Replication in music psychology. (September 2013)

• Record Type: Journal Article
• Authors: Frieler, Klaus; Müllensiefen, Daniel; Fischinger, Timo; Schlemmer, Kathrin; Jakubowski, Kelly; Lothwesen, Kai
• Other Names: Fischinger Timo guest-editor.
• Is Part Of: Musicae scientiae. Volume 17:Number 3(2013)
• Page Start: 265
• Page End: 276
• Publication Date: 2013-09
• Subjects: belief system -- decline effect -- music psychology -- publication bias -- reliability -- replication -- quality management; Music -- Psychological aspects -- Periodicals; 780
• DOI: 10.1177/1029864913495404
• Languages: English
• ISSNs: 1029-8649

Abstract

In this article, we address the current state and general role of replication in empirical sciences in general and music psychology in particular. We argue that replication should be an integral part of the quality management of science because it helps to improve and maintain the general benefit of empirical sciences by enhancing the confidence in scientific phenomena and theories. Replicating empirical experiments has two major benefits: (1) It increases the sheer number of observations and (2) it provides independent evidence which works as a safety net against methodological fallacies, causally influential but unknown (i.e., random) factors, researcher degrees of freedom, and outright fraud. Furthermore, we argue that for low-gain/low-cost sciences such as music psychology, measures to ensure quality standards, in particular the amount of replication experiments conducted, can be expected to be lower than in high gain/high cost sciences. These lower expectations stem from the general acknowledgments that in low-gain/low-cost sciences (1) research resources are normally scarce and (2) the consequences of inadequate theories are relatively harmless. We argue that the view of music psychology as a low-cost/low-gain science can explain the striking lack of replication studies and meta-analyses. We also discuss possible counter-measures to enhance the reliability of music-psychological knowledge.