Error Rates of Multiple F Tests in Factorial ANOVA Designs.

MF-$0.75 HC-$1.50 PLUS POSTAGE *Analysis of Variance; Computer Programs; Error Patterns; *Hypothesis Testing; Sampling; *Tests of Significance The primary purpose of the present study was to investigate empirically the effect of multiple hypothesis testing on error rates in factorial ANOVA designs under a variety of controlled conditions. The per comparison, per experiment, and experimenthise error rates were investigated for three hypothesis testing procedures. The specific conditions manipulated included: (1) the number of factors in the design, (2) the number of levels of each factor, (3) the number of observations per cell, and (4) the population values of each null hypothesis (magnitude of the effect). A Monte Carlo computer simulation procedure was used for generation of the data. Type I and Type II errors were tabulated where appropriate fof the three error rates. (Author) U S (DEPARTMENT OF HEALTH EDUCATION 8. ELFARE NA TONAL INSTITUTE OF EDUCATION ERROR RATES OF MULTIPLE F TESTS IN FACTORIAL ANOVA DESIGNS Judith S. Halderson National Assessment of Educational Progress