ERIC Number: ED233067
Record Type: RIE
Publication Date: 1983-Apr-14
Reference Count: 0
Analysis of Balanced 2 [to the kth power] Mirror-Image Designs.
Schafer, William D.; Dayton, C. Mitchell
A 2-to-the-kth-power mirror-image design is defined as one in which repeated observations of subjects occurs among the levels of a usual 2-to-the-kth-power design, but there is the restriction that no subject may receive a given level of any factor more than once. Such a restriction might arise, for example, if a subject's response is expected to be affected artifically if any given level of a factor is presented two or more times. This paper describes an analysis for data which arise from such designs. Only the balanced case is treated. Subjects in a 2-to-the-kth-power mirror-image design may be divided into 2-to-the-k-minus-1 groups, each of which receives a specific, unique pair of treatments. This grouping arises from the restriction in the repetition of levels admissible in the design. Each subject thus yields two scores under two combinations of distinct levels of the factors. It is assumed that the populations the groups represent are homogeneous bivariate normal. Significance tests are developed in the paper for the usual main and interaction effects associated with a 2-to-the-kth-power design. A fully worked example is included. (Author)
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Authoring Institution: N/A
Identifiers: Mirror Image Designs; Repeated Measures Design
Note: Paper presented at the Annual Meeting of the American Educational Research Association (67th, Montreal, Quebec, April 11-15, 1983).