Advances in Consumer Research Volume 9, 1982 Pages 271-276
RECYCLING THE FAMILY LIFE CYCLE: A PROPOSAL FOR REDEFINITION
Mary C. Gilly, Southern Methodist University
Ben M. Enis, University of Missouri-Columbia
The family life cycle, as traditionally defined and recently "modernized," fails to recognize the changing role of women in contemporary society and the impact of such changes upon the types and composition of families. This paper presents a redefined family life cycle which accounts for the increasing number of single-person house-holds, cohabitation by non-legally married adults, delayed parenting, and rising divorce rates. The proposed concept is compared with other FLC definitions, and consumer behavior associated with the various redefined life cycle stages is discussed.
It stands to reason that family life cycle stage is a useful indicator of consumption behavior. Though changes in attitudes and behavior of interest to family and consumer researchers occur as individuals grow older, many of these changes are associated less with the biological process of aging than with changes in family status. Thus, dates marking changes in purchase patterns may not be birthdays, but dates of marriage, birth of children, dissolution of marriage, departure of children.
Reason notwithstanding, there is little empirical support in consumer research literature for the Family Life Cycle (FLC) as a predictor of consumption behavior. The authors contend that this relative void stems from inadequate precision in defining the concept. This paper offers a more comprehensive operationalization of the family life cycle concept than previous FLC definitions.
THE FAMILY LIFE CYCLE IN CONSUMER ANALYSIS
The family life cycle concept has been well-accepted in sociology for fifty years (Loomis, 1936) and in marketing for over twenty-five years (Clark, 1955). It is recognized in most sociology and marketing textbooks, primarily as a useful basis for segmentation. Engel, Kollat and Blackwell ( 1978) comment, "because the life cycle concept combines trends in earning power with demands placed on income, it is one of the most powerful ways of classifying and segmenting individuals and families."
Both marketers and sociologists have long used the family life cycle as a predictor of various types of behavior. These attempts have been made at different levels of sophistication and have had varying degrees of success. Lansing and Kish (1957) compared the variables of age and life cycle with respect to six aspects of the family's consumption pat tern: income, indebtedness, whether the wife works, home ownership, purchase of new cars and purchase of television sets. Life cycle was found to be a better discriminator than age in all six categories and, in fact, provided some information that analysis by age group concealed. However, the rho values for FLC ranged from .01 to .15, indicating that, while a better predictor than age, FLC explained very little of the variation in the dependent variables.
Perhaps the definitive statement of the traditional family life cycle in marketing was offered by Wells and Gubar (1966). Table 1, excerpted from their work, shows the stages in the traditional life cycle, and typical buying or behavior patterns for each stage. They also compared age and life cycle stage as predictors of expenditure on various products. For half the goods and services, here was little difference between age and life cycle. However, for 231 goods or services, there were substantial differences with 54 favoring age and 177 favoring life cycle. No indication was given as to how much of the variation in the expenditure for these products was explained by family life cycle stage.
Hisrich and Peters (1974) examined family life cycle and three other commonly used segmentation variables (income, social class and age) as correlates of two aspects of purchase behavior: choice of entertainment activities during the past year, and frequency of use of these entertainment activities. Findings revealed that income and family life cycle stage were more highly related to the use of all the entertainment activities than age or social class. However, all four variables showed strong associations with the frequency of use of all the entertainment activities. Stage in life cycle was able to explain 40 percent of the variation in use and 50 percent of the variation in the frequency of use.
Family life cycle was also used as an independent variable by Cox (1975) with length of marriage, in explaining the process of adjustment in family purchase decision-making. Analysis of variance of the relationship between the stage in the family life cycle and two operational definitions of adjustment revealed rho values of .15 and .36. These rho values were higher than the corresponding coefficients of determination using length of marriage, suggesting that family life cycle is the more powerful independent variable
These studies show that the use of family cycle stage as an independent variable has been limited in scope and has had mixed success. As Wells and Gubar (1966) note, the question of how to define the different stages is not a trivial one. In her standard sociological work on the FLC, Fischer (1955) recognized that "there are, of course, deviations from this [traditional] pattern that need hardly be mentioned, such as failures to marry, divorce, childless couple, early death of one spouse, remarriage, and so on. She dismissed these exceptions as inconsequential; indeed they may have been in the 1950s. Wells and Gubar estimated that such households accounted for five to ten percent of the data in 1966.
