Type I error rates of t-test and T1 statistic under balanced and unbalanced designs
The t-test used prominently for comparing means between two groups is usually restricted with the assumptions of normality and variance homogeneity. However, the violation of the assumptions occurs in many real world data. In this study, two methods of comparing the means of two samples were conduct...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Conference or Workshop Item |
| Language: | English |
| Published: |
2010
|
| Subjects: | |
| Online Access: | http://repo.uum.edu.my/2216/1/Zahayu_Md_Yusuf%2C_Suhaida_Abdullah%2C_Sharipah_Soaad.pdf http://repo.uum.edu.my/2216/ http://skskm.spus.biz/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | The t-test used prominently for comparing means between two groups is usually restricted with the assumptions of normality and variance homogeneity. However, the violation of the assumptions occurs in many real world data. In this study, two methods of comparing the means of two samples were conducted. The first method used the traditional t-test while the second method used the TI statistic for
comparing means. The performances of the t-test and the TI statistic were evaluated under different conditions. They were sample sizes, type of distributions (normal or non-normal), and unequal group variances. The Type I error rates of the two test statistics were obtained and compared. Based on the investigations, the TI statistic was able to produce good Type I error rates with values near to the nominal level, a = 0.05. |
|---|