As you might expect from something called the National Crime Victimization Survey, it's quite detailed. Initial interviews are conducted in-person. There's a sc

As you might expect from something called the National Crime Victimization Survey, it's quite detailed. Initial interviews are conducted in-person. There's a screening questionnaire that asks extensively about a bunch of specific kinds of crime: https://bjs.ojp.gov/content/pub/pdf/ncvs22_bsq.pdf https://bjs.ojp.gov/content/pub/pdf/ncvs22_bsq.pdf For instance, for theft, respondents are asked separately about 8 different categories of things that might have been stolen. Then, if any crimes are mentioned, they're asked an entire follow-up about each incident, including 35 pages of questions: https://bjs.ojp.gov/content/pub/pdf/ncvs22_cir.pdf https://bjs.ojp.gov/content/pub/pdf/ncvs22_cir.pdf To account for memory problems with dates (people are bad at remembering how long ago something happened), respondents have follow-up interviews every six months to ask about events that happened in between interviews, so the results can be compared to their previous survey six months ago. You can find more details and reports on their site here: https://bjs.ojp.gov/data-collection/ncvs https://bjs.ojp.gov/data-collection/ncvs