Working Wife Vs. Non-Working Wife Families: a Basis For Segmenting Grocery Markets?

Susan P. Douglas, Centre d'Enseignement Superieur des Affaires, Jouy-en-Josas, France
ABSTRACT - The growth in the number of working wife families is widely acknowledged to be one of the most important social trends of the era. Yet little attention has been paid to examining its implications for shopping behavior patterns and marketing strategies. The study outlined in this paper was intended to explore this issue, and in particular whether the wife's employment status is likely to provide a basis for segmenting markets for grocery products.
[ to cite ]:
Susan P. Douglas (1976) ,"Working Wife Vs. Non-Working Wife Families: a Basis For Segmenting Grocery Markets?", in NA - Advances in Consumer Research Volume 03, eds. Beverlee B. Anderson, Cincinnati, OH : Association for Consumer Research, Pages: 191-198.

Advances in Consumer Research Volume 3, 1976      Pages 191-198

WORKING WIFE VS. NON-WORKING WIFE FAMILIES: A BASIS FOR SEGMENTING GROCERY MARKETS?

Susan P. Douglas, Centre d'Enseignement Superieur des Affaires, Jouy-en-Josas, France

[Financial support for this project was provided by the Centre d'Enseignement Superieur des Affaires, Jouy-en-Josas, and the Marketing Science Institute, Cambridge, Mass. The author wishes to thank Professor Frank Carmone (University of Hong Kong), Professor Yoram Wind (University of Pennsylvania), and Professor Bernard Dubois (C.E.S.A., Jouy-en-Josas) for their assistance in this project, and MS Jane Ross (Marketing Science Institute) for performing the computer runs associated with the study.]

ABSTRACT -

The growth in the number of working wife families is widely acknowledged to be one of the most important social trends of the era. Yet little attention has been paid to examining its implications for shopping behavior patterns and marketing strategies. The study outlined in this paper was intended to explore this issue, and in particular whether the wife's employment status is likely to provide a basis for segmenting markets for grocery products.

INTRODUCTION

Over the past half century the proportion of working wife families has increased by leaps and bounds. Whereas in 1910, only one out of five wives worked, today the proportion is close to three out of five (Ferris, 1971). The consequences for family behavior and life-style patterns once the wife's main working hours are removed from the home are overwhelming in magnitude (Blood and Hamblin 1958, Hedges and Barnett 1972, Heer 1958, Linden 1973, Nye and Hoffman 1963). As the Presidential Commission on Manpower in 1972 suggested, "perhaps no other single change in family life has affected so many families in as relatively short a time as has the movement of married woman into the labor force" (U.S. Department of Labor 1972). Despite the importance of this social trend, scant attention appears to have been paid to examining its implications for marketing strategies. Relatively little interest seems, for example, to have been shown in assessing whether working wife families have specific needs and interests which provide opportunities for developing new products and services, or for emphasizing certain benefits and appeals.

There are two major reasons for thinking that this neglect of the working wife families is unwarranted. In the first place, time constraints on the performance of household duties such as shopping, cooking, cleaning, and in particular on the time at which these tasks are performed in working wife families, suggest that different strategies are likely to be developed to cope with these problems. One hypothesis is that working wives will show greater interest than non-working wives in time-saving products and services such as take-out dinners, laundry services, and instant dusting spray (Management Review 1967).

Secondly, differences in attitudes among working and non-working wives towards female roles are likely to influence their behavior. The dual roles of the wife in her employment and in the home, imply that the degree of involvement in various "homemaker" roles such as "Chief Cook", "Mother", "Housekeeper", will differ from the non-working wife for whom such roles are the focal point of her existence (Hartley 1960). This does not necessarily imply rejection of homemaker roles by the working wife, but rather that she seeks other opportunities for self-fulfillment in activities outside the home (Bailyn 1970, Powell 1960). Consequently, the time and effort devoted to homemaker roles, as well as concepts of appropriate behavior in these roles, may differ from those of the non-working wife. Stimulation received from involvement in a job may also generate greater interest in new and different products, and emphasis on different product benefits; for example, more concern with originality, or less concern with the preference of family members. Nonetheless, working wives, and equally non-working wives, may not all have similar attitudes toward homemaking. Motivations for working or for not working differ (Hoffman 1963b). Some women work primarily from financial necessity and remain predominantly involved in homemaker roles. Equally some non-working women solve frustrations with the domestic routine by involvement in non-paid social and charitable activities rather than paid employment. Consequently, they are open to an equally wide sphere of contacts and relationships as working women, and may have as little time or energy to devote to household duties.

