Premium
Optimal Bandwidth Selection in Observed‐Score Kernel Equating
Author(s) -
Häggström Jenny,
Wiberg Marie
Publication year - 2014
Publication title -
journal of educational measurement
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.917
H-Index - 47
eISSN - 1745-3984
pISSN - 0022-0655
DOI - 10.1111/jedm.12042
Subject(s) - equating , smoothing , bandwidth (computing) , kernel (algebra) , selection (genetic algorithm) , statistics , kernel smoother , computer science , mathematics , econometrics , kernel method , machine learning , telecommunications , radial basis function kernel , support vector machine , combinatorics , rasch model
The selection of bandwidth in kernel equating is important because it has a direct impact on the equated test scores. The aim of this article is to examine the use of double smoothing when selecting bandwidths in kernel equating and to compare double smoothing with the commonly used penalty method. This comparison was made using both an equivalent groups design and a nonequivalent group with anchor test design. The performance of the methods was evaluated through simulation studies using both symmetric and skewed score distributions. In addition, the bandwidth selection methods were applied to real data from a college admissions test. The results show that the traditional penalty method works well although double smoothing is a viable alternative because it performs reasonably well compared to the traditional method.