This study compared two mock search engines, one delivering search results in 250 ms and a slower search engine delivering search results in either 2, 3, 4, or 5 seconds. The key findings are:
• User perception, satisfaction, stated preference, and choice (observed preference) are moderately correlated.
• Regardless of slow search engine latency, user stated preference is inconclusive.
• Regardless of slow search engine latency, the difference in user satisfaction scores between the search engines is inconclusive.
• When the slower search engine latency is 5 seconds, some users state they perceive the faster engine as faster.
• When the slower search engine latency is 4 or 5 seconds, some users choose to use the faster engine more often.
• Based on pooling data for 2 s and 3 s, once latency ex- ceeds 3 seconds for the slower engine, users are 1.5 times as likely to choose the faster engine.