Advances in Consumer Research Volume 2, 1975 Pages 71-80
FAMILY ROLE STRUCTURE AND HOUSING DECISIONS
Donald J. Hempel, University of Connecticut
[This research was partially supported by funds from the University of Connecticut Research Foundation and the Marketing Department at the University of Lancaster, England. This assistance is gratefully acknowledged.]
[Donald J. Hempel is Professor of Marketing and Acting Head of the Marketing Department at the University of Connecticut.]
This study explores the determinants and effects of family role structures in house buying decisions. It extends the earlier work of the author by examining the relative predictive value of selected family characteristics which other studies have identified as important determinants of role structure. It deals with the question of whether it makes any difference how roles are allocated within the family by analyzing the correlation of role structure with decision process variables. Comparisons are presented of both husband and wife role perceptions in two different cultural settings.
Family roles in buying decisions have received a great deal of attention from consumer researchers in the past year (e.g., Ferber and Lee, 1974; Davis and Rigaux, 1974; Hempel, 1974; Cunningham and Green, 1974). The empirical basis for developing models of family buying behavior is expanding rapidly, and there have been several recent efforts to formulate a theoretical framework for incorporating this information (Sheth, 1974; Granbois, 1972). Much of the evidence is available only in fragments however, and there is need for a more integrated coverage of family buying decisions as a behavioral system. More specifically, empirical research is needed concerning the relationship among three sets of variables: (1) the patterns of individual responsibility and influence within the decision-making unit--i.e., role structures; (2) predictors of the role structures which are likely to operate in specific buying situations--i.e., determinants; and (3) implications of different role allocations for the economic behavior of the household--i.e.. effects.
Despite the extensive literature dealing with family decision making, there are significant limitations to this information base. As researchers in other disciplines have noted (e.g., Safilios-Rothschild, 1970), the available research findings concerning family role differentiation are too often based upon simple measures of general influence derived from the perceptions of wives. Many of the studies dealing with determinants of role structure were conducted more than a decade ago. Some recent studies have raised concerns about the reliability of this evidence for developing current theories of family buying decisions (e.g., Cunningham and Green, 1974; Ferber and Nicosia, 1972; and Davis, 1971). The growing emphasis upon practical implications of consumer research has also increased sensitivity to the fact that relatively little is known about the relationship of marital roles to other decision process variables (Sheldon, 1973).
RELEVANT RESEARCH ON FAMILY DECISION MAKING
Much of the empirical research concerning family buying decisions has been summarized in a series of review articles published within the last three years (Ferber, 1973; Granbois, 1972; and Sheth, 1971). It is sufficient to note here some of the more recent findings concerning husband-wife influence in family housing decisions. Home buying decisions receive special attention because they are considered to be a theoretical extremity of the "degree of deliberation" continuum on which a typology of buyer behavior might be constructed.
There is substantial evidence that the perceptions of both husband and wives should be considered in efforts to explain or predict family buying decisions (Hempel, 1974; Ferber, 1973; Davis, 1970). In a recent study of the asset accumulation behavior of young couples, Ferber and Nicosia (1972) stress the need to consider both the separate influences of each spouse and how they interact. Their findings concerning real estate investments (including home ownership) indicate that husband and wife influences may operate in opposite directions, perhaps reflecting intrafamily conflict on spending priorities. For example, attitudes toward the priority of savings were significantly related to real estate investment behavior in all but one instance, but the regression coefficients for husband and wife had opposite signs in every instance'
One of the more interesting developments in recent studies of family buying decisions is the growing conviction that husband-wife roles vary with the type of decision and with the product category (Ferber and Nicosia, 1972; Davis, 1970; Jaffee and Senft, 1966). The marital role allocations by decision area for 25 different product categories were examined by Davis and Rigaux (1974) in their study of Belgium couples. They found that housing decisions were highly syncratic across all three stages of problem recognition, search, and final decision. Contrary to the other products considered, housing decisions did not become less specialized between the initial and final stages. They reported little variability in role consensus among the three phases of the decision process for most products, but again housing was an exception. The evidence presented shows relatively high intrafamily role consensus for housing decisions in the problem recognition and final decision stages, while agreement at the search stage was substantially lower (81%, 82%, and 60% respectively). These findings support treatment of housing as a special type of purchase decision in the development of a typology for marital role differentiation. The position of housing decisions at the center of the "syncratic" classification in Davis and Rigaux's feasibility triangle suggests that it might be useful to consider this product class as particularly representative of the role category.
