Intergenerational Influences in Adult Buying Behaviors: an Examination of Moderating Factors

Susan E. Heckler, University of Michigan
Terry L. Childers, University of Minnesota
Ramesh Arunachalam, University of Minnesota
ABSTRACT - One interesting area of study which has received relatively little attention in the consumer behavior literature is the degree to which family influence carries over into our consumption activities as adults. Two studies have been conducted in an effort to begin to develop an understanding of the extent of intergenerational transfer (transmission of attitudes, values and behaviors from parents to children) and the impact of possible moderating factors in the process across a variety of choices. Hypotheses regarding intergenerational effects are developed from past consumer and sociological literature and are tested using data collected from two separate populations. Results of the two studies are presented, and the paper concludes by introducing a conceptual framework which integrates similarities and differences seen in the two samples.
[ to cite ]:
Susan E. Heckler, Terry L. Childers, and Ramesh Arunachalam (1989) ,"Intergenerational Influences in Adult Buying Behaviors: an Examination of Moderating Factors", in NA - Advances in Consumer Research Volume 16, eds. Thomas K. Srull, Provo, UT : Association for Consumer Research, Pages: 276-284.

Advances in Consumer Research Volume 16, 1989      Pages 276-284

INTERGENERATIONAL INFLUENCES IN ADULT BUYING BEHAVIORS: AN EXAMINATION OF MODERATING FACTORS

Susan E. Heckler, University of Michigan

Terry L. Childers, University of Minnesota

Ramesh Arunachalam, University of Minnesota

ABSTRACT -

One interesting area of study which has received relatively little attention in the consumer behavior literature is the degree to which family influence carries over into our consumption activities as adults. Two studies have been conducted in an effort to begin to develop an understanding of the extent of intergenerational transfer (transmission of attitudes, values and behaviors from parents to children) and the impact of possible moderating factors in the process across a variety of choices. Hypotheses regarding intergenerational effects are developed from past consumer and sociological literature and are tested using data collected from two separate populations. Results of the two studies are presented, and the paper concludes by introducing a conceptual framework which integrates similarities and differences seen in the two samples.

The family has long been identified as the primary socialization agent for each new generation. Included in that socialization process is the development of a large set of skills and knowledge relevant to acting as successful consumers in a complex marketplace. It is through the family that we first learn skills such as budgeting or bargaining, that we are first exposed to the huge variety of products available and that we first develop attitudes toward and preferences for those product ;. One interesting area of study which has received relatively little attention in the consumer behavior literature is the degree to which this family influence carries over into our consumption activities as adults.

The transmission of attitudes, values and behaviors from parents to children is generally termed intergenerational transfer. Although it might be expected that evidence of such transfer would be strong, past sociological research has demonstrated only a modest relationship between parent and child attitudes and values (McBroom, et al 1985). However, this sociological research examines the transmission of general social values and norms. Almost no research has been conducted which examines such intergenerational effects in a consumer setting. The purpose of the present study is to examine the extent of intergenerational transfer of brand and store choices across a wide variety of products, and to examine possible moderating variables which may affect the duration of intergenerational influences through adulthood

BACKGROUND

Consumer socialization has Seen defined as the processes by which young people acquire skills, knowledge, and attitudes relevant to their functioning as consumers in the marketplace (Ward 1974). Agents affecting these processes include family, peers, media and public institutions (e.g., government, schools). Very early discussions of consumption related socialization were offered by sociologists in discussions of the development of "conspicuous consumption" patterns. Speculations were offered, for example. that children learn "rational" aspects of consumption from parents, "expressive" aspects from peers and mass media, and broader, "social role" aspects from schools or government (Parsons, et al 1953). Unfortunately, these essays on socialization processes were not accompanied by any empirical examination of the phenomena being hypothesized.

