A Review on Official Survey Item Classification for Mixed-Mode Effects Adjustment

Abstract

The COVID-19 pandemic has had a direct impact on the development, production, and dissemination of official statistics. This situation led National Statistics Institutes (NSIs) to make methodological and practical choices for survey collection without the need for the direct contact of interviewing staff (i.e. remote survey data collection). Mixing telephone interviews (CATI) and computer-assisted web interviewing (CAWI) with direct contact of interviewing constitute a new way for data collection at the time COVID-19 crisis. This paper presents a literature review to summarize the role of statistical classification and design weights to control coverage errors and non-response bias in mixed-mode questionnaire design. We identified 289 research articles with a computerized search over two databases, Scopus andWeb of Science. It was found that, although employing mixed-mode surveys could be considered as a substitution of traditional face-to-face interviews (CAPI), proper statistical classification of survey items and responders is important to control the nonresponse rates and coverage error risk.