01(8)_논문 08.pdf457.6KB

Ⅰ. Introduction

Rural communities are important to the expansion plans of several national chain retailers in the U.S., such as Dollar General, Family Dollar, and grocery chains such as Kroger (Horton, 2010). Wal-Mart began their national chain store in rural areas and has successfully expanded throughout rural and suburban America as the store selling not only grocery but also low-end apparel products (Stone, 1997). However, despite the importance of rural areas to the retail industry, research involving such communities is scarce even in the face of rapid infiltration of national chain players into rural communities and increased broadband penetration in rural areas in the past decade. 

 Several early studies discuss the differences between rural and urban consumers in terms of their behaviors, attitudes, and lifestyles (Glenn & Hill, 1977; Smith & Krannich, 2000). Rural people are more likely to be traditional, prejudiced, and ethnocentric, while people in urban areas are prone to favor change (Lowe & Peek, 1974). Due to these differences in behaviors and attitudes between rural and urban consumers, it is expected that the two groups will also vary in their use of criteria in building loyalty toward a specific retailer in which to make purchases. Furthermore, the unique nature of rural communities such as their dense and closed networks suggests that shopping behaviors of rural consumers may differ from those of urban shoppers (Kim, 2008). Knowledge of the differences in determinants of loyalty between rural and urban consumers is important to help apparel retailers use their limited marketing dollars effectively to increase their customers’ loyalty, both in rural and urban communities.

 Past research shows that there are several store attributes that affect store loyalty (Lumpkin et al., 1985), such as service quality (Harrison-Walker, 2001; Odekerken-schroder et al., 2001) and product assortment (Matos et al., 2008). In addition, early studies provide ample evidence that customers who are satisfied with store attributes are likely to intend to repurchase (Baker et al., 2002; Pan & Zinkhan, 2006) and participate in positive WOM (Bitner, 1990; Anderson, 1998). Similar to findings of previous research, this study expects that customers who are satisfied with store attributes will be more likely to repurchase from the store and be more likely to engage in positive WOM about the store, implying a relationship between store attribute factors and loyalty. 

According to Infomat Research (2010), women’s wear accounts for approximately 60 percent of the apparel market in the U.S. currently. Since women’s apparel has an influential role in the apparel market as well as the retail market, this study explores rural and urban consumers’ shopping behaviors specifically for women’s apparel. Therefore, this study examines differences in women’s clothing shopping behaviors for rural versus urban consumers, focusing on key store attributes as determinants of loyalty toward a specific retailer. 

Ⅱ. Review of Literature

1. Store attributes as determinants of store loyalty

  Loyalty refers to repeated purchase behavior over a period of time based on favorable feelings toward the subject in terms of both attitudinal and behavioral aspects (Keller, 1993). Oliver (1997) defined loyalty as “a deeply held commitment to rebuy or patronize a preferred product/service consistently in the future, thereby causing repetitive same-brand or same brand-set purchasing despite situational influences and marketing efforts having the potential to cause switching behavior.” The definition places emphasis on two aspects of loyalty, which are attitudinal and behavioral. Many previous studies have argued that there must be both behavioral and attitudinal commitment (e.g., favorable set of stated beliefs) for true loyalty to exist (e.g., Jacoby & Chestnut, 1978). The behavioral aspect indicates repeated purchases, while the attitudinal component includes a degree of dispositional commitment in terms of some unique value associated with the brand or retailer (Chaudhuri & Holbrook, 2001). Attitudinal loyalty is often measured by customers’ recommendations to others (i.e., WOM) (Bowen & Chen, 2001), since customers who have a favorable attitude towards a retailer or brand are likely to be engaged in WOM (Dick & Basu, 1994; Bowen & Chen, 2001).

