Consumer Mobility: a Life History Approach

Linda L. Golden, University of Texas at Austin
W. Thomas Anderson, Jr., University of Texas at Austin
Nancy M. Ridgway, University of Texas at Austin
ABSTRACT - While Americans are traditionally characterized by geographical mobility, consumer behavior differences among mobility groups have received little attention. This study identified three consumer segments varying in long-term geographic mobility which differed according to consumption behavior, demographic and psychographic variables. Differences among mobility segments highlight the potential relevance of geographical mobility to consumer behavior and strategy development.
[ to cite ]:
Linda L. Golden, W. Thomas Anderson, Jr., and Nancy M. Ridgway (1980) ,"Consumer Mobility: a Life History Approach", in NA - Advances in Consumer Research Volume 07, eds. Jerry C. Olson, Ann Abor, MI : Association for Consumer Research, Pages: 460-465.

Advances in Consumer Research Volume 7, 1980     Pages 460-465

CONSUMER MOBILITY: A LIFE HISTORY APPROACH

Linda L. Golden, University of Texas at Austin

W. Thomas Anderson, Jr., University of Texas at Austin

Nancy M. Ridgway, University of Texas at Austin

[The research was conducted under grants from the University of Texas Research Institute and the Institute for Constructive Capitalism, The Graduate School of Business, the University of Texas at Austin.]

[Linda L. Golden and W. Thomas Anderson, Jr. are Associate Professors of Marketing. Nancy M. Ridgway is a doctoral student in marketing.]

ABSTRACT -

While Americans are traditionally characterized by geographical mobility, consumer behavior differences among mobility groups have received little attention. This study identified three consumer segments varying in long-term geographic mobility which differed according to consumption behavior, demographic and psychographic variables. Differences among mobility segments highlight the potential relevance of geographical mobility to consumer behavior and strategy development.

INTRODUCTION

On average, Americans move once every five years (U.S. Bureau of Census 1972) with approximately twenty percent of Americans moving each year. While most of these moves are within the same county, and a hypermobile segment appears to account for a disproportionate amount of the changes in residence (Morrison 1971), the incidence of American mobility is substantial enough to potentially transform existing markets and open new ones.

Investigating mobility is not a matter of simply identifying "movers". Complicating factors are distance and time span. The distance continuum may be dichotomized to local mobility (in the same community) and longer distance (geographical) mobility. In addition, perspective may be focused on recent mobility (e.g. "new arrivals") or on the longer term pattern of mobility over a particular time span.

Most all of the research into consumer mobility has focused on "new arrivals" to an area (Andreasen 1966, Andreasen and Durkson 1968, Bell 1970, Bell 1969, Kleimhagen 1968). Only the most recent consumer mobility research (Darden, French and Howell 1979) used a sample not restricted to "new arrivals", Darden, French and Howell (1979) investigated local and geographic mobility over a five year period, as well as predisposition to move. They identified five groups varying in amount and type of mobility: Stayers, Local Movers, Typicals (average movers), Mobiles (above average movers) and Hypermobiles. Results indicated that mobile types had distinctive lifestyles (Activities, Interests and Opinions) and media habits, a conclusion consistent with all previous consumer mobility research: Mobiles constitute a distinct market segment.

The present research goes beyond previous research in three critical areas. First, similar to the Darden, French and Howell (1979) study, the sample is not restricted to "new arrivals", although the research focus is limited to geographical mobility. While past mobility behavior was elicited in some of the "new arrival" investigations (e.g. Andreasen 1966), this study compares mobile and nonmobile consumers on certain demographic and behavioral variables previously investigated only for "new arrivals". Second, Darden, French and Howell (1979) restricted their investigation to lifestyle and media habit variables while this study investigates consumer behavior, demographic and psychographic variables for consumers varying in degree of mobility. Thus, this research contrasts consumers varying in degree of geographic mobility on variables for which mobiles and nonmobiles have not previously been compared. Finally, while Darden, French and Howell (1979) identified mobility groups on a five year time span, this research focuses on the entire life history of geographical mobility. The focus here is an accumulation of geographical mobility experience rather than on relatively recent experiences.

