Measuring Social Desirability Using a Novel Machine Learning Approach Based on EEG Data

Authors

Hermann Baumgartl, Aalen UniversityFollow
Philipp Roessler, Aalen UniversityFollow
Daniel Sauter, Aalen UniversityFollow
Ricardo Buettner, Aalen UniversityFollow

Abstract

While social desirability is a long-standing problem in information systems research and which is difficult to measure in questionnaires, we propose a more objective measure of social desirability based on electroencephalographic data. Using a novel machine learning approach analyzing specific fine-graded electroencephalographic sub-bands, we achieve an accuracy of over 75% on completely unseen evaluation data, which is a methodological landmark in measuring social desirability. Our results have theoretical, methodological and practical implications.

Recommended Citation

Baumgartl, Hermann; Roessler, Philipp; Sauter, Daniel; and Buettner, Ricardo, "Measuring Social Desirability Using a Novel Machine Learning Approach Based on EEG Data" (2020). PACIS 2020 Proceedings. 100.
https://aisel.aisnet.org/pacis2020/100