Printer Friendly

Contributions of social capital theory in predicting rural community inshopping behavior.

N.J. Miller (*)


The social environment in which the retailer conducts business is not often measured for its fundamental influences on consumers' local purchasing behavior. This study, using social capital theory as a theoretical framework, examines whether reciprocal actions exist between community members as consumers and retailers and if these actions are persuasive in predicting the economic activity regarded as consumer inshopping. Determinants of inshopping behavior are analyzed from the community member's perspective in a study of the rural community marketplace. The sample population consisted of consumers living in two rural Iowa communities with populations less than 10,000, agricultural-based economies, and retail mixtures of locally owned and operated small-sized businesses as well as national chain and discount organizations. Structural equation modeling estimated the causal patterns among consumers' attachment to community with two endogenous variables regarding reciprocity and inshopping behavior. Findings of fer supporting evidence that social relationships aid in predicting rural marketplace relationships. [C] 2002 Elsevier Science Inc. All rights reserved.

1. Introduction

Rural market environments are noted for the increasing number of businesses that are failing to thrive or survive (Stone, 1995; Strange 1996). Especially devastating to the small-sized independent retailer has been an omission in considering the unique social as well as economic factors inherent in the rural community environment. Competition from larger regional markets and discount chain stores has enticed local residents to spend their dollars beyond their community market. Under these circumstances transaction costs arise because local retailers sustain losses in both sales dollars and trust in their community's support, and the community as a whole forgoes sales tax dollars that could be used for improving local services.

Consumer behavior research examining why consumers leave their community to shop has found that dissatisfaction with local retailer products and services often leads consumers to outshop (Riecken & Yavas, 1988; Samli, Riecken, & Yavas, 1983). Younger residents and higher income groups are closely associated with outshopping but mixed results can be found for demographic variables involving tenure of residence in the community (Thompson, 1971), or residents' life-cycle stage (Herrmann & Beik, 1968). Attempts to predict outshopping behaviors have generally reported weak relationships and a lack of generalizability across a variety of product categories (Hozier & Stem, 1985). This suggests the existence of another set of related attitudinal variables which have yet to be evaluated.

Shopping within the local community marketplace is called inshopping. Inherent in many of the previous studies is the belief that by applying outshopping findings retailers can develop marketing strategies for increasing local inshopping behavior. It is not known if this is a correct assumption or if varying levels of inshopping results from other forces not yet examined. The purpose of this study was to examine determinants of rural consumers' inshopping behavior, testing social capital's theoretical constructs of community attachment and reciprocity.

Social capital theory has been a useful framework for explaining social connections or relationships that can generate collective actions advantageous to the group (Coleman, 1988, Putnam, 1995). Etzioni's (1988) view of the 'responsive community' is one in which individuals make decisions based on normative commitments and affective involvements, and because these individuals have at least some degree of involvement in their community, these decisions are based on varying levels of both the individual's needs ('I') and on the needs of the community ('We').

In the rural community, often social relationships merge with commerce. As Etzioni (1988) describes, the market is but a subset of society. Consumers and retailers are also community members with histories that extend beyond the marketplace (Granovetter, 1985). Positive relationships forged in the community's social environment may contribute to stronger inshopping or local patronage behavior. Hozier and Stem (1985) found consumers' overall beliefs about a community significantly predicted outshopping behavior, and Cowell and Green (1994) reported evidence that consumers' sentiment toward their community influenced where they shopped.

According to social capital theory, community sentiment and involvement partially represent the resident's degree of community attachment. Theoretically, the stronger the individual' s attachment to their community the more likely they are to work together toward the common good of the community (Coleman, 1988). Attitudinal variables embodied in the concept of community attachment have been measured in the rural community social environment (Stinner, Van Loon, Chung, & Byun, 1990) but these variables along with concepts embodied in the norm of reciprocity have not been tested for their influence on local consumer behavior.

Social capital theory also posits that aspects of reciprocity operating in a social group can contribute or inhibit the group's cohesiveness and willingness to work together. As a norm, reciprocity obliges an individual to offer some form of goodness for a goodness that is received (Gouldner, 1960). Putnam (1995) considered the reciprocal norms of behavior and obligations found in the intricate web of relationships as necessary for enhancing social capital.

Reciprocal actions are said to occur at the interpersonal level between partners who are strongly attached through some form of social relationship, and at the institutional level between partners who share less attachment and therefore tend to calculate or quantify the costs versus the benefits of reciprocating (Bums, 1973). Reciprocal actions between consumers and retailers may represent a unique mix of interpersonal and institutional levels, and may explain the customer dissatisfaction that in past studies has been attributed to outshopping (Riecken & Yavas, 1988; Samli, Riecken, & Yavas, 1983). Or, consumer shopping may reflect the national decline in reciprocity along with the level of commitment to one's community and the need to conform to the expectations of others, as has been suggested by Stanfield and Stanfield (1997).

Exploratory research in the rural community marketplace environment by Miller and Kean (1995, 1997) found that reciprocity partially explained consumers' intentions to shop with local retailers. However, in these studies, researchers measured only consumers' inshopping intentions, not their perceived inshopping behavior, nor were variables related to community attachment included in the analyses. These earlier studies examined the relationship between reciprocity and inshopping intentions as further explanation of exchange theory, in that reciprocity was considered necessary in a between-party exchange relationship. The social environment in which the economic exchange occurred was not evaluated from a social capital theoretical perspective.

In sum, there has been limited measurement of factors suggested to predict why consumers choose to shop within their community marketplace. For economists trying to measure the effects of social capital on economic growth, describing how social capital works has not yet been accomplished (Pennar, 1997). It is not clearly understood whether concepts affiliated with social capital theory, particularly community attachment and reciprocity, can generate influence on consumers' purchasing activities. The purpose of this study, therefore, is to investigate determinants of rural consumers' inshopping behavior using social capital theory as a theoretical framework. This can be accomplished by incorporating inshopping behavior, reciprocity, and community attachment into a comprehensive model (see Fig. 1). The model illustrates the effects of community attachment on reciprocity, and, ultimately their influence on rural consumers' inshopping behavior for the product categories of apparel and home furnishings.

