The Professional Counselor | Volume 14, Issue 1 107 The assessment developers have reported internal reliability for the CSAES as a calculation for each of the four subscales and the second-order factor of Suicide Assessment using Cronbach’s α: General Suicide Assessment α = .882, Assessment of Personal Characteristics α = .88, Assessment of Suicide History α = .81, Suicide Intervention α = .83, and Suicide Assessment α = .93 (Douglas & Wachter Morris, 2015). The CSAES was noted for showing “structural aspects of validity and sensitivity to detect differing levels of self-efficacy” (Douglas & Wachter Morris, 2015) based on the utilization of a fourfactor model that was cross-validated. The scale was validated with a total of 324 participants. Of the participants, 258 (79.63%) were female. Unfortunately, a limitation of the instrument is that the diversity of participants was unknown due to ethnicity inadvertently being left off the demographic questionnaire while it was being developed (Douglas & Wachter Morris, 2015). The current study assessed the internal reliability of the CSAES using Cronbach’s α and omega (ω) using test-retest reliability because the measure of consistency was between two measurements of the same construct to the same group at two different times. The overall CSAES has strong reliability (α = .978; ω = .978), and each subscale had the following reliability scores: General Suicide Assessment (α = .943; ω = .946); Assessment of Personal Characteristics (α = .947; ω = .947); Assessment of Suicide History (α = .896; ω = .890); and Suicide Intervention (α = .920; ω = .915). All subscale reliability coefficients were high when looking at both α and ω, meaning that good internal consistency was found among items of the scale (Green & Salkind, 2014). Data Analysis Before running any analyses, we screened the data using SPSS version 26.0.0.1 software to check for (a) missing data, (b) average expected scores per the outcome variables, (c) standard deviations within range, and (d) normality of data (Green & Salkind, 2014). The initial sample included 108 surveys; however, only 60 of these could be paired with posttest data, using the username data point. Therefore, individuals with their missing pair were deleted along with any others that included missing data (i.e., listwise deletion). The final sample included 60 respondents. A power analysis using a statistical power analysis program (G*Power 3.1) for paired t-test showed an N of 54 was needed for 95% power with an alpha level of .05 and a moderate effect size of .5. Next, scales were computed, and univariate testing occurred. Data met all assumptions for conducting a paired t-test (i.e., dependent variable was continuous, normally distributed, and without outliers and the observations were independent of one another). The paired t-test was used to test the effectiveness of training for suicide assessment and intervention in a basic skills class using a single-group (pretest/ posttest) design, including demographics and the CSAES. To ensure that all students received the same teaching activities and assignments, we decided not to utilize a control group. We wanted to ensure that all students were equally prepared and trained to address the topic of suicide upon entering their clinical courses and upon graduation. Results The purpose of the study was to compare counseling students’ pretest and posttest self-efficacy assessments when taking a basic skills course that highly emphasized skills related to suicide training and assessment. The results of a paired-samples t-test on the General Suicide Assessment subscale indicate that on average, students scored significantly higher (Mpost = 28.57, SD = 4.80) after taking the basic skills class (Mpre = 21.20, SD = 7.80), t(59) = −9.15, p < .001. A large effect was found (d = 1.182, 95% CI [−1.51, −.85]). The results of a paired-samples t-test on the Assessment of Personal Characteristics subscale indicate that on average, students scored significantly higher (Mpost = 41.17, SD = 6.59) after taking
RkJQdWJsaXNoZXIy NDU5MTM1