Publication Date
In 2025 | 0 |
Since 2024 | 0 |
Since 2021 (last 5 years) | 0 |
Since 2016 (last 10 years) | 0 |
Since 2006 (last 20 years) | 1 |
Descriptor
Source
Journal of Experimental… | 4 |
Author
Guo, Jiin-Huarng | 2 |
Luh, Wei-Ming | 2 |
Bartram, Dave | 1 |
Fidalgo, Angel M. | 1 |
Hashimoto, Kanako | 1 |
Kennedy, Eugene | 1 |
Muniz, Jose | 1 |
Publication Type
Journal Articles | 4 |
Reports - Evaluative | 2 |
Reports - Research | 2 |
Education Level
Audience
Location
Laws, Policies, & Programs
Assessments and Surveys
What Works Clearinghouse Rating
Fidalgo, Angel M.; Hashimoto, Kanako; Bartram, Dave; Muniz, Jose – Journal of Experimental Education, 2007
In this study, the authors assess several strategies created on the basis of the Mantel-Haenszel (MH) procedure for conducting differential item functioning (DIF) analysis with small samples. One of the analytical strategies is a loss function (LF) that uses empirical Bayes Mantel-Haenszel estimators, whereas the other strategies use the classical…
Descriptors: Bayesian Statistics, Test Bias, Statistical Analysis, Sample Size

Luh, Wei-Ming; Guo, Jiin-Huarng – Journal of Experimental Education, 2002
Used Johnson's transformation (N. Johnson, 1978) with approximate test statistics to test the homogeneity of simple linear regression slopes in the presence of nonnormality and Type I, Type II or complete heteroscedasticity. Computer simulations show that the proposed techniques can control Type I error under various circumstances. (SLD)
Descriptors: Computer Simulation, Error of Measurement, Regression (Statistics)

Luh, Wei-Ming; Guo, Jiin-Huarng – Journal of Experimental Education, 2005
To deal with nonnormal and heterogeneous data for the one-way fixed effect analysis of variance model, the authors adopted a trimmed means method in conjunction with Hall's invertible transformation into a heteroscedastic test statistic (Alexander-Govern test or Welch test). The results of simulation experiments showed that the proposed technique…
Descriptors: Robustness (Statistics), Computer Simulation, Educational Research, Error Patterns

Kennedy, Eugene – Journal of Experimental Education, 1988
Ridge estimates (REs) of population beta weights were compared to ordinary least squares (OLS) estimates through computer simulation to evaluate the use of REs in explanatory research. With fixed predictors, there was some question of the consistency of ridge regression, but with random predictors, REs were superior to OLS. (SLD)
Descriptors: Computer Simulation, Error of Measurement, Estimation (Mathematics), Least Squares Statistics