Effect of Training in Verbal Self-Guidance on Performance Effectiveness in a MBA Program, The

Canadian Journal of Behavioural Science, Jan 2006 by Brown, Travor C, Latham, Gary P

Performance. The behavioural measure of an individual's interpersonal skills was anonymously assessed by peers. Specifically, peers rated the frequency with which they observed a team member performing each behaviour using a 5-point scale (0 = almost never and 4 = almost always). These peers worked with the person throughout the entire semester. Thus, peers were arguably the most important source of information regarding a person's behaviour. The outcome measure of a person's performance (i.e., GPA) was measured by faculty who were neither aware of the purpose of the study, nor the experimental condition to which a person had been randomly assigned. The composite criterion, performance effectiveness, was created by summing each person's z-score for GPA and BOS.

Results

VSG Manipulation Checks

Responses indicated that the VSG participants were satisfied with the training (M = 39.46; SD = 5.02; maximum score = 50, a = .79). Moreover, they reported greater use of VSG skills (M = 19.28; SD = 3.51; maximum score = 25; α = .91) than those in the control (M = 15.43; SD = 4.91) condition, F(1, 108) = 22.00, p

Goal-Setting Manipulation Checks

Goal-setting manipulation checks revealed that the mean goal set by goal-setting participants was 47.07 (SD = 4.08; maximum score = 56), that these participants were highly committed to their goals (M = 21.12; SD = 2.94; α = .71; maximum score = 25), and that they perceived their goals to be both specific (M = 11.42; SD = 2.30; α = .66; maximum score = 15) and difficult (M = 7.98; SD = 1.44; a = .65; maximum score = 10). ANOVA confirmed that there were no significant differences in these four measures between the goal-setting participants in the VSG condition and those in the control condition, F(1, 58) = 1.50, p > .05; F(1, 57) = .15, p >. 05; F(1, 48) = .03, p > .05; F(1, 48) = .01, p > .05, respectively.

Performance

GPA. GPA was available for 121 participants; 5 had left the MBA program. GPA ranged from 2.68 to 3.89 (M = 3.44; SD = .21; maximum score = 4). Consistent with the first hypothesis, a 2 × 2 ANOVA revealed a main effect for VSG on GPA, F(1, 117) = 6.60, p .05. The means and standard deviations are shown in Table 2. The interaction effect is presented in Figure 1.

Bobko (1986) argued that a spurious main effect can result in a 2 × 2 ANOVA when an ordinal interaction is present. Given our hypotheses that the participants in the VSG / goal-setting condition would have the highest performance level, we tested the interaction using Bobko's ordinal interaction technique. This technique involves two planned contrasts: 1) testing the equality of the means of the three noninteraction conditions (i.e., VSG/DYB; Control/ Goal Setting; Control/DYB) using a one-way analysis; and 2) comparing the means of the interaction (i.e., VSG/goal setting) versus the average of the remaining three conditions using a planned t-test. Contrast one was not significant, F(2, 90) = 2.01, p > .05); thus, there was no significant difference among the means of the three conditions. Contrast two was significant, t(119) = 2.51, p

Interpersonal skills. For the behavioural measure of performance, BOS scores were obtained for 117 participants (response rate = 92.86%). The level of agreement among raters was calculated using the average interrater agreement statistic, rwg, (James, Demaree, & Wolf, 1993). The rwg, for the 11 behavioural items ranged from .71 to .84; the average level of agreement across the 11 items was .77. The median number of peer ratings per person was four. Cronbach's alpha coefficient for the 11-item BOS was .90. The grand mean of the 11-item BOS was 36.26 (SD = 4.17; maximum score = 44).

The correlation between GPA and interpersonal skills was significant (r = .43, p .05; F(1, 113) = .22, p > .05, respectively. The means and standard deviations of this ANOVA are shown in Table 2; the interaction effect is presented in Figure 2. Again, Bobko's (1986) ordinal interaction analysis was conducted. For contrast one, ANOVA revealed no significant differences among the three conditions of VSG only, goal setting only, and Control/DYB, F(2, 88) = .99, p > .05. In addition, there was no significant difference between the means of the VSG/goal setting and the mean of the other three conditions, t(115) = 1.41, p = .16.

Composite criterion. A 2 × 2 ANOVA of the composite criterion of performance effectiveness revealed a main effect for VSG, F(1, 111) = 4.11, p .05. The means and standard deviations are shown in Table 2. The interaction effect is presented in Figure 3.

Again, Bobko's (1986) ordinal interaction analysis was conducted to see if the main effect for VSG was spurious. For contrast one, ANOVA revealed no significant differences among the three conditions of VSG only, goal setting only, and Control /DYB, F(2, 87) = .88, p > .05. For contrast two, there was a significant difference between the mean of the VSG/goal-setting condition versus the mean of the other three conditions, t(113) = 2.66, p

Discussion

The theoretical and practical significance of this experiment is three-fold. First, of significance to both goal-setting theory and training in VSG is the interaction effect found for VSG and goal setting on performance. The lack of a main effect for goal setting alone is consistent with previous findings that when people are in a learning mode, setting a specific, difficult performance goal does not improve performance (Kanfer & Ackerman, 1989; Seijts & Latham, 2001). In the present study, the task of mastering the MBA curriculum and the requisite interpersonal skills, required more than motivation in terms of choice, effort, and persistence; it required learning interpersonal skills. This finding supports Locke's (2000) conclusion that the highest level of performance occurs when participants have high task knowledge and high motivation. In the present study, performance was highest in the condition where participants were trained to use VSG and to set a high performance goal that motivated them to apply their newly acquired knowledge and skill.