Applying Age-Gender Theory From Social Gerontology to Understand the Consumer Well-Being of the Elderly

Elaine Sherman, Hofstra University
Leon G. Schiffman, Baruch College - CUNY
ABSTRACT - This paper explores two competing social gerontological explanations for shifts in subjective well-being that occur with advancing age. One of the two viewpoints, leveling, suggests that as men age, their life situation deteriorates at a more rapid rate than is the case for aging women. In contrast, double jeopardy proposes that the greater social inequalities experienced by women worsen with age. Suggestions as to possible consumer behavior applications are offered.
[ to cite ]:
Elaine Sherman and Leon G. Schiffman (1984) ,"Applying Age-Gender Theory From Social Gerontology to Understand the Consumer Well-Being of the Elderly", in NA - Advances in Consumer Research Volume 11, eds. Thomas C. Kinnear, Provo, UT : Association for Consumer Research, Pages: 569-573.

Advances in Consumer Research Volume 11, 1984      Pages 569-573


Elaine Sherman, Hofstra University

Leon G. Schiffman, Baruch College - CUNY


This paper explores two competing social gerontological explanations for shifts in subjective well-being that occur with advancing age. One of the two viewpoints, leveling, suggests that as men age, their life situation deteriorates at a more rapid rate than is the case for aging women. In contrast, double jeopardy proposes that the greater social inequalities experienced by women worsen with age. Suggestions as to possible consumer behavior applications are offered.


In consumer and marketing research there is a long tradition of using demographic variables to understand and profile consumer segments (Roscoe, et al., 1977). However, consumer researchers have mate little progress in developing conceptual frameworks for selecting ant/or combining common demographic variables (Roscoe, et al., 1977). Possible exceptions are some of the research dealing with alternative measures of social class (e.g., Jain, 1975; Shimp & Yokum, 1981) and alternative measures of the family life cycle (e.g., Derrick & Lehfeld, 1980; Murphy & Staples, 1979).

In keeping with these exceptions, the purpose of the present paper is to describe and offer relevant suggestions about a small stream of social gerontological research that focuses on age-gender differences. It is felt that this research has the capacity to serve as a frame of reference for conducting research on the behavior and attitudes of the elderly as consumers. Equally important, the borrowed research provides a rationale for using age and gender in a conceptual rather than a purely descriptive fashion. This is desirable, since a goal in consumer behavior research should be to raise the application of as many demographic variables as possible from the merely descriptive. to the richer ground of conceptualization.


The examination of patterns of inequalities is of basic interest to all behavioral science disciplines. For instance, within the context of consumer behavior, researchers have often employed variables such as social class standing that have innate inequality characteristics (e.g., Jain, 1975; Myers & Mount, 1973; Myers, et al., 1971; Shimp & Yokum, 1981). Indeed, most to son demographics (e.g., sex, age, employment status, income, occupation, education) measure inequalities within a particular context. To illustrate, higher income is almost always a better position to be in than lower income; membership in certain religious, ethnic and racial groups may be an advantage or disadvantage for their members; and it is generally felt to be better to be younger than older. Thus, common demographics have a function beyond serving merely as descriptor variables; that is, within an appropriate context they function as a barometer of inequality, status and stigma.

In a similar fashion, social gerontologists have in recent years become interested in certain types of inequalities that would seem to be of potential importance to consumer behavior, especially as they pertain to studying and segmenting the increasingly important elderly consumer market. In particular, they have explored two competing "hypotheses" that offer very different conclusions as to how men and women respond to the aging process. First, there is the leveling hypothesis which suggests that as men age, their life situations deteriorate at a more rapid rate than is the case for aging women. In contrast, the double jeopardy hypothesis suggests that the greater inequalities experienced by women in society worsen with advancing age.

Each of these two conflicting hypotheses will be described, and then the small amount of available empirical literature will be considered.

The Leveling Hypothesis

The leveling hypothesis suggests that the transition to old age is more disconcerting and traumatic an experience for aging men. This view proposes that advancing age has an equalizing effect on male-female inequalities that have existed in mid-life. One argument in support of the leveling hypothesis is that women tend to have "smoother" individual life cycles; that is, their central roles tend to remain basically unchanged from girlhood to death (i.e., women are basically dependent on men financially and are also responsible for domestic matters). In contrast, men suffer abrupt losses of their "core" roles (as workers and providers) at retirement.

