Family Decision Making Role Perceptions Among Mexican-Americanand Anglo Wives: a Cross Cultural Comparison

ABSTRACT - This investigation compared Mexican-American and Anglo wives' perceptions of the manner in which their families make purchasing decisions involving major durables. Analysis of these perceptions indicate that Mexican-American families tend to be more husband dominant than Anglo families where the purchase of major durables are concerned. Anglo families also appear to engage in more joint purchase decision making. These findings present interesting and potentially valuable implications for marketers wishing to optimize their performance in an increasingly heterogenous marketplace.

Citation:

Giovanna Imperia, Thomas C. O'Guinn, and Elizabeth A. MacAdams (1985) ,"Family Decision Making Role Perceptions Among Mexican-Americanand Anglo Wives: a Cross Cultural Comparison", in NA - Advances in Consumer Research Volume 12, eds. Elizabeth C. Hirschman and Moris B. Holbrook, Provo, UT : Association for Consumer Research, Pages: 71-74.

Advances in Consumer Research Volume 12, 1985      Pages 71-74

FAMILY DECISION MAKING ROLE PERCEPTIONS AMONG MEXICAN-AMERICANAND ANGLO WIVES: A CROSS CULTURAL COMPARISON

Giovanna Imperia, Loyola University-New Orleans

Thomas C. O'Guinn, University of Illinois at Urbana-Champaign

Elizabeth A. MacAdams, University of Illinois at Urbana-Champaign

ABSTRACT -

This investigation compared Mexican-American and Anglo wives' perceptions of the manner in which their families make purchasing decisions involving major durables. Analysis of these perceptions indicate that Mexican-American families tend to be more husband dominant than Anglo families where the purchase of major durables are concerned. Anglo families also appear to engage in more joint purchase decision making. These findings present interesting and potentially valuable implications for marketers wishing to optimize their performance in an increasingly heterogenous marketplace.

Marketers have long recognized the potential influence of such factors as social class, ethnicity, and culture on buyer behavior. Yet, investigations of these factors have generally taken a largely macro perspective, and have only recently turned significant attention to the most fundamental of all consumption groups, the family. This is indeed an oversight since the family represents the primary environment in which roles related to these very important socioeconomic factors are learned and expressed. One of their most significant forms of expression is buyer behavior.

While the influence of culture and subculture on family purchasing has received some attention (Douglas 1976; Safilios-Rothschild 1969), it remains, however, a largely unexplored area of buyer behavior. This is particularly true in the case of comparisons involving subcultures. There, the need for investigation is acute. The manner and degree to which subcultural identification influences family decision making roles and processes has simply not been the subject of significant empirical research.

The investigation reported here sought to compare the manner in which Mexican-American and Anglo wives perceive family decision making roles when the purchase of major durables is involved. It was, therefore, a buyer behavioral manifestation of Mexican-American ethnicity expressed within the family decision making role structure that this investigation sought to demonstrate by means of a direct comparison with the dominant Anglo culture.

PREVIOUS RESEARCH

While the study of family decision making has come to represent a significant research area in buyer behavior, there has been little effort to explore subcultural differences in family decision making. This is unfortunate in that the expression of so many of the characteristics and attributes of an ethnic subculture are mediated through the family (Mindel and Habenstein 1976). In many ways, families serve as a cultural reservoir of norms and values which may help ethnic groups to maintain a distinct physionomy, some elements of which may be expressed in terms or buyer behavior. This may very well be the case of family purchasing decision roles.

Previous research has indicated that the Mexican-American family has several traits in common with the traditional Mexican family. These include father dominance, masculine superiority, sharply defined separation of sexes, and submission to authority (Howard 1952; Madsen 1964; Padilla and Ruiz 1973). In comparison to the dominant Anglo culture of the United States, the Mexican-American subculture seems more patriarchial, and particularly resistant to family role deviance. Murillo (1971) has suggested that this may be due to the fact that Mexican-American Latin values are more clearly defined and behavioral patterns more closely adhered to than is common in the dominant Anglo culture. Even though many Mexican-Americans may be in the process of cultural assimilation, and are neither Anglo nor Mexican, one must still suspect a continued reliance on the parent culture. Such a reliance seems even more likely given the formal and institutionalized bilingual/bicultural policy of both the public (bilingual education) and private (bilingual mass media programming and advertising) sectors. This reliance may manifest itself in many ways, including buyer behavior, perhaps playing a very significant role in the manner in which families decide on the purchase of specific goods and services.

PRESENT STUDY

This investigation attempted to demonstrate hypothesized differences in the manner in which Mexican-American and Anglo wives perceive family decision making roles. Given a traditionally patriarchal family structure, one would suspect greater husband dominance within Mexican-American families than within Anglo families (H1), and conversely, fewer wife dominant decisions within Mexican-American families as compared to Anglo families (H2). Of course, one other likely effect of such an orientation would be the presence of significantly fewer joint decisions within Mexican-American families than within Anglo families (H3). These three hypotheses are formally presented below.