Scant attention has been paid to examining the impact of the wife's employment status on purchase behavior. Some comparisons have been made of purchase behavior in working and non-working wife families. These have found, for example, that working wives tend to be more efficient in organizing household tasks; they make fewer shopping trips, and are more likely to be accompanied by their husbands than non-working wives (Anderson 1972, Baldwin and Lunn 1972, Hoffman 1963a, Lunn, Baldwin and Dickens 1972). On the other hand, there appears to be little evidence to indicate that working wives are more interested in convenience products than non-working wives (Anderson 1972).

Such studies barely scratch the surface of the working wife phenomenon. Typically, they compare working and non-working wife families without making a distinction between part-time and full-time working wives, nor taking into account the interaction between employment status and other factors such as family income, age, number of young children. Nonetheless, these are likely to have an important bearing on the wife's time availability and interest in convenience products. If, for example, non-working wives have several young children at home, or are heavily involved in activities outside the home, they are likely to have equivalent or greater time pressures to part-time working wives; equally, the older non-working wives may have greater discretionary income to spend on convenience products and services than young working wives. In addition, little research has been undertaken to exam/ne differences in attitudes among working and non-working wives, and the extent to which these influence their purchase behavior.

The purpose of the exploratory study described in this paper was to investigate the effect of the wife's employment status on grocery purchasing behavior, taking into consideration differences in other socioeconomic and demographic characteristics among working and non-working wives and in attitudes toward various women's roles. The two major research questions were as follows:

1. Do working wives appear to have substantially different shopping behavior patterns from non-working wives likely to provide an adequate basis for segmenting grocery markets? Or are observed differences merely a reflection of differences with regard to other factors such as education, income, or the number of children at home?

2. Are there important attitudinal differences within the two groups which lead to different purchase behavior patterns, and suggest that a finer level of analysis based on sub-segments within each group is likely to be more relevant for management purposes?

First the data base and the research design for the study are briefly presented, followed by a discussion of the research findings relating to the two key issues. Some issues in the investigation of the impact of the wife's employment status on purchase behavior are then examined and guidelines for further research proposed.

RESEARCH METHODOLOGY

Data Base

In the study a questionnaire was administered by personal interview to 106 married women in the Philadelphia and Boston areas. All respondents had been married at least two years and were between the ages of 25 and 50. Approximately half the sample (49) was not gainfully employed outside the home, while the other half (57) was composed of women who had been employed full-time for at least eighteen months. Their jobs included a variety of occupations such as lawyers, doctors, researchers, secretaries, waitresses and sales clerks. The sample was selected by a quota procedure and both the working and non-working groups were stratified by age, income and number of children. The objective was thus to obtain two groups "matched" in terms of background characteristics, rather than two groups which were representative of the working and non-working wife populations.

The structure of the sample and its small size clearly limits the extent to which any substantive conclusions can be drawn concerning differences in purchase behavior of working and non-working wife families, particularly insofar as differences within the two groups are concerned. The purpose of the study was not, however, to compare these differences, but rather to isolate the impact of the wife's employment status from that of other related variables, such as income, stage in family life-cycle, number of young children at home. Consequently, given research budget constraints, the alternative of small matched samples controlling for these variables was preferred to that of larger but less comparable .samples.

In the questionnaire, respondents were asked about grocery purchase behavior; attitudes to female roles; and background characteristics of the family. The grocery purchase questions examined 1) interest in various product benefits when purchasing grocery items; 2) frequency of purchasing selected convenience products and services; and 3) the organization of shopping activities, for example, frequent, of shopping in different types of stores and husband participation in shopping activities. Attitudes toward four aspects of a woman's role were examined: a) home and family orientation, b) cooking, shopping and other household tasks, c) self-perceptions and concepts, and d) social interaction. The background questions included a number describing the home, such as the type of household appliances owned, as well as the standard family background questions relating to income, age, and education of husband and wife, etc.

Data Analysis

Following the two major research questions the analysis was divided into two parts (Figure 1). First, overall differences in grocery purchase behavior between the two groups were examined. Then, differences in attitudes towards female roles within each group were investigated, together with the purchasing behavior patterns of different attitudinal sub-groups.

Overall differences between working and non-working women. After a preliminary series of Chi square tests of all three sets of grocery purchase variables, two, use of convenience products and grocery shopping, were submitted to a multivariate analysis of covariance (BMD 12V, Green and Tull 1975) to test whether apparent differences between working and non-working wives were significant after adjustment for differences in 11 socio-economic characteristics of the two groups (the covariates).