  Many previous studies have investigated the antecedents of loyalty (e.g., Baker et al., 2002; Pan & Zinkhan, 2006) and have found that store attributes are significant factors that lead to customers’ loyalty towards a certain store. Previous studies have supported the widely accepted notion that customers who are satisfied with store attributes are likely to intend to repurchase (Baker et al., 2002; Pan & Zinkhan, 2006) and participate in positive WOM (Bitner, 1990; Anderson, 1998). Pan and Zinkhan (2006) used meta-analysis to identify predictors of retail patronage that have frequently been reported in previous literature. Product quality, price, product selection/assortment, convenience, service quality, friendliness of salespeople, store image, store atmosphere, fast checkout, and personal relationships were common predictors of store loyalty. They found that among the various predictor variables investigated, wide selection, service quality and product quality were highly associated with satisfaction and patronage behavior. Another study investigated service quality perceptions and found that they were positively associated with store loyalty intentions (Baker et al., 2002). Specifically, interpersonal service quality had a positive influence on store patronage intention, and merchandise quality and monetary price perception had a positive effect on store patronage intentions via merchandise value perception. Other studies examining elements of the store environment found that service quality and merchandise value had a significant influence on store loyalty intentions (Baker et al., 2002). Regarding attitudinal loyalty, Harrison-Walker (2001) found that service quality was a significant antecedent of WOM activity, and high service quality led to positive WOM activities among customers. Common factors across previous studies were convenience, price, merchandise assortment, service quality, and store atmosphere (Baker et al., 2002; Pan & Zinkhan, 2006).

 In addition to the common store attributes that lead to customers’ store loyalty, Sullivan, Savitt, Zheng, and Cui (2002) studied the specific context of selecting an apparel retailer and found convenience, price, merchandise assortment, value for money, fashionability of current styles, merchandising, and sales service were key determinants of store choice. In contrast, store promotion did not have a significant effect on customers’ loyalty towards the retailer. According to Gedenk and Neslin’s (1999) study, promotion had no significant influence on future brand loyalty. Specifically, price promotions were negatively related to future brand loyalty, while non-price promotion had no effect on the future brand loyalty.

 Among the various determinants of retail store loyalty suggested by previous literature, this study will focus on a variety of attributes that we expect to represent fashionability; promotion; shopping environment including service quality; and retail basics including convenience, price and merchandise assortment Therefore, this study proposes that consumer satisfaction with store attributes related to fashionability, shopping environment, and retail basics will have a positive influence on loyalty towards women’s clothing retailers, whereas promotion will not have a significant influence on loyalty among women’s clothing shoppers. The hypotheses are as follows.

 H1: Satisfaction with fashionability related store attributes is positively associated with loyalty among women’s clothing shoppers.

 H2: Satisfaction with promotion related store attributes is not associated with loyalty among women’s clothing shoppers.

 H3: Satisfaction with shopping environment attributes is positively associated with loyalty among women’s clothing shoppers.

 H4: Satisfaction with retail basic attributes is positively associated with loyalty among women’s clothing shoppers.

2. Urban versus rural community

 Early research defined a rural community as comprising farm residents and residents of towns with a population under 2,500 (Lowe & Peek, 1974), while an urban community includes residents of cities of 50,000 population or above, together with residents of their suburbs. Kean et al. (1998) suggested three criteria in regards to defining a rural community: 1) economic base: at least 20 percent of earned income attributed to agriculture, manufacturing, or tourism; 2) population size: 2,500 - 9,999 people; and 3) distance from the nearest Standard Metropolitan Statistical Area (SMSA) excluding communities that were inside a SMSA. More recently, according to the rural-urban continuum code from United States Department of Agriculture (USDA, 2003), an urban area is defined as: 1) counties in metro areas of one million population or more; 2) counties in metro areas of 250,000 to 1 million population; or 3) counties in metro areas of fewer than 250,000 population. Rural counties include those: 1) with urban population of 20,000 or more, adjacent to a metro area; 2) with urban population of 20,000 or more, not adjacent to a metro area; 3) with urban population of 2,500 to 19,999, adjacent to a metro area, 4) with urban population of 2,500 to 19,999, not adjacent to a metro area; 5) completely rural or less than 2,500 urban population, adjacent to a metro area; or 6) completely rural or less than 2,500 urban population, not adjacent to a metro area.

