TPC-Journal-V5-Issue1

The Professional Counselor /Volume 5, Issue 1 179 Data Analysis A longitudinal study design was employed for this investigation. After completion of the data collection process, participants’ responses were analyzed using descriptive data analysis, one-way analysis of variance (ANOVA), repeated measures ANOVA, paired-samples t-test and mixed between/within-subjects ANOVA. Prior to analysis, the data were screened for outliers using the outlier labeling method (Hoaglin & Iglewicz, 1987; Hoaglin, Iglewicz, & Tukey, 1986), which resulted in identifying 11 cases with outliers. Therefore, Windsorized means were calculated based on adjacent data points to replace the outliers (Barnett & Lewis, 1994; Osborne & Overbay, 2004). The resulting data were checked for statistical assumptions and no violations were found. A sample size of 179 graduate counseling students was deemed appropriate for identifying a medium effect size (power = .80) at the .01 level for the employed data analysis procedures (Cohen, 1992). Results Counseling Trainees’ Self-Efficacy Several one-way between-groups ANOVAs were conducted to examine the impact of each trainee’s age, gender, ethnicity and program track (i.e., SC, MHC or MCFC) on his or her level of self-efficacy at each of the three data collection points. There was no statistically significant relationship between self-efficacy and trainees’ age at the NSO data collection point ( F [3, 178] = 1.35, p = .26), at the CPO data collection point ( F [3, 178] = .39, p = .76) or at the FIGSM data collection point ( F [3, 178] = .71, p = .55). Similarly, there was no statistically significant relationship between self-efficacy and trainees’ gender at the NSO data collection point ( F [1, 178] = .48, p = .49), at the CPO data collection point ( F [1, 178] = .02, p = .88) or at the FIGSM data collection point ( F [1, 178] = .001, p = .97). There was no statistically significant relationship between self-efficacy and trainees’ ethnicity at the NSO data collection point ( F [4, 178] = 1.03, p = .39), at the CPO data collection point ( F [4, 178] = .82, p = .51) or at the FIGSM data collection point ( F [4, 178] = .03, p = .97). Finally, there was no statistically significant relationship between self-efficacy and trainees’ program track at the NSO data collection point ( F [2, 178] = .03, p = .97), at the CPO data collection point ( F [2, 178] = .40, p = .67) or at the FIGSM data collection point ( F [2, 178] = .04, p = .96). Counseling Trainees’ Self-Efficacy Over the Course of the Program A one-way within-subjects repeated measures ANOVA was conducted to examine participants’ ( N = 179) CSES scores at the three data points (i.e., NSO, CPO, FIGSM). Table 1 presents the descriptive statistics. Mauchley’s Test indicated that the assumption of sphericity was violated, χ 2 (2) = .53, p < .001; therefore, the within-subjects effects were analyzed using the Greenhouse-Geisser correction (Greenhouse & Geisser, 1959). There was a statistically significant effect of time, F (1.3, 242.79) = 404.52, p < .001, Partial η 2 = .69 on participants’ CSES scores. Sixty-nine percent of the variance in CSES scores can be accounted for by the time participants spent in the program (large effect size; Sink & Stroh, 2006; Streiner, 2003). Therefore, trainees scored higher on the CSES at each interval during their counselor preparation program. Table 1 Descriptive Statistics for Self-Efficacy Across Data Collection Points Data Collection Point M SD Mdn Mode Range New student orientation 57.09 14.42 59 58 23–84 (61) Clinical practicum orientation 77.43 8.53 78 79 53–99 (46) Final internship group supervision meeting 83.04 6.80 84 76 66–95 (33) Note. N = 179.