Within group analysis. In the second stage of the analysis, sixteen key attitudinal variables (four from each of the four sets of attitudes) were identified by means of a series of factor analyses [First a separate factor analysis for each of the four sets of additudinal characteristics was conducted (BMD 08M). The variable with the highest loading on the first four factors in each factor analysis was then retained for the cluster analysis. The total number of factors identified precluded the use of all the factors, but the first four in each case accounted for between 60.8 - 80.2% of the total variance. For further discussion of the rationale for this procedure see Susan Douglas and Patrick Le Maire (1974). The factor analyses were conducted jointly for working and non-working groups after preliminary analysis based on separate factor analyses resulted in the identification of highly similar underlying structures suggesting that the joint analysis could appropriately be used to summarize the data for both groups.] (Table 1). Cluster analyses of these variables using the Howard Harris clustering program (Howard and Harris 1966) were then conducted separately for the working and non-working wife groups; two major sub-groups were identified among working wives and three among non-working wives. Differences in grocery purchase behavior of these groups were then examined by means of discriminant analyses.

FIGURE 1

FLOW CHART OF ANALYSIS

TABLE 1

THE 16 KEY ATTITUDINAL STATEMENTS

RESULTS

Differences Between Working and Non-Working Wives

Some differences emerged in grocery purchasing behavior of the working and the non-working wives but these were not always along the lines hypothesized.

Grocery shopping patterns. As Table 2 indicates there are some slight differences between the two groups in the way shopping activities are organized; particularly insofar as the working wives go less frequently to neighborhood stores, (possibly implying fewer midweek trips), and tend to make somewhat more use of husbands. These differences were substantially more marked in the unadjusted means, and tend to be affected by the wife's age and the socio-economic status of the family.

Use of convenience products and services. Here the results confirm those of other studies insofar as the working wives do not have any systematic tendency to use time-saving products and services more frequently than the non-working wives. In fact the latter appear to be heavier users of take-out dinners and several other products such as instant dusting spray and baked goods.

Reasons for buying grocery products. The working and non-working wives also appear to consider similar factors important when buying grocery products for regular meals. The only significant difference occurs in relation to "something different" which, contrary to expectations, is more important to the non-working than working wives (Table 3). Reasons for buying when entertaining company are also similar, but are not reported here.

TABLE 2

RESULTS OF THE MULTIVARIATE ANALYSES OF COVARIANCE FOR GROCERY SHOPPING AND USE OF CONVENIENCE PRODUCTS AND SERVICES (ADJUSTED MEANS AND APPROXIMATE F STATISTICS FOR THE MULTIVARIATE ANALYSIS)

[A five point scale was used to rate frequency of shopping and usage of convenience products; the higher the mean, the more frequent the activity or usage.]

TABLE 3

A COMPARISON OF REASONS FOR SELECTING GROCERY PRODUCTS AND OF OWNERSHIP OF HOUSEHOLD APPLIANCES FOR WORKING AND NON-WORKING WIVES

Household equipment and domestic help. There is also little indication that the working wives rely more on the use of domestic help or on mechanical labor saving devices to aid in their cooking and housekeeping chores. On the contrary, the non-working wives appear to be better equipped than their working counterparts, especially with washing machines and dryers, freezers, electric carving knives and self-defrosting refrigerators (Table 3). Perhaps, since non-working women spend mote time in the home, they attach more importance to possessing these appliances than do working women.

Although the small sample size precludes any definitive conclusions, the wife's employment status appears to affect the way in which shopping activities are organized but not the time devoted to meal preparation and other household tasks. Time pressure problems appear to be solved either by a sharing of tasks with other family members or simply by the wife's absorption of the increased work load rather than by recourse to outside sources of help and shifting of the burden to the commercial sector.

Consequently, families in which the wife is employed are unlikely to be prime targets for time-saving convenience products and services or pioneers in the use of such products. Wives who do not work appear to be equally interested in such items, possibly because they too are pressured for time due to involvement in social or neighborhood activities. On the other hand, since the working wives appear to seek solutions for saving time spent shopping, emphasis by retailers on speed of service and home delivery services could prove an attractive selling point for them.

The increased participation of husbands in working wife families in grocery shopping and particularly in assuming the major responsibility for these tasks, also suggests that the relevant decision-making unit and hence appropriate promotional target is different when the wife is employed. While in non-working wife families, focus predominantly on the wife is likely to be appropriate, in working wife families inclusion of both husband and wife appears desirable. This does, however, depend on the extent to which the husband influences product and brand decisions. Examination of this aspect is therefore needed before any definitive conclusions can be reached.

