In July 1998, surveys were mailed to agencies in both the representative and comparative samples. Surveys were addressed to the respondent from the previous year or to the chief of police or the sheriff. Within the first few months, the response rate was approximately 50 percent, with surveys being received either through surface mail or fax (a self-addressed, postage-paid envelope and toll-free fax number were provided to each survey recipient). After followup calls were conducted, the response rate increased to 92 percent for the representative sample and 81 percent for the comparative sample. Response rates varied by area type for the representative sample, but not significantly (see figure 1).

In a few cases, respondents failed to answer one or more survey questions. In these instances, the agency was excluded from the analysis of the question or questions. Evaluation of the number of agencies that answered different questions revealed that respondents were less likely to answer if they were asked to indicate a percentage or raw number. In contrast, respondents were more likely to answer questions that asked them to select from a list. There did not appear to be any other survey design or methodological problems that contributed to the likelihood that a respondent would or would not answer a question.