2. Theoretical conceptualization and hypotheses

The social environment of the community is not often appraised for the positives that could be accessed for local retail market strategy formation. Sociological research indicates that some rural communities have been successful in stabilizing their business enterprises by stimulating local investment in the community's infrastructure, and producing an optimistic community spirit (Flora, 1995). The shared feature of these successful communities is a dense network of relationships or attachments and strong norms of reciprocity between community members. These concepts are embodied in the theory of social capital (Coleman, 1988; Putnam, 1995). The hypothesized relationships among community attachment and reciprocity variables are discussed along with relationships posited to exist with the dependent variable addressing rural community inshopping behavior.

2.1. Community attachment

Community attachment is a multidimensional notion conveying a variety of linkages that individuals may have within their community. The structure of these linkages can facilitate certain actions of individuals or groups within the community making possible the production of social capital (Coleman, 1988). Linkages or relationships theoretically strengthen the norm of reciprocity (Sahlins, 1965). Two concepts are often used in measuring community attachment, namely, involvement and sentiment (Goudy, 1990; Kasarda & Janowitz, 1974; Stinner et al., 1990). Involvement concerns the frequency of membership in both formal and informal associations within the community, while sentiment is measured by level of satisfaction with global or overall aspects of the community. In sum, community attachment is strengthened by the development of extensive friendships, community-wide association ties, and recognition of the community as a satisfactory place to live (Kasarda & Janowitz, 1974).

Presently investigations concerning community attachment's relationship with shopping behavior are limited. Miller (1997 a) found that rural communities with higher levels of community attachment also had higher levels of local shopping. Cowel and Green's (1994) investigation of household spending patterns found purchasing of apparel and products for the home was significantly related to strong levels of community attachment. This study will extend aspects of the Cowel and Green research by attempting to measure the strength of rural community attachment as a factor affecting consumers' level of reciprocity as well as their inshopping behavior.

2.2. Reciprocity

Reciprocity is said to be deeply ingrained in our moral beings and meant to stabilize exchanges in the daily tasks of living (Sahlins, 1965). Justification for reciprocal action is obscure but, as Becker states, "The mere recognition of a benefit seems to generate a sense of obligation to repay" (1986, p. 73). Reciprocity is also an important tenet in Homans' (1961) treatise regarding social exchange behavior. Homans posited that no exchange relationship would continue unless both partners in the relationship gained profit of some type.

Burns (1973) contrasted the characteristics of interpersonal versus institutional reciprocating activity. According to Burns' theory, interpersonal reciprocity exists in social partnerships where strong levels of concern for others and attachments are found, for example, between family members or friends. Whereas, institutional reciprocity is found in situations where partnerships are formed on less personal foundations; thus, there is a need to predetermine the payoff value for what has been offered rather than leaving the obligation open for interpretation. There are interpersonal and institutional reciprocal relationships occurring at different levels among community members.

Partnerships formed between consumers and retailers as community members may be influenced by their level of satisfaction with either of these two forms of reciprocity, or perhaps even by a third type of reciprocity characterized as having interactive aspects of interpersonal with institutional reciprocal activities and occurring specifically in the retailers' motivation for selecting specific products and services. Specifically, for consumers to reciprocate by purchasing from a retailer, the retailer is expected to have the community residents' wants and needs in mind when making their product/service selection. This form of reciprocity addresses the consumers' perceptions about the product/service qualities selected by the local retailers, as well perceptions about the retailers' attempts to make the local shopping environment beneficial to the consumer, (e.g., exciting and time saving).

Granovetter's (1985) theory of embeddedness argues that most economic behavior is embedded in social networks. Embeddedness is said to occur when behaviors are so bound by ongoing social relations that they cannot be considered independent of each other. Thus it appears that interpersonal relationships may exist socially that alter institutional relationships established for conducting economic exchange; in this case, necessitating that retailers consider consumers' needs on a personal rather than a collective level. Reciprocation is the honoring of one's goodness by returning goodness. Since their customers are well-known as a form of 'goodness' the priority for purchasing of products/services would seem to be an attempt to meet the known needs of the community resident. In the rural market, making a profit is not an acceptable form of goodness, and is not often reported as a priority for conducting rural business (Miller, Kean, & Littrel, 1999a).

In theoretical support of this hypothesized form of reciprocity in the rural community, previous studies addressing the outshopping phenomenon were further examined. Samli (1989) determined that a significant causal relationship existed between consumers' level of dissatisfaction with small community retail establishments and outshopping behavior. Specifically, these consumers were dissatisfied with product selection and quality of goods the local retailers offered in exchange for perceived high prices. Lack of equitable reciprocation prevented or inhibited the development of economic partnerships between many local residents and retailers. Furthermore, due to the absence of perceived reciprocating activities, there were strong inducements to shop beyond the local market (Samli, 1989; Samli & Uhr, 1974).

Work by Babin, Darden, and Griffin (1994) suggests that much more than the products' characteristics were under consideration in that consumers often choose to shop where they can experience the entertainment and social aspects the site has to offer. Aspects of store image including friendliness of personnel have long been considered important determinants of shopping behavior (Martineau, 1958). In Hildebrandt's (1988) structural equation testing of store image, attributes of the merchandise, as well as the fairness and services of the store employees, were found to significantly predict store performance. In addition, research by Berry (1993) found that consumers' attitudes about retailers' trustworthiness were important in forming patronage relationships. These findings imply that satisfaction with reciprocity may exist in an institutional form, involving the retailers' personal/professional characteristics such as fairness in conducting business, as well as in a more generalized form, involving the retaile rs' selection of desirable products or services offered for sale in a pleasant local shopping environment.