Both the long- and short-term temporal perspectives on mobility are potentially relevant to marketers. There is likely to be a continuing flow of "new arrivals" to many areas, and identification of differences between these "mobiles" and "nonmobiles" is useful for strategy development to cultivate patronage of "new arrivals". However, immediate adjustments in behavior for "new arrivals", or recent movers, are likely to be transitory. By operationally defining mobility on the basis of accumulation of experience within different geographical areas, differences between "mobiles" and "non-mobiles" may highlight continuing behavioral contrasts. Consumers who have been exposed to relatively frequent changes in consumption environment over their lifetime may develop "permanent" and distinctive behaviors for coping with change. And, preferences developed in previous consumption environments may be carried to new environments such that a behavioral repertoire is formed reflecting aspects of a diversity of past consumption environments.

Tentative Theory of Mobility

In his research, Andreasen (1966) proposed an "opportunity-willingness theory" of mobility. Specifically, "A central postulate of the theory is that geographic mobility for a given household is a product of objectively determined opportunities presented by the environment, and a household's subjective willingness to move" (Andreasen 1966, p.342). His results tend to confirm the hypotheses developed from the theory.

This research has a "theoretical" framework which explicitly broadens Andreasen's "opportunity-willingness theory" based on a life history perspective. The central postulate of the "life history opportunity-priority theory" is that mobility is a function of life opportunities and lifestyle priorities determined by and reflected in life history. "Life opportunities" corresponds to Andreasen's "objectively determined opportunities'' and may be circumscribed by many variables, such as current place of residence, past places of residence and occupation. Lifestyle priorities include Andreasen's "subjective willingness to move;" however, lifestyle priorities would be mirrored in "subjective willingness to move." For example, an individual who is family-centered in lifestyle priorities may be unwilling to uproot to follow job or career opportunities.

The life history approach postulates that an individual's mobility is determined by and reflected in life history. A person's life history of mobility may influence propensity to move in the future; thus, mobility itself may be a lifestyle, such as with Morrison's hypermobiles (1971). Further, the elements of the central postulate would be interactive: life opportunities and lifestyles could potentially influence each other and both be reflected in life history. The life history approach suggests that the effects of mobility on an individual may be evolutionary and a person's life history may be a vital link in understanding and identifying mobility segments.

The "life history opportunity-priority theory" is advanced to provide a theoretical perspective on geographical mobility, but the research purpose is not to formally test what is loosely called a theory. It is expected that a life history approach will result in the identification of groups varying in degree of mobility which evidence unique consumption behaviors, demographic and psychographic profiles.

Research Focus

Darden, French and Howell (1979, p.73) suggest that mobility research focus on outshopping differences between mobile and nonmobile groups and usage rates of products. Since varied consumer environments may expand the assortment of goods mobiles have been exposed to, it is proposed that high mobiles will be more likely to engage in outshopping and related behaviors than low mobiles.

Further, due to a tendency to be similar to innovators (Bell 1969, Bell 1970, Darden, French and Howell 1979), high mobiles may be more frequent patrons of convenience stores. Although previous research does not empirically compare mobiles and nonmobiles (Andreasen 1966, Andreasen and Durkson 1968, Bell 1970, Bell 1969) results suggest that high mobiles are more likely to be the sole decision maker and use personal sources than are low mobiles. Mobiles are also expected to have more frequent and varied shopping patterns than low mobiles.

While methodologies and foci differ between this and previous research, results suggest that mobiles are younger and of a higher socioeconomic status than lower mobiles (Andreasen 1966, Andreasen and Durkson 1968, Bell 1970, Bell 1969, Darden, French and Howell 1979). This research is anticipated to confirm previous research. In addition, it is expected that high mobiles will have lived a shorter time in their current town, tend to move for job-related reasons and be less satisfied with their current town than low mobiles.

Four psychographic variables are investigated: life satisfaction, locus of control, family ideology and orientation toward the past, present and future. It is anticipated that high mobiles will have the lower life satisfaction, be more externally oriented, have a less traditional family ideology, and be more future-oriented than low geographic mobiles.

