The main purpose of the content evaluation was to provide a measure of the accuracy of the census items employment, payroll, and receipts. The universe of the project was a subset of establishments which responded to the census. Nonrespondents to the census and nonmail cases were not included in this study. Data were collected by means of personal visit interviews at national samples of establishments drawn from each trade area. Estimates of the total for each census item were prepared from re-interview data (assumed to be accurate) and from tabulation data (values actually used in census tabulations). The ratio of these two estimates, shown in the table below, is the measure of accuracy of the portion of the published census total that was tabulated from census respondents.