More recent efforts by consumer researchers have empirically examined the socialization process, particularly regarding the development of consumer skills of children and adolescents (c.f. Ward and Wackman (1973); Moschis (1979)). However, relatively little research has focussed on the impact of intergenerational effects on consumer decision making (Wilkie 1987; Moschis 1985), and even less has examined the impact of these family influences on one's behaviors as an adult. Of particular interest is the development of consumer preferences 3>nd choice making skills during early adulthood. Research which has been conducted has examined a variety of moderating factors thought to impact the socialization process. Moschis and Churchill (1978), for example, concluded that the family's social class affects intergenerational transfer of consumer skills. They assert that adolescents from lower class families do not have the number of opportunities to participate in consumption decisions that middle and upper class children have, and additionally, lower class families may not engage in discussions relevant to consumer socialization as frequently as middle or upper class families. In another study examining socio-economic factors, Moschis, et al (1983) suggest that increased consumer knowledge is transferred in middle class families over those of other socio-economic status. Other studies have examined factors such as age (Moschis, et al 1986), the effects of parental communication styles (Moschis and Moore 1984: Moschis 1985), 3rd gender (Moschis, et al 1977; Moschis and Churchill 1979).

In none of these studies is an attempt made to assess the importance of intergenerational influences in explaining adult behavior, particularly early adult behavior. In fact, almost no research has examined the influence of parents' consumption decisions on subsequent choices made by their adult offspring for the same product. The research that has been conducted includes one study by Woodsen, Childers and Winn (1976) which showed that intergenerational influences were strong for decisions about insurance companies and identified age as a moderator of the effects. Another study found significant relationships between undergraduate college students and their parents regarding favorite stores, brand loyalty, opinion leadership and innovativeness (Arndt, 1971). While these studies have identified the presence of intergenerational transfer, certainly more effort is needed to clarify how intergenerational influences are utilized by adults in their decision making processes and for what marketing situations these influences are likely to be important. The purpose of the studies to be described below, is to begin to develop a broader understanding of the contexts in which intergenerational influences occur and of the factors moderating such influences.

HYPOTHESES

As indicated above, very little research has been done which examines the effects of intergenerational transfer on adult consumption behavior. However, in an extensive review and integration of research examining family communication patterns and consumer socialization of adolescents, Moschis (1985) sets forth a number of propositions describing potential intergenerational effects and moderators of such effects. While the propositions have been developed to explain adolescent behavior and attitude development, when combined with past research introduced above, they can serve as a basis for expectations regarding adult behavior. The propositions Moschis sets forth are too numerous to explore in a single study, and so the present research focuses on possible .moderating effects which may inhibit or enhance intergenerational transfer of preferences. In the section which follows, hypotheses are developed which describe the expected effect of a selection of these moderating factors on intergenerational carryover for product and store choices of adults. The factors have been selected primarily due to their presence in past studies of family decision making or in the conceptual framework offered by Moschis. The descriptive nature of the study is appropriate given the early stage of theoretical development which characterizes the intergenerational transfer effect in consumer decision making.

Marketing Variables

Past research suggests that intergenerational influence will vary for different types of products. For example, Moschis and Moore (1979) found that the perceived risk associated with a product-choice mediated the extent to which adolescents accepted parental guidance. Other research has suggested that intergenerational transmission of product preferences is likely to be greater and longer lasting for shopping goods and products of high perceived risk and weaker and of shorter duration for convenience and specialty goods (Woodsen, Childers and Winn 1976; Moschis 1985). These results lead to the hypothesis that:

H1a: Intergenerational influence will be greater for shopping goods than for convenience goods.

It would also be expected that store choice decisions would follow a similar pattern.

H1b: Store choices for shopping goods will demonstrate a stronger intergenerational transfer than stores al which convenience goods are found.

Age

Many studies have shown that age is an important moderating factor in the degree to which parents' preferences or attitudes impact on their children's choices (Vener 1957; Moschis et al 1977; Moschis and Moore 1979). Additionally, Woodsen, Childers and Winn (1976) found age to moderate intergenerational influences in adults' decisions about insurance. They report that while early adult insurance choices are strongly affected by parental choice, as the consumers get older, their decisions are less similar to that of their parents. Based upon their findings it is hypothesized that:

H2: Intergenerational influence will decrease with age for both product and store choices.