Differences Among Working and Non-Working Wife Sub-Groups

In the second phase of the analysis attention was centered on examining differences in attitudes towards various female roles and related purchase behavior patterns within the two groups. Again the small size of the groups identified implies that findings should be viewed as exploratory, and that considerable further investigation and validation of apparent differences is required.

Working wives. Among the working wives two highly distinct groups were identified; their profiles are shown in Table 4. The "progressive egalitarian" group was characterized predominantly by their liberated attitude toward a woman's role. Although not house-proud, they enjoyed cooking, but were less innovative and fashion-conscious than the other group. The second group of "fashionable traditionalists" had traditional attitudes toward female roles, and were houseproud. While interested in fashion they tended to feel some-. what guilty about the amount they spent on clothes.

TABLE 4

KEY DISCRIMINATING ATTITUDINAL VARIABLES FOR THE WORKING WOMEN SUB-GROUPS: (MEANS ON 6 POINT LIKERT SCALE)

A comparison of the grocery purchase behavior of these two groups revealed some differences (Table 5), though not necessarily what one might expect from their attitudes. The "liberated" views of the "egalitarians" suggest that they may attach less importance to being a "good" housewife than other working wives and hence devote less effort to housekeeping activities. On the contrary, however, they appear to make significantly less use of all types of convenience products and services, and to shop more frequently in the time-consuming farmers' markets than the 'traditionalists''. Perhaps this results from their greater concern with economy.

Some partial explanation for these patterns can be found in the socio-economic characteristics of the two groups. Members of the "progressive" group are generally better educated and have higher status jobs than the "traditionalist" group (Table 6). Consequently they may have higher achievement levels and attach greater importance to performing effectively in both domestic and career roles. The "traditionalists", on the other hand, may work predominantly for financial reasons, and feel little need to compensate for the time spent away from home by devoting effort to meal preparation and other household chores.

Non-working wives. The profiles of the three attitudinal sub-groups identified among non-working wives were not as distinctive as those of the working wife groups, but again differed primarily in attitudes to a woman's role and interest in clothes (Table 7). One group - "the disorganized homemaker" - was characterized primarily by her tendency always to feel rushed. She also had somewhat conservative role perceptions and showed little interest in clothes or food products. The "liberated socializer", on the other hand, rejected the traditional view of a woman's role and enjoyed reading and going to parties but tended to have guilt feelings about her expenditures on clothes. The "conservative homebody" had highly traditional role perceptions but was outgoing and interested in fashion and new food products.

Examination of the grocery shopping behavior of the three groups showed some differences, particularly in the frequency of shopping, and husband participation in shopping, but the patterns were not as clear-cut as for the working wife groups (Table 8).

Although further investigation and validation of both the working and non-working sub-groups is clearly necessary, attitudes toward a woman's role emerge as an important factor differentiating among women in both groups. Attention might therefore usefully be focused on further investigating the role of such factors and their relationship to interest in different types of products as well as other aspects of grocery shopping behavior among working and non-working wives.

TABLE 5

KEY DISCRIMINATING GROCERY SHOPPING VARIABLES FOR TWO WORKING WOMEN ATTITUDINAL GROUPS

[Variables for which the means are similar are omitted.]

TABLE 6

DIFFERENCES IN OCCUPATIONAL STATUS OF THE TWO WORKING WIFE GROUPS

TABLE 7

KEY DISCRIMINATING ATTITUDINAL VARIABLES FOR THE THREE NON-WORKING WIFE GROUPS: (MEANS ON 6 PT. LIKERT SCALE)

TABLE 8

KEY DISCRIMINATING GROCERY PURCHASE VARIABLES FOR THE THREE NON-WORKING GROUPS

[Variables for which the means are similar are omitted.]

CONCLUSIONS

Although at this stage it would clearly be premature to draw any substantive conclusions, the findings suggest a number of guidelines for future research.

In the first place, the results to a large extent confirm those of previous studies which suggest that working wives do not differ significantly from non-working wives in the use of convenience products and services, nor in interest in different product benefits when buying grocery products. This is particularly interesting insofar as the hours worked by the wife, her age and number of young children were controlled for, and hence the lack of differences between working and non-working wife families cannot be attributed to such factors. On the other hand, certain differences emerge in the organization of shopping activities in working and non-working wife families which could usefully be further investigated, particularly in relation to the impact of the wife's employment on husband involvement in shopping activities.