In statistical support of this hypothesized third form of reciprocity, a prior research study by Miller (1997b) applied structural equation modeling to test the ability of reciprocity and satisfaction with retail establishments, as two separate latent variables, to significantly predict retailer inshopping intentions. For the sample of 953 rural residents, findings suggested that the norms of reciprocity were strongly related to their level of satisfaction with local retailers and their product/services. Correlations between interpersonal and institutional reciprocity satisfaction with retailer and product/service satisfaction were higher than correlations between retailer and product/service satisfaction (0.73, 0.83 vs. 0.70, respectively). Modification indices provided by LISREL VII suggested the fit could be improved if the latent variable satisfaction with local retail facilities was dropped from the model since it was essentially measuring concepts embodied in the reciprocity indicators of interpersonal and institutional satisfaction. Subsequent analyses were conducted and the trimmed model was assessed as having a better fit with the data.

Rather than eliminate the measure of consumers' satisfaction with retailers and their products/services as a latent variable with potential for predicting inshopping behavior, consideration of this variable as a third indicator of reciprocity may provide additional statistical insight and enhance understanding of reciprocity as a complex variable not frequently measured.

Findings could then address how satisfactory the exchange of money for goods and service between consumers and retailers are as partners in commerce as well as neighbors, friends, or relatives in the same community. Discovering what retail characteristics or qualities induce strong levels of customer satisfaction and encourage reciprocity could offer implications for retail market plan development. In an effort to explore suggested relationships from a social capital theoretical perspective, three indicators addressing satisfaction with interpersonal, institutional and retail reciprocity were examined for their contribution to consumer inshopping behavior.

2.3. Inshopping behavior

Consumers' inshopping/outshopping activity has been measured in many ways. Herrmann and Beik's (1968) early work on consumers' shopping movements defined inshoppers as those who had not shopped beyond a five-mile radius of the downtown area in the previous year. The amount of inshopping differed by product categories, with children's clothing and housewares the categories most frequently purchased out-of-town. More recent work on inshopping by Pinkerton, Hassinger, and O'Brien (1995) surveyed consumers in two small midwest communities. Respondents were asked if they or another member of the household had purchased any of the listed 29 goods and services. For those goods and services reported as having been purchased, the respondent was then asked how often they bought the item in their local marketplace. The items revealed as least frequently purchased in the community were shoes and clothing for men, women, and children.

Given the conceptual considerations described thus far, three hypotheses are proposed for examining the impact of seven latent variables on rural consumers' inshopping behavior. The proposed relationships among predictors of inshopping behavior are diagrammed in the theoretical model (see Fig. 1). Accordingly, the following three hypotheses are proposed:

Hypothesis 1: Rural consumers' community attachment will have a positive relationship with their level of satisfaction with reciprocity.

Hypothesis 2: Rural consumers' community attachment will have a positive relationship with their inshopping behavior.

Hypothesis 3: Rural consumers' satisfaction with reciprocity will have a positive relationship with their inshopping behavior.

3. Research methods

3.1. Sample

Questionnaires were mailed to a randomly selected sample of consumers ages 18 years and older, residing in two rural Iowa communities or within a twenty-mile radius surround ing the two communities. To qualify as rural communities, Iowa towns were required to meet four criteria: location in a non-SMSA county; total population of less than 10,000; an agricultural economy; and a retail store mix including locally owned, chain, and discount stores. The locations of these two communities were in counties defined by the Bureau of the Census (1992a) as rural counties adjacent to metropolitan counties. Though one community was located in the northern half of the state and the second in the southern half, both communities were within 75 miles of metropolitan communities, thus consumers were provided with a variety of shopping alternatives.

From each community, a random list of 1,000 household addresses were generated. A total of 2,000 adult consumers were contacted using three separate mailings. Employing a modified version of the Dillman Total Design Method (1985), a follow-up postcard was mailed two weeks after the first mailing, and a third mailing of an additional questionnaire was generated two weeks following the postcard mailing. Of the 2,000 surveys mailed, 803 were returned providing a 40% overall response rate.

For purposes of employing statistical testing with linear structural equations (LISREL), a smaller sampling of the total responses was needed. In LISREL, a larger sample size can result in a larger [chi square] goodness-of-fit statistic suggesting the model has a better fit with the data than when a smaller sample size is used (Bollen, 1989; Tanaka, 1993). Therefore, cases were randomly selected, proportionate to the original number of responses from each community, producing a total sample of 312 cases from consumers in both rural communities. Additionally, employed were two screening questions to insure that rural residents subsequently analyzed in this study were consumers of the designate products and that the designated products were perceived as available within the community marketplace. Analyses were therefore conducted on a sample of 246 cases with 39% from Community A (n = 95) and 61% from rural Community B (n = 151) (see Table 1). Testing determined no significant demographic or attitudinal diffe rences between the total original respondents and the smaller randomly selected and screened respondents. Subsequent analyses were performed on the communities combined.

Comparison of community profiles in Table 1 suggests strong similarities among the sample population demographics. This rather homogeneous group was largely Caucasian (99%), middle-aged (53 years old), married (66%), Protestant (61%), and had 13 years of education or less (59%) with a average income level in the $30,000 to $39,999 range. These figures are fairly representative of this rural midwestern state (1).

A few differences were noted when comparing the participants age and level of education with that of the state. The average participant was older, and less likely to have posthigh school education than the state's average for these two characteristics. When examining community-level data, the percentage of residents over age 65 is higher in Community A and B than for the state (25% and 20% respectively compared to 15%). For each community participating in this study, 29% of the respondents reported their age as 65 years or older. Fewer of these older participants finished high school or worked toward a posthigh school degree. Additionally, though more females participated in this study, this may be partially explained by the fact that proportionally more females than males reside in Community A and B (53.3% and 55.% respectively).

