Survey Sampling, Administration, and Analysis

The use of surveys in social science research is abundant and may appear straightforward, but can involve a complex set of procedures. Total survey error refers to all of the errors that could occur in researchers’ attempts to gain valid information about a population with the use of surveys. These errors have been placed into five categories: coverage error, nonresponse error, editing and processing errors, measurement error, and sampling errors (Groves, 1989). Coverage error refers to the failure to provide some members of the population the chance to be selected into the sample, nonresponse error refers to the failure to obtain responses from all members of the selected sample, editing and process errors refer to the failure to capture data accurately from the respondents, measurement error represents the failure of the observed response to reflect the true opinion of the sample member, and sampling error refers to the fact that sample statistics are not expected to exactly reflect population parameters.