Cognitive Age: a Nonchronological Age Variable

Benny Barak, Rutgers-The State University of New Jersey
Leon G. Schiffman, Baruch College (CUNY)
ABSTRACT - The determination and measurement of self perceived age as an alternative to chronological age has received almost no attention in consumer behavior and marketing research. This paper discusses a newly developed self-perceived age measure entitled "cognitive age", and presents some results concerning its reliability and its response patterns.
[ to cite ]:
Benny Barak and Leon G. Schiffman (1981) ,"Cognitive Age: a Nonchronological Age Variable", in NA - Advances in Consumer Research Volume 08, eds. Kent B. Monroe, Ann Abor, MI : Association for Consumer Research, Pages: 602-606.

Advances in Consumer Research Volume 8, 1981      Pages 602-606


Benny Barak, Rutgers-The State University of New Jersey

Leon G. Schiffman, Baruch College (CUNY)


The determination and measurement of self perceived age as an alternative to chronological age has received almost no attention in consumer behavior and marketing research. This paper discusses a newly developed self-perceived age measure entitled "cognitive age", and presents some results concerning its reliability and its response patterns.


While demographic variables have been a mainstay of marketing and consumer behavior research, this group of variables is typically selected and operationally defined quite automatically and usually without much imagination. Moreover, there has been a general lack of attention given to the development of new forms of demographic and "demographic-like" variables. Within a consumer behavior context, Roscoe, LeClaire and Schiffman (1977) have been sensitive to this problem when they suggested the need to refine existing demographics and develop new ones. In particular, they proposed that the age variable, the variable of concern in this paper, should be broadened so that it reflects such age-related factors as: the age of the household, age at birth of first child, age of siblings, birth order, age at first awareness of a product (or brand), age at first trial of a product and perceived age (i.e., youthfulness).

In the spirit of the Roscoe, LeClaire and Schiffman (1977) recommendations, this paper is concerned with the need for both refined and new measures of age, especially non-chronological measures of age. More specifically, this paper will endeavor to: (1) set out some of the major limitations of chronological age, (2) review the major types of nonchronological age variables, (3) propose a new perceived age variable, "cognitive age," and present some initial results pertaining to its reliability and response patterns in comparison to chronological age, and (4) offer our thinking on the future use of this age variable in various types of research, especially consumer behavior research.


Chronological age is usually defined as either the number of years lived (Hendricks and Hendricks 1976), or as the distance from birth (Jarvik 1975). As a demographic variable, chronological age stands out from all other variables in terms of frequency of its use. In consumer behavior research it is often employed in descriptive consumer behavior studies, or in efforts to segment consumer markets.

Despite its great popularity, the use of chronological age is problematic for researchers interested in age-related research, particularly research that examines the attitudinal or behavioral patterns of the elderly. More precisely, chronological age does not lend itself well to functioning as a dependent variable; that is, it is exceedingly difficult to justify employing almost any behavioral variable of interest to consumer researchers as a predictor of chronological age. Stated differently, the unique antecedent character of chronological age restricts its usefulness to being employed as a predictor variable.

Still further, from the perspective of the present paper, and consumer behavior research in general, the overriding shortcoming of chronological age would seem to be that it does not take into account the fact that people frequently perceive themselves to be at an age other then their birth age, and that this self-perceived or cognitive age seems to influence purchase behavior. There have been few references in the marketing and consumer behavior literature to the influence of self-perceived age on consumer behavior. A noteworthy exception is the Ford Motor Company's eventual recognition of the importance of self-perceived age in positioning its Mustang automobile:

The car was designed to appeal to young people who wanted an inexpensive sporty automobile. Ford found to its surprise that the car was being purchased by all age groups. It then realized that its target market was not the chronological young, but those who were psychologically young (Kotler 1976, p. 147).

Following this line of thinking, it might be expected that consumers would tend to consume many products according to their perceived age, and not according to their chronological age. This suggests that an individual's identity (and behavior) may depend, as much, if not more, on perceived or felt age than upon chronological age. Thus, a flexible and versatile perceived age measure would provide consumer researchers, marketers and public policy makers with an attractive alternative to relying on a chronological measure of age; and more importantly, it might provide greater insights into the patterns of aging and the consumer behavior of the elderly.


