The proposed theoretical and methodological innovations and the data cleaning tools we will create, can be applied widely in the social sciences and humanities.
This project will:
• contribute to the community of web survey practitioners, providing evidence and knowledge on maximized response rate and response quality by investigating best practices to maximize data quality via different web experiments with respect to splitting a long questionnaire into shorter parts, exploring the optimal survey length and the optimal duration of the breaks between the sub-questionnaires, and applying data cleaning methods relying on response quality indicators.
• demonstrate that social media data can either supplement or even substitute some web survey data by cross-validating social media data (Facebook and Twitter) with web-survey data (Hellenic Candidate Study, Hellenic Voter Study, Expert Survey). This innovative supersession of more “traditional” data sources (web surveys) with big data (produced by social media posts) could help to reduce the length of web survey questionnaires: since we will be able to collect some of the information from their social media posts, we will not have to ask them to provide the same information on the questionnaire.
• promote the combined use of social media data, online questionnaire data and printed data to build knowledge about the attitudes and opinions of social and political actors on major issues like populism and euroscepticism.
• The methods we propose to exploit social media data could be utilized by many data producers (including private firms).
• Our methods for data cleaning and the improvement of the quality of web survey data, will lead to cost savings for many data producers who can have a dataset of the same or even better quality for a small fraction of this cost.
• Combining of two different methods of coding: human and machine coding we plan to demonstrate that machine coding of Greek language texts can be an efficient method for coding text for specific purposes. The combination of the two methods will result to a significant cost effect as researchers and companies will be able to save time and money.
• High quality data and big data permits access to a vast amount of information which allows the comprehensive mapping of the particular discursive forms populism and euroscepticism take, inclusive of operational typologies of populist and euroscepticist discourses. Organized in an open access digital handbook, these will potentially refashion future debates on sovereignty and democracy and furnish stakeholders with a more accurate view of the issues at hand to benefit mutual trust.
• PopEUCompass, the Compass for Populism and Euroscepticism will be available to the voters for free, helping them find out where they stand when it comes to populism and to euroscepticism.