Cumming (1961), in comparing the effect of retirement on men with the effect of widowhood on women, proposes that the transition to widowhood for women may be easier because widows have many peers in similar positions that can provide emotional support. It has also been suggested that aging is easier for women because it begins earlier and lasts longer and is, therefore, more of a gradual process (Beeson, 1975). Similarly, Blau (1973) proposes that retirement for men is more demoralizing because it reduces men's daily social contacts and represents an abrupt severance with a previously critical role.

Also supporting the leveling hypothesis is the alternative view that women may better adjust to growing older because they are socialized to adjust to repeated role "discontinuities" at each stage of their life cycle (Kline, 1975). This perspective is in rather sharp contrast to the portrayal of women as acting out a highly stable, continuous role throughout life. According to Atchley (1972), women are under more stress to assume conflicting priority roles of "worker," "housewife," and "mother;" whereas "worker" is men's primary role for most of their lives. Thus, proponents of this alternative view argue that "role inconsistency" rather than "role consistency" may account for why women have greater resiliency in old age (and therefore the leveling hypothesis).

There are still other age-gender viewpoints that support a leveling hypothesis. For instance, it has been suggested that not only do women undergo role changes throughout their life cycle, but that there is an apparent role reversal process occurring between older men and older women. More precisely, Borgese (1963) and Guttman (1974) argue that men become less active and more dependent with increasing age. In contrast, women take a more active posture in controlling their social environment in old age (Guttman. 1974).

To sum up, there seem to be three distinctly different views that support the leveling hypothesis: (1) that women have a smoother individual life cycle, while men's adjustment to a restricted lifestyle is more difficult; (2) that women have more varied life-roles, while men's adjustment to retirement (their only critical role) is very difficult; and (3) that a form of role reversal occurs in later life, where women gain control and men lose control.

The Jeopardy Hypothesis

In direct contrast to the leveling hypothesis, the jeopardy hypothesis predicts that along with advancing age there is a worsening of any inequalities existing between males and females in earlier life stages. While jeopardy can be thought of in terms of age-gender inequalities, it actually has a much broader horizon. In the gerontological literature, for example, there are frequent references to "double," "triple," and "multiple" jeopardy (e.g., Palmore & Manton, 1973; Hendricks & Hendricks, 1977; Chappell & Havens, 1981a). These descriptive forms of jeopardy refer to additive-negative effects of membership in more than one disadvantaged or stigmatized social category or grouping.

Double jeopardy, the most commonly cited form of jeopardy, refers to the occupation of any two stigmatized statuses which are characterized by greater negative consequences than the occupation of either one alone (Dowd & Bengtson, 1978; Chappell & Havens, 1981a). Within the context of age-gender differences (the issue of primary concern), the double-jeopardy hypothesis would exist if older males were found to be better off than older females as measured by such variables as life satisfaction (Larson, 1978), and mental health functioning (Kahn, et al., 1960). The double-jeopardy hypothesis also requires that evidence be provided that shows that as females age their conditions deteriorate and that these differences between the sexes did not exist to such an extent at a younger age, or that females were in a better position than males when they were younger and this situation has changed as they aged.

There is a considerable literature describing the various forms of inequalities existing between men and women that tends to implicitly support the jeopardy hypothesis (deBeauvoir, 1953; Epstein, 1973; Payne & Whittington, 1976). The major theme of this research stresses that women's positions in society are based on "derived status;" that is, women's roles in society are established largely in terms of the position or status of the men with whom they are associated, and that women are associated with "being" while men are more oriented towards "doing" (Beeson, 1975). As deBeauvoir (1973) indicates, women are often viewed as sex objects, while men are evaluated by achievements in their personal or professional lives.

Still further, the generally poor health assessment (Payne a Whittington, 1976), greater poverty (Butler & Lewis 1973), increased prevalence of living alone (Brotman, 1972), disadvantages with respect to remarriage (Butler & Lewis, 1973), and the lack of social support systems for widows (Barak, 1982), are all cited as illustrations of the special stigmatized state of older women in American society and are additional evidence favoring the double jeopardy hypothesis.

Empirical Research: Leveling Versus Jeopardy

One of the first empirical studies to directly test the leveling versus jeopardy hypothesis examined the status of minority aged to see if the differences in selected "quality of life" variables between individuals with minority and majority status (i.e., blacks and Mexican-Americans versus whites) decreases or increases from middle to old age (Dowd & Bengtson. 1978). The results reveal that for health and income, the gap between minority aged and their white counterparts increased significantly with increasing age -- with the greatest disparity occurring among those aged 65 gears or older. These findings support the double-jeopardy hypothesis. Yet, when it came to measures of life satisfaction and frequency of contact with relatives, the extent of ethnic variation declined across age strata, indicating support of the leveling hypothesis.