Gerontological researchers have suggested a variety of non-chronological age variables. The three broad categories of nonchronological age which are most frequently championed are briefly considered here, i.e., biological age, social age and social-psychological age.

Biological Age

Biological age is an estimate of an individual's present position with respect to his or her potential life-span (Birren and Renner 1977; Jarvik 1975). The measurement of biological age is difficult to accomplish and tends to take the approach recommended by Bell (1972): namely, the measurement of biochemical age through assays of blood serum and urine. Moreover, in discussing biological age, Bromley (1974) has pointed out that the body's organs are made up of different types of cells, so that an estimate of the biological age of any particular organ is extremely difficult to establish. In addition, the overall effectiveness of a human body is determined by the least efficient part of the system required to keep the body functioning and this varies among humans.

Social Age

Social age is the age of an individual as defined in terms of social roles and habits (Birren and Tenner 1977). It implies that age expresses an individual's place in the social structure; which is indexed by such variables as socioeconomic status, occupation, education, race and sex (Bengston, Kasschau and Ragan 1977).

Social aging is also concerned with the different roles a person takes while passing through the life cycle. There is a continuous role change that takes place and the patterned sequencing of these roles reflects some of the changes in an individual's life (Bleu 1973). Tied to this role alteration is a subjective perception of appropriate and inappropriate age-specific, society-determined norms that are an integral part of the varied roles (Bengston, Kasschau and Ragan 1977).

Social-Psychological Age

Three major types of social-psychological age measures have received special attention: (1) subjective age, (2) personal age, and (3) other-perceived age.

Subjective Age

Subjective (or identity) age measures an individual's self-perception in terms of reference age groups, i.e., "middle-aged," "elderly," or "old" (Blau 1956, 1973; Peters 1971; Rosow 1967, 1974; Ward 1977). It subjectively establishes how a person feels about such reference age groups.

Several conclusions can be dream from the findings of research examining subjective age:

1. The majority of elderly have a strong tendency to see themselves as considerably younger than their chronological age (Bleu 1956, 1973; Peters 1971; Rosow 1967, 1974).

2. The self-identification with a younger age group varies in terms of social class standing (Bengston, Kasschau and Ragan 1977; Peters 1971; Rosow 1967).

3. Women are more sensitive to the negative stereotypes associated with "elderly" and "old," and tend to see their age differently from their male counterparts (Bengston, Kasschau and Ragan 1977; Peters 1971).

4. Loss of critical roles and status also have a differential effect on subjective age perception; specifically, Neugarten (1977) suggests that especially "off-schedule" crises causing adaptational problems bring about change in the subjective perception of age.

5. Elderly who perceive themselves as younger are more likely to be innovative (Bleu 1973).

6. Elderly voters who perceive themselves to be younger tend to have a more liberal and less traditional outlook on life (Bengston and Cutler 1976).

7. Those who perceive themselves to be younger are more likely to have had more education than those who perceive themselves as older (Rosow 1967, 1974; Peter 1971).

8. Finally, research indicates that subjective age is related to subjective well-being (i.e., life satisfaction or morale) and self-confidence (Bengston, Kasschau and Ragan 1977; Peters 1971).

A major problem with the subjective age construct has been the ambiguity surrounding how it has been defined and measured. Subjective age has been measured through self-rating scales, expressed in terms of some form of nominal age-reference categorization (i.e., "young," "middle-aged," "elderly," "old"). This kind of rating is suspect since one is never sure Just what such a nominal age-reference categorization means to the respondents in terms of commonly accepted units such as years.

Personal Age

Personal age is a different type of self-perceived age than subjective age. It is established by means of a self-report of an individual's age perceptions measured in terms of units of years. As operationalized by Kastenbaum, Derbin, Sabatini and Artt (1972), personal age consists of four major age dimensions: (1) feel-age (how old a person feels), (2) look-age (how old a person looks), (3) do-age (how involved a person is in doing "things" favored by members of a certain age group), and (4) interest-age (how similar a person's interests are to members of a certain age group).

