| Power Analysis in Experimental Designs with t test Analysis |
|
Kang, Jeong-Hee;Bang, Kyung-Sook;Ko, Sung-Hee; |
1College of Nursing, Chonbuk National University
2College of Nursing, Ajou University
3College of Nursing, Chonbuk National University |
| t 검정 실험 설계 시 표본 크기 결정에 대한 논의 |
|
강정희, 방경숙, 고성희 |
1전북대학교 간호대학
2아주대학교 간호대학
3전북대학교 간호대학 |
|
|
| Abstract |
Purpose
In the literature power analysis in experimental studies is often executed and reported falsely. This descriptive study was done to promote the correct application of the Cohen(1988)'s power analysis method. Method: Articles of experimental studies from a nursing journal were selected and reviewed to examine the uses of and the reports on the power analysis process. Also, the appropriate power analysis process was discussed with an example of the most common experimental design, an independent two-group design with t test analysis plan. Result: Around half the experimental studies examined reported that they carried out power analysis. Cohen's method was the most frequently utilized but with accuracy in question. Conclusion: Power analysis to estimate sample size is the interplay between alpha, power, and effect size, and other factors in the case of t test analysis. Researchers should have a clear understanding of how to apply the Cohen's power analysis method so they do not produce poorly estimated sample sizes. |
| Key Words:
Experimental design, t test, power analysis, sample size |
| |
|
PDF Links
Full text via DOI
Download Citation