Chappell and Havens (1980, 1981b), in even more recent empirical research, have examined the leveling versus double-jeopardy hypotheses. In this case, however, they have operationalized the testing of these two competing hypotheses in terms of gender-related variations in aging - which is more in keeping with the present research interests. To serve as criterion variables, they selected two different "quality of life" indicators. The first criterion variable, mental health functioning, they labeled an "objective" indicator; whereas the second criterion variable, life satisfaction, they labeled a "subjective" indicator. Their sample was divided by age and gender, with 75 years of age serving as the division point.

The findings support the existence of double jeopardy for the objective indicator, but not for the subjective indicator (Chappell & Ravens, 1980). Specifically, the old/old (those 75+ years of age) were found to have worse mental health functioning than the young/old (those less than 75 years of age), women had worse mental health functioning than men, and older women were found to evidence worse mental health than older men.

In contrast, for life satisfaction (the subjective indicator), there were no significant differences between the various age-gender groups examined. Chappell and Havens (1980) suggest that the acceptance of the double-jeopardy hypothesis in the case of the objective indicator, and not in the case of the subjective indicator, supports Dulude's (1978) contention that women are less likely than men to internalize their negative subjective situations.

A second level of analysis, employing a number of behavioral factors (i.e., perceived mental health, socioeconomic status, frequency of contact with friends), which might contribute to objective mental health disjunctioning, revealed that these factors tended to only contribute to explaining the deteriorated mental health of old/old women. In isolating older women, these results support the relevance of combining age/sex into composite categories. over the simple age or sex groups so frequently found in behavioral science research.

Sot all research dealing with these two competing hypotheses has utilized two stigmatized variables. For instance, Palmore and Manton (1973) investigated the combined effects of three stigmatized variables (i.e., "triple jeopardy") -- age, gender and race. Similarly, Chappell and Havens (1981b) have also extended the analysis of their earlier research on double jeopardy to the exploration of a third stigmatizing variable (i.e., Canadian ethnic group membership). The results of measuring the effects of various combinations of disadvantaged statuses revealed that older women were particularly disadvantaged, and that older minority women were most disadvantaged.

It is fair to conclude that the empirical research that directly tests the leveling-jeopardy hypotheses supports the existence of a jeopardy effect of aging. However, this conclusion is limited by the one or two cases that suggest the existence of leveling, and the several cases that do not support either hypothesis. Even the discrepancy existing between objective and subjective measures of well-being tends to cloud the conclusions.


The underlying conditions which are necessary to accept (or reject) each of these two competing phenomena need to be considered. In order for leveling to be judged to exist, the analysis would have to reveal the following condition:

If the difference in a dependent variable between males and females, in an older age group, is not as great as the difference between males and females, in a younger age group, this would indicate that age is likely to be exerting a leveling effect.

On the other hand, in order for double jeopardy to be judged to exist, the analysis would have to indicate the following condition:

If the difference in a dependent variable between males and females, in an older age group, is greater than the difference between males and females, in a younger age group, this would indicate that age is likely to be exerting a double jeopardy effect.

In conducting research designed to test the leveling versus double jeopardy hypotheses within a consumer behavior context, the dependent variable should probably be some established life satisfaction index or even a more specific consumer satisfaction index or a consumer usage variable.

Subjective well-being is a well established summary construct within social gerontological research. Under its umbrella have been included such specific measures as: "life satisfaction," "happiness," "contentment," "adjustment," and "morale;" each with numerous specific operational variations. Cast in this central role, it is no wonder that subjective well-being is at once an important and fuzzy construct, one with boundaries quite difficult to delimit.

Multi-item and multidimensional measures of subjective well-being have been used as dependent variables in a large number of studies concerned with various aspects of the personal and social adjustment of older people (e.g., Chappell & Havens, 1981a; Dowd & Bengtson, 1978). Of the many measures, the Neugarten et al (1961) 20-item Life Satisfaction Index A (LSIA), as well as several revisions (i.e., Adams, 1969; Woods et al, 1969), has been the most widely used in the study of older individuals (e.g., Hoyt, et al, 1980; Markides & Martin, 1979). These LSIA derived scales have even been employed in several recent studies dealing with the subjective well-being of the elderly as consumers (e.g., Bearden, et al., 1978; La Forge, 1980).