In addition to the four age dimensions, respondents are also required to answer questions dealing with their individual feelings about age in general and how the four personal age dimensions compare with the respondents' chronological age. Thus, personal age is measured as a basic part of a complex procedure designed to investigate people's feelings about age. In the format employed by Kastenbaum, Derbin, Sabatini and Artt (1972), the personal ago measure requires lengthy personal interviews that do not lend themselves to the types of survey research normally conducted by consumer and marketing researchers. Even though Kastenbaum's personal age measure has these limitations, it nevertheless provides a cornerstone-framework for the development of the cognitive age measure to be reported on here.

Other-Perceived Age

This final social-psychological age measure is concerned with the subjective evaluation of the age status of an individual as assessed by others. The measure is especially appropriate for examining stereotyping of age groupings and seems to be largely based on perceived physical looks and the perceived social roles of the individual(s) being observed (Lawrence 1974). This type of perceived age, while not as yet dealt with in the consumer behavior or communication-mass media literature, would seem especially ripe for exploring how consumers' perception of another person's age (e.g., a model in an advertisement) interacts with the product or brand, the usage situation, and other relevant environmental or situational factors.


In this section we will: (1) discuss how we operationally defined our nonchronological age measure, (2) describe the sample of elderly consumers that served as the subjects for the initial examination of our self-perceived age measure, (3) present the results of our effort to examine the reliability of the measure, and (4) contrast soma of the basic characteristics of our survey results to the findings reported by Kastenbaum, Derbin, Sabatini and Artt (1972).

The Measurement and Analysis of Cognitive Age

Our self-perceived age measure, which we have labeled "cognitive age," was operationally defined in terms of four questions which were designed to correspond to the four dimensions of personal age suggested by Kastenbaum, Derbin, Sabatini and Artt (1972). The introductory statement which is either read by an interviewer, or read directly by the respondent, and the questions and response mode for the cognitive age variable are displayed in the Appendix. To prepare survey results for analysis, responses to each of the four age dimensions (i.e., feel-age, look-age, do-age and interest-age) can either be scored separately, or an overall or composite score can be derived. As a first step, each respondent's score, for each of the four dimensions, is assigned a midpoint value (e.g., a response of "50's" was recoded to be "55"). This procedure provided an opportunity to set a numerical value (in terms of years) for a person's cognitive age (either for each of the four dimensions or a composite score), and it also allowed us to compare cognitive age and chronological age. In our preliminary research the composite score for each respondent was formed by the simple average of the midpoint values of the four age dimensions.

The Sample

The initial data gathered on the cognitive age measure was collected as part of a study that questioned some 324 elderly consumers who met the following qualifications: (1) they were female, (2) they reported their exact chronological age, (3) they were chronologically 55 years or older, and (4) they responded to all four cognitive age questions.

The respondents who were all residents of the counties that comprised a major Northeastern city, were all personally interviewed by specially trained interviewers. The questionnaire, which required about 45 minutes to be administered, focused on hair care product behavior and attitudes. Some of the major consumer behavior and consumer-related variables measured were: venturesomeness, self-confidence, dogmatism, opinion leadership, life satisfaction, club membership, a range of media habit questions, a battery of product/brand usage and experience questions, and a selected number of demographic items.

Estimation of Reliability

The cognitive age variable was subjected to three widely employed measures of reliability: test-retest reliability, Guttman's Lambda Test, and a split-half reliability test. The test-retest reliability was based on data collected during two interview sessions (separated by a three week period), conducted with the same small sample of 15 respondents. The resulting test-retest coefficient was .88. The Guttman Lambda and Spearman-Brown split-half reliability tests were respectively .86 and .85. On the basis of these three reliability estimates, it seems reasonable to suggest that the cognitive age scale has good internal consistency and is reliable.

Cognitive Age: Patterns and Regularities

In proposing that a nonchronological age variable such as cognitive age be employed along with, or in certain cases as a substitute for the traditional chronological age variable, it seems critical to attempt to ascertain the extent to which the cognitive and chronological age variables are measuring unique aspects of age. To this end, we elected to compare our findings with those reported by Kastenbaum, Derbin, Sabatini and Artt (1972) who undertook the original exploratory work which examined the four age dimensions.

