Applying modern psychometric techniques to melodic discrimination testing: Item response theory, computerised adaptive testing, and automatic item generation

被引:0
|
作者
Peter M. C. Harrison
Tom Collins
Daniel Müllensiefen
机构
[1] Queen Mary University of London,
[2] School of Electronic Engineering and Computer Science,undefined
[3] Goldsmiths,undefined
[4] University of London,undefined
[5] Department of Psychology,undefined
[6] Lehigh University,undefined
[7] Department of Psychology,undefined
[8] Music Artificial Intelligence Algorithms,undefined
[9] Inc.,undefined
来源
关键词
D O I
暂无
中图分类号
学科分类号
摘要
Modern psychometric theory provides many useful tools for ability testing, such as item response theory, computerised adaptive testing, and automatic item generation. However, these techniques have yet to be integrated into mainstream psychological practice. This is unfortunate, because modern psychometric techniques can bring many benefits, including sophisticated reliability measures, improved construct validity, avoidance of exposure effects, and improved efficiency. In the present research we therefore use these techniques to develop a new test of a well-studied psychological capacity: melodic discrimination, the ability to detect differences between melodies. We calibrate and validate this test in a series of studies. Studies 1 and 2 respectively calibrate and validate an initial test version, while Studies 3 and 4 calibrate and validate an updated test version incorporating additional easy items. The results support the new test’s viability, with evidence for strong reliability and construct validity. We discuss how these modern psychometric techniques may also be profitably applied to other areas of music psychology and psychological science in general.
引用
收藏
相关论文
共 50 条