Training program evaluation. A field experiment was conducted at a not-for-profit research and development organization to examine the expectations, attitudes, and decisions of employees with regard to training programs (Academy of Management Journal, September 1987). In particular, the study was aimed at determining how managers' evaluations of a training program were affected by the prior information they received and by the degree of choice they had in entering the program. These two factors, prior information and degree of choice, were each varied at two levels. The prior information managers received about the training program was either a realistic preview of the program and its benefits or a traditional announcement that tended to exaggerate the workshop's benefits. Degree of choice was either low (mandatory attendance) or high (little pressure from supervisors to attend). Twentyone managers were randomly assigned to each

of the 2 × 2 = 4 experimental conditions; thus, a 2 × 2 factorial design was employed. At the end of the training program, each manager was asked to rate his or her satisfaction with the workshop on a 7-point scale (1 = no satisfaction, 7 = extremely satisfied). The ratings were subjected to an analysis of variance, with the results shown in the partial ANOVA summary table.

(a) Complete the ANOVA summary table.

(b) Conduct the appropriate ANOVA F-tests (use α = .05). Interpret the results.

(c) The sample mean satisfaction ratings of managers for the four combinations of prior information and degree of choice are shown in the table below. Use Tukey's method to rank the four means. Use α = .06.

(d) Use the Scheffe method to perform all ´ pairwise comparisons of the four treatment means. Use α = .05.

(e) Use the Bonferroni approach to perform all pairwise comparisons of the four treatment means. Use α = .05.

(f) Compare the results, parts c-e.