Table 1 presents the percentage of our respondents who reported their chronological age in response to the four cognitive age questions. As Kastenbaum, Derbin, Sabatini and Artt (1972) found with their data, our results indicate also only a moderate degree of correspondence between the four cognitive age dimensions and chronological age. Specifically, none of the 16 percentages reported in Table 1 reach the 40 percent agreement mark. The greatest degree of agreement was 44 percent (i.e., look-age at chronological age 50's) and the least was 16 percent (i.e., interest-age at chronological age 80's).



These results reveal that for the majority of our elderly subjects their chronological age does not adequately correspond to their perceived age as reflected by any of the four cognitive age dimensions. More importantly, the results can be interpreted as suggesting that the cognitive dimensions of age capture distinctly different aspects of age than is reflected by chronological age.

An important point that needs to be determined is the basic directionality of the decline in mean percentage of agreements that occur with advancing decades. What we want to know here is the extent to which our elderly subjects tend to view themselves as younger or older than their chronological age. In this regard, Table 2 shows the percentage of respondents who perceived themselves to be at a younger age grouping than their chronological age for each of the four dimensions of our cognitive age variable. The results reveal that our elderly respondents are considerably more likely to identify their age related feelings and actions with a younger age group than the one which is consistent with their chronological age.



Also, the results indicate (i.e., the difference between the results in Tables 1 and 2) that the percentage of respondents who identify with an older age group than the one corresponding to their own chronological age was exceedingly small (zero for interest-age at chronological age 70's, and at the high end only five percent for feel-age at chronological age 60's).

Still further, the mean percentages appearing in the last column of Table 2 reveal that as our respondents' chronological age increases, they are more likely to identify with a younger cognitive age grouping.

It is also important to look at the degree of consistency between the four age dimensions that constitute the cognitive age variable. Table 3 reveals that the number of cases in which each of the four cognitive age dimensions received the same decade-response as one of the other cognitive age dimensions ranges between a low of 45 percent (for interest/look age) to a high of 57 percent (for do/ interest age). Performing a similar type of analysis, Kastenbaum, Derbin, Sabatini and Artt (1972) found somewhat lover levels of agreement (i.e., all were less than 50 percent -- the range was 28 percent for look/interest age and 49 percent for feel/do age). Nevertheless, the extent of inter-dimension agreement in our data is still quite low, suggesting that the four dimensions tend to reflect separate aspects of this perceived age variable.



Finally, the mean percentages in the last column of Table 3 reveal that when our respondents reported their perceived look-age, they were providing a response that had only a 33 percent mean correspondence with the other three cognitive age dimensions. In contrast, the other three cognitive age dimensions each have mean percentage agreement scores that clustered slightly above 50 percent. That the look-age had the lowest level of mean percentage agreement is consistent with the Kastenbaum, Derbin, Sabatini and Artt (1972) results. They found a 36 percent mean agree-merit score for look-age and the other three items that make up their personal age variable.

These results indicate that the four dimensions of our cognitive age variable are capturing aspects of age that are not adequately reflected in an individual's chronological age. Moreover, the analysis reveals the importance of examining each of the four cognitive age dimensions, for the composite measure that combines the dimensions is likely to mask some of the differences reflected by each of dimensions. Still further, the four age dimensions related to each other and to chronological age in a manner quite similar to patterns found by Kastenbaum and his associates (1972).


It is our feeling that the cognitive age variable, as well as other nonchronological age measures, will enrich the process of studying the impact of age on consumer behavior (and vice versa). Also, it is likely that nonchronological age variables will provide information not generally possible when chronological age is related to consumer behavior.

The present results also provide additional support for the increasingly popular view that the elderly should not be viewed as if they are a single group or market segment, with uniform attitudes and behavior. For instance, a consumer who is in her sixties might perceive of herself as being in her forties and identify with role models of that age. In such a case, the possibility exists that she really belongs to a different target market than indicated by her chronological age; that is, if she perceives herself to be younger, she actually might belong to a younger target market, even though her chronological age is older. It would seem desirable that research be undertaken to examine the relationship between cognitive age (and chronological age) and a cross-section of different product categories (some likely to be age sensitive and some not) to see if the product category makes a difference in terms of age-related appeals and age-related role models.

