PDF pending restoration

**ERIC Number:**ED191910

**Record Type:**RIE

**Publication Date:**1980-Apr

**Pages:**251

**Abstractor:**N/A

**Reference Count:**0

**ISBN:**N/A

**ISSN:**N/A

Statistical Power with Group Mean as the Unit of Analysis.

Barcikowski, Robert S.

Focusing on the probability that differences will be found among sample treatment means when such differences do exist among the population means, this paper discusses the power of the F-test. The basic question concerns the use of analysis of variance (ANOVA) and the number of units to be randomly sampled to have a high probability of detecting meaningful differences among the treatment level means, where the units of analysis are independent group means found by averaging over related scores. The Fisher's F-test is used to establish differences among treatment level means. The unit of analysis is the group mean where there is a relationship among members of the group (e.g. in a classroom situation). Methods used to calculate power for independent entities (e.g. students) are reviewed, as well as a fabricated example problem to provide insight into the group formulas which are developed. (RL)

**Publication Type:**Reports - Research

**Education Level:**N/A

**Audience:**N/A

**Language:**English

**Sponsor:**National Inst. of Education (DHEW), Washington, DC.

**Authoring Institution:**Ohio Univ., Athens.

**Identifiers:**F Test; Generalizability Theory; Unit of Analysis Problems

**Note:**Some of the computer printouts contain light print.