The implications of role structure for buying behavior were considered in a recent study by Ferber and Lee (1974). They found that families who assign a high priority to savings in the form of real estate are more likely to allocate decision making responsibility (i.e., the "family financial officer" role) to one member rather than both. Their findings indicate that the probability of buying a house during the early years of marriage was greater when the husband was the dominant decision maker. This tendency toward specialization in the financial management roles among young home buyers may encourage some compensatory allocation of other roles in the home buying decision process.
Most of the empirical studies concerning role allocations in the purchase of major durables indicate that joint decision making is the modal case. One very recent study suggests that shared decisions are becoming more common for these product categories, perhaps as a consequence of major social changes (Cunningham and Green, 1974). As the summary of findings presented in Table 1 evidence, both of these generalizations appear to be appropriate for housing decisions. The apparent trend toward the merging of husband and wife roles may be affected by respondent efforts to present family role distributions which appear equitable in an environment of growing social concern for equality. Some evidence of a compensating pattern of role allocations was presented by Hempel (1974). Efforts to characterize a complicated decision process in a general statement of husband-wife influence may represent attitudes about how roles should be allocated as much as how the roles were actually distributed. To the extent that this response behavior is associated with a convergence of future purchase behavior toward the attitude, the net effect will be to improve the predictive value of the data. The measures of role structure therefore may have desirable reliability qualities even though they might be lacking validity.
Despite some possible confounding of attitude and behavior in the measures of role structure, they can be considered potentially useful bases for market segmentation. Evaluation of this potential requires more information about the linkage between role structure and more readily observable characteristics of the household. It also requires information about the implications of role structure for other aspects of consumer decision processes, such as search, purchase behavior, postpurchase satisfaction. The results presented here are a preliminary attempt to address these needs.
The data for this study were obtained from two surveys of recent home buyers conducted during the summers of 1968 and 1971. Both investigations incorporated probability samples of households who purchased either a new or previously occupied house and recorded their ownership during the first six months of the study year. Personal interviews and mail questionnaires were used to obtain information from both husbands and wives in 206 households from the Hartford metropolitan area of Connecticut, and 317 households in the Preston-Lancaster area of Northwestern England. The bases for analysis of the mail-questionnaire data presented here was reduced considerably by nonreturns and the elimination of all observations with missing information. Details concerning the research design and questionnaires have beer. published elsewhere (Hempel, 1970).
DETERMINANTS OF FAMILY ROLE STRUCTURE
The bulk of research on the determinants of family role allocations in purchase decisions has focused upon the socioeconomic characteristics of the household. There are several general hypotheses which have been advanced to explain the extent of joint decision-making in buying decisions: (1) the greater the resources, contribution, and status of the individual, relative to his or her spouse, the greater the influence--e.g., education, occupational prestige, income contribution, and employment status; (2) more extensive experience as a decision-making unit is associated with a reduction in joint-decisions--e.g., age, years married, later stages of family life cycle, and income; and (3) greater "connectedness" of the family's social network and social distance from the middle class are inversely related to the degree of joint-decision making. Discussions of the empirical studies which serve as a basis for these generalizations can be found in several recent publications (Ferber, 1973; Engel, Kollat, and Blackwell, 1973; and Granbois, 1972.
MAJOR STUDIES OF FAMILY ROLE STRUCTURE IN HOUSING DECISIONS
Table 2 shows the correlation between four measures of marital roles in housing decisions and several sets of determinants. The first two role measures are dichotomized variables which represent the classification scheme proposed by Herbst (1954) applied to the housing decision patterns described by Hempel (1974). The third role measure represents the extent of joint decision making as the total number of shared decisions reported by either spouse. The overall measure of the husband-wife influence perceived by each spouse was constructed by aggregating the decision scores (l=wife, 2=joint, 3=husband) over five decision areas (neighborhood, style, price, timing of purchase, and mortgage source). More elegant statistical results could be developed by calculating biserial correlations for the first two role structure measures, but the product moment correlations presented here provide a consistent and useful first approximation for evaluating the relative importance of different determinants.
