COMPARISON OF SOME EFFECT SIZE MEASURES IN SIMPLE AND MULTIPLE LINEAR REGRESSION MODELS
[ X ]
Tarih
2021
Yazarlar
Dergi Başlığı
Dergi ISSN
Cilt Başlığı
Yayıncı
Erişim Hakkı
info:eu-repo/semantics/openAccess
Özet
It is very important that the results of the statistical analysis are understandable. Therefore, while reporting the results ofstatistical analysis, some effect size measures should be given along with the P-value. In this study, some effect size measure(R2(Eta-Squared), R?2(Epsilon-Squared) and R?2(Omega-Squared)) were compared in terms of their performance (bias) insimple and multiple linear regression models. For this purpose, Monte Carlo simulation technique was used. Results ofsimulation runs showed that very small sample sizes caused the predictions of all three effect size measures to deviatesignificantly. On the other hand, deviations of all three effect size measures gradually decreased due to increased sample sizes.However, the R?2and R?2estimates were quite unbiased when compared to R2regardless of experimental conditions. Thus, itcould be concluded that reporting R?2or R?2is more appropriate to evaluate the practical significance along with P-values insimple and multiple linear regression models.
Açıklama
Anahtar Kelimeler
Matematik, İstatistik ve Olasılık
Kaynak
Eskişehir Technical University Journal of Science and and Technology A- Applied Sciences and Engineering
WoS Q Değeri
Scopus Q Değeri
Cilt
22
Sayı
1