The proportions are much larger today. Nearly three-quarters of all households in 1960 included the head and spouse; this figure had dropped to two-thirds by 1966, and to 62 percent by 1978. During the 15 years from 1960 to 1975, the number of one-person households approximately doubled, from 6.9 million to 13.9 million; this represents an increase from 13 percent in 1960 to 20 percent in 1975. Men and women under the age of 30 accounted for half the growth in this type of household since 1972.
There were one million divorces and two million marriages in the United States in 1975. Two and one half million men and four million women were currently divorced in 1975. In addition, more marriages were being postponed. The proportion of single persons 20 to 24 years old (the range of years in which the median age at first marriage for both men and women falls) increased from 28 percent in 1960 to 48 percent in 1978 for women and from 53 percent in 1960 to 66 percent in 1978 for men.
AN OVERVIEW OF THE LIFE CYCLE
These figures demonstrate the fact that the trend in the United States today is away from the "traditional" family life cycle of marriage, birth of children, departure of children, and death of spouse. Many of these changes are related to the growth number of life style choices available to women today. There appear to be three major issues:
1. Childless and Delayed-Child Marriages. Choice and timing have traditionally been primarily the women's prerogative. More and more women are postponing or eschewing childbearing. More women are remaining single (23.9 percent in 1978 compared to 19 percent in 1960). More marriages are childless; in 1974, about 28 percent of all married couples with the husband under 35 had no children under 18 years of age at home, as compared with the 1969 level of 24 percent. And more married couples are postponing the birth of the first child. The median age of the mother at birth of first child was 22.1 in 1972 compared to 21.8 in 1960.
By and large, these decisions are a function of the broadening perspective of women today. New careers are open to women, more women are obtaining higher educations and/or are finding other ways to lead fulfilling lives. Raising children is no longer considered to be the predominant role of women in modern society. A greater variety of life styles exists today than the traditional FLC recognizes.
2. Single-parent Households. More than 90 percent of the single-parent households in the United States today are heated by women. There were eight million such households in 1978. Assuming only the mean poverty line income for each, these households represent almost 32 billion dollars in purchasing power. The traditional life cycle concept does not recognize single-parent households.
3. Male Head-of-Household. Until recently, in official U.S. Census Bureau designations, the age-of-head-of-household statistic employed was that of the male. In terms of childbearing, however, it is the age of the woman in the household that is important. Children have a significant effect on the family's purchasing behavior and therefore, their presence is of interest to consumer researchers. Because children are an important determinant of life cycle stage, FLC stage classified by male age-of-head-of-household tends to distort significant age milestones. For example, most experts agree that a woman 35 years of age or older faces significantly higher risk of bearing a defective child than does a woman under 35 years of age. (Carro, 1976; Daniels and Weingarten, 1979).
In short, the family life cycle concept is potentially useful for consumer analysis. The authors believe that its relatively sparse utilization in recent years can plausibly be attributed to distortions caused by minimizing the variety of roles open to women (and men) in the contemporary family. It is possible to correct this distortion by redefining the FLC.
REDEFINING THE FAMILY LIFE CYCLE CONCEPT
Other researchers (Norton, 1974; Glich 1977; Murphy and Staples, 1979) have offered similar proposals, the last being perhaps the most comprehensive and well-known. The present work was begun independently of Murphy and Staples; however, similarities do exist. The authors' redefined family life cycle concept, like that of Murphy and Staples, has more categories than does the traditional concept as defined by Wells and Gubar (1966). Therefore, the proposed concept will be explained in relation to the Murphy and Staples "modernized" FLC and the Wells and Gubar "traditional" FLC.
Family life cycle stages are based upon three characteristics (1) age, (2) marital status, and (3) ages and presence/absence of children in the household. There is considerable discussion today about "proper" age categories. The proposed family life cycle has three categories: young (under 35 years of age), middle-aged (ages 35 to 64 years of age), and elderly (over 64). These age categories are based upon age of the female in the household (if there is one). As noted above, thirty-five seems to be the age at which childbearing becomes more risky. It is recognized that these age categories are somewhat arbitrary; however, the same categories are used by Murphy and Staples.
The concept of "marriage" is also more complex than first appears. In the proposed FLC redefinition, a single person is anyone presently unmarried regardless of past marital status. The "married" category includes any couple residing together who intend to share a long-term relationship. Thus roommates are not "married." On the other hand, a marriage license is not required. A man and woman living together would be considered married if they established or intend to establish a household. This is a significant change from the traditional family life cycle concept, but it reflects modern society. For example, surveys in California reveal that there are more people in the 21-to-30 age group in the state living together than are actually married (Time, 1977). The traditional concept would classify each of these couples as two bachelors rather than recognizing their "married" status. In fact, a homosexual couple would also be "married" within the context of the present discussion if they had established or intended to establish a relationship expected to be permanent.