METHODOLOGY

A four-part questionnaire was developed and extensively pretested for clarity in the field. Part one of the questionnaire obtained information about the respondents' shopping patterns and consumption behavior. Three products were chosen to represent a continuum of frequently purchased convenience/shopping goods to infrequently purchased shopping/specialty goods: groceries, clothing, and furniture and appliances (including TV and stereo). For each product category separately the questions in part one provided information regarding: frequency of shopping, shopping companions, the community where the respondent shops most frequently, other types of products shopped for at the same time, criteria and information sources for store selection, number of stores visited per shopping trip, and the decision-agent for store selection. In addition, the respondent was asked to list all other activities engaged in when shopping outside the community of residence, frequency of shopping at convenience stores, and the types of products purchased at convenience stores.

Part two of the questionnaire measured mobility across life history. In order to provide respondents with a manageable frame of reference the mobility data were gathered by age categories chosen to represent stages during which critical life decisions and transitions may occur: 17 years of age or younger, 18-23, 24-34, 35-49, 50-64, and 65 years of age and over. Respondents indicated how many times they had changed towns during each life stage. This section also obtained residential profile information: years of residence in current community, why respondents moved there, how satisfied/dissatisfied they are, why they are satisfied or dissatisfied, and why they would expect to move. Parts three and four of the questionnaire ascertained standard demographic information and measured four psychographic variables: James' Internal-External Locus of Control (Rotter 1966); Life Satisfaction (Robinson 1977); Attitude toward the Past, Present, and Future (Rokeach 1956); and Traditional Family Ideology (Levenson and Huffman 1955).

The questionnaire was mailed to a sample of 4500 individuals in Eastern Oklahoma. Names were selected randomly from recently issued telephone directories of five trading areas representing an urban-rural geographic continuum. The areas sampled were: Tulsa (n = 1500), suburbs of Tulsa (n = 1500), Muskogee (n = 500), Checotah (n = 500), and Stigler (n = 500). The last three communities, respectively, represented increasing isolation from urban Tulsa, and were the rural communities sampled. The cover letter requested that the questionnaire be completed by the person who does most of the shopping in the household. A total of 833 questionnaires were returned, resulting in 807 usable questionnaires, with urban respondents being slightly over-represented in the sample.

ANALYSIS AND RESULTS

The first phase in the analysis involved identifying mobility groups. Mobility scores ranged from zero to nine, with the average respondent moving approximately one and one-half times per stage across life history stages (x=1.41). The mobility data were divided into tertiles to represent high, medium and low mobility consumers. The low mobility group consisted of 261 respondents with mobility scores ranging from zero to .6000 (x=.21). The medium mobility group contained 280 respondents with mobility scores ranging from .667 to 1.600 (x=1.03). And, the 266 high mobiles had mobility scores ranging from 1,667 to 9.000 (x=2.75). Thus, on average, the low mobiles tended to never or almost never change towns during their lifetime, the medium mobiles had moved approximately once per stage across life history stages, and the high mobiles had moved almost three times per stage across life history stages. T-tests revealed significant (alpha less than .00) differences between mean scores for all pairs of mobility groups.

Consumption Behavior Variables

To reduce redundancy in the set of behavioral variables and make the data set more manageable for further analysis, the twenty-four behavioral variables were factor analyzed with varimax rotation and Kaiser normalization (Nie, et. al. 1975, p. 468) using the conventional cutoff criterion of eigenvalues greater than one. Nine factors emerged explaining 60.2 percent of trace and constituting a conceptually reasonable set of factors (see Table 1). Factor scores were subsequently derived for each subject on each of the nine factors using the obtained factor weights.

TABLE 1

FACTORS AND LOADINGS FOR BEHAVIORAL VARIABLES

To determine the relationship between the mobility groups and consumption behavior, factor scores for each respondent on each factor were input into a discriminant analysis (Nie, et. al. 1975, p. 434). Table 2 displays the factors which discriminated among the groups at or within an alpha level of five percent. The outshopping and convenience store patronage factors emerged significant. As mobility increased so did the incidence of outshopping across products investigated. Further, as mobility increased so did the frequency of convenience store patronage and the number of products purchased at convenience stores. Thus, high mobiles were more likely to outshop, patronize convenience stores, and buy more products at convenience stores than medium or low mobiles.

Nine behavioral variables were excluded from the factor analysis as they could not be intervally scaled. These variables were first and second most important store choice criteria and information sources for each of the product categories. Significant chi square results (Nie, et. al. 1975, p.218) are reported in Table 3.