3.2. Measures

3.2.1. Product categories

Previous studies of rural consumer shopping behavior found that intensity of outshopping varied by product category. Because consumers' shopping patterns are product specific, product categories are usually selected and identified when measuring outshopping or inshopping behavior (Darden & Perreault, 1976; Hozier & Stem, 1985; Miller & Kean, 1997; Reynolds & Darden, 1972).

Items with strong social visibility, such as apparel and home furnishings, have been found to be the most frequently outshopped product categories (Papadopoulos, 1980). Yet, when levels of community attachment were high, apparel and products for the home were frequently inshopped categories (Cowel & Green, 1994). In an effort to understand more about apparel and home furnishing shopping behavior, this study focused on these two categories. The following definitions, given in the directions heading each section of the survey, were provided to explain these categories and to narrow the range of products: 'Home Furnishing' products include items like -- small appliances, bedding, draperies, televisions, furniture, paintings, mirrors, lamps, and craft items for use in the home; 'Apparel' products include items like -- shoes, clothing, and accessories.

3.2.2. Scale development

Overall, in developing scales for measuring inshopping, reciprocity, and community attachment, existing valid scales were first examined for their relevance or adaptability. Where no compatible instruments were discovered new scales were developed by using a combination of original items and modified items from previously developed scales. All scales were evaluated after pretesting the initial instrument with 91 sophomore-level business students in an introductory marketing class who described their hometown population size as less than 10,000.

To achieve an internally consistent instrument, a two-step analytical procedure was employed for measuring the following latent variables: community attachment (sentiment and involvement); and reciprocity (interpersonal, institutional and retail). First, exploratory factor analysis was performed on the original items using principal components factoring without iteration followed by varimax rotation. In a second step, item-total correlation and Cronbach's coefficient alpha were computed for each of the scales. Factor loadings and interitem correlations were examined for each scale item. Poor performing items were discarded in an effort to reduce the survey length and to improve reliability (Churchill, 1979). Magnitudes of alpha coefficients obtained in this study were above the average 0.70 level as documented by Peterson's (1994) review of consumer research. The final instrument is offered in the Appendix.

3.2.3. Community attachment

Community attachment was measured by two indicators regarding the consumers' sentiment and their involvement in the community. The modified items for the community attachment variables of sentiment and involvement were borrowed from Stinner et al., (1990). A mixture of semantic differential, incremental, and Likert-type questions were used. The measure of sentiment was a composite score created from the sum of three items with an internal reliability score of 0.80, and involvement was a composite score of six items summed with an internal reliability score of 0.72.

3.2.4. Reciprocity

Samli, Riecken and Yavas' (1983) scale provided modified items for measuring satisfaction with reciprocity in relation to retail establishments, and Miller and Kean's (1995) work provided items for measuring satisfaction with institutional and interpersonal reciprocity. All questions were structured in a 5-point Likert-like format with '1' for very dissatisfied and '5' for very satisfied. The three composite scores were created by summing six items to measure satisfaction with interpersonal reciprocity, five items to measure satisfaction with institutional reciprocity, and seven items to measure satisfaction with retail reciprocity. Internal reliability was assessed at 0.81 for interpersonal, 0.87 for institutional, and 0.91 for retail reciprocity.

3.2.5. Inshopping

Items for measuring inshopping were developed by the researcher based on previous measurement of rural marketplace behavior (Miller, Kean, & Littrell, 1999b). For this investigation, inshopping behavior was measured by asking consumers to respond to a series of three questions for each product category. For example, consumers were directed to consider adult clothing as a product category and then to answer the following questions: (a) Please indicate whether or not this product or service is available in this community (yes/no); (b) Do you buy it? (yes/no); (c) If you buy it, please indicate how frequently you buy it in your community (4-point range from Never '1' to Always '4'). Those respondent who stated the particular product or service was not available in their community, or who stated they were not purchasers of the particular product or service were not included in the final analysis. The variable inshopping behavior was therefore a measure of how frequently consumers shopped in town for products and services that were locally available and likely to be purchased. Higher scores indicated the shopping was more frequently conducted in town (inshopped), whereas lower scores suggested the shopping was conducted out of town (outshopped). Specific product categories analyzed for this study included clothing for adults and children, shoes for adults and children, items for the home such as furniture and appliance, and home improvement related products or services. A composite inshopping behavior score for five specific apparel product categories and for three specific home furnishings product categories were each calculated from 246 cases and used in data analysis.

4. Results

4.1. Assessment of the overall model fit

Structural equation modeling (SEM) using maximum likelihood estimation (LISREL VII: Joreskog & Sorbom, 1989) was used for analysis of the theorized structural model. A two-step process was adopted following Anderson and Gerbing's (1988) recommended approach. First confirmatory factor analysis was performed to specify the relations of the observed measures posited as underlying constructs relative to social capital theory. Second, an estimation of the measurement model's fit was conducted and assessed with multiple goodness-of-fit indices.

Results of the confirmatory factor analysis suggested convergent validity in that each of the indicators' estimated path coefficients on their posited underlying construct were twice their standard error and therefore significant (Anderson & Gerbing, 1988). Thus, the seven indicators were found to have significant paths to their respective latent variables for this sample population (N = 240). The measurement model indices were: [X.sup.2] = 27.07, df = 11, p = .004, Goodness-of-Fit Index (GFI) = 0.970, Adjusted Goodness-of-Fit (AGFI) = 0.923, and standardized Root Mean Squared Residual (RMSR) = 0.043.

Initial analysis of the structural model indicated the data fit the theorized model fairly well ([X.sup.2] = 27.15, df = 11, p = .004, GFI = 0.970). However, the direct path from community attachment ([[xi].sub.1]) to Inshopping Behavior ([[eta].sub.2]) was not found to be significant at the 0.05 level for this sample population (N = 240). The direct effect posited in Hypothesis 2 was therefore not supported. The structural model was respecified to ascertain if by removing the insignificant path, the solution could be improved. Anderson and Gerbing (1988) considered it proper in theory testing to respecify a model by deleting a posited relationship or path between latent variables in an effort to improve the model's overall fit only if the deletion made theoretical sense. Previous tests have found a significant direct relationship between community attachment and shopping location behavior; however, these tests have not included reciprocity in the model. Given that the latent variable community attachment wa s significantly related in an indirect path with consumers' inshopping behavior, through the reciprocity, removing this path was considered both theoretically and statistically appropriate for further analysis.