Gender

Researchers examining adolescent behavior have concluded that the degree of parental influence on consumer decisions is affected by the sex of the child. Because girls, especially teenagers, exhibit an earlier and relatively higher need for conformity to peer group norms, they are more likely to make decisions relevant to personal appearance independently of their family (Saunders, et al 1973; Moschis, et al 1977). As a result, less intergenerational influence is demonstrated for such products. This earlier pattern of consumer independence may also influence female decision strategies as they reach adulthood. Because they began making consumption decisions at an earlier age, intergenerational influences may be weaker generally for females than for males. Therefore it is hypothesized that:

H3: Intergenerational influence will be stronger for males than for females in both product and store selections.

Family Relationships

Moschis (1985) reports that families' communication structures have an effect on the types of parental influences demonstrated in adolescent behaviors. The relationships he discusses are quite interactive - involving variables such as frequency, content and structure of the communication, with respect to the congruency of gender between parent and child, in addition to o{her non-family influences. However, a general conclusion which can be inferred from his discussion is that families that demonstrate more frequent communication have a higher likelihood of displaying intergenerational transfer. One possible indicator of such communication and family environment is the level of family orientation expressed by the offspring. It seems plausible that when individuals place more importance on the family and on parental authority, intergenerational transfer will be longer lasting and will be demonstrated for a broader set of contexts. It is therefore hypothesized that:

H4 Increased intergenerational influence for both product and store choices will be demonstrated when family orientation is stronger.

Additional Offspring Characteristics

A number of other factors may influence the strength of intergenerational transfer, including the individual's education, income and marital status and whether the household contains an extended family. For example, it would be expected that as level of education increases, individuals acquire increasing amounts of exposure to various models of decision making, and to consumer education generally. As a result the person would be less likely to mimic parental choices and to demonstrate instead an independent style of decision making. It is thus hypothesized that:

H5: As education increases intergenerational influence for both product and store choices decreases.

Similarly, as an adult offspring becomes more financially secure, or if the offspring income exceeds that of the parent, it would be expected that consumer choices would become more disparate from those of the parents, particularly for shopping goods which "represent" the status of the offspring. Therefore, it is hypothesized that:

H6a: As income increases intergenerational influence in product and store choices decreases.

H6b: The decrease will be more pronounced for shopping goods.

Marital status is another potentially important moderator of intergenerational influences The strongest demonstration of intergenerational influence would be expected for offspring who have never been married. Once the offspring participates in a cohabitation situation, factors such as resolution of conflicts regarding product or store preferences while adapting to the budget situation in the new household would be expected to decrease the level of intergenerational influence demonstrated. As a result it is hypothesized that:

H7: Intergenerational influences will be stronger for individuals who have never been married.

Finally, if the adult offspring are living with their parents, it is expected that product and store choices will be more similar to those of their parents than for those living away from their parents. This may be due to continued parental pressure in consumption decisions, or to a lack of environmental changes which may lead to more independent decisions. Stated formally, it is hypothesized that:

H8: Intergenerational influences will be stronger for individuals whose household includes an extended family.

METHODOLOGY

A number of the hypotheses developed above contain elements which might be expected to change over time. In order to examine adult behaviors as thoroughly as possible, and to consider the early adult period as well as adulthood generally, two separate surveys were conducted using distinct populations. Because the content of the surveys was identical, it will be outlined before the populations are described.

