The objective of this research was to find out how the two search engines, Google and Bing, perform when users work freely on pre-defined tasks, and judge the relevance of the results immediately after finishing their search session. In a user study, 64 participants conducted two search tasks each, and then judged the results on the following: (1) the quality of the results they selected in their search sessions, (2) the quality of the results they were presented with in their search sessions (but which they did not click on), (3) the quality of the results from the competing search engine for their queries (which they did not see in their search session). We found that users heavily relied on Google, that Google produced more relevant results than Bing, that users were well able to select relevant results from the results lists and that users judged the relevance of results lower when they regarded a task as difficult and did not find the correct information.
KEYWORDSsearch engines, evaluation, results quality, interactive information retrieval, task-based user studies, retrieval effectiveness, Google, Bing
LITERATURE REVIEW Users' satisfaction with search engine resultsWhen users are asked, the majority states that they are satisfied with the results quality of search engines (Purcell, Brenner, & Raine, 2012), and "91% of search engine users say they always or most of the time find the information they are seeking when they use search engines" (Purcell et al., 2012, p. 3). A search engine's ranking of results is even considered as a criterion for credibility (Westerwick, 2013). Furthermore, users usually do not reflect on the relevance calculations made by search engine algorithms and the resulting results ordering (Tremel, 2010). Users most often choose only from the first results page, and they prefer the first few results listed . Petrescu (2014) reports that more than two-thirds of all clicks go to the first five positions, and the result ranked first alone accounts for 31% of all clicks. In a large-scale study analyzing millions of queries from the Yahoo search engine, Goel at al. (2010) found that only 10,000 different websites account for approx. 80% of clicks on the search engine results pages (SERPs). Users are generally satisfied with the first few results, even when the results positions are mixed, and therefore, less relevant results are shown on the first position(s) (Keane, O'Brien, & Smyth, 2008;Pan et al., 2007). Users generally prefer Google's results to those from other search engines. This may