ERIC Number: ED463306
Record Type: Non-Journal
Publication Date: 2001-Apr
Reference Count: N/A
Optimizing Experimental Designs Relative to Costs and Effect Sizes.
Headrick, Todd C.; Zumbo, Bruno D.
A general model is derived for the purpose of efficiently allocating integral numbers of units in multi-level designs given prespecified power levels. The derivation of the model is based on a constrained optimization problem that maximizes a general form of a ratio of expected mean squares subject to a budget constraint. This model provides more general closed form solutions than other available formulas. The proposed methodology allows for the determination of the optimal numbers of units for studies that involve more complex designs. In addition, the proposed model makes use of effect sizes and the estimates of variance components to optimize multi-level designs during the budget formulating stages. Thus, researchers are able to estimate the amount of currency needed to adjust power to a targeted level while precluding them from requesting too much or too little money in the budget planning phase of the study design. A procedure is also described for optimizing designs when estimates of variance components are unavailable. Case studies illustrate the methodology. Three appendixes contain supplemental calculations. (Contains 8 tables and 26 references.) (Author/SLD)
Publication Type: Reports - Descriptive; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A