Reliability of Data

All economic census results are subject to nonsampling error. Nonsampling error can be attributed to many sources during the development or execution of the economic census:

  • inability to identify all cases in the actual universe;
  • definition and classification difficulties;
  • differences in the interpretation of questions;
  • errors in recording or coding the data obtained and
  • other errors of collection, response, coverage, processing and estimation for missing or misreported data.

Although nonsampling error is not measured directly, the Census Bureau employs quality control procedures in all phases of the collection, processing, and tabulation of the data to minimize the effects of nonsampling error.

Selected results from the economic census are also subject to sampling error as well as nonsampling error. Sampling error occurs because data are requested from a sample instead of a complete enumeration of establishments in the population. The sample selected for each sector is one of many probability samples that could have been selected under identical circumstances. Each of the possible samples would yield a different set of estimates. Common measures of the variability among these estimates are the sampling variance, the standard error, and the relative standard error (RSE). A relative standard error is an expression of the standard error as a percent of the quantity being estimated. No measures of sampling variability are provided for sample-based estimates derived from the economic census, except for the Construction sector. For additional information about sampling and nonsampling errors, see the methodology text for the sector of interest.

Source: U.S. Census Bureau | Economic Planning and Coordination Division | 1-877-790-1876 | Last Revised: March 05, 2015