Product Tape Ratings

Of primary interest in this study is the examination of intergenerational influence across a wider array of consumer choices than has been previously documented. In order to support he categorizations of convenience versus shopping goods, the products used in the survey were rated by a separate sample of twenty-six students on three characteristics. Using a commonly accepted definition of product types (c.f, Kotler, p. 466) a questionnaire was developed which required that the respondents rate each product on "time spent shopping for the product' (1= Almost no time; 5 = Great deal of time); "frequency of purchasing the product" (1 = Very Infrequently; 5 = Very Frequently); and, "comparisons made before purchasing the product" (1 = no comparisons made; 5 = Always made comparisons). After summing across the categories of convenience versus shopping goods, mean scores were shown to be significantly different on all three characteristics. Specifically, time spent shopping was greater for shopping products (Mean = 3.9) than for convenience products (Mean = 1.7; t-value = 12.25, p < .01); convenience products (Mean = 3.0) were purchased more frequently than shopping products (Mean = 1.7; t-value = 11.05, p < .01); and more comparisons we.e made for shopping products (Mean = 3.9) than for convenience products (Mean = 2.8, t-value = 5.32, p <.01).

Purchase Pattern Survey

Using the information described above, a survey was designed which required respondents to indicate whether they generally bought the same brand as their parents for 22 different convenience products and 13 different shopping products. Responses for the items included whether the respondents currently: buy the same brand as parer,is, buy different brand from parents, don t know or don't buy. Next, the same type of responses were given for 9 different store choices - five for shopping products and four for convenience products.

Respondents were then asked to complete an 11-item filler scale measuring internal-external locus of control, followed by a 9-item scale designed to measure their family orientation. This measure of family orientation was developed by Bales and Couch (1969) as part of a large set of items designed to measure values related to interpersonal relations. In the nine item sub-scale titled Acceptance of Authority, individuals' attitudes toward parental guidance and family participation are measured. Increasing scores on the scale indicate more acceptance of parental authority by the offspring and increasing importance of family.

A series of demographic questions was then completed which included age, education, marital status, income of respondent, whether they lived with their parents and whether their parents lived in the same metropolitan area.

Respondent Populations

As indicated above, two different populations were sampled in order to collect information from a broad section of the adult population. In the first study the survey was given to a convenience sample of undergraduate ar,d graduate business students at a major midwestern university. Distribution and completion occurred with the class periods, thus eliminating problems of survey nonresponse. Usable questionnaires were collected from 123 students, ninety-one percent of whom were under the age of 30. Other characteristics of the sample included seventy-two percent never having been married, forty-three percent female, fifty-three percent undergraduate and ninety-three percent with household incomes of less than 51 5000/year.

The second study was conducted in order to examine the hypotheses using a population which represented a more diverse array of demographic characteristics &man was possible using students as respondents. The population utilized in this study was defined as the non-faculty staff of a major midwestern university. A random sample of 300 was selected to receive the survey, from which 209 usable questionnaires were returned. Intra-university mail was utilized to send and return the surveys, but all respondent replies were anonymous. The range of demographic characteristics was much broader for this group and as a result is believed to more adequately represent a cross-section of consumers than does the group of business students. For example, the ages of respondents ranged from twenty-one to sixty-four, with about fifty percent of the sample being less than thirty. Household incomes ranged from less than $10000 to more than $90000, with twenty-eight percent of the respondents having never been married.

Dependent Measure Definitions

In order to examine the hypotheses developed above, an index was created which measures the proportion of consumption decisions made for which the brand purchased was the sa,me as was purchased by the respondents' parents. Specifically, the number of responses "Same as Parents" was used as the numerator and the total number of products/stores for which intergenerational transfer was assessed was used as the denominator. The six indices which were calculated included a measure for convenience products, shopping products, total products, convenience stores, shopping stores and total stores. In order to assure that each of the indices was an adequate measure for the hypothesis tests, single sample Z-tests were conducted to evaluate whether the proportions calculated were statistically greater than zero. Each of the indices for both the student and non-student samples was found to be statistically different from zero. The actual values of the indices ranged from approximately twenty-nine (z = 4.91, p < .05) to forty-three percent (z = 25.1, p < .01) in the student sample and from approximately sixteen (z = 11.4, p < .01) to twenty-nine percent (z = 20.4, p < .01) in the non-student sample.

In the results sections which follow, the two sets of respondent data will be evaluated. In order to examine the moderating effects of the various stimulus and individual characteristics, t-tests were conducted when categorical variables were utilized. Overall indices of product and store choices are utilized for these tests because the indices for convenience versus shopping products or stores produced the same pattern of results. Correlations were calculated and evaluated when interval or ratio level descriptors were collected. Correlations are reported for both convenience and shopping categories because of differences seen across the categories.