If predictive value is used as the criterion for judging importance, education and occupational status appear to be the most significant determinants of role structure in house buying decisions. The best single predictor across role measures and markets was the ratio of husband/wife occupational status. This indicator of relative resources tended to be a better predictor of the role perceptions reported by husbands than those of wives. The husband's education also was closely related to the role perceptions of both spouses in Connecticut, but the educational variable was considerably less significant in England. It is noteworthy that the wife's occupational status appears to have much more influence upon the husband's perceptions of role structure than her own. The measures of the wife's employment status, particularly the number of hours worked, were also correlated with the husband's perceptions, but these variables were not significant determinants for the wives
The direction of relationship was generally consistent with that found in previous studies. Age, years married, occupational status of the wife, and her employment status were all inversely related to husband dominance of the buying decisions. Education, the husband's occupational status, and family income were positively correlated with the measures of husband dominance. The curvilinear hypothesis that joint decision making is less common in the lower and upper social classes was not confirmed for the lower class. This apparent discrepancy can be reconciled with the recognition that few members of the lower class are likely to appear in a sample of home buyers. The "lower" educational and occupational groups in this study can more appropriately be regarded as members of the lower-middle social classes.
In general, the husband dominant patterns can be predicted more readily from the determinants considered here than either the syncratic classification or the total number of joint decisions. A multiple regression analysis of the different role structure measures indicates that five or six of the determinants can explain" about 25% of the variance in either dominance measure, whereas they will account for less than 20% of variance in the other two measures. In either case, there is need for consideration of other determinants of role structure in addition to those presented here.
POSSIBLE DETERMINANTS OF FAMILY ROLE STRUCTURE IN HOUSE BUYING DECISIONS AMONG CONNECTICUT (N=132) AND ENGLISH (N=106) HOUSEHOLDS
The selection of appropriate role structure measurements is complicated by the finding that the predictive value of the determinants appears to vary by culture and sex of the respondent. In Connecticut, the family characteristics generally predict the husband's perceptions of role structure much better. The relative importance of determinants varies less by sex in England, but the role structure measures also are less predictable. Perhaps this pattern reflects a cultural setting in which the marital role perceptions and expectations are more independent of the socioeconomic characteristics which often are used in our culture to characterize the family as a decision making unit.
DECISION PROCESS IMPLICATIONS OF ROLE STRUCTURE
Does the pattern of role allocations within family buying decisions influence other aspects of buyer behavior? This question has not been investigated in much detail, but one recent study by Ferber and Lee (1974) suggests that such effects may be appreciable. The linkages of role structure to other behavioral variables should be a basic concern of both the pragmatist involved in strategy formulation and the theoretician oriented toward model building.
Table 3 presents the correlation of the role structures perceived by husbands and wives with measures of search behavior, expenditures and postpurchase satisfaction. These results indicate that the main effects of role allocations in housing decisions may show up in the interrelationships with expenditures for other goods. The other decision process variables were related to role structure in some cases, but the correlations were much less significant.
The amount which the family spent for complementary goods (furniture, appliances, decorating, etc.) during the two month period centered around the occupancy of their house was related to all of the role structure measures for English households. In Connecticut, this expenditure behavior emerges only in the wife dominant and syncratic households. A prevalence of joint decision making was associated with lower expenditures for complementary goods in both countries, as was the wife dominant case in England. Husband dominance and autonomic role allocations were directly related to the level of expenditures among English families. This pattern suggests that the wife's participation in house buying decisions may generate some tradeoffs in the buying decisions for other products. More role specialization in housing decisions may occur because the family is concurrently engaged in buying decisions for other major products.
The time required for the decision process (i.e., the decision span) was positively correlated with the role perceptions of both spouses in the husband dominant English households, but only with the wives' perceptions in similar Connecticut households. This pattern is repeated for the syncratic households, but the direction of relationship is negative. These results seem to suggest that when the wife is jointly involved in the house buying decisions, less time is required for the decision process. The impact of marital roles upon the extent of search (i,e., the number of houses inspected) appears to be negligible in all but one case--wife dominant English households were likely to examine more housing alternatives.
