Towards Validation of the Preference for Automation Types Scale in the Context of a Chatbot writing Essays

In a world where automated systems and people are often in contact, their relationship could be improved by better understanding the psychological mechanisms of interaction with machines. To reach this goal many psychological constructs, such as trust in automation or affinity for technology interaction, have been researched and accompanying questionnaires developed. This study examined the novel construct preference for automation types and partly validated the new Preference for Automation Types Scale developed at the Institute for Multimedia and Interactive Systems at the University of Lübeck. In addition, preference for automation types was compared before and after usage of a chatbot as an automated system. The fit between the model proposed by the Preference for Automation Types Scale and the gathered data (N = 81) was judged by the results of exploratory and confirmatory factor analyses. Additional data collection served the purpose of construct- and face-validating the Preference for Automation Types scale. While results showed a poor model fit, construct and face validity were indicated. No difference in preference for automation types was found between before and after usage of a chatbot.