The impact of questionnaire design on measurements in surveys 1
|Convenor||Dr Natalja Menold (GESIS )|
|Coordinator 1||Ms Kathrin Bogner (GESIS)|
In social surveys “total net household income” is an indicator of the socio-economic status. It is used as an explanatory variable in mobility studies and as a social-demographic background item in inequality research. In social sciences, the income brackets are usually good enough for a comparative analysis of social structures.
The question design and the answer categories have to fulfill quality requirements: all possible payments accruing to a household and all its members must be reported in references; all households in the survey’s universe must be represented in the statistics used to detect the the answer categories.
The presentation analyses the influence of readability of European Social Survey 2010 (ESS) questionnaire items on item non-response in two countries and in two languages - in the English language for Great Britain and in Estonian language for Estonia from 2010. The comparison of questionnaires of several languages enables finding more universal results and avoiding results that are specific for just one language or one country.
We found clear evidence of the influence of questionnaire readability on item non-response. Results also allow assuming, that the relationships between item readability characteristics and item non-response might be society -dependent.
Despite some previous evidence that numeric labels in rating scales might be associated with lower data quality, numeric labels have been broadly used in social science surveys. In the present paper the differences in respondents’ burden und reliability between the rating scales with numeric and verbal labels in web surveys is addressed. The results of the first eye-tracking study show that respondents’ burden were higher in the case of numeric than in the case of verbal labels. In the second study lower reliabilities were obtained for rating scales with numeric labels than for those with verbal labels.