Pop Groove Ground-Truth Dataset (PGGT)

This dataset is the outcome of Luzie Ahrens' (2024) master thesis "Entwicklung einer Ground-Truth Datenbasis für die automatische Groove-Klassifikation von Popmusiktiteln" in the Audio Communication & Technology M.Sc. program at Technische Universität Berlin, supervised by Dr. Steffen Lepa and Prof. Reinhard Kopiez. Data is available in RDS format for R as well as in SAV format for SPSS and contains n = 1775 valid observations, from m = 178 respondents with sociodemographic and music-related information (genre liking and musical expertise in terms of GOLD-MSI, Schaal et al. 20214). Variable and category labels are provided in the SAV version. All data stems from an online survey that was performed in spring 2024, participant recruitment was conducted via social media. The respondents evaluated k = 448 pop song excerpts, 68% of which were familiar to them, on the German-language EGQ (Experience of Groove Questionnaire, Düvel et al. 2024), as well as on the German DGA (Dimensions of Groove Affordances, Lepa et al. 2025) survey instrument. Broken down by genre, 215 assessments are available for country, 275 for electronic, 73 for R'n'B, 721 for hip hop, 21 for Latin, 376 for pop and 94 for rock. The uneven distribution results, on the one hand, from the number of cases of the genres in the stimulus pool and, on the other hand, from the random selection of tracks from the pool during the online survey.