A case study on attitude towards online auction use applying quantile regression analysis

This study uses the technology acceptance model (TAM) as a basis, extended by conducting quantile regression to examine the factors influencing the use of an online auction Website and then comparing the results to the conventional linear regression-type approaches. Specifically, the TAM has been widely assessed in various research contexts by using ordinary least squares regression model and structural equation modelling. However, these two methods focus on describing the ‘average’ behaviour of a conditional distribution, thus overlooking extreme value and outlier effect. Following this, this study applied quantile regression model to examine the TAM for online auctions, thereby complementing previous empirical findings about the TAM. Using survey data from 476 online auction users with experience selling items on Yahoo!Auction, the results indicate that perceived ease of use is most important for customers with a more favourable attitude towards the site, whereas perceived enjoyment plays a more crucial role in improving attitudes towards using the site among customers with less favourable attitudes towards the site. Findings also suggest that perceived usefulness is more effective in improving attitude towards using the site among customers with moderately favourable attitudes than among customers with unfavourable and very favourable attitudes.