Looping questions are used to collect data about several similar events, such as employmentspells, retirement accounts, or marriages. The loops gather information about the number ofevents experienced as well as details about each one. The questions require respondents to thinkhard to recall each event and are often lengthy and repetitive. Looping questions can be askedin two formats, and which format a survey uses may affect the quality of the data collected.We develop hypotheses about the effects of format on measurement error in looping questionsand test the hypotheses using experimental data from a web survey with a link to administrativerecords. Results show that one format collects more accurate event reports, but the other formatprovides higher quality data to the follow-up questions. We conclude with guidance for thosewho write survey questions and those who rely on survey data for substantive analyses.