Project Details
Projekt
Modular Questionnaire Designs for Social Surveys: Statistical Modelling of Design Missingness
Subject Area
Empirical Social Research
Term
from 2018 to 2023
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 407454818
Surveys have become an indispensable source of information on social and political circumstances in modern societies. Analyses of such surveys oftentimes feed directly or indirectly into policy advice. Thus, data from social surveys have to be valid and reliable, and ever larger datasets containing ever more complex data structures are required.Unfortunately, together with decreasing response rates and increasing fieldwork efforts, the higher requirements regarding data quality lead to surging survey costs. As a result, institutes responsible for survey implementations are looking for new ways to collect adequate data at lower costs. This has led to a rapid increase in online surveys in the past years, which promise cost savings when compared to traditional ways of data collection, such as face-to-face interviews. However, online surveys place an upper limit on questionnaire length that is markedly lower than in face-to-face surveys. Thus, until now online surveys have not been able to substitute these traditional data collection methods. For that reason, an approach is needed which allows for compressing longer questionnaires in shorter online survey formats. Such an approach is the modular questionnaire design, which breaks with the tradition of presenting all questions in a questionnaire to all respondents. Instead, respondents only receive parts of the complete questionnaire, and practical ways have to be established to impute the resulting missing values reliably and efficiently. This is the main aim of the project proposed here. In detail, the project will investigate whether these modular questionnaire designs and imputation methods, leading to shorter online surveys, can be an alternative to large-scale, face-to-face social surveys. Besides diminishing the costs, modular questionnaire designs may offer many other benefits resulting from the reduced questionnaire length, such as an increase in data quality owing to the avoidance of respondent boredom and fatigue. Such approaches, however, are not well investigated with regard to the application in social surveys and demand a comprehensive investigation.The project follows a three-step procedure to achieve its main aim. Firstly, module-building strategies are developed with regard to criteria that are highly relevant for social surveys. Secondly, these strategies together with the imputation methods are evaluated by using a simulation study. Thirdly, data stemming from a modular questionnaire design are analysed regarding data quality and cost-saving benefits. Within this scope, imputation methods und models are evaluated and developed regarding these data.
DFG Programme
Research Grants
International Connection
Iceland, Netherlands, Switzerland, USA
