Time-of-day and day-of-week variations in Amazon Mechanical Turk survey responses

Social science research studies are frequently conducted on Amazon Mechanical Turk (MTurk). I use data from four previous surveys of inflation expectations on MTurk, with a total of 2780 observations, to study how participant characteristics and behaviors depend on the day of the week and time of day of participation. Saturday participants are older, less educated, and more likely to have low income. Even controlling for demographics, Saturday participants are more likely to answer objective knowledge questions correctly and to provide reasonable inflation forecasts, and less likely to provide “don’t know” responses. They are also less likely to provide nonsensical responses when asked for open-ended comments. Time-of-day effects are less notable. Standard data cleaning procedures typically neither reduce nor exacerbate these patterns. I add data from several other studies in political science and psychology, and find similar Saturday effects with a sample size of over 8000. I discuss implications for various research designs, especially in macroeconomics applications.