Minimum Sample Size Requirements for Nonparametric Item Response Theory With Continuous Items

• In Event: 52.034 - Single-Case Designs and Small Sample Issues

Abstract

For a statistical method, it is always critical to find out the minimally sufficient sample size required for practical applications. Straat, van der Ark, and Sijtsma (2014) had conducted a numerical study to provide guidelines for test practitioners in determining sample size while analyzing tests with nonparametric item response theory (NIRT). The study further extends their study to NIRTC, the NIRT for continuous items proposed recently (Chen, 2015). A simulation study is conducted with several critical independent variables (e.g., test length and latent variables’ correlation) to represent common test situations. According to the simulation outcomes, guidelines of minimum sample size requirements for NIRTC applications were drawn, and issues about further establishing the methodology for NIRTC were also illustrated.

Authors

• Jyun-Hong Chen, National Sun Yat-sen University

• Ching-Lin Shih, National Sun Yat-sen Universit

• Shu-Ying Chen, National Chung Cheng University