 Several studies have investigated the differences between rural and urban communities in terms of their residents’ behaviors, attitudes, and lifestyles (Glenn & Hill, 1977; Smith & Krannich, 2000).Rural people are more likely to be traditional, work-oriented, puritanical, prejudiced, ethnocentric, and isolationist, whereas people in metropolitan areas tend to favor change, be less lenient towards the treatment of criminals, and be more lenient toward divorce than people in rural areas (Lowe & Peek, 1974). Previous researchers found that these different attitudes and behaviors were attributable to different characteristics of rural and urban societies including occupation, age, religion-ethnicity, the degree of population concentration, the size of community (Glenn & Hill, 1977), education of the citizens (Smith & Krannich, 2000), and the lifestyles of the people living within the community (Lowe & Peek, 1974).

 In addition to attitudinal and demographic differences, there are differences in interpersonal relationships of rural and urban customers due to the differing sizes of the two types of communities (House et al., 1988; Beggs et al., 1996). House, Umberson, and Landis (1988) found that the level of social integration is higher in small communities (i.e., rural towns) than in large urban areas. Beggs, Haines, and Hurlbert (1996) and Miller and Kean (1997) also found that, relative to urban dwellers, rural residents have a larger proportion of long-term and strong relationships including kin, neighbors, and kin-neighbors and a smaller proportion of shortterm relationships.

3. Moderating effect of rural versus urban customers

 Although several studies have compared general characteristics, lifestyles, or interpersonal relationships between rural and urban residents (Glenn & Hill, 1977; House et al., 1988; Beggs et al., 1996; Smith & Krannich, 2000), very few studies have taken into account comparisons between urban and rural customers’ shopping behavior (Sun & Wu, 2004). In particular, no studies could be identified that study differences in urban and rural apparel shopping behavior in the United States.

 Sun and Wu (2004) compared urban and rural consumers in China, and found that metropolitan customers were more sophisticated than non-metropolitan customers. Urban customers were more likely to shop at popular and large-scale retailers and stores with better customer service. The authors argued that urban residents have more access to alternative retailers, while rural customers do not have diverse retail options from which to select. In contrast, consumers in rural areas were more price-conscious than consumers in cities; rural customers tended to compare or negotiate the price when shopping while urban consumers accepted the marked price, perhaps due to the different economic status of rural and urban customers. Rural shoppers were more likely to purchase inexpensive, functional, and less innovative products, while urban customers were more likely to buy trendy, innovative, and unique products.

 Based on previous studies, since rural customers tend to place a high priority on retail basic factors such as price and product, rural customers would be more likely to build a loyal relationship with a retailer where they are satisfied with store attributes relating to price and product assortment. On the other hand, since urban customers tend to place emphasis on fashionability of product and shopping experience factors such as service and store environment, the fashionability and shopping environment would play an important role for the urban customers to decide whether or not they patronize a certain women’s clothing retailer.

 H5: Retail basics (e.g., product selection and price) have greater influence on rural customers’ loyalty than urban shoppers’ loyalty toward women’s clothing retailers.

 H6: Fashionability of clothing has greater influence on urban customers’ loyalty than rural shoppers’ loyalty toward women’s clothing retailers.

 H7: Shopping environment has greater influence on urban customers’ loyalty than rural shoppers’ loyalty toward women’s clothing retailers.

Ⅲ. Research Method

1. Data collection

 The secondary data used for this study was collected by BIG research in their Consumer Intentions and Actions Study. Their panel survey comprises data from 8,197 panel members surveyed from April 6 to April 13, 2010. The panel is distributed across the entire population of the United States and is selected and contacted by BIG research. Compared to other previous studies, the sample of this study is based on a random sample reflecting the general U.S. population. Thus, this data from BIG research would provide generalizations for U.S. consumers considering differences between rural and urban shoppers. The survey assesses consumer shopping behaviors for a variety of products each month over a twelve month period and includes categories such as clothing, cell phones, fast food, and restaurants. A different product is examined in-depth each month. This study uses the survey data for women’s clothing.