RESULTS

Study 1 Results

The first set of hypotheses predict that the intergenerational transfer of brand or store choice will be greater for shopping goods than for convenience goods. In the sample of undergraduate and graduate students, the reverse was found. For both product and store choice, a greater percentage of convenience goods (Mean = 43.2) or stores (Mean = 40.1) were reported to be the same as parents than were shopping goods (Mean = 29.6) or stores (Mean = 29.0). Each of these differences was statistically significant (t:product = 6.07, p < .01; Store = 3.02, p < .01).

The results of analyses of the various individual characteristics are found in Tables 1 and 2. Table 1 contains the mean index values for the categorical characteristics and Table 2 includes the correlations calculated between the indices and the interval or ratio level individual characteristic measures. Each of the tables will be referred to below, as the results of the remaining hypothesis tests are presented-.

The second hypothesis predicts that intergenerational influence decreases with the age of the decision maker. The results of the hypothesis tests using data collected from the student sample are mixed. While all of the correlations are in the predicted direction (Table 2), the correlation between age and the convenience product index is not statistically significant. That is, while there does appear to be an inverse relationship between age and intergenerational influence in shopping product and both shopping and convenience store choices, the relationship is not statistically significant for convenience products.

The present study does not provide any evidence of differences between males and females for either product or store choice. No significant differences are seen between men and women in their indices of intergenerational transfer (Table 1), and so, hypothesis three is not supported.

TABLE 1

MEAN INTERGENERATIONAL EFFECT INDICES STUDY 1 (N=123) AND STUDY 2 (N=209)

The moderating effect of family orientation is examined in the lest of hypothesis four. As predicted, the effect of intergenerational transfer increases with increasing perceived importance of family for both convenience and shopping product choices (Table 2). The effect is also present for the selection of shopping; goods stores, but is not significant for convenience store choices (Table 2). Overall, hypothesis four is supported in three of the four analyses of the student respondents.

The final four hypotheses examine demographic characteristics not previously examined in the literature, but seemingly important in the understanding of intergenerational influence. Hypothesis five predicts that added years of education will diminish the intergenerational transfer effect*. The results support the hypothesis for both product and store choice (Table 1). A related factor, income, is examined in the test of hypothesis six, where mixed results are found (Table 2). As was evident in the examination of age effects, all correlations are in the predicted direction, and significant for both store choice and shopping goods. However, the correlation between income and the index of convenience product choices is not significant.

The impact of marital status on the intergenerational effect is examined in the test of hypothesis seven. As was predict d, the effect is stronger for both product and store choices when respondents have never been married (Table 1). However, in the examination of the effects of living in one's parents home, the difference in product choice indices is not significant. Those respondents living with their parents do show a significantly greater proportion of similar store choices (Table 1). Overall, the results of the test of hypothesis eight are mixed.

TABLE 2

CORRELATIONS OF INTERGENERATIONAL EFFECT INDICES WITH DEMOGRAPHIC AND FAMILY OREIENTATION VARIABLES

In summary, the results of the first study provide support for several of the hypotheses developed above. However, there appear to be some complex relationships which develop for product versus store choices, or for convenience versus shopping products. Potential explanations of such complexities are offered in the section which follows.

Discussion of Study 1

The test of hypothesis one showed significant but reverse effects than were predicted by past literature. However, the results support an alternative argument for adult behavior, based upon the notion of time commitment. Using the logic that individuals are more willing to spend time choosing shopping goods, it might be expected that intergenerational influence would be strongest for convenience goods or "negative" goods (such as insurance, funeral parlors) where parental choices are used as a time-saving heuristic in decision making processes, especially when consumers have little motivation to use complex decision making processes.