A second analysis, conducted without the direct path from community attachment to inshopping behavior revealed a small increase in [chi square]. The ratio of [chi square] to degrees of freedom was reduced from 2.47 to 2.28 suggesting a more parsimonious fit (Bagozzi & Yi, 1988; Bollen, 1989). The model fit the data sufficiently well to justify interpretation ([X.sup.2] = 22.37, df. = 12, p = .007, Goodness-of-Fit Index (GFI) = 0.970, Adjusted Goodness-of-Fit (AGFI) = 0.930, and standardized Root Mean Squared Residual (RMSR) = 0.044). Most importantly, this model made substantive sense in support of the guiding theory (Bollen & Long, 1993) (see Fig. 2).

In a final assessment of the structural model's fit, the coefficient of determination for all the observed variables jointly was 0.343. The magnitude of variance explained for the structural equation was reasonably strong for the mediating variable, reciprocity, [R.sup.2] = 0.34, and the for the dependent variable, inshopping, [R.sup.2] = 0.47. Parameter estimates and t-values for the final model are given in Table 2. The decomposition of direct and indirect effects can be found in Table 3. Results suggest that community attachment had a positive direct effect on reciprocity ([H.sub.1]) and that reciprocity had a positive direct effect on consumers' inshopping behavior ([H.sub.3]). In terms of total effects, community attachment also had a significant indirect effect on inshopping behavior.

4. Discussion

4.1. Theory development

Social capital theory was supported by evidence that consumers' level of satisfaction with reciprocity was a significant determinant of rural consumers' level of shopping with local retailers. Following Putnam's (1995) and Coleman's (1988) belief that reciprocal norms of behavior and obligations were necessary in the formation of social capital, this study found that positive aspects of social capital contributed in explaining positive inshopping behavior.

The inclusion of reciprocity with community attachment has not been considered in prior investigations of the rural community. However, the findings from this study do support the results of earlier individual studies on social theory (Coleman, 1988; Putnam, 1995), social embeddedness (Granovetter, 1985), reciprocity as defined in exchange theory (Gouldner, 1960; Homans, 1961; Sahlins, 1965), and in some of the fundamental concepts embodied in the theory of relationship marketing (Morgan & Hunt, 1994). Discovering that reciprocity acted as a mediating variable between community attachment and inshopping behavior was an important finding contributing to social capital theory development.

Community attachment was also found to have a significant influence on reciprocity. This finding offers empirical support for Sahlins' (1965) premise that favorable linkages or attachments enhance reciprocal behavior. Social bonds formed from positive interactions among friends and community members are known to be particularly important indicators of sense of community and sentiments about the community (Goudy, 1990). This finding also supports Granovetter' s (1985) theory of embeddedness which argues that most economic behavior is embedded in social networks, and Etzioni's (1988) economic paradigm regarding the balance between the 'I' and the 'We' in the responsive community.

Embeddedness is said to occur when behaviors are so bound by ongoing social relations that they cannot be considered independent of each other. Indeed, rural consumers held expectations that local merchants would be fair and honest in their business dealings with community residents (institutional reciprocity), and that, since the merchants usually knew their customers very well (interpersonal reciprocity), they would select products or services with the specific needs and wants of residents in mind (retail reciprocity). Path coefficients indicate that consumers' attachment to their community and their beliefs about reciprocation have pervasive influences over the multidimensional nature of rural marketplace exchange behavior.

4.2. Limitations and suggestions for future research

Further refinement through retesting and evaluation is warranted for this exploratory model. It is not presently well understood how interpersonal and institutional forms of reciprocity interact with residents' feelings about their local retailers. There is a strong indication that satisfaction with local retail establishments is entangled with residents' expectations about reciprocal behavior.

This study is also limited by the selection of two product categories for consideration. Previous research regarding outshopping activity suggests that shopping behavior differs by product category. Perhaps consumers expect more personalized retailer selection of products and services when items are closely identified with individual tastes, as is inherent in apparel and home furnishings, than when generic products are involved in consideration of shopping location. There are a variety of consumer product categories warranting future consideration in the retesting of this model.

Though several steps were taken to ascertain that inshopping was measured on the basis of products/services that were perceived as locally available and that only rural residents who were consumers of that product were measured for their level of shopping in town, improvements to the measure of inshoppng level are still needed. Recall methods requiring consumers to indicate their purchase location and approximate price for the most recent purchase in multiple product classes is considered problematic, as is asking consumers to estimate the general proportion of in-town to out-of-town purchasing. Perhaps journaling information relative to all purchasing of specific products or services for a set period might be used in subsequent studies in generating a more factual account of shopping location behavior.

Two additional limitations result from the nature of this sample population. Participants were fairly homogeneous in race, religion, and income, and the age mode for participants from each community was 65 years of age. In future studies of more diverse populations, perhaps the experiential aspects of inshopping behavior should be assessed for social biases. Furthermore, findings based on a sample population from the rural midwest may not be generalizable to other geographic areas.

4.3. Implications for rural retailers

Findings do offer strong implications for the rural community market place. Preserving the rural community requires preserving the community's economic viability. Results of this investigation suggest that ingredients required for economic viability, are social linkages enhancing reciprocal forms of exchange between consumers and retailers as cooperative members of a community. Building and maintaining social relationships should be encouraged between the local retailer and consumers, other merchants in the community, and community members in general. From the perspective of social capital, all of these potential partners in the community should perceive that both parties are attempting to make some type of contribution benefiting community members. Retailers should assess their own professional qualities as well as their employees' to determine if all are courteous, fair, and helpful in approaching their customers. A fundamental social aspect of conducting business requires that merchants greet customers by name, make efforts to locate needed products, and in general strive to create a pleasant environment for conducting marketplace exchange.