While respondents indicated first and second most important store choice criteria for each product category, only the second most important choice criterion for furniture and appliance was significant. Price was the second most important criterion for all groups; however, medium mobiles were the most price-conscious and were more interested in selection and service than expected. Low mobiles appear to be more likely to be interested in product quality and other attributes, such as convenience, friendly personnel, etc., than expected. High mobiles were the least price conscious and were more oriented toward product selection and quality than expected.

Differences in information sources for store selection were significant for two products: groceries and clothes. The results were generally consistent between the two products, except for low mobiles. For groceries, low mobiles relied on in-store exposure, friends and relatives and multiple information sources more often than expected, but for clothes they relied on media sources (direct mail, magazines, radio and TV) and multiple sources more than expected. Medium mobiles relied on in-store exposure and newspaper sources more than expected for both products, and high mobiles relied on friends and relatives and media sources more often than expected. While all groups relied more on in-store exposure and the newspaper than any other sources for both products, high mobiles tended to rely relatively more on personal sources.

TABLE 2

DISCRIMINANT ANALYSIS RESULTS FOR SIGNIFICANT BEHAVIORAL FACTORS

TABLE 3

ACTUAL AND EXPECTED FREQUENCIES AND PERCENTAGES FOR SIGNIFICANT CONSUMPTION BEHAVIOR VARIABLES BY CONSUMER MOBILITY CATEGORIES

Demographic Variables

Nine demographic variables were input to a discriminant analysis (Nie, et. al. 1975, p.434): sex; marital status; wife's employment status; chief wage earner's age, education and occupation; household income; number of automobiles; and area of residence (urban, suburban or rural). As Table 4 indicates, age and education of the chief wage earner significantly discriminated the three mobility groups. Age decreased and education increased as mobility increased across the three groups. Thus, high mobiles tended to be younger and better educated and low mobiles were older and less educated, with medium mobiles between.

TABLE 4

DISCRIMINANT ANALYSIS RESULTS FOR SIGNIFICANT DEMOGRAPHIC VARIABLES

Because most of the residential profile variables were noninterval, a contingency table analysis was performed on these variables (Nie, et. al. 1975, p.218). As Table 5 indicates, the only residential profile variable which was not significant was the reason for satisfaction or dissatisfaction with current town.

TABLE 5

ACTUAL AND EXPECTED FREQUENCIES AND PERCENTAGES FOR SIGNIFICANT RESIDENTIAL VARIABLES BY CONSUMER MOBILITY CATEGORIES

Years of residence in the current town appears to be inversely related to mobility, as anticipated. More than expected, low mobiles lived in their current town for twenty-one years or longer and medium mobiles ranged between three and thirty years more than expected. High mobiles had lived in their current town for categories indicating twenty years or fewer more than expected, with thirty-eight percent of high mobiles having lived in their current town five years or less.

More than expected, low mobiles moved to their current town for job- or marriage-related reasons. Medium mobiles moved to their current town for marriage, relatives, or rural-preference reasons more than expected, and high mobiles moved for Job or other reasons (health, religious, retirement, etc.) more than expected. High mobility consumers appear to be the most job-oriented, followed by medium and low mobiles, respectively, and are most likely to move for this reason.

While most of the respondents for all three mobility groups reported high satisfaction with their current town of residence, satisfaction appears to decrease as mobility increases. More frequently than expected, low and medium mobiles indicated they were very satisfied, whereas high mobiles reported being very satisfied less frequently than expected.

Low and medium mobiles did not tend to expect to move, whereas high mobiles appear to generally expect to move. Low mobiles anticipated moving to a rural environment more frequently than expected and high mobiles anticipated moving for job-related or other (personal, retirement, etc.) reasons more frequently than expected.

Psychographic Variables

As indicated in Table 6, life satisfaction and family ideology were the two significant psychographics. Life satisfaction decreased as degree of mobility increased, and family ideology became less traditional as mobility increased. Thus, high mobiles were least satisfied with life and had the least traditional family ideology, while low mobiles were most satisfied with life and had the most traditional family ideology.

TABLE 6

DISCRIMINANT ANALYSIS RESULTS FOR SIGNIFICANT PSYCHOGRAPHIC VARIABLES

CONCLUSIONS AND IMPLICATIONS

While restricted to the characteristics of the sample and the specifics of the research, the life history approach permitted identification of three mobility groups having distinct consumption, demographic and psychographic profiles. The summary contrasts below are presented for high and low mobility consumers, as medium consumers consistently fell between.

