Digital footprints of sensation seeking
Journal
Big Data in Psychology
ISSN
2190-8370
ISSN-Digital
2151-2604
Type
journal article
Date Issued
2018
Author(s)
Schoedel, R.
;
Au, Q.
;
Völkel, S. T.
;
Lehmann, F.
;
Becker, D.
;
Bühner, M.
;
Bischl, B.
;
Hussmann, H.
;
Abstract
The increasing usage of new technologies implies changes for personality research. First, human behavior becomes measurable by digital data, and second, digital manifestations to some extent replace conventional behavior in the analog world. This offers the opportunity to investigate personality traits by means of digital footprints. In this context, the investigation of the personality trait sensation seeking attracted our attention as objective behavioral correlates have been missing so far. By collecting behavioral markers (e.g., communication or app usage) via Android smartphones, we examined whether self-reported sensation seeking scores can be reliably predicted. Overall, 260 subjects participated in our 30-day real-life data logging study. Using a machine learning approach, we evaluated cross-validated model fit based on how accurate sensation seeking scores can be predicted in unseen samples. Our findings highlight the potential of mobile sensing techniques in personality research and show exemplarily how prediction approaches can help to foster an increased understanding of human behavior.
Language
English
Keywords
sensation seeking
machine learning
big data
behaviour
smartphone sensing
Refereed
Yes
Publisher
Hogrefe
Volume
226
Number
4
Start page
232
End page
245
Official URL
Subject(s)
Division(s)
Eprints ID
264550
File(s)![Thumbnail Image]()
Name
DigitalFootprints_Schoedel.pdf
Size
373.91 KB
Format
Adobe PDF
Checksum (MD5)
646fdc42883152d5efd0348c4f95fb1c