As indicated above, many of the hypotheses regarding the moderating effects of individual characteristics on intergenerational influence were supported. The lack of a gender effect may be caused by complexities not considered relevant to the present analysis. For example, previous research showing gender effects has limited the product type being examined to what might be considered more "female relevant" products of clothing and personal care items (Moschis, 1985). In the present analysis the indices include many examples of more generalized categories. Because the purpose of the study is to aid in the development of a more general conceptual framework, the absence of a gender effect appears to support the notion that intergenerational transfer is of importance in understanding, consumption decisions of both males and females.

Two variables for which hypothesis tests received only partial support were age and income. For each of these variables, all correlations we,e in the predicted, negative direction, but the correlations between the individual characteristics and choice indices were not significant for convenience products. One explanation of these results is that, as discussed above, intergenerational influence serves as a heuristic for these choices. As a result, the impact of the influence would be expected to be less strongly moderated by age or income. Convenience product and store preferences acquired from the family appear to have a more enduring effect for adults and to be relatively independent of individuals' discretionary spending potential. An alternative explanation may be that the correlations have been impacted by the constrained range of the age and income variables in 1 sample of student populations. The results of study two should add insight regarding these two potential explanations of the non-significant correlations.

Study 2 Results

In study two, as in the sample of students, the reverse of the hypothesized effect is found for the indices of convenience versus shopping product and store choices. For both product and store choice, a greater percentage of convenience goods (Mean = 29.3) or stores (Mean = 26.4) were reported to be the same as parents than were shopping goods (Mean = 16.2) or stores (Mean = 16.1). Each of these differences was statistically significant (tproduct = 9.15, p < .01; tStore = 4.24, p <.01).

Tables 1 and 2 contain the results of analyses of the various individual characteristics. Again Table 1 contains the mean index values for the categorical characteristics and Table 2 presents the correlations calculated between the indices and individual characteristic measures. One difference in the reported analyses of the second study is due to the four response categories utilized to measure education in study two. Because of the increased number of categories an analysis of variance was used to perform the hypothesis test. Once again, each of the tables will be referred to below, as the results of the remaining hypothesis tests are presented.

The results of the tests of hypothesis two are consistent across the product and store indices using data collected from the nonstudent sample. All of the correlations are in the predicted direction and are statistically significant (Table 2). That is, the inverse relationship between age and intergenerational influence is present for both convenience and shopping product and store choices.

As in study one, there is no evidence of differences between males and females for either product or store choice. No significant differences are found between men and women in their indices of intergenerational transfer (Table 1), and so, hypothesis three is, again, not supported.

The moderating effect of family orientation is less clear in the analysis of the Study 2 data. As predicted in hypothesis four, the effect of intergenerational transfer increases with increasing perceived importance of family for convenience product choices (Table 2). However, the effect is not significant for the shopping product or either category of store choice (Table 2). Overall, hypothesis four receives only weak support when examining the nonstudent response data.

Other demographic characteristics also appear to provide mixed evidence of moderating intergenerational influence. Hypothesis five is not supported in Study 2. Varying years of education appear to have no significant effect on intergenerational transfer for either product or store choices (Table 1). However, as in the first study, the related income factor shows mixed results (Table 2). Al correlations are in the direction predicted in hypothesis six, but the correlation between income and the indices of both convenience product and store choices are not significant.

Finally, as was predicted in hypothesis seven, the intergenerational transfer effect is stronger for both product and store choices when respondents have never been married (Table 1). Additionally, the same effects are seen when the respondent is living with their parents. Those respondents living with their parents show a significantly greater proportion of similar product and store choices (Table 1). So, in Study 2, both hypotheses seven and eight are supported. One additional question asked of the nonstudent sample was whether they currently lived in the same metropolitan area as their parents. As seen in Table 1, this geographic proximity variable does significantly moderate intergenerational influence for both product and store choices.

Discussion of Study 2

The tests of H1 once again showed significant but reverse effects than were predicted by past literature. These results, combined with those of Study 1, appear to offer strong support for the alternative argument based upon the notion of time commitment. Again, this explanation suggests that for adults, the intergenerational influence is strongest and most enduring 'or convenience goods or "negative" goods (such as insurance, funeral parlors). Alternatively, adults are more willing to spend time selecting shopping goods and are more likely to utilize more information in making such decisions than would be provided by simply examining their parents' choices.