 From the 8,197 panel-member responses, respondents with missing values for county code were removed since county code was used to categorize each member as rural or urban. This left a sample size of 7,975, including 1,270 rural customers and 6,705 urban customers. For analysis purposes, similar sub-sample sizes are desirable, so a sample of 1,270 urban consumers was drawn via systematic random sampling routine in PASW (SPSS 18.0) and used for analysis. Thus, the total sample was 2,540, comprising 1,270 each of rural and urban panel members. 

2. Measurement

 The BIG research survey contains a total of 157 items including demographics (e.g., age, gender, race, education, employment status, and county code) and general consumption behaviors (e.g., how much the respondents spend for shopping this month, which shop they go to the most, why they select the shop, why they switch stores, and intention for future purchasing) for various products and services (e.g., furniture, fast food, full-service restaurant, cell phone, women’s clothing, men’s clothing, children clothing, toy, shoes, sports goods, bedding, hardware, and car). Among these, 16 items were related to women’s clothing consumption behavior.

 This study addressed factors underlying store attributes and loyalty in terms of women’s clothing. The respondents were asked where they shop for women’s clothing most often and they chose one retailer out of 41 choices, including 39 retailers (e.g., Wal-Mart, Kohl’s, JC Penney, Macy’s, Target, Nordstrom, Victoria’s Secret, and Gap), others, and no preference. The following items regarding store attributes and loyalty were related to the specific retailer that they choose in the previous question.

 Loyalty was measured by both behavioral and attitudinal aspects. For behavioral loyalty, the respondents were asked approximately how many years they have shopped most often at the store mentioned above for women’s apparel (one retailer out of 39 retailers that they selected as the place they shop for women’s clothing the most often); the item was a 25-point scale (1= 1 year, 25= 25 or more years). For attitudinal loyalty, the likeliness of giving a recommendation was measured with: “How likely is it that you would recommend this store for Women's Clothing to a friend or colleague?” Respondents used an 11-point Likert type scale anchored on ‘extremely likely’ to ‘not at all likely’ (0=not at all likely/10=extremely likely).

 Store attributes were measured by answers to the question: “What are the reasons why you buy your Women's Clothing there?” Respondents were requested to check all that apply among “price”, “selection”, “location”, “quality”, “service”, “advertising”, “fashion ideas”, “newest styles”, “newest fabrics”, “in-store experience”, “knowledgeable sales people”, “store appearance”, “store layout”, “store credit card”, “store loyalty card”, “coupons and special sales”, and “trustworthy retailer” (No=0, Yes=1).

To identify respondents as living in a rural or urban community, the Rural-Urban Continuum Code (2003) developed by the United States Department of Agriculture (USDA) was used. The 2003 Ruralurban Continuum Codes classifies metropolitan counties by size and nonmetropolitan counties by degree of urbanization and proximity to metro areas, and each of the 3103 counties was assigned one of the nine Rural-Urban Continuum Codes: 1 is counties in metro areas of one million population or more; 2 is counties in metro areas of 250,000 to one million population; 3 is counties in metro areas of fewer than 250,000 population; urban population of 20,000 or more, adjacent to a metro area; 4 is urban population of 20,000 or more, adjacent to a metro area; 5 is urban population of 20,000 or more, not adjacent to a metro area; 6 is urban population of 2,500 to 19,999, adjacent to a metro area; 7 is urban population of 2,500 to 19,999, not adjacent to a metro area; 8 is completely rural or less than 2,500 urban population, adjacent to a metro area; and 9 is completely rural or less than 2,500 urban population, not adjacent to a metro area. Using the Rural-Urban Continuum Codes, the codes were collapsed into two categories: metro counties and non-metro (rural) counties; codes from 1 through 3 are metro counties and from 4 through 9 are non-metro counties, which considers population size as well as distance from a metro area. 

