The Mediating Effect of Perceived Service Risk on Perceived Value of Internet Apparel Shopping: From the Quality-Risk-Value Approach

Abstract

Abstract The significance of service quality of an Internet retailer and consumers’ risk perceptions in Internet shopping have received much investigation. However, little research effort has been paid to examine the dynamics of consumer behavior in Internet shopping from a value perception approach, which accounts for service quality, sacrifice, and risk as precursors. Groth (1995) asserts that perceived value and perceived risk are both pertinent factors in purchase and consumption decisions. Consumers evaluate risk involved in shopping activities as well as service quality and sacrifice to determine the overall value derived from shopping from the retailer (Groth 1995). Therefore, the purpose of this study was to develop insights into the interrelationships among perceived serve quality, perceived service sacrifices, perceived service risks of an Internet retailer, and perceived value of Internet shopping with the retailer. A total of 532 female students were recruited from two large, Midwestern universities, and 361 U.S. female college students provided usable responses to this experiment study. As experiment stimuli, two mock Internet apparel retail web sites (higher vs. lower service quality), were created to closely mimic “real” apparel retailer sites using Microsoft® FrontPage® and Macromedia® Dreamweaver® using the elements identified from focus group interview results. Participants were first asked about their previous Internet shopping experience and the expenditure on the clothing items purchased via the Internet. The second part of the questionnaire contained measures of four research constructs. We used descriptive statistics, t-tests, and structural equation modeling to test research hypotheses and the proposed model. An initial pilot test was conducted with eight female college students to examine the clarity and appropriateness of wording of the questionnaire. The second pilot test of the experimental treatments of levels of service quality was conducted. This between-subject manipulation check revealed that the higher service quality site had a significantly higher mean score for perceived quality (6.46) than did the lower service quality site (4.33) on a 7-point Likert scale (t=4.38, p<.01). Most participants were between 18 and 23 years old (96.7%) with a mean age of 20.8 years, and White or European American (84.8%). Majors varied; however, about three-quarters of the participants (75.9%) were majoring in merchandising and apparel related fields. About three-quarters of the participants were seniors or juniors (75.1%). Participants reported they had varied levels of previous Internet apparel shopping experience. T-test showed no significant differences between the two geographic groups in (1) participants’ demographic characteristics and previous Internet shopping experiences, and (2) their responses on research variables. Therefore, the entire sample was analyzed as a whole. We assessed the dimensionality of the measurement scales of research constructs using factor analysis. Exploratory factor analysis using principal components and varimax rotation was conducted to determine whether multiple indicators for each research variable comprised one factor dimension. A manipulation of the Internet retailer's service quality level had significant treatment effects on all research variables - perceived service quality of an Internet retailer, perceived service sacrifice of apparel shopping with the Internet retailer, perceived service risk of apparel shopping with the Internet retailer, and perceived value of apparel shopping with the Internet retailer -using a two-tailed test (p<.01). For hypotheses testing, we employed structural equation modeling (SEM) analyses. SEM analyses revealed that both perceived service quality and perceived service sacrifice had significant impacts on consumer perception of service risk of shopping at an Internet retailer. In turn, perceived service risk negatively influenced consumer perception of Internet shopping value on an Internet retailer web site. All, except one, proposed hypotheses in the model were statistically supported in this study. Findings of decomposition of effects analyses showed the partial mediating effect of the perceived service risk of an Internet retailer for both treatment situations (higher vs. lower service quality provided by an Internet retailer). The findings are limited to U.S. female consumers, who were recruited using convenience sampling; therefore, the demographics of the sample are not representative. Future studies may extend this study by adopting the random sampling method to increase the generalizability of the research findings. This paper offers an understanding of perceived service risk as an important mediator in the Internet retailing context. In addition, the present paper contributes to scholarship of Internet shopping by expanding knowledge of the quality-risk-value framework applied to Internet retailing to better understand consumer online purchasing behavior.