The Implications of a push to web experiment in the context of a longitudinal study
Jul 26, 13:45
A fieldwork experiment is currently being implemented with the aim of examining the findings of a push-to-web (P2W) within a new round of the Generations and Gender Survey (GGS) data collection. The objective of the paper is to present an overview of the challenges and opportunities that a P2W approach can have in terms of response rates and data quality. The GGS is a cross-national longitudinal survey that provides open access data for researchers on a broad array of topics including partnerships, fertility, work-life balance, gender relations, transition to adulthood and intergenerational exchanges. In this paper, we describe the experiment being conducted in three countries: Croatia, Germany and Portugal (in each country a region was selected for conducting the experiment). The region includes urban as well as rural areas and it offers suboptimal conditions for survey research, such as low provision of high speed internet. The study compares respondents who are approached using a P2W approach to those who are interviewed using the traditional face-to-face (F2F) methodology that was used in the first round of the GGS. In every country, respondents are randomly assigned to the F2F-only condition, or a P2W condition. In addition, there are country-specific methodological experiments that are useful to further investigate how a P2W design can be implemented in future GGS rounds: in Germany, different strategies for providing incentives are tested (five treatment groups were formed with various combinations of incentives - unconditional / conditional); in Croatia, there are different strategies for sending out reminders (one group receives two reminders and no phone contact is allowed, while the other group receives three reminders and telephone contacts are also used); in Portugal, strategies for selecting a respondent from the people living at the same address (random route sampling) are tested. After the fieldwork is completed, respondents are invited to respond online to a follow-up module. This stage is important to gain a clearer idea of what response rates can be achieved in these circumstances. Moreover, the paradata collected provides key information on user experience and context. The results of the experiment are being used to inform the model and guidelines of future GGS data collection.