3. Analysis

 A preliminary exploratory factor analysis (EFA) was conducted to identify the store attribute factors. EFA was selected since it is better suited for identifying factor structure while principal component analysis is primarily used for data reduction (Norris & Lecavalier, 2009). EFA was conducted with Comprehensive Exploratory Factor Analysis (CEFA) software (Browne et al., 1998), and oblique Crawford- Ferguson(CF)-Quartimax rotation method was used to accommodate the binomial answers of the store attribute questions. Also, the factors of store attributes would probably correlate with one another, this study selected oblique rotation, which does not have the restriction that the rotated factors are not required to be uncorrelated. The Ordinary Least Squares (OLS) discrepancy function was used to estimate the model fit, since this method is preferable to the discrepancy function used in Maximum Likelihood (ML) when the data is not normally distributed (Norris & Lecavalier, 2009).

 To test the hypotheses, SPSS 18.0 was used for multiple regression analysis. To analyze the relationship between store attributes and loyalty and the moderating effect of urban versus rural customers, hierarchical regression analysis was used in this study. The first step consisted of entering independent variables (i.e., store attribute factors) as one block into the model. Next, a moderating variable (i.e., rural versus urban) was added to the model as the second block. For the final step, an interaction term was entered, which was calculated as the product of store attribute factors and rural versus urban.

Ⅳ. Results

 In preliminary analysis, exploratory factor analysis (EFA) followed by an oblique CF-Quartimax rotation was used to identify determinants of customers’ choice of retailer. The factor loadings of all items were at least 0.5 in size. Based on the factor analysis, four store choice factors were identified with each comprising multiple items: “fashionability” (newest styles, newest fabrics, and fashion ideas); “promotion” (coupons and special sales, store loyalty card, and store credit card); “shopping environment” (store appearance, knowledgeable sales people, store layout, service, location, and product selection) (see Table 1). The average of the scale items for each factor was used for further analysis.

<Table 1> Factor loadings for store attributes

To test the hypotheses of this study, hierarchical multiple regression analysis was used. Table 2 illustrates the results of the hierarchical multiple regression for loyalty. The store attribute factors, rural versus urban customers, and the interaction term were used as predictors, and loyalty was the dependent variable. 

<Table 2> Hierarchical multiple regression analysis for loyalty

 In step1, the four dimensions of store attributes were entered in the model. All four dimensions of store attributes were found to be significant. The beta values showed that fashionability (β=.073, p<.000), promotion (β=.066, p<.000), shopping environment (β=.066, p<.000), and retail basics factors (β=.547, p<.000) were significant predictors of loyalty. Positive beta values suggested that customers who selected a women’s apparel retailer based on fashionability (i.e., newest style, fabric and fashion ideas), special promotions (i.e., coupons, loyalty card), shopping environment (i.e., service, sales person, and store appearance), or retail basics (i.e., price, product selection, location) were more likely to show loyalty behavior (i.e., long term relationship and recommend the retailer to others) toward their preferred retailer. In other words, all criteria for selecting retailers had a positive influence on loyalty. Among these four criteria, the retail basics factor had the strongest impact on loyalty. Therefore, hypothesis 1, 3 and 4 were supported while hypothesis 2 was not supported.

In step 2, the rural versus urban factor was added to the model. The increase in R2 was significant (p=.001). Finally, in step 3, the moderation effects of rural versus urban on the relationship between the store attributes and loyalty were examined. The increase in R2 from model 2 to model 3 was significant (p=.046). The results showed that there was significant moderation effect of rural versus urban on the relationship between fashionability of store attributes and loyalty (β= .057, p=.022). Since the rural versus urban was  coded as a dummy variable (urban=0; rural=1), the negative beta value of the interaction term between fashionabaility and rural versus urban indicates that fashionability had a greater influence on loyalty among the urban customers than the rural customers. In addition, the rural versus urban significantly moderated the relationship between retail basics and loyalty (β=.070, p=.046). In other words, the impact of retail basics on loyalty was stronger for the rural shoppers than urban shoppers. Therefore, hypotheses 5 and 6 were supported while hypothesis 7 was not supported.