There is also a tradition in social gerontology to search for an association between established well-being scales and more focused, objective and readily available behavioral and socioeconomic variables. A primary objective of such research seems to be to identify a variety of more direct and easy to measure behavioral and socioeconomic factors that may serve as suitable substitutes or proxies for the more abstract multi-item well-being measures.

While many specific variables have been explored as possible correlates of subjective well-being, it is possible to categorize most of these factors into three broad categories - health factors, socioeconomic factors and social-activity factors (Larson, 1978).

It is more than likely that various established consumer behavior inventories such as consumer satisfaction indices and consumer alienation scores would function as appropriate dependent variables in testing the leveling-jeopardy controversy within a consumer behavior context. Moreover, it would also seem desirable to test these paradoxical hypotheses using various consumer-specific constructs such as venturesomeness, innovativeness, perceived risk, and opinion leadership as dependent variables, especially when they are cast within a product-category specific purchase situation.



In designing consumer behavior research that tests the leveling-jeopardy hypothesis, the independent variable might appropriately consist of six age-gender categories as indicated in the Table. These six categories are not arbitrarily selected, rather they are based on 8 careful review of the available social gerontological literature. Indeed, it is noteworthy that some prior research testing the leveling-double jeopardy hypotheses has contrasted the 55-64 age group (i.e., "mature-adults") and the 65-plus age group (Dowd & Bengtson, 1978); whereas other research has contrasted the 65-74 age group (i.e., "young-elderly") and the 75-plus age group (i.e., "old-elderly") (Chappell & Havens, 1980). However, in a study with a sample of sufficient size and dimension, it would be desirable to accommodate all three major age categories (i.e., 55-65, 65-74 and 75-plus).


From both a marketing and public policy perspective, the stream of research reviewed here confirms the importance of partitioning or segmenting the elderly market; and thereby, provides some additional support that it is a mistake to treat the elderly as if they represented a single distinct market group. In particular, this paper suggests the usefulness of segmenting the older portion of the population in terms of age-gender categories and then looking at each segment's life satisfaction, consumer satisfaction or consumer behavior.

Taking advantage of the present literature review, the following are some Possible avenues for future research:

1. It would be desirable to extend the present age-gender research by employing established measures of subjective well-being and relating them to various measures of consumer satisfaction.

2. It would also be useful to further explore the development and testing of consumer behavior-specific life satisfaction scales

3. Similarly, research should explore how age-gender factors tend to influence a wide variety of consumer behavior-relevant issues, e.g., attitude formation and change, levels of perceived risk, brand and patronage loyalty, and consumer-related informal communications patterns.

4. Finally, it would seem highly instructive to determine if specific geographic areas, with possible distinctly different traits, expressly influence the relationships to be explored. It is anticipated that such geographic analysis could provide a dual function: (1) as a form of market segmentation analysis of the firm's product and service customers -- based on the different age-gender categories, and (2) as a descriptive tool to get behind the initial overall analysis where the research might provide evidence favoring either leveling or double jeopardy.


Abu-Labon, S. (1981), "Women and Aging: A Futuristic Perspective," Psychology of Women Quarterly, 16, 53-56.

Adams, D. L. (1969), "Analysis of a Life Satisfaction Index," Journal of Gerontology, 24, 470-474.

Atchley, R. C. (1972), The Social Forces in Later Life; An Introduction to Social Gerontology. Belmont, California: Wadsworth.

Barak, B. (1982), "Elderly Solitary Survivor and Social Policy: The Case for Widows," in Advances in Consumer Re- search, Vol. 9, et. Andrew Michell, Ann Arbor: Association for Consumer Research, 27-30.

Bearden, William O., et al. (1978), "A Path-Analytic Investigation of Life Satisfaction Among Elderly Consumers," in Advances in Consumer Research, Vol. 6, et. William L. Wilkie, Ann Arbor: Association of Consumer Research, 386-391.

Beeson, D. (1975), "Women in Studies of Aging: A Critique and Suggestion," Social Problems, 23, 52-59.

Blau, Zena Smith (1973), Old Age in a Changing Society. New York: New Viewpoints.

Borgese, E. (1963), The Ascent of Women. New York: George Braxillen.

Brotman, E. (1972), "The Fastest Growing Minority: The Aging," American Journal of Public Health, 64, 257-263.

Butler, R., and M. Lewis (1973), Aging and Mental Health positive Psychosocial APProaches. St. Louis, Missouri: C. V. Mosby.