Secondly, focus on different sub-groups among working and non-working wives appears to be a promising line for future research. In addition to examining attitudinal differences, the impact of factors such as occupation and education level on purchase behavior needs to be considered particularly in the case of working wives. In this respect investigation of differences in the purchasing behavior of women in different status jobs, and with different work motivations, for example, between professional "career" women and women with lower status jobs merits attention.

A third issue concerns the relationship between attitudes toward female roles and purchase behavior. While groups with different attitudes appear to be characterized by different purchase behavior, the nature of the relationship between attitudes and behavior is not clear. Intuitively, one would expect traditional female role perceptions to lead to a rejection of convenience products and services, yet the reverse relationship occurs both among working and among non-working wives. Possibly home-oriented women have lower levels of self-achievement, and are more ready to accept the easy solution. Further exploration of this issue is also required.

In many respects the study raises more questions than it solves. Yet it helps to pinpoint some key issues, and to demonstrate the complexity of the working wife phenomenon and its implications for shopping behavior. The fact that a woman is involved in employment outside the home does not necessarily imply that she will devote less effort to homemaking activities than her non-working counterpart; conversely, a non-working wife is not necessarily highly committed to her domestic role. Nonetheless involvement in a full-time job imposes time constraints on the performance of household duties which have consequences for purchase behavior. In addition although involvement in a job is associated with different attitudes to homemaker roles, these vary among different categories of working wives; and the precise impact on purchase behavior remains unclear.

Further clarification of these issues should be a major concern at this time. As the women's liberation movement gains more widely based support, the invasion by women of upper status jobs is likely to continue. In addition, the shift of married women into the labor force shows no signs of abating. Mastering the implications of these major social forces is thus likely to be critical to understanding future developments in grocery shopping patterns.

REFERENCES

Beverlee Anderson, "Working vs. Non-Working Women: A Comparison or Shopping Behavior", Proceedings of the A.M.A. Fall Conference, (Chicago: A.M.A. 1972).

E. S. Baldwin and J. A. Lunn, "Researching the Changing Consumer", Proceedings of the ESOMAR Congress (Cannes: ESOMAR 1972).

Lotte Bailyn, "Career and Family Orientations of Husbands and Wives in Relation to Marital Happiness", Human Relations, 23 (1970), pp. 97-114.

Robert O. Blood and Robert L. Hamblin, "The Effect of the Wife's Employment on Family Power Structure", Social Forces, 36 (may 1958), pp. 347-352.

Robert O. Blood and Donald Wolfe, Husbands and Wives (New York: The Free Press, 1960).

Susan Douglas and Patrick Le Maire, "Improving the Quality and Efficiency of Life-Style Research", Proceedings of the XXIII ESOMAR Congress (Hamburg: ESOMAR 1974).

Abbott L. Ferris, Indicators of Trends in the Social Status of American Women (New York: Russell Sage, 1971).

Paul E. Green and Donald Tull, Research for Marketing Decisions, third edition, (Englewood Cliffs, N.J.: Prentice Hall, in press).

Ruth E. Hartley, "Some Implications of Current Changes in Sex and Role Patterns", Merrill-Palmer Quarterly, April 1960), pp. 153-164.

Janice Hedges and Jeanne Barnett, "Working Women and the Division of Household Tasks", Monthly Labor Review, 95 (April 1972), pp. 9-14.

David Heer, "Dominance and the Working Wife", Social Forces, 3b (May 1958), pp. 341-347.

Lois W. Hoffman, (a; "Parental Power Relations and the Division of Household Tasks"; (b) "The Decision to Work", in F. Ivan Nye and Lois Hoffman's The Employed Mother in America, (Chicago: Rand McNally 1963).

H. Howard and Britt Harris, "A Hierarchical Grouping Routine", IBM J60/65 Fortran Program, University of Pennsylvania Computer Center, 1966.

Fabian Linden, Women: A Demographic, Social and Economic Presentation, (New York: Conference Board 1973).

Tony Lunn, Sally Baldwin and Jackie Dickens, "Monitoring Consumer Life-Styles", Admap 8, (November 1973).

F. Ivan Nye and Lois Hoffman, The Employed Mother in America (Chicago: Rand McNally, 1963).

Kathryn S. Powell, "Family Variables", Marriage and Family Living, 22 (February 1960), p. 22-35.

U.S. Dept. of Labor, Report of the President's Commission on Man-Power, 1972.

"Working Waves", Management Review, (April 1967), pp. 4-21.

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