This definition of "single" and "married" also differs from the Murphy and Staples modernized FLC categories. They recognize divorced individuals with and without children, but not widowed or single individuals with and without children. There is no reason to believe that the consumption behavior of a widowed person differs significantly from that of a divorced individual. Murphy and Staples justify the exclusion of individuals who remain single throughout their lives by stating they "are by definition no: included in forming a family." But these individuals represent consumption units, and therefore should be of interest to consumer behaviorists. Further, if "young single," "middle aged divorced without children," and "older unmarried" are recognized as valid family life cycle stages, individuals who remain single throughout their lives should also be included.
In the redefined family life cycle, as in its traditional and modernized counterparts, the presence or absence of children is important. Family life cycle stages are distinguished by the age of the youngest child: a child under six years of age and therefore not in school has very different consumption (and other) requirements than does a child of six or older. The age of children is also a determinant of the amount of freedom enjoyed by mothers, e.g. , it is easier for a mother to work if all her children are in school.
In short, the proposed family life cycle is based upon chronological age (of the female member of the household if appropriate) interrupted by two types of critical points. The first type is marriage and separation by divorce or death of one member of the household. The second is the arrival of the first child and the departure of the last child. Thus, the redefined family life cycle accounts for four household types over three age groups through a person's lifetime: the one adult household, the two adult household, the two adult plus children household, and the one aduLt plus children household.
These household types are traced through the age groups of young, middle-aged, and elderLy, with the following modifications. In the last age group, children tend to leave home so that only the one adult and two adult households are represented in this age group. In the young age group, the two-parent household is divided into Full Nest I (youngest child under six years of age) and Full Nest II (youngest child six years or over). This division also applies to the single parent category. Similarly, in the middle-age group the two parent family is divided into Delayed Full Nest (middle-aged couples with youngest child under six) and Full Nest III (all children six years of age or older). The Delayed Full Nest accounts for an increasing number of couples who wait to have children. The Empty Nest and Solitary Survivor categories can be divided to account for households with at least one working member and those in which all are retired. The result is a total of thirteen categories in the proposed redefined family life cycle (as compared with the seven [or nine if retirement is considered] categories in the traditional FLC). Figure I summarizes the various stages.
The critical points representing major changes in a person' s life (and therefore his/her behavior as a consumer and .n other roles), have been superimposed upon a chronological age base. The critical points are marriage and divorce/ death, and addition/departure of children. The horizontal lines indicate passage of time, while the vertical lines indicate transition from one household type to another. Only the more common paths are shown. For example, it is possible for a single person to have a child while never having been married. If this event occurred, this individual would move from a bachelor stage to a single parent stage. For simplicity's sake, this line was not included. Other unusual family life cycles could be followed that are not expressly drawn in this figure. However, the stages do exist to classify almost all individuals and households.
COMPARISONS WITH OTHER FLC DEFINITIONS
As noted above, other researchers have attempted FLC redefinition. Table 2 was prepared to facilitate comparison of the present approach with two oft-cited alternatives: Murphy and Staples (1979) and Wells and Gubar (1966). Perhaps the major difference between the present redefinition of the family life cycle and both of the alternatives is that the present approach strives to classify all households, e.g., to remove the "other" category which can distort analysis. The present approach partitions the set of all households along three dimensions: (1) three mutually exclusive age categories, (2) whether or not there is more than one adult in the household, and (3) whether or not there are children. Thus, this approach strives to be mutually exclusive and collectively exhaustive. Of course, exceptions will always exist, such as young children residing with grandparents, but the redefined FLC cycle does take into consideration a larger percentage of households than do either of the alternatives discussed. Table 2 displays the data.
THE REDEFINED FAMILY LIFE CYCLE
Using the same adjustments on the data as do Murphy and Staples, the redefined FLC reduces the "other" category to 2.8 per cent. This improves the concept by reducing the unclassified group from 1/6 of the total to 1/30 - a fivefold reduction to a negligible fraction of the population. Recalculation of the proportions when 1980 census data become available is not expected to significantly alter the magnitude of the improvement.