Chappell, N., and B. Havens (1980), "Old & Female: Testing the Double Jeopardy Hypothesis," Sociological Quarterly. 21, 157-71.

Chappell, N., and B. Havens (1981a), "Rural-Urban Differences Among Older Women." Paper presented at the Midwest Sociological Society, Minneapolis, Minnesota.

Chappell, M. L., and B. Havens (1981b), "Triple Jeopardy, Age, Sex and Ethnicity." Paper presented at Gerontological Conference Toronto. Canada.

Cumming, F., and W. Henry ( 1961), Growing Old: The Process of Disengagement. New York: Basin Books.

deBeauvoir, S. (1973), The Coming of Age, translated by Patrick O'Brin. New York: Warner Paperback Library.

Derrick, F. W. and A. R. Lehfeld (1980), "The Family Life Cycle: An Alternative Approach," Journal of Consumer Research, 7, 214-17.

Dowd J., and V. Bengtson (1978), "Aging in Minority Populations: An Examination of the Double Jeopardy Hypothesis," Journal of Gerontology, 33, 427-36.

Dulude, L. (1978), Woman and Aging: A RePort on the Rest of Our Lives. Ottawa: Advisory Council on Status on Women.

Epstein, C. F. (1973), "Black and Female, The Double Whammy," Psychology Today, 7, 57.

Guttman, N. (1974), The Country of Old Men. Culture and Personality, B. Levine, et. Chicago, Illinois: Aldine.

Hendricks, J., and C. Hendricks (1977), Aging in Mass Society. Cambridge, Massachusetts: Winthrop Publishers.

Hoyt, D., M. Raiser, G. Peters, and S. Babchuk (1980), "Life Satisfaction and Activity Theory: A Multidimensional Approach," Journal of Gerontology, 35, 935-41.

Jain, A. (1975), "A Method for Investigating and Representing Implicit Social Class Theory," Journal of Consumer Research, 2, 53-59.

Kahn, R. L., A. T. Goldfarb, H. Pollack, and A. Peck (1960), "Brief Objective Measures for the Determination of Mental Status in the Aged," American Journal of Psychiatry, 107, 326-28.

Kline, C. (1975), "The Socialization Process of Women," The Gerontologist, 15, 15-23.

LaForge, M. (1980), An Analysis of the RelationshiP Between Patterns of Adjustment in Aging and Selected Aspects of Consumer Behavior, Doctoral dissertation, University of Georgia.

Larson, R. (1978), "Thirty Years of Research on the Subjective Well-Being of Older Americans," Journal of Gerontology, 33, 109-25.

Markides, K., and E. Martin (1979), "A Causal Hotel of Life Satisfaction Among the Elderly," Journal of Gerontology, 34, 86-93.

Murphy, P. E. and W. A. Staples (1979), "A Modernized Family Life Cycle," Journal of Consumer Research, 6, 12-22.

Myers, J., and J. Mount (1973), "More on Social Class vs. Income as Correlates of Buying Behavior," Journal of Marketing. 37. 71-73.

Myers, J., R. Stanton, and A. Haus (1971), "Correlates of Buying Behavior: Social Class vs. Income," Journal of Marketing, 35, 8-15.

Neugarten, B. L., R. J. Havighurst, and S. S. Tobin (1961), "The Measurement of Life Satisfaction," Journal of Gerontology, 16, 134-43.

Palmore, E., and t. Manton (1973), "Ageism Compared to Racism and Sexism," Journal of Gerontology, 28, 363-69.

Payne, B., and F. Whittington (1976), "Older Women: An Examination of Popular Stereotypes and Research Evidence," Social Problems, 23, 486-504.

Roscoe, Jr., A. M., A. LeClaire, Jr., and L. G. Schiffman (1977), "Theory and Management Applications of Demographics in Buyer Behavior." in eds. A. D. Woodside, J. M. Sheth, and P. D. Bennett, Consumer and Industrial Buying Behavior, New York: Elsevier-North Holland.

Shimp, T., and J. T. Yokum (1981), "Extensions of the Basic Social Class Model Employed in Consumer Research." in et. Rent B. Monroe, Advances in Consumer Research, Vol. 8 102-107.

Wood, V., M. Wylie, and G. Sheafer (1969), "An Analysis of a Short Self-Report Measure of Life Satisfaction: Correlation with Rater Judgment," Journal of Gerontology, 24, 465-469.