H1: Mexican-American wives will report significantly more husband dominant purchasing decisions involving durables than will Anglo wives.

H2: Mexican-American wives will report significantly fewer wife dominant purchasing decisions involving durables than will Anglo wives.

H3: Mexican-American wives will report significantly fewer joint purchasing decisions involving durables than will Anglo wives.

METHOD

In order to test these hypotheses, data were collected from a sample of Mexican-American and Anglo wives residing in a medium size city in the Southwestern United States. Data were gathered from 125 Mexican-American and 159 Anglo wives. One should remember that the focus of this investigation is the identification of cross cultural differences as perceived by the wives or the two respective cultures. Wives were chosen as the focus because they seem central to the differences which may exist between the two cultures, one patriarchal, the other less so. While it is generally considered preferable to collect data from both husband and wife, there is no reason to suspect the presence of any type of systematic error capable of either obscuring or producing the hypothesized subcultural differences.

Measures

Decision dominance was assessed by asking the respondents to identify the spouse who normally makes certain decisions regarding the purchase of four different categories of durables, or to identify those decisions that are best categorized as joint. The decision components represented three to six common and relevant dimensions of each purchase, eg., price, store, model, size, etc. The exact list varied with the specific item, and is presented in Table I. The four product categories were furniture, appliances, automobile, and house or apartment.

TABLE I

DECISION COMPONENTS

A full range of demographic measures was also collected in order to more fairly and comprehensively compare these two groups. Demographic measures included age, national origin, years at present address, employment status, occupation, husband's occupation, years of education for both husband and wife, income, home ownership, and number of children. Both a Spanish and an English language instrument were developed, back-translated, pre-tested, and checked for semantic congruity.

Sample

Data were collected via a multistage disproportionately stratified sample. A simple random sample of equivalent size would not have provided an adequate basis for determining whether emerging differences were due to ethnic origin or to some other socioeconomic factors. Factors such as income, education level, and employment status are often correlated with race, and for this very reason the sample was stratified by both income and race. A stratification by both seems more appropriate than a stratification by race alone since it increases the heterogeneity of the Mexican-American subsample and reduces its skewness toward lower income levels. It was assumed that differences in employment status and education are also reflected in income, thus eliminating the need for further stratification.

Stratification was accomplished by selecting census tracts in those areas where the association of race and income was at its lowest. One such tract was then selected within each of four income strata: less than $10,000 annual family income, $10,000 to $15,000, $15,001 to $20,000, and $20,001 to $30,000. In the next stage, city blocks were randomly selected within each of the census tracts. Finally, three respondents per block were systematically selected. Five-hundred questionnaires were originally distributed to 250 Mexican-American wives and to 250 Anglo wives. A total of 285 were returned in usable condition, thus yielding an overall response rate of 57.0 percent. Mexican-American wives returned 126 of the 250 questionnaires, or 50.4 percent. Anglo wives returned 158 or 63.2 percent of the questionnaires.

Respondents were contacted in their homes and asked to fill out a questionnaire at their convenience. Arrangements were then made to pick up the completed questionnaires. This drop-off technique incorporates the advantages of both personal interviews and self-administered questionnaires, and is relatively inexpensive. It also has the advantage of yielding a higher response rate than mail questionnaires, since it allows the researcher to select beforehand suitable respondents who are willing to complete the questionnaire (Stover and Stone 1978).

Analysis

The three interval level dependent measures were created by summing separately the number of husband dominant, the number of wife dominant, and the number of joint decisions for all the decision components (e.g., price, model, store, etc.) involved in the selection of the four durables. Separate univariate F-Tests were then used to compare the mean number of husband dominant, wife dominant, and joint decisions of the Mexican-American and Anglo groups. This exactly replicates the analytical procedure used by Green and Cunningham in their 1974 study of feminine role perceptions. In order to test for the possible non-independence of the dependent measures a three-way analysis of variance using race, income and employment status was also performed.

FINDINGS

The results indicate two important differences in the way in which Mexican-American and Anglo wives perceive decision making roles within their families when the purchase of major durables is involved. In the initial phase of the analysis, three univariate F-tests compared the mean number of husband dominant, wife dominant and joint purchasing decisions for Mexican-American and Anglo families. As Table II indicates, two of these differences were statistically significant (p < .05).

TABLE II

MEAN NUMBER OF PURCHASING DECISIONS FOR MEXICAN-AMERICAN AND ANGLO FAMILIES

In the first instance, the mean number of husband dominant durable product purchase decisions was significantly greater in the Mexican-American group than in the Anglo group, thus confirming H1. This, of course, indicates the perception of greater male dominance in Mexican-American families when the purchase of durables are involved. Table II also reveals a mean number of Mexican-American joint decisions significantly smaller than that of the Anglo group, thus indicating fewer shared decisions in Mexican-American families. This confirms hypothesis H3. There was, however, no statistically significant difference in the number of wife dominant decisions between the Mexican-American and Anglo groups. Hypothesis H2 was, therefore, rejected.