This logical classification scheme is made possible by a rather broad -- perhaps controversial -- definition of "family". The authors recognize that including in the definition unmarried cohabitants, particularly those of the same sex, is a variance with the traditional view of the conjugal family, and may even be offensive to some in contemporary society. But this definition does reflect the realities of modern times. Much of consumer behavior is household-oriented. Thus, it is necessary for research purposes to eschew normative judgments and define family life cycle stages so that all households in society are included.
The obvious next step in this research is empirical testing Although many firms routinely use various combinations of the variables which define the FLC (cf. Reynolds and Wells, 1977), the authors do not agree with the approach suggested by Derrick and Lehfeld (1980). Simply collecting data and searching them for empirical relationships will not suffice. Empirical work should be guided by conceptual models, so that theory can be developed.
The family life cycle, as redefined, can perhaps provide the foundation for theory development in consumer behavior. To do so, a number of methodological issues must be addressed. The authors hope that conference discussion and/ or reader comment will help them and stimulate others to empirically examine the redefined family life cycle.
COMPARISONS OF FAMILY LIFE CYCLE DEFINITIONS
(June 16, 1980), "A Family Meeting Turns into a Feud," Report on the White House Conference on the Family. Newsweek, 31.
(June 16, 1980), "All in the Family," Report on the White House Conference on the Family, Time, 31.
Barton, S. G. (1955), "The Life Cycle and Buying Patterns" in Consumer Behavior, Vol. II, ed. Lincoln H. Clark, 53-7.
Bernard, Jessie (1975), "Notes on Changing Life Styles" 1970-74 Journals of Marriage and Family, 37, 582-600.
Bomball, Mark R., Walter, J. and Pursell Donald E. (1975), "Forecasting stage 2 of the Family Life Cycle," Journal of Business, 48, 65-73.
Carro, Geraldine (March, 1976), "Childbirth after 35: Cutting the Risks," Ladies Home Journal, 93, 108.
Clark, Lincoln H. ed. (1955), Consumer Behavior, Vol. II, (New York, New York University Press).
Cox, Eli P. (1975), "Family Purchase Decision Making and the Process of Adjustment," Journal of Marketing Research, 12, 189-195.
Daniels, Pamela and Weingarten, Kathy (Spring, 1979), "A New Look at the Medical Risks in Late Childbearing," Women and Health, 4, 76-91.
Danziger, Carl and Greenwald, Matthew (1977), "An Overview of Unmarried Heterosexual Cohabitation and Suggested Marketing Implications" in Advances in Consumer Research, Vol. 4, ed. William D. Perreault, Jr. (Atlanta- Association for Consumer Research, 324-30.
Davids, Leo (October, 1971), "North American Marriage: 1990" The Futurist, 5. 190-4.
Derrick, F. W. and Lehfeld, A. R. (September, 1980), "The Family Life Cycle: An Alternative Approach," Journal of Consumer Research, 214-7.
Engel, James F. Kollat, David T. and Blackwell, Roger D. (1978), Consumer Behavior (Hinsdale, IL: Dryden Press.
Feldman, Harold and Feldman, Margaret (1975), "The Family Life Cycle: Some Suggesting for Recycling," Journal of Marriage and Family, 37, 277-84.
Fisher, Janet A. (1955), "Family Life Cycle Analysis in Research on Consumer Behavior," in Consumer Behavior, Volume II, ed. Lincoln H. Clark (New York: New York University Press, 28-35.
Fritzsche, D. J. (May, 1981), "An Analysis of Energy Consumption Patterns by Stage of Family Life Cycle," Journal of Marketing Research, 227-32.
Glick, Paul C. (April, 1947), "The Family Life Cycle," American Sociological Review, 12, 164-74.
Glick, Paul C. (1977), "Updating the Life Cycle of the Family," Journal of Marriage and the Family, 39. 5-13.
Glick, Paul C. and Norton, Arthur J. (1977), "Marrying, Divorcing, and Living Together in the U.S. Today," Population Bulletin, 32, (Washington, DC: Population Reference Bureau, Inc.
Glick, Paul C. Parke, Robert Jr. (1965), "New Approaches in Studying the Life Cycle of the Family," Demography 2, 187-202.
Gove, Walter R., Grimm, James W., Motz, Susan C. and Thompson, James D. (1974), "The Family Life Cycle: Internal Dynamics and Social Consequences," Sociology and Social Research, 57, 182-95.
Haley, Russell I. (July, 1968), "Benefit Segmentation: A Decision-Oriented Research Tool," Journal of Marketing, 32, 30-5.
Hisrich, Robert D. and Peters, Michael P. (July, 1974), "Selecting the Superior Segmentation Correlate," Journal of Marketing, 38, 60-3.