Ⅴ. Discussion

 This study examined the relationship between the store attributes and loyalty towards women’s clothing retailers. This study found that store attributes including fashionability, promotion, shopping environment, and retail basics influenced store loyalty for women’s clothing retailers. Regarding the positive influences of fashionability, shopping environment, and retail basics on loyalty, the results correspond to many previous studies that have investigated the antecedents of store loyalty (e.g., Baker et al., 2002; Pan & Zinkhan, 2006). In contrast, the results showed a positive relationship between promotion and loyalty, which is different from the previous study of Gedenk and Neslin (1999). This might be because the current study included in the promotion factor price promotions as well as loyalty cards while Gedenk and Neslin (1999) considered only gifts or price promotions. The store loyalty card and store credit card might play a role in increasing loyalty among women’s clothing shoppers.

 Furthermore, this research found that there was a moderating effect of rural versus urban women’s  clothing shoppers on the relationship between store attributes and store loyalty. Specifically, store attributes of fashionability had a greater influence on loyalty for women’s clothing shoppers in urban areas than for women’s clothing shoppers in rural areas. This result corresponds to Sun and Wu’s (2004) finding that urban customers were more likely than rural shoppers to buy trendy, innovative, and unique products. This might be due to differences in characteristics of consumers living in rural versus urban communities in terms of age and occupation (Lowe & Peek, 1974; Glenn & Hill, 1977; Smith & Krannich, 2000), and these different characteristics would lead urban customers to place emphasis on current fashion style as a key determinant of loyalty to a women’s clothing retailer.

 Finally, the retail basics factor, which includes price, location, and product selection, was the most significant predictor among store attribute factors on loyalty for urban and rural women’s clothing shoppers. This finding illustrates the importance of good practice in terms of pricing, location and merchandise assortment. Despite the great emphasis on retail basic factors for both rural and urban women’s clothing shoppers, the rural shoppers are likely to place more emphasis on retail basic factors than urban shoppers when they choose women’s clothing retailers. This result corresponds to Sun and Wu’s (2004) findings that rural shoppers were likely to be more price-conscious and buy less expensive and more functional products than urban shoppers.

Ⅵ. Managerial Implications

 Results of the study have some managerial implications. First, this study suggests that apparel retailers need to pay attention to retail basic factors such as price, location, and product selection to build loyalty with customers. Although all the store attributes had a positive influence on loyalty, the retail basics were the most significant predictor of loyalty. Reinforcing efforts to provide enhanced retail basic factors would be the most significant keys to build customer loyalty and encourage customers’ recommendation to others regardless of a consumer’s location in a rural or urban community. Since WOM has been found to be one of the most powerful and influential sources of information affecting purchasing behavior (Brown & Reingen, 1987; Wangenheim & Bayon, 2004), strong retail basic store attributes will not only result in a long-term relationships with existing customers but also in attracting new customers to the store through WOM.

 Second, this study suggests that women’s clothing retailers need to use different strategies to attract metropolitan versus non-metropolitan customers. The women’s clothing retailers in metropolitan areas might need to provide the newest styles and trendy clothing to meet urban customers’ needs and retain their loyalty. On the other hand, apparel retailers targeting rural customers might need to focus on reasonable price, convenient location, and diverse product selection to build customers’ loyalty. This strategy might allow local retailers to strengthen their business in rural areas where business can be tough to come by (Bhat & Fox, 1996).

Ⅶ. Recommendations for Future Study

 Future studies examining store attribute preferences should consider implementing questions using scaletype responses in order to measure the differences in magnitude among the store attributes. This would enable researchers to compare and contrast various store attributes in relation to each factor’s relative importance.

 Considering women’s clothing segmentation whether formal wear or casual wear, customers would use different criteria to select the store to shop for different clothing type. There is possibility that rural customers would wear casual clothes more than urban customers. Future studies should consider clothing type as a control variable to get better understanding of comparison between rural women’s clothing shoppers and urban shoppers.

 Furthermore, similar studies should be conducted using different types of retailers. This study focused specifically on women’s clothing retailers. Future studies should be conducted in other shopping environment contexts to enable comparisons between different shopping environmental settings.