Talk: Internet-based research methods: Challenges and solutions (19 July 2019)
Ulf-Dietrich Reips presented at Linguistic Modeling and its Interfaces lecture series at the University of Tübingen July 19:
“Internet-based research methods: Challenges and solutions”
- Friday, 19 July 2019 10:15 AM
- University of Tübingen, Blochbau, Room 1.13 (Wilhelmstr. 19, 72074 Tübingen)
ABSTRACT: In this presentation I will provide an overview of guidelines and techniques, methods, and tools for Internet-based experimentation and big data and social media research, including solutions to many of the methodological challenges in data collection via the Internet. The challenges discussed here include issues in design, security, selection of and access to social media platforms, recruitment, multiple submissions, measurement scales (e.g. visual analogue scales), response time measurement, dropout, and data quality. Some widely used tools and practices like Amazon Mechanical Turk, the cognitive reflection test, and forced responding are critically reviewed. Among other methods that have been developed in Web methodology and Internet science, I will explain the one-item-one-screen (OIOS) design, the seriousness check, option bars, the multiple site entry technique, and our recently published guidelines to conduct Google Ngram studies (Younes & Reips, 2019, PLoS One). Picked from our iScience Server at http://iscience.eu I will demonstrate a new version of our Web experiment generator WEXTOR (https://wextor.eu) as a tool that automatically implements optimal solutions. Another example is Social Lab, our “Open Source Facebook” available at http://sociallab.es that can be played with in learning about privacy at http://en.sociallab.es. The presentation will conclude with a discussion of failed visions, problematic trends, and chances in evolving scenarios of using social media and online data in scientific work. Publications are available from http://www.researchgate.net/profile/Ulf-Dietrich_Reips and http://tinyurl.com/reipspub .
