A Mixed Mode Approach on Polling’s Sampling Composition in Greek Context

The selection of quality samples and their subsequently impact on polling results constitutes an under-researched topic in Greek Politics. In essence, recent public debates regarding polling accuracy and research quality undermine the methodological discussion and the need for adopting new strategies and novel data-driven approaches. In this study, we explore how different modes of data collection affect data quality. In particular, by comparing sampling designs based on social media recruitment, traditional CATI approaches and combined methods, we shed light on the data quality and contribute to the ongoing debate. The primary scope of this proposal is to identify the core questions that should accompany the debate on data quality and provide new insights on the under-researched area within the Greek context. Our findings derived from the comparison of various modes of data collection offer an original contribution to the field. Methodologically, we will present data collected by the Public Opinion Research Unit of the University of Macedonia Research Institute (www.poru.eu) . These datasets utilize the combination of the following modes of data collection: social media recruitment, mobile only recruitment, mixed mode recruitment.