TPC Journal-Vol 9 Issue 3-FULL

The Professional Counselor | Volume 9, Issue 3 233 consultation training experiences reported significantly higher consultation practice proficiency perceptions ( M = 38.39, SD = 6.40) when compared to participants without consultation training ( M = 26.47, SD = 8.96) or with one to two consultation training experiences ( M = 32.62, SD = 9.67). Finally, showing incremental validity, the CSS-S also explained an additional 7% of the variance in participants’ MIIC scores ( R 2 = .132, p = .000), above and beyond the independent variables in the first (i.e., age, gender, and years of experience; R 2 = .004, p > .05) and second (i.e., SWAI; R 2 = .058, p = .000) blocks. Reliability Analyses The reliability analyses results showed satisfactory support for the CSS-S. For the present study, Cronbach’s alpha coefficient for the CSS was .96; no items appeared to reduce the reliability coefficient of the scale. We also examined the Spearman-Brown coefficient for the split-half reliability and obtained .96. Discussion In this study, we obtained strong results for the single-factor structure as well as validity and reliability properties for a shorter version of the CSS-S in a sample of counselors. Our results also revealed further validation for consultation as a distinct area of practice. Psychometric Qualities of the CSS-S CFA results revealed that the CSS-S is a psychometrically sound unidimensional instrument, measuring counselors’ consultation skills as a distinct modality relative to other professional counseling practice domains. In the current sample, upon poor fit of the initial single-factor solution, we further utilized modification indices and eliminated items. As a result, different than the original 17-item instrument (Moe et al., 2018), we obtained a shorter version of the CSS with eight items indicating a good CFA solution fit. We further obtained significant results for the CSS-S via convergent, divergent, concurrent, and incremental validity procedures. For convergent validity, we found that counselors’ CSS-S scores were moderately related to the subscales of the SWAI (i.e., Client Focus, Rapport, and Identification). These relationships revealed that the CSS-S measured a similar but different competency area compared to the area of supervision. We also found that counselors’ CSS-S scores were unrelated to gender or ethno-cultural identification and were weakly related to the MIIC, establishing divergent validity. In other words, counselors’ consultation skills were distinctly different than their gender or ethno-cultural identification and separate from their interdisciplinary collaboration ability. Supporting concurrent validity, counselors’ CSS-S scores got higher as they had more consultation training (i.e., three or more consultation training experiences vs. one to two or no consultation training experiences). Beyond other variables (i.e., years of experience and supervisory working alliance), counselors’ consultation skills significantly contributed to their interdisciplinary collaboration ability, indicating incremental validity of the CSS-S. Finally, we examined the reliability of the CSS-S by observing the internal consistency across the items. Both Cronbach’s alpha and split-half reliability results were strong, demonstrating satisfactory results for the CSS-S. The CSS-S appears to possess useful validity and reliability characteristics for assessing counselors’ perceptions of their own abilities to practice consultation and may help scholars develop more empirically grounded scholarship on consultation as a distinct mode of practice.