Findings suggest that merchants should be actively involved within their community's social structure. Not only is the merchant enhancing the perception of concern for community, but a better understanding of consumers' wants and needs can be gained when merchants participate or observe their patrons' social and professional activities. Products that are closely related to self-identity, such as apparel and home furnishings, should be carefully selected with individual consumers in mind. Collectively buying for the general targeted market is not suggested as a sound strategy in rural communities. The ability of local retailers to appraise and adjust to customers as individuals offers a competitive advantage over out-of-town discount and chain store rivals. All of these socially derived conditions can inhibit or enhance rural consumer inshopping behavior and should be considered when formulating marketing plans.


This study was supported by the Iowa Agriculture and Home Economics Experiment Station, and the Family and Consumer Sciences Research Institute, College of Family and Consumer Sciences, Iowa State University, Ames, Iowa 50011

Journal paper No. 18348 of the Iowa Agriculture and Home Economics Experiment Station, Ames, Iowa, Project No. 3343 and supported Hatch Act and the State of Iowa Funds.

(*.) Tel.: + 1-515-294-1930; fax: + 1-515-294-6364.

E-mail address: (N.J. Miller).


(1.) State population demographics suggest 96% of the people are Caucasian, 52% are female, and 80% are high school graduates or higher. The average age, considering all residents, is 34 years old; however, it should be noted that 25.8% of the state's population is under age 18 and 15.3% is age 65 years or older. The median family income is $31,659. The Census designates 39% of the state's population as residing in rural areas. The two communities under study are defined by the Census as 100% rural populations (Bureau of the Census, 1992b).


Anderson, J.C., Gergin, D.W., 1988. Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin 103(3), 411-423.

Babin, B.J., Darden, W.R., Griffin, M., 1994. An empirical comparison of alternative conceptualizations of postconsumption reactions. Journal of Consumer Satisfaction/Dissatisfaction and Complaining Behavior 7 172-182.

Berry, L.L., 1993. Playing fair in retailing. Arthur Anderson Retailing Issues Letter 5 (2), Center for Retailing Studies, College Station, TX.

Bagozzi, R., Yi, Y., 1988. On the evaluation of structural equation models. Academy of Marketing Science 16(1), 74-94.

Becker, L.C., 1986. Reciprocity. University of Chicago Press, Chicago, IL.

Bollen, K.A., 1989. Structural equations with latent variable. Wiley, New York, NY.

Bollen, K.A., Long, S. J., 1993. Testing structural equation models. Sage Publications, Newbury Park, CA.

Bureau of the Census, 1992a. 1990 Census of population and housing, 17. U.S. Department of Commerce, Economics and Statistics Administration, Washington D.C.

Bureau of the Census, 1992b. 1990 Census of population and housing, Summary Tape, File 3 on CD-ROM. [Machine readable data files], The Bureau [producer and distributor], Washington D.C. Available: [February 15, 1998].

Burns, T., 1973. A structural theory of social exchange. ACTA Sociologica 16(3), 188-208.

Churchill, G.A., 1979. A paradigm for developing better measures of marketing constructs. Journal of Marketing Research 16(2), 64-73.

Coleman, J.S., 1988. Social capital in the creation of human capital. American Journal of Sociology 94, (Supplement) S95-S120.

Cowell, D.K., Green, G.P., 1994. Community attachment and spending location: The importance of place in household consumption. Social Science Quarterly 75(9), 637-655.

Darden, W.R., Perreault, W.D., 1976. Identifying interurban shoppers: Multiproduct purchase patterns and segmentation profiles. Journal of Marketing Research 13 (2), 51-60.

Dillman, D., 1985. Mail and telephone survey. Wily and Sons, New York, NY.

Etzioni, A., 1988. The moral dimension: Toward a new economics. New York, Free Press.

Flora, C.B., 1995. Social capital and sustainability: Agriculture and communities in the great plains and the corn belt. Research in Rural Sociology and Development: A Research Annual 6, 227-246.

Goudy, W.J., 1990. Community attachment in a rural region. Rural Sociology 55(Summer), 178-198.

Gouldner, A.W., 1960. The norm of reciprocity: A preliminary statement. The American Sociological Review 25(4), 61-178.

Granovetter, M., 1985. Economic action and social structure: The problem of embeddedness, American Journal of Sociology 91(11), 481-510.

Herrmann, R.O., Beik, L., K. (1968). Shoppers' movements outside their local retail area. Journal of Marketing 32(10), 45-51.

Hildebrandt, L., 1988. Store image and the prediction of performance in retailing. Journal of Business Research 17(8), 91-100.

Homans, G.C., 1961. Social behavior: Its elementary forms. Harcourt Brace, New York, NY.

Hozier, G.C., Stem, D. E., 1985. General retail patronage loyalty as a determinant of consumer outshopping behavior. Journal of the Academy of Marketing Science 13 (Winter), 32-45.

Joreskog, K.G., Sorbom, D., 1989. LISREL VII: A guide to the program and applications. SPSS Inc, Chicago, IL.

Kasarda, J.D., Janowitz, M., 1974. Community attachment in mass society. American Sociological Review 39(6), 328-39.

Martineau, P., 1958. Personality of the retail store. Harvard Business Review 36(1-2), 145-165.

Maslow, A.H., 1970. Motivation and personality. Harper and Row, New York, NY.

Miller, N. J., 1997a. Probing exchange theory: A comparison of consumers in three rural communities. In: Brucks, M., MacInnis D. (Eds). Advances in Consumer Research, vol. 24 (p. 119). Association for Consumer Research, Provo, UT.