Jain, Subhash C. (1975), "Life Cycle Revisited: Applications in Consumer Research," Advances in Consumer Research, Vol. 2, ed., Mary Jane Schlinger, Association for Consumer Research, 39-49.
Kerckhoff, Alan C. (1976), "Patterns of Marriage and Family Formation and Dissolution," Journal of Consumer Research, 2, 261-75.
Lansing, John B., Kish, Leslie (1957), "Family Life Cycle as an Independent Variable," American Sociological Review, 22, 512-9.
Lansing, John B. and Morgan, James N. (1955), "Consumer Finances Over the Life Cycle" in Consumer Behavior, Vol. II, ed. Lincoln H. Clark, (New York: New York University Press, 36-51.
Linden, Fabien (November, 1977), "The Six Ages of Economic Man," Across the Board, Vol. XIV, No. 11, The Conference Board.
Loomis, Charles P. (1936), "The Study of the Life Cycle of Families," Rural Sociology, 1, 180-99.
Miller, Donald L. (1955), "The Life Cycle and the Impact of Advertising," in Consumer Behavior, Vol. II, ed. Lincoln H. Clark, (New York: New York University Press 61-65.
Murphy, Patrick E. and Staples, William A. (1979), "A Modernized Family Life Cycle," Journal of Consumer Research, 6, 12-22.
Norton, Arthur J. (1974), "The Family Life Cycle Updated: Components and Uses," in Selected Studies in Marriage and the Family, ed., Robert F. Winch and Graham B. Spencer, (New York: Holt, Rinehart and Winston, 162-170.
(October, 1980), "One Parent Families: Going It Alone Creates a New Market," Sales and Marketing Measurement, 6, 47.
Oppenheimer, Valerie R. (1974) "The Life Cycle Squeeze: The Interaction of Men's Occupational and Family Life Cycles," Demography, 11, 227-45.
Parke, Robert Jr. and Glick, Paul C. (1967) "Prospective Changes in Marriage and the Family," Journal of Marriage and the Family, 29, 249-56.
Reynolds, Fred D. and Neter, John (1979), "Age Classification," Marketing Science Institute.
Reynolds, Fred D. and Wells, William D. (1977)s Consumer Behavior, (New York: McGraw-Hill Book Company).
Rich, Stuart U. and Jain, Subhash C. (1968), "SociAl Class and Life Cycle as Predictors of Shopping Behavior," Journal of Marketing Research 5, 41-9.
Rollins, Boyd C., and Feldman, Harold (February, 1970), "Marital Satisfaction Over the Family Life Cycle," Journal of Marriage and the Family, 32, 20-8.
Satow, Ray (1977) "Some Comments on Changing Life Styles Among Single Young Adults," in Advances in Consumer Research, Vol. 4, ed. William D. Perreault, Jr., (Atlanta: Association for Consumer Research, 324-9.
Spanier, Graham B., Lewis, Robert A. and Cole, Charles (1975), "Marital Adjustment Over the Family Life Cycle: The Issue of Curvilinearity," Journal of Marriage and the Family, 37, 263-75.
(January, 1977) Time, 10, 39.
(1979), "The American Family in Transition," A Collection of Articles Reprinted from Psychology Today, (New York: Ziff-Davis Publishing Company).
Trost, Jan (Spring, 1974) "This Family Life Cycle--An Impossible Concept?" International Journal of Sociology of the Family, 37-47.
Uhlenberg, Peter (1974), "Cohort Variations in Family Life Cycle Experiences of U.S. Females," Journal of Marriage and the Family.
(May, 1973), U. S. Bureau of the Census, Family Composition, Subject Report of the 1970 Census, Washington, DC: U.S. Government Printing Office.
(1972), U.S. Bureau of the Census, Marriage, Divorce, and Remarriage by Year of Birth: June, 1971. Current Population Reports: Population Characteristics, Series P-20, No. 239, (Washington, DC: U.S. Government Printing Office).
VanDusen, Roxann A, and Sheldon, Eleanor B. (February, 1976), "The Changing Status of American Women: A Life Cycle Perspective," American Psychologist, 106-16.
Wells, William C. and Gubar, George (1966), "Life Cycle Concept in Marketing Research," Journal of Marketing Research, 3, 355-63.
Wortzel, Lawrence H. (1977), "Young Adults: Single People and Single Person Households," in Advances in Consumer Research, Vol. 4, ed. William D. Perreault, Jr., (Atlanta: Association for Consumer Research. 321-9.