of the husbands said the same thing (Roberts, Kirshbaum and Cooper, forthcoming). In subsequent analysis, I will be able to compare these "home/family" oriented females and males in terms of grocery shopping attitudes and behavior and 03 other types of household task performance. I believe there will be much richer insights from this type of analysis than from the more traditional "all males/all females" type of comparison.

Secondly, I would suggest that we look for gender effects in dyadic or other small group relationships where results can be managerially actionable. Little research exists in a business or industry setting that deals with female salesperson/male prospect or male salesperson/female prospect dyads.

The literature in organizational behavior suggests that there is little difference in attitudes or behaviors between males and females in the same type of position or the same profession (see, for example, Harlan and Weiss, 1980). Consequently, we may find little differences in either process or outcome being caused by the insertion of a female into a previously all-male dyad or group.

Since many male managers seem to cling to the belief that their female counterparts are somehow "different," this is an area in which support for the null hypothesis of "no difference" would be important. If differences are found, managers could adjust communications strategies, either sales strategies or internal communications strategies, to deal with them.

Methodological Approaches

If you believe, as I do, that psychological gender differences are minimal, then the weakness of our findings is not the result of poor methodology but is merely reflective of reality. Even so, the subject of measuring instruments is worth discussion.

These papers used several different scales, each of which has advantages and disadvantages.

The Osmond and Martin scale and the Spence, Helmrich and Stapp Personal Attributes Questionnaire have been used frequently in behavioral sciences research and less frequently, but with good results, in consumer behavior research. The Bem androgyny scale, however, is a different matter. Many of us had high hopes for the androgyny concept, which has stimulated much interest and controversy among behavioral sciences researchers. Results to date, however, including the present study by Gentry and Haley, have been disappointing (Gentry and Doering, 1977; Martin and Roberts. 1983).

It is important to recognize that none of these scales is unidimensional. Both Osmond and Martin (1975) and Spence and Helmrich (1978) factor analyzed their instruments. The resulting factors can either be used separately or the factor scores can be summed for each respondent to given equal weight to each of the underlying constructs.

The initial controversy over scoring of the Bem androgyny scale has been resolved. It should be scored into four categories as Gentry and Haley did, not three as Bem originally proposed. Failure to recognize that failure to recognize that this scale is multidimensional could be a reason for the weak results it has produced in consumer behavior studies. However, in a paper presented last year at ACR, Martin and I found that, while our factor analysis of the Bem scale replicated those of other researchers (Whetton and Swindells, 1977) quite well, our results were not greatly improved by treating it as a multi-dimensional construct rather than as a uni-dimensional one.

Of all these measuring instruments the Markus approach to scoring the BSRI (1982), especially when combined with a latency measure, may hold the greatest promise. It will be interesting to see it used in future studies.

RESEARCH INTO HOUSEHOLD DECISION-MAKING

While Qualls' paper represents a distinct advance in our approach to household decision-making research, it also highlights additional developments, both conceptual and methodological, which are needed.

Conceptual

The limited conceptual model which Qualls uses is indicative of the fact that we have no satisfactory model of household decision making. Sheth's (1974) is the only general model which exists. Its inclusion of conflict is useful, but otherwise it is static and not easily operationalized. An over-riding need in this whole area of research is for a commonly-accepted model which can accommodate both various types of household units and changes in decision-making processes over the household life cycle. Such a model would encourage researchers to use a common framework for their research and would help to eliminate some of the semantic inconsistencies in the extant research. This would help to bring order to what now tends to be extremely fragmented research efforts.

This concern over lack of a common model also points to one trend which is beginning to occur and which must be encouraged. Researchers are beginning, conceptually at least, to take note of the dramatic changes which have taken place and are continuing to take place in household composition. Murphy and Staples' Modernized Family Life Cycle (1979) and Gilly and Ennis's further extension (1981) push us in that critical direction.

Yet, I doubt that the last word has been said conceptually on the subject of changing household composition. Additionally, the conceptual advances have not yet been reflected in most of the empirical research.