Miller, N.J., 1997b. Explaining rural consumers' inshopping behavior. In: King, R.L. (Ed.) Retailing: End of a century and a look to the future. Special Conference Series, vol. 8, (pp. 117-121). Academy of Marketing Science and the American Collegiate Retailing Association Triennial Conference, Richmond, VA.

Miller, N.J., Kean, R.C., 1997, Reciprocal exchange in rural communities. Psychology and Marketing 14(10), 637-661

Miller, N.J., Kean, R,C., 1995. Reciprocal consumer behavior intentions in rural communities: Inducements to shop with local retailers, paper presented at the Rural Retailing: Emerging Community Issues Conference, Salt Lake City, UT.

Miller, N.J., Kean, R.C., Littrell, M.A., 1999a. Exploring consumer and retailer exchange in rural communities: Part I. Family and Consumer Sciences Research Journal, 28(1), 71-98.

Miller, N.J., Kean, R.C., Littrell, M.A., 1999b. Exploring consumer and retailer exchange in rural communities: Part II. Family and Consumer Sciences Research Journal. 28(1), 99-121.

Morgan, R.M., Hunt, S.D., 1994. The commitment-trust theory of relationship marketing. Journal of Marketing 58(7), 20-38.

Papadopoulos, N.G., 1980. Consumer outshopping research: Review and extension. Journal of Retailing 56(Winter), 41-59.

Pennar, K., 1997. The ties that lead to prosperity. Business Week (December 15) 153-155.

Peterson, R.A., 1994. A meta-analysis of Cronbach's coefficient alpha. Journal of Consumer Research 21(9), 381-391.

Pinkerton, J,R., Hassinger, E.W., O'Brien, D.J., 1995. Inshopping by residents of small communities. Rural Sociology 60 (Fall), 467-480.

Putnam, R.D., 1995. Tuning in, tuning out: The strange disappearance of social capital in America, PS: Political Science and Politics, (December) 664-683.

Reynolds, F.D., Darden, W.R., 1972. Intermarket patronage: A psychographic study of consumer outshoppers. Journal of Marketing, (10) 50-54,

Riecken, G., Yavas, U., 1988. A taxonomy of outbuyers: A new perspective. International Journal of Retailing 3(1), 5-15.

Sahlins, M.D., 1965. On the sociology of primitive exchange. In: Sahlins, M.D. (Ed.), The Relevance of Models for Social Anthropology, ASA Monograths, 1 (pp. 139-187). Travistock Publications, London, England.

Samli, A.C., 1989. Retail marketing strateg. Quorum Books, New York, NY.

Samli, A.C., Uhr, E. B., 1974. The outshopping spectrum: Key for analyzing intermarket leakage. Journal of Retailing 50(Summer), 70-79.

Samli, A.C., Riecken, G., Yavas, U., 1983. Intermarket shopping behavior and the small community: Problems and prospects of a widespread phenomenon. Journal of the Academy of Marketing Science 11 (Winter-Spring), 1-14.

Stanfield, J.R., Standfield, J.B., 1997. Where has love gone? Reciprocity, redistribution, and the nurturance gap. Journal of Socio-Economics 26(2), 111-126.

Stinner, W.F., Van Loon, M., Chung, S.-W, Byun, Y.C., 1990. Community size, individual social position, and community attachment. Rural Sociology 55(Winter), 494-521.

Stone, K.E., 1995. Competing with the retail giants: How to survive in the new retail landscap. John Wiley & Sons, New York, NY.

Strange, M., 1996. Transforming the rot belt. Des Moines Sunday Register (February 25), C1-C2.

Tanaka, J.S., 1993. Multifaceted conceptions of fit in structural equation models. In: Bollen, K.A., Long, J.S. (Eds.), Testing structural equation models (pp. 10-39). Sage Publications, Newbury Park, CA.

Thompson, J.R., 1971. Characteristics and behavior of outshopping consumers. Journal of Retailing 47(Spring), 70-80.
Table 1

Responses from two rural communities

Community                              A            B

General Community Information
 Location in state                     Southwest    Northwest
 Town population (a)                   6264         6026
 Local apparel retailers (b)           5            10
 Local home furnishings retailers (b)  6            7
 Total local retail stores (b)         266          304
Respondent demographics (c)            n = 95       n = 151
 Median age                            51.8 years   53.4 years
 Median years lived in town            34.4 years   33.9 years
  Male                                 34.7%        38.6%
  Female                               65.3%        61.4%
  [lesser than or equal to]13 years    60.8%        56.8%
  >13 years                            39.1%        43.2%
  Protestant                           54.0%        72.0%
  Catholic                             24.0%        14.7%
Median Income Range                    $30--39,999  $30--39,999

Community                              Combined

General Community Information
 Location in state
 Town population (a)
 Local apparel retailers (b)
 Local home furnishings retailers (b)
 Total local retail stores (b)
Respondent demographics (c)            n = 246
 Median age                            52.8 years
 Median years lived in town            34.1 years
  Male                                 37.1%
  Female                               62.9%
  [lesser than or equal to]13 years    59.2%
  >13 years                            40.8%
  Protestant                           61.4%
  Catholic                             20.3%
Median Income Range                    $30--39,999

(a)Information from 1990 Census of Population and Housing, 17, (1992)
U.S. Department of Commerce, Economics and Statistics Administration,
Bureau of the Census.