Satisfaction with the house purchased was not consistently related to family role structure. There is some evidence that family satisfaction is greater if neither spouse is perceived as dominant. The perception of role specialization (i.e., autonomic structure) among Connecticut wives was associated with lower levels of postpurchase satisfaction.
CORRELATION OF ROLE STRUCTURE MEASURES WITH SELECTED DECISION PROCESS VARIABLES
The results of this study indicate that role structure in family house buying decisions is affected by household characteristics which are often used as bases for market segmentation. Education, occupational prestige, and the wife's employment status appear to be the most useful predictors. There was considerable variation in the relative importance of different determinants across sex, culture, and role structure measure.
The family characteristics and other influences upon role allocations considered here are an insufficient set of role structure determinants. Other variables such as attitude, experience, and decision making constraints (e.g., time pressure) should be incorporated in future studies to provide a more complete basis for models of family buying decisions. Measures of the attitude structure and personal involvement of each spouse, and their interplay with product experience variables appear to be particularly attractive areas for further research on housing decisions.
The evidence concerning role structure effects was rather disappointing. Though empirical research concerning these relationships is very scanty, the available conceptual models suggest a stronger association with search behavior and postpurchase satisfaction than the data indicate. These results suggest that role structure has relatively little direct influence upon other types of behavior within the decision process. The findings about the interrelationships between roles in housing decisions and expenditures for related goods are more promising, and they will be examined further for insights into the interdependency among buying decision processes.
Blood, R.O. & Wolfe, D.M. Husbands and wives The dynamics of married living. Chicago: The Free Press, 1960.
Cunningham, I. 5. M. & Green, R. T. Purchasing roles in the U.S. family, 1955 and 1973. Journal of Marketing, 1974, 38(4), 61-64.
Davis, H. L. Measurement of husband-wife influence in consumer purchase decisions. Journal of Marketing Research, 1971, 8, 305-312.
Davis, H. L. Dimensions of marital roles in consumer decision making. Journal of Marketing Research, 1970, 7, 168-77.
Davis, H. L. & Rigaux, B. P. Perception of marital roles in decision processes. Journal of Consumer Research, 1974, 1(1), 51-62.
Engel, J. F., Kollat, D. T., & Blackwell, R. O. Consumer behavior. (2nd Ed.) New York: Holt, Rinehart, 1973.
Ferber, R. Family decision making and economic behavior. In E. B. Sheldon (Ed.), Family economic behavior: Problems and prospects. Philadelphia: J. B. Lippincott Co., 1973.
Ferber, R. & Nicosia, F. Newly married couples and their asset accumulation decisions. In B. Strumpel, J. Morgan & E. Zahn (Eds.), Human behavior in economic affairs. San Francisco: Jossey-Bass, 1972.
Ferber, R. & Lee, L. C. Husband-wife influence in family purchasing behavior. Journal of Consumer Research, 1974, 1(1), 43-50.
Granbois, D. H. Decision processes for major durable goods. In G. Fisk (Ed.), New essays in marketing theory. Boston: Allyn and Bacon, 1972.
Hempel, D. J. A comParative study of the home buying process in two Connecticut housing markets. Storrs, Connecticut: Center for Real Estate and Urban Economic Studies. The University of Connecticut. 1970.
Hempel, D. J. Family buying decisions: A cross cultural perspective. Journal of Marketing Research, 1974, 11, 295-302.
Jaffee, L. J. & Senft, H. The roles of husbands and wives in purchasing decisions. In L. Adler & I. Crespi (Eds.), Attitude research at sea. Chicago: American Marketing Association, 1966.
Sheldon, E. G. (Ed.), Family economic behavior: Problems and Prospects. Philadelphia: J. G. Lippincott Co., 1973.
Sheth, J. N. A theory of family buying decisions. In P. Pelleman, (Ed.), InSights in consumer and market behavior. Belgium: Namus University, 1971.
Sheth, J. N. Models of buyer behavior: Conceptual quantitative, and empirical. New York: Harper & Row, 1974.