TPC Journal V8, Issue 4- FULL ISSUE

304 The Professional Counselor | Volume 8, Issue 4 A CFA based on structural equation modeling was computed using IBM SPSS Amos version 25 to test the psychometric properties of the revised 14-item scale with adults in the United States (RQ1). A number of goodness-of-fit (GOF) indices recommended by Byrne (2016) were investigated to determine model fit. The Chi Square CMIN absolute fit index was statistically significant: χ 2 (74) = 3.54, p < 0.001. More suitable GOF indices for large sample sizes ( N > 200) were examined and revealed adequate model fit: comparative fit index (CFI = .96); root mean square error of approximation (RMSEA = .07); 90% confidence interval [.06, .08]; standardized root mean square residual (SRMR = .038); incremental fit index (IFI = .96); and normed fit index (NFI = .94). Collectively, the GOF indices above demonstrated adequate model fit based on the guidelines provided by Byrne. The path model with standardized coefficients is displayed in Figure 1. Tests of internal consistency reliability (Cronbach’s Alpha) revealed strong reliability coefficients for all three FSV subscales: α = .90, α = .91, and α = .87, respectively. An investigation of the path model coefficients (see Figure 1) revealed a moderate to strong association between the FSV barriers. Consequently, researchers computed a follow-up CFA to test if a single-factor model solution for the FSV Scale was a better fit with the data. Results revealed a poor model fit for the single-factor solution, suggesting that retaining the 3-factor model was appropriate for the data. Figure 1. Confirmatory Factor Analysis Path Model (N = 410)