(b)Information from Kenneth Stone and Scott Baumler (1994) Retail Trade
Analysis, Iowa State University Cooperative Extension Publication, Ames

(c)Due to incomplete responses, the total percentage may not sum to
Table 2

Parameter estimates for the structural model

Endogenous factor loadings for [[eta].sub.1]
 [y.sub.1]                                    [[lambda].sub.11]
 [y.sub.2]                                    [[lambda].sub.21]
 [y.sub.3]                                    [[lambda].sub.31]
Endogenous factor loadings for [[eta].sub.2]
 [y.sub.4]                                    [[lambda].sub.42]
 [y.sub.5]                                    [[lambda].sub.52]
Exogenous factor loading for [[xi].sub.1]
 [x.sub.1]                                    [[lambda].sub.11]
 [x.sub.2]                                    [[lambda].sub.21]
Exogenous factor explaining [[eta].sub.1]
 [[xi].sub.1]                                 [[gamma].sub.11]
Endogenous factor explaining [[eta].sub.2]
 [[eta].sub.1]                                [[beta].sub.21]
 [[eta].sub.1]                                [R.sup.2] =
 [[eta].sub.2]                                [R.sup.2] =
                                              [X.sup.2] =
                                              df =
                                              p value =
                                              Standardized RMSR =
                                              GRI =
                                              AGFI =

                                               LISREL   t-value

Endogenous factor loadings for [[eta].sub.1]
 [y.sub.1]                                     1.000
 [y.sub.2]                                     0.923    12.034
 [y.sub.3]                                     0.897    11.727
Endogenous factor loadings for [[eta].sub.2]
 [y.sub.4]                                     1.000
 [y.sub.5]                                     0.728     3.078
Exogenous factor loading for [[xi].sub.1]
 [x.sub.1]                                     1.000
 [x.sub.2]                                     0.644     5.578
Exogenous factor explaining [[eta].sub.1]
 [[xi].sub.1]                                  0.548     5.330
Endogenous factor explaining [[eta].sub.2]
 [[eta].sub.1]                                 0.399     4.885
 [[eta].sub.1]                                 0.343
 [[eta].sub.2]                                 0.469

Note: t-values of >2.0 were significant at the p < .05 level.
Table 3

Decomposition of effects for inshopping behavior

                                      Direct    t-value  Indirect

[[eta].sub.1] = RECIPROCITY
   [[xi].sub.4] = Community
   attachment                        .545 (*)    5.320
   [[xi].sub.4] = Community
   attachment                                            0.219 (*)
   [[eta].sub.1] = Reciprocity       0.399 (*)   4.866

                                     t-value  Total  t-value

[[eta].sub.1] = RECIPROCITY
   [[xi].sub.4] = Community
   attachment                                 0.545   5.320
   [[xi].sub.4] = Community
   attachment                         3.712   0.219   3.712
   [[eta].sub.1] = Reciprocity                0.399   4.866

(*)Significant at the 0.5 level.

RELATED ARTICLE: Appendix A. Instrument

A.1. Inshopping behavior

Listed below is a variety of goods and services.

(a) First indicate whether or not this product or service is available in the community. (b) Do you buy it? (c) If you buy it, please indicate how frequently you buy it in your community.

(a) Yes/No (b) Yes/No (c) Never, Once in a while, Most times, Always

Clothes for adults

Shoes for adults

Clothes for children

Shoes for children

Home improvement products/services



A.2. Community attachment


1. How would you rate your involvement in community-wide activities? Semantic Differential Scale-Not Involved (five spaces) Very Involved

2. What interest do you have in knowing what goes on in this community? 5-point Likert-type-'1' No Interest ... '5' Great Interest

3. On a scale of 1 to 5 with 1 being 'poorly' and 5 being 'well', how well do you feel that you fit into your community?

4. On a scale of 1 to 5 with 1 being 'nothing' and 5 being 'everything', how much do you have in common with most of the people in your community?

5. What proportion of local people do you know? '1' for None, '2' Half or Less, '3' More than Half '4' All

6. What percentage of adults in this community would you say that you know on a first name basis? '1' for Less than 25%, '2' for 25-50%, '3' for 51-75%, '4' for 76 -100%.


1. Please indicate your degree of satisfaction with this community if '5' represents the best possible place you could live and '1' indicates the worst.

2. Suppose for some reason you had to move away from this community, how sorry would you be to leave? (R) '1' Very pleased, '2' A little pleased, '3' It wouldn't make any difference one way or another, '4' A little sorry, '5' Very sorry.

3. Would you say you feel at home in this community? '1' Definitely not, '2' Probably not, '3' probably, '4' Definitely

A.3. Reciprocity

Please indicate the extent to which you are dissatisfied or satisfied with each of the following statements with '1' for Very Dissatisfied to '5' for Very Satisfied.

Satisfaction with interpersonal reciprocity

1. How satisfied are you with amount of support you get from the community?

2. How satisfied are you that local merchants support other local merchants?

3. How satisfied are you with the amount of interest people in this community have in helping each other?

4. How satisfied are you with the amount of sacrifice local people make for the good of the community?

5. How satisfied are you with the amount of sacrifice you make for the good of the community?

6. How satisfied are you with the level of loyalty demonstrated by people in this community?

Satisfaction with institutional reciprocity

1. How satisfied are you that local merchants make an effort to provide quality merchandise or service?

2. How satisfied are you that local retailers are fair in their dealings with you?

3. How satisfied are you with the way local merchants do business in the community?

4. How satisfied are you that merchants in this community listen to consumer complaints?

5. How satisfied are you that local retailers will stand behind their products or services?

Satisfaction with retail reciprocity

1. How satisfied are you with the excitement level of shopping in this community?

2. How satisfied are you with the degree of merchandise style local merchants offer?

3. How satisfied are you with the local merchants' price offerings?

4. How satisfied are you in finding what you need in local stores?

5. How satisfied are that local retailers really try to meet consumers' needs?

6. How satisfied are you with the amount of selection local apparel merchants offer?

7. How satisfied are you with the amount of time you save while shopping locally?
COPYRIGHT 2001 JAI Press, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2001 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Miller, N.J.
Publication:The Journal of Socio-Economics
Article Type:Statistical Data Included
Geographic Code:1USA
Date:Nov 1, 2001
Previous Article:Richard Hattwick: Founder and editor of the Journal of socio-economics. (Editorial).
Next Article:A new look at prices and money: The Kelsonian binary model for achieving rapid growth without inflation.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters