|Page tools: Print Page Print All|
2.10. In addition to sampling errors, other inaccuracies may occur because of insufficient coverage, inadequacies in the source of information, imperfections in answers provided by respondents and errors made in the coding and processing of data. Inaccuracies of this kind are referred to as non-sampling error and may occur in any statistical collection. Every effort is made to minimise non-sampling error by employing careful questionnaire design, trying to obtain responses from all selected enterprises and employing efficient operating procedures. Many mechanisms are employed to verify data provided by companies and individuals. These include form evaluation studies in which data providers are observed completing forms; data confrontation studies in which the same or similar data items provided in different collections are compared and reconciled on a case by case basis; and more traditional validation procedures where data provided on individual forms are checked for internal logical and empirical consistency, comparability with past data, etc. Where errors are detected, they are corrected in consultation with the provider.