Of course, the cognitive age variable should be employed in studies that do not solely focus on the elderly; that is, it should be used in studies that concentrate on other age groups, as well as a cross-section of age groups. For example, this research could include investigations of age perception among teenagers as compared to the elderly. (Teenagers might like to perceive of themselves as being older than their chronological age, while the elderly perceive of themselves as being younger than their chronological age.)

As a final suggestion, the cognitive age measure should lend itself to cross-cultural consumer behavior studies. Specifically, studies that explore the scope and nature of cultural differences in the perception of age, and how these differences influence various aspects of consumer behavior. It would also be desirable to assess the effect of subcultural and situational factors on the response to the cognitive age measure.



Most people seem to have other 'ages' besides their official or 'date of birth' age. The questions which follow have been developed to find out about your 'unofficial' age. Please specify which age group you FEEL you really belong to: twenties, thirties, forties, fifties, sixties, seventies, or eighties.



Bell, Benjamin (1972), "Significance of Functional Age For Interdisciplinary and Longitudinal Research in Aging," Aging and Human Development, 3, 145-147.

Bengston, Vern L. and Cutler, Neal E. (1977), "Generations and Intergenerational Relations: Perspectives on Age Groups and Social Change," in Robert H. Binstock and Ethel Shanas, Eds., Handbook of Aging and the Social Sciences, N.Y.: Van Nostrand Reinhold Company, 130-159.

Bengston, Vern L., Kasschau, Patricia L. and Ragan, Pauline K. (1977), "The Impact of Social Structure on Aging Individuals,'' in James E. Bitten and K. Warner Schaie, eds., Handbook of the Psychology of Aging, N.Y.: Van Nostrand Reinhold Company, 327-359.

Birren, James E. and Renner, V. Jayne (1977), "Research on the Psychology of Aging: Principles and Experimentation," in James E. Birren and K. Warner Schaie, eds., Handbook of the Psychology of Aging, N.Y.: Van Nostrand Reinhold Company, 3-38.

Blau, Zena S. (1956), "Changes in Status and Age Identification,'' American Sociological Review, 20, 198-202.

Blau, Zena S. (1973), Old Age in A Changing Society, N.Y.: Franklin Watts, Inc.

Bromley, Dennis B. (1974), The Psychology of Human Aging, 2nd ed. Harmondsworth, Middlesex, England, Penguin Books, Inc.

Hendricks, C. Davis and Hendricks, Jon (1936), "Concepts of Time and Temporal Construction Among the Aged, with Implications for Research," in Jaber F. Gubrium, ed., Time, Roles and Self in Old Age, N.Y.: Human Sciences Press, 13-49.

Jarvik, Lissy F. (1975), "Thought on the Psychobiology of Aging," American Psychologist, 30, 576-583.

Kastenbaum, Robert, Derbin, Valerie, Sabatini, Paul and Artt, Steven (1972), "'The Ages of Me' Toward Personal and Interpersonal Definitions of Functional Aging," Aging and Human Development, 3, 197-211.

Kotler, Philip (1976), Marketing Management, Analysis, Planning and Control, 3rd ed. Englewood Cliffs, NJ: Prentice-Hall, Inc.

Lawrence, J., "The Effect of Perceived Age on Initial Impressions and Normative Role Expectations," International Journal of Aging and Human Development, 5, 369-391.

Neugarten, Bernice L. (1977), "Personality and Aging," in James E. Birren and K. Warner Schaie, eds., Handbook of the Psychology of Aging, N.Y.: Van Nostrand Reinhold Company, 626-649.

Peters, G. R. (1971), "Self-Conceptions of the Aged, Age Identification and Aging," The Gerontologist, 11, 69-73.

Roscoe, Jr., A. Marvin, LeClaire, Jr., Arthur and Schiffman, Leon G. (1977), "Theory and Management Applications of Demographics in Buyer Behavior," in Arch G. Woodside, Jagdish Sheth and Peter D. Bennet, eds., Consumer and Industrial Buying Behavior, N.Y.: Elsevier North-Holland, Inc., 67-76.

Rosow, Irving (1967), Social Integration of the Aged, N.Y.: The Free Press.

Rosow, Irving (1974), Socialization to Old Age, Berkeley, CA.: University of California Press.

Ward, Russel A. (1977), "The Impact of Subjective Age and Stigma on Older Persons," Journal of Gerontology, 32, 227-232.