Fewer of the hypotheses regarding the moderating effects of individual characteristics on intergenerational influence were supported in the second study. One consistent finding was the absence of gender differences in intergenerational transfer for either products or stores. As stated earlier, this may be due to the lack of gender specific product choices which have been examined in past research (Moschis, 1985). In the present analysis the indices include many examples of more generalized categories. The consistency of the finding across the studies does support he notion that intergenerational transfer is an important concept in developing an understanding of consumption decisions of both males and females.

Two variables for which hypothesis tests received differential support across the two studies were age and income. The role of age as a moderator of intergenerational transfer is seen more strongly in the second study. In this study, all correlations were significant and in the predicted direction. That is, it appears that all choices reflect less intergenerational influence as age increases. As suggested earlier, this may support the notion that Study 1 findings represent a constrained age range in the sample of student populations where the large majority of the respondents had spent limited time away from their parents' home.

Interestingly, convenience product and store preferences acquired from the family do appear to have a more e-,during affect for adults across their discretionary spending potential. This may be due to the fact that convenience goods and stores are less socially visible (so heuristics are used in decision making), whereas shopping good choices are seen as more reflective of the shopper, and therefore parents' choices are seen as less relevant to the decision as more discretionary money becomes available.

Another difference in the two studies is seen in the relationship of family orientation with intergenerational influence. The magnitude of the correlations decreases substantially in the second study, with the only significant relationship indicating that intergenerational influence in convenience product choices increases with family orientation. One possible explanation for this inconsistency between the studies, is a change in the respondents' perspectives when completing the "Family Orientation" scale. The scale format does not specify which family group should be considered when responding, so, whereas the student sample (72% Never Married) would be expected to answer the questions based upon their experiences as children many of the respondents in the second study may have responded as parents. As a result, the measure is not necessarily representing the relationship which would be relevant to intergenerational influence in the present study. This is also supported by the findings in both studies regarding the negative impact on intergenerational influence of ever being married. Any future research examining these relationships must clearly indicate to adult respondents which family experiences they must consider when completing such scales, and may also want to consider the impact of multiple sets of parents when studying married respondents.

FIGURE 1

CONCEPTUALIZATION OF INTERGENERATIONAL INFLUENCE

In summary, a number of differences have been identified in the two studies. The discussions offered above have suggested some explanations for these differences; however, in some cases the discrepancies themselves appear to be informative. In the final section the findings of the two studies will be integrated and used to develop a model describing the nature of moderating effects on intergenerational transfer

CONCLUDING DISCUSSION

Figure 1 presents a simple conceptualization of the intergenerational transfer process, based upon the results presented above. The definition of intergenerational transfer, as the impact of parental influence on the behavior of offspring as adults, is represented as a precursor to choices regarding product and store choices. (Note that this is not to be construed as a complete model of choice, it represents only the nature of the intergenerational transfer process.) Moderating variables, which affect the transfer process are seen as falling into one of two categories. The first category represents variables which impact the strength of the influence. These variables include age, income, education and family orientation. For each of these variables, parental influence decreases as the opportunity for shopping experience increases. One situation for which parental influence continues to endure to a greater extent across these experiences is in the selection of convenience products. The best explanation for this effect seems to be that knowledge of parental choices is used as a time-saving heuristic when the products are not complex or involving.

The second category of moderators reflect the opportunity for parental influence. Variables such as marital status, living with parents and geographic proximity to parents all relate to the opportunity for continued interaction with parents. In general it appears that continued opportunity for interaction prolongs or emphasizes the degree to which adults' choices are tied to those of their parents. Effects of one of the variables in this category - marital status may be more complex than can be identified in the present study. Although being married clearly serves to reduce the influence of the respondents' parents in this survey, the nature of the questions did not allow the examination of possible influences of the spousal parents.

