Measuring the Credibility of Recommender Systems KyungHyan Yoo a , Ulrike Gretzel a a Laboratory for Intelligent Systems in Tourism Department of Recreation, Park & Tourism Sciences Texas A&M University, USA {toinette, ugretzel }@tamu.edu Abstract Recommender systems promise to support travelers in complex decision-making processes; however, whether a recommendation is seen as credible advice and actually taken into account not only depends on travelers' perceptions of the recommendation but also the system as the advice-giver. Trust has been identified as an important factor in online relationships but it is often focused on security issues that are of lesser relevance for recommender systems. It is argued that conceptualizations of trust in terms of credibility are more important to evaluate the persuasiveness of recommender systems. The findings of a study to develop and test credibility measures for recommender systems are presented and implications for future research are discussed. Keywords: recommender systems; credibility; expertise; trustworthiness; scale development. 1 Introduction The introduction of the Internet has led to an explosion of information and fundamentally changed information search behavior, especially in the realm of travel and tourism. The number of people who search information using the Internet has dramatically increased and recent survey results indicate that Internet searches are overtaking personal recommendations as the preferred means for obtaining travel information (eMarketer, 2005). However, it is often difficult to find information in digital environments and too much information can easily lead to confusion and information overload (Henry, 2005). Electronic agents promise to manage information acquisition for consumers, thus reducing cognitive effort and freeing cognitive capacity needed for engaging with the information obtained (Dholakia & Bagozzi, 2001). Recommender systems are one type of electronic agent that is expected to play an increasingly important role in helping consumers find what they need and want (Kim & Kim, 2001; Barwise, Hammond & Elberse, 2002). Häubl and Trifts (2000) define recommender systems as software tools that make