Year: 2017
Type: report
Abstract: This paper focuses on data quality within the Afrobarometer project, which conducts surveys in a wide number of African countries to investigate citizens’ beliefs about democracy and political institutions, as well as gain information on their political behaviour. An empirical assessment of interviewer variance is conducted for several key items of the Afrobarometer surveys, for 12 countries and across 3 survey rounds. Interviewer variances are estimated by making use of two-level linear and logistic models, as well as cross-classified models to distinguish between interviewer and area or sampling unit effects. The analyses indicate that interviewer variances are substantial in Afrobarometer data and require further attention. Several recommendations are made to improve Afrobarometer data quality and direct further research.
Source: Lirias
Author: Demarest, Leila
Institution: KU Leuven
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W6989434312