TPC Journal V8, Issue 2 - FULL ISSUE

The Professional Counselor | Volume 8, Issue 2 183 with faculty respondents. Internal consistency indices for the overall measure (α = .78) as well as for the three scales (engagement, α = .82; knowledge, α = .75; fear scale, α = .74) were adequate for an attitudinal questionnaire. Figure 1. Confirmatory Factor Analysis Path Model for Undergraduate Student Sample ( N = 312) Phase 3: Hierarchical Logistic Regression Analyses The guidelines for HLR assumption checking were followed (Field, 2018). Items were winsorized to remove extreme outliers. Skewness and kurtosis values (see Table 1) were largely within the acceptable range (± 1.00) for both samples. Pearson product correlations were computed between the independent variable scores, revealing no multicollinearity. Box and Tidwell’s (1962) procedure revealed that the assumption of linearity was met for both samples (i.e., the logit of the criterion variable was linearly related to all continuous predictor variables). Faculty members. HLR analyses were computed to investigate the predictive validity of the CMHPCS (research question 3). Specifically, researchers aimed to determine the extent to which