Due to the fact that income and employment status are often related to race, and that both can presumably affect family decision making, three separate three-way ANOVAs were performed, one for each of the three dependent measures. While income was a five level ordinal measure, race and employment status had only two levels. In the case of the latter, the two levels corresponded to employment of least 20 hours a week, or unemployed. Tables III and IV present the two analyses in which significant effects were found.

TABLE III

ANALYSIS OF VARIANCE

HUSBAND DOMINANT DECISIONS BY RACE, EMPLOYMENT STATUS AND INCOME

TABLE IV

ANALYSIS OF VARIANCE

JOINT DECISIONS BY RACE, EMPLOYMENT STATUS AND INCOME

Suspicions that the subcultural differences in family purchasing behavior observed in the initial analysis were simply a statistical artifact attributabLe to employment status and/or disparate income levels, were not borne out by this additional analysis. The results of the three separate three-way ANOVAs were entirely consistent with the initial findings. Just as before, the main effect or race was statistically lp < .05) significant when the number of husband dominant and joint decisions were considered as dependent measures. Likewise, no significant effects, main or interaction, were detected when the number of wife dominant decisions served as the dependent variable. Most important, however, was the fact that there were no statistically significant interaction effects, two-way or three-way, detected in any of the three analyses, thus confirming the relative independence of the dependent measures.

DISCUSSION AND IMPLICATIONS

The findings of this investigation revealed two major differences in the manner in which Mexican-American and Anglo wives perceive family decision making roles when the purchase of major durables are involved.

First, the findings clearly indicate a stronger pattern of husband dominance in Mexican-American families than in Anglo families. These findings also suggest that Mexican-American families engage in significantly less joint decision making than do Anglo families, at least when major durables are involved. In both cases, these cross cultural differences appear to be genuine and not merely the product of disparate income levels or differences in employment status. It should also be noted that the number of wife dominant decisions in both Mexican-American and Anglo homes was extremely low. Not only does this explain why H2 was not confirmed, but it also suggests that wife dominant decisions on major durables are simply rare in both cultures. The major difference between the two cultures is that more Anglo wives have reached equality in decision making, not that they have supplanted their husbands as the primary decision maker.

The findings seem to suggest several potentially valuable implications for marketers. Although our findings show that Hispanic husbands are more dominant in decision making than Anglo husbands, marketers should avoid believing in the stereotype of the autocratic Hispanic husband. For both Anglo and Mexican-American families, joint decision making was the dominant mode. Thus advertising and promotion for durables cannot effectively be geared to just one sex. However, an emphasis in copy which stresses points that are more important to one spouse than the other might vary for ads aimed at different cultural groups. For example, ads targeting Hispanics might place attributes which males race as important in the headline or earlier in the copy than chose rated highly by females. Such actions could help make advertising targeted to Hispanics more culturally relevant, and thus more effective.

While these findings are interesting and yield potentially valuable implications, they were derived from an investigation which was clearly exploratory in nature and simple in design. Future investigations should attempt to go beyond this basic conceptualization and methodology. They should attempt to determine how sex related differences among different ethnic groups vary through the steps of the decision process. For example, who first recognizes or articulates the need for I he product? Once the need is expressed, are there consistent sex differences in terms of the perceived importance of product attributes? Are there clear and consistent cultural differences at the actual time of purchase? We need to know if and how cultural differences affect each of these questions in order to effectively influence purchasing decisions among different cultural groups.

REFERENCES

Cunningham, Isabella C.M. and Robert T. Green (1974), "Purchasing Roles in the U.S. Family, 1955-1973," Journal of Marketing, 38, (Oct.) 61-81.

Douglas, Susan P. (1976), "Cross-National Comparisons of Consumer Stereotypes: A Case Study of Working and Nonworking Wives in the U.S. and France," The Journal of Marriage and the Family, 31 (May), 296-301.

Howard, R. G. (1952), "Acculturation and Social Mobility Among Latin Americans in Reseca City," Unpublished M.A. Thesis, University of Texas at Austin.

Madsen, W. (1964), The Mexican-American of South Texas, New York: Holt, Rinehart, and Winston.

Mindel, C. H. and R. W. Habenstein (1976), Ethnic Families in America, New York: Elsevier.

Murillo, N. (1971), "The Mexican-American Family," in N. N. Wagner and Haug, M. J., (eds.): Chicanos: Social and Psychological Perspectives, St. Louis: Mosby.

Padilla, A. M. and R. A. Ruiz (1973), Latino Mental Health, Baltimore: National Institute of Mental Health.

Safilios-Rothschild, Constantina (1969), "Family Sociology or Wives' Sociology? A Cross-Cultural Examination of Decision-Making," Journal of Marriage and the Family 31 (May), 269-301.

Stover, R. V. and W. J. Stone (1978): "Hand Delivery of Self-Administered Questionnaires," Public Opinion Quarterly, 42, 284-287.

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