The purpose of this study was to examine intergenerational influence present in product and store decisions and to identify the nature of moderating effects in the intergenerational transfer process. The discussions above have presented specific comments regarding such moderating effects, as well as to develop a conceptual framework with which to further categorize and describe such effects. This represents only the first step in family understanding the role of parental influence on adults' consumption decisions. Further research is needed to identify the processes by which some of the effects seen in this study occur. For example, the present study does not explore the manner in which intergenerational effects are integrated into a marital situation. Understanding how a couple reconciles inconsistent parental experiences, or how parental influence is considered when an adult (or couple) moves to a different geographic area is beyond the scope of this research, but may help to further delineate the intergenerational transfer process. While many questions remain to be asked and answered regarding intergenerational effects, this paper has demonstrated the existence of such effects in a broad cross section of the population for both product and store choices.

REFERENCES

Arndt, J. (1971) "A Research Note on Intergenerational Overlap of Selected Consumer Variables," Markeds Komunikasion, Vol. 3, 1-8.

Bales, R. and Couch, A. (1969) "The Value Profile: A Factor Analytic Study of Value Statements," Sociological Inquiry, Vol. 39, 3-17.

McBroom, N.H., F.W. Reed, C.L. Burns, J.L. Hargraves and M. A. Trankel (1985) "Intergenerational Transmission of Values: A Data-Based Assessment," Social Psychology Quarterly, Vol. 48, No. 2, 150-163.

Moschis, G.P. (1979) "Decision Making Among the Young: A Socialization Perspective" Journal of Consumer Research, Vol. 6, No. 4; 101-112.

Moschis, G.P. (1935) "The Role of Family Communication in Consumer Socialization of Children and Adolescents" Journal of Consumer Research, Vol. 11, no. 4, 898-913.

Moschis, G.P. and G.A. Churchill, Jr. (19/8) "Consumer Socialization: A Theoretical and Empirical Analysis," Journal of Marketing Research, Vol. 15, November, 599-609.

Moschis, G.P., and R.L. Moore (1979) Decision Making Among the Young: A Socialization Perspective" Journal of Consumer Research, Vol. 6, September, 101-112.

Moschis, G.P., and R.L. Moore (1984) "Racial and Socioeconomic Influences on the Development of Consumer Behavior," in Advances in Consumer Research, E. Hirschman and M. Holbrook eds, Vol. 12, (Ann Arbor, MI: Association for Consumer Research), 525-531.

Moschis, G.P., R.L. Moore and R.B. Smith (1983) "The Impact of Family Communication on Adolescent Consumers," in Advances in Consumer Research, T.C. Kinnear, ed., Vol. 11, ( Chicago: Association for Consumer Research), 314-319.

Moschis, G.P., R.L. Moore and L.F. Stephens (1977) "Purchasing: Patterns of Adolescent Consumers, Journal of Retailing, 53, Spring, 17-26, 92

Parsons, T. R.F. Bales, and E.A. Shils (1953) Working Papers in the Theory of Action, (Glencoe, IL: The Free Press).

Saunders, J.R., A.C. Salmi, and E.F. Tozier (1973) "Congruence and Conflict in Buying Decisions of Mothers and Daughters," Journal of Retailing, Vol. 49, Fall, 3-; 3.

Vener, A.M. (1957) "Adolescent Orientation to Clothing: A Social Psychological Interpretation," unpublished doctoral dissertation, Michigan State University, East Lansing MI 48823.

Ward, S. (1974) "Consumer Socialization," Journal of Consumer Research, Vol. 1, No. 2. 1-16.

Ward, S. and D. Wackman (1973) Effects of Television Advertising on Consumer Socialization (Cambridge, MA: Marketing Science Institute).

Wilkie, W. (1987) Consumer Behavior (New York: John Wiley and Sons),178-181.

Woodsen, L. T.L. Childers and P.R.Winn (1976) "Intergenerational Influences in the Purchase of Auto Insurance" in Business Conference Proceedings (W. Locander, ed), Chicago IL: American Marketing Association, 43-49.

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