Dataset for "Development and Validation of a Multilevel Deep-Learning Framework for Individualized Prediction in Clustered Data"

DOI

We present the dataset accompanying the article "Development and Validation of a Multilevel Deep-Learning Framework for Individualized Prediction in Clustered Data". The data were collected as part of the Smartphone Sensing Panel Study and include 508 participants, 6,771 participant-days, 37 person-level features, and 560 day-level features. The dataset is intended for developing and evaluating machine learning methods for individualized prediction in clustered data. To protect participant privacy, demographic and technical variables that were not used in the analyses have been removed. These variables can be made available upon reasonable request. The dataset is documented in an accompanying codebook. Additional materials, including the analysis code, are available at https://osf.io/pxdwj.

Identifier
DOI https://doi.org/10.23668/psycharchives.22353
Metadata Access https://api.datacite.org/dois/10.23668/psycharchives.22353
Provenance
Creator Zhu, Ningzhe; Schoedel, Ramona; Sust, Larissa; Bühner, Markus; Terhorst, Yannik
Publisher PsychArchives
Contributor Leibniz Institut für Psychologie (ZPID)
Publication Year 2026
OpenAccess true
Representation
Language English
Resource Type Dataset; researchData
Discipline Social Sciences