Consumer Acceptance of Online Customization for Apparel

As online markets grow and become increasingly competitive, customization is becoming recognized as an important tool to satisfy and retain e-commerce customers. By adopting customization, the apparel industry may have a greater potential than ever before of being successful in e-commerce. While previous studies have presented different observations about consumer responses toward online customization, no studies have clearly identified the key determinants and the path to build consumers' positive responses and future intentions toward online customization system for apparel. Thus, the purpose of this study was to explain consumers' acceptance of online customization for apparel by identifying factors that influence the future intention of using a customizing system and to investigate the relationships between the factors. The specific objectives of this study were 1) to determine whether two major determinants of the Technology Acceptance Model (TAM) are valid to explain consumer acceptance of online customization, 2) to investigate the role of trust in the acceptance of online customization for apparel, and 3) to investigate how complexity levels of apparel customization impact consumer perceptions of websites. A research model was constructed based on TAM and on the multi-dimensional trust model from Structural Equation Modeling (SEM) techniques. The model hypothesized the relationships between eight latent variables with a total of 37 indicators. An online survey was used to collect data based on respondents' experiences for customizing jeans in different complexity levels (high, medium, and low) of the shopping procedure. Three types of surveys were developed to present one shopping simulation in each survey. Samples were 300 female adults (100 for each survey) secured by random (nationwide) and quota sampling methods. A pilot test was conducted prior to performing the main study. The hypothesis was tested by AMOS 7.0 and the assumptions for SEM and descriptive statistics were analyzed by SPSS 15.0. Compared with the demographic characteristics of the nationwide population reported by the U.S. Census Bureau, the characteristics of the samples were younger and had a higher education and a slightly higher income. Little difference existed between the groups reported from the U.S. Census Bureau and this study in terms of the race and area distributions. Two latent variables were combined as one and seven indicators were dropped by Confirmatory Factor Analysis (CFA). Thus, overall model fit tests were based on seven latent variables and a total of 30 indicators. The initial model was not acceptable and was revised by adding a path from Perceived Ease of Use to Perceived Safety at first, and then by adding a path from Perceived Usefulness to Perceived Safety. Interestingly, the accepted model presented a negative relationship between Perceived Ease of Use and Intention, which is not consistent with previous studies. Additionally, the final model showed the importance of Perceived Safety which mediated the relationships between either Perceived Usefulness or Perceived Ease of Use and Trust. The older age group (over 40) indicated that the usefulness of a website with a low level of complexity is not significantly different from that of a website with a high level of complexity. The results suggested that TAM might be useful in identifying factors that influence consumer acceptance of online customization, but the interpretation needs to be different. This study also supported the need for extending variables for the TAM to explain consumer acceptance of online customization for apparel. Finally, this study provided a revised insight for antecedents of trust in initial relationships. Implications were suggested to marketers, followed by limitations and future research.