Other conceptual developments are being put forward in the area of household decision-making. A number of them, which I cannot cover in this paper, are discussed in a forthcoming book on the changing household (Roberts and Wortzel, forthcoming).

Methodological

This forthcoming book also highlights a number of needed methodological developments. In particular, it stresses the need for treating the household as a decision-making unit as opposed to merely conducting analyses at the individual level. Much more work is needed in this area.

For the purpose of this discussion, however, I would like to return to an issue which was at least implicit in my critique of the measuring instruments used in these three studies.

Just as we have no commonly-accepted model of household decision-making processes, we lack commonly-accepted scales for measuring key variables, especially sex-role orientation. We simply borrow from the behavioral sciences. The borrowing appears to be based on personal preference or convenience or both.

All the previous comments about lack of comparability of our research efforts apply here also. The development of measuring instruments which are deemed worthy of consistent use by researchers in this field would make studies much more comparable. This seems especially critical for sex-role orientation. The scales we are now using may do a good job of measuring social attitudes, but these attitudes do not seem to represent the critical aspects of sex-role orientation as they relate to household decision-making and consumption processes.

CONCLUSION

Let me simply summarize these comments by saying that I am less sanguine about the likelihood of truly meaningful results, either theoretical or empirical, emerging from research into gender differences than I am about the developments that await us in household decision-making research. The latter field offers the prospect of genuinely worthwhile developments which can have both social and managerial significance.

Whichever area each of us chooses to pursue in our own individual research, I urge our thoughtful attention to conceptual and methodological problems in addition to concern for both theoretical and managerial relevance. This will bring important research initiatives into the preeminent position they deserve in consumer behavior research.

REFERENCES

Bernstein, Bennie, Lew Bank, and Lissy F. Javik (1980), 'Sex Differences in Cognitive Functioning, Evidence, Determinants, Implications," Human Development, 23, 289-313.

Caulder, Bobby J., Lynn W. Phillips, and Alice M. Tybout (1981), "Designing Research for Application," Journal of Consumer Research, September, 197-207.

Gentry, James W., and Mildren Doering (1977), "Masculinity-Femininity Related to Consumer Choice, AMA Educators' Conference Proceedings, ed. by Barnett A. Greenberg and Danny N. Bellinger, Chicago. 423-427.

Gilly, Mary C. and Ben M. Enis (1982), "Recycling the Family Life Cycle: A Proposal for Redefinition," Advances in Consumer Research, Vol. IX, ed. by Andrew Mitchell, Ann Arbor, 271-276.

Harlan, Anne and Carol Weiss (1980), "Sex Differences in Factors Affecting Managerial Career Advancement," Working Paper No. 56, Center for Research on Women, Wellesley College, Wellesley, MA.

Markus, Hazel, Moire Crane, Stan Bernstein, and Michael Siladi (1982), "Self-Schema and Gender," Journal of Personality and Social Psychology," 42, 38-50.

Martin, John, and Mary Lou Roberts (1983), "Effect of Sex of Owner and personal Circumstances on Attitudes Toward a Service Establishment," Advances in Consumer Research, Vol. X, ed. by Richard P. Bagozzi and Alice M. Tybout, Ann Arbor, 339-344.

Murphy, Patrick E., and William A. Staples (1979), "A Modernized Family Life Cycle," Journal of Consumer Research, June, 12-22.

Osmond, Marie W., and Patricia Y. Martin (1975), "Sex and Sexism: A Comparison of Male and Female Sex-Role Attitudes," Journal of Marriage and the Family, November, 744-758.

Spence, Janet L . and Robert L. Helmrich (1978), Masculinity and Femininity, Austin, University of Texas Press.

Roberts, Mary Lou, Lawrence J. Kirshbaum, and Linda R. Cooper, "How Two-Income FamiLies Live and Work," American Demographics (forthcoming).

Roberts, Mary Lou, and Lawrence H. Wortzel, eds. (forthcoming), Marketing to the Changing Household: Management and Research Perspectives, Cambridge, MA: Ballinger Publishing Company.

Whetton, C. and T. Swindells (1977), "A Factor Analysis of the Bem Sex Role Inventory," Journal of Clinical Psychology, 33, 150-153.

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