TABLE

In addition, the three mobility groups appear to use slightly different furniture store choice criteria and information sources for grocery and clothing store selection. Medium mobiles tend to be the most price-oriented in their selection of a furniture and appliance store. And, consistent with previous research, high mobiles exhibit a slight tendency toward more reliance on personal sources of information than other mobility groups (Andreasen and Durkson 1968, Bell 1970, Bell 1969).

Consistent with the perspective offered by the "life history opportunity-priority theory," high mobiles had lived in their current town for shorter periods (implying a past history of mobility), were more likely to have moved for job-related reasons, and were more likely to expect to move for job-related reasons. This implies that the job may present the opportunity or necessity for mobility (even if the move is to search for a job in an environment offering more opportunities) and the importance of "career" to lifestyle priorities. In addition, high mobiles appear to be less likely to be satisfied with their current place of residence than either medium or low mobile groups.

The use of life history as a basis for identifying mobility segments suggests that the behavioral variables significant in this study may be those for which long-term differences exist between mobility groups. The strategy implication is that convenience store patronage and outshopping may be a way-of-life for geographic mobiles and that convenience store marketers or marketers wishing to capture outshoppers may benefit from orienting marketing strategy to the characteristics of high mobiles.

It is interesting to note that the only other study not focusing on "new arrivals" (Darden, French and Howell 1979) also found only two significant demographic contrasts: age and education. This implies that these variables may be the best demographic predictor variables of long-term geographic mobility and the most useful for segmentation purposes.

The overall implication of the body of mobility research is that consumers varying in degree of mobility do constitute distinct market segments. Future research should attempt to develop and test theories of mobility relevant to consumer behavior and provide additional comparative primary data on mobiles and nonmobiles. Specifically, additional research is needed focusing on the potential implications of outshopping, outshopping store choice criteria, and differences in brand loyalty and products purchased among mobility groups. One way of reducing the risk of moving to a new community may be brand or store loyalty. And, due to varied environmental exposure, the potential assortment of products purchased by mobiles may be different from lower mobility groups, such that outshopping is necessitated. This may further be moderated by the current community of residence and product availability within this community. The mobility patterns of Americans have potentially broad ranging implications for consumer behavior and marketing strategy.

REFERENCES

Andreasen, Alan R. (1966), "Geographic Mobility and Market Segmentation," Journal of Marketing, 3, 341-48.

Andreasen, Alan R., and Durkson, Peter G. (1968), "Market Learning of New Arrivals," Journal of Marketing Research, 5, 166-76.

Bell, James E., Jr. (1969), "Mobiles--A Neglected Market Segment," Journal of Marketing, 33, 37-44.

Bell, James E., Jr. (1970), "Mobiles--A Possible Segment for Retail Cultivation," Journal of Retailing, 46, 3-15.

Kleimhagen, Arno K., and Stampfli, Ronald W. (1968), "A 'Principle of Drift' for Institutional Patronage?" Journal of Retailing, 44, 3-12.

Levenson, Daniel, and Huffman, Phyllis (1955), "Traditional Family Ideology and Its Relation to Personality," Journal of Personality, 24, 251-73.

Morrison, Peter A. (1971), "Chronic Movers and the Future Redistribution of Population: A Longitudinal Analysis,'' Demography, 8, 171-83.

Nie, Norman H., Hull, C. Hadlai, Jenkins, Jean G. Steinbrenner, Karin, and Bent, Dale H. (1975), Statistical Package for the Social Sciences, Second Edition, New York: McGraw Hill.

Robinson, John P. (1977), How Americans Use Time: A Social-Psychological Analysis of Everyday Behavior, New York: Praeger Press.

Rokeach, Milton (1956), "Political and Religious Dogmatism: An Alternative to the Authoritarian Personality," Psychological Monographs, 70, No. 425.

Rotter, J. B. (1965), "Generalized Expectancies for Internal versus External Control of Reinforcement," Psychological Monographs, 80, No. 609.

U.S. Bureau of the Census (1972), Current Population Reports, Series P-20, Mobility of the Population of the United States: 1970-1971, No. 235, U.S. Government Printing Office, Washington, D.C.

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