The Professional Counselor | Volume 14, Issue 2 203 “Feeling down, depressed, or hopeless”; Kalkbrenner, 2022). Maroufizadeh et al. (2019) demonstrated convergent validity evidence of PHQ-9 scores through moderate to strong correlations with measures assessing similar constructs. Internal structure validity of PHQ-9 scores were found through factorial invariance testing by J. S. Patel et al. (2019). Scores on the PHQ-9 have demonstrated moderate to strong internal consistency reliability scores when investigated by a variety of researchers including α = .90 (Dosovitsky et al., 2021), α = .85 (Maroufizadeh et al., 2019), and α = .78 (Dajpratham et al., 2020). Similarly, we found strong internal consistency reliability evidence of PHQ-9 scores with the present sample of U.S. adults (α = .926, 95% CI [.917, .934]; ω = .927, 95% CI [.917, .935]). Data Analysis Two single-order CFAs were computed to test the dimensionality of a national sample of U.S. adults’ scores on the unidimensional and two-dimensional IWI models. We referred to Dimitrov (2012) and Schreiber et al. (2006) for cutoff scores to interpret model fit, including chi-square absolute fit index (CMIN, non-significant p-value or χ2 to df < 3), the comparative fit index (CFI, .90 to .95 = acceptable fit and > .95 = strong fit), standardized root mean square residual (SRMR < .08 = acceptable fit and < .06 = strong fit), and root mean square error of approximation (RMSEA < .08 = acceptable fit and < .06 = strong fit;). Pending acceptable fit, one or both models will be tested for factorial invariance of scores. Based on our review of the extant literature, the following sociodemographic variables were tested for factorial invariance: gender, help-seeking history, ethnoracial identity, income, and education (Kalkbrenner, 2022; Kobayashi et al., 2021; Patrick et al., 2020). Meade and Kroustalis (2006) recommended that comparison groups for multiple-group confirmatory factor analysis (MCFA) should be comprised of at least 200 participants. Accordingly, the levels of the gender and ethnoracial identity variables were coded into the following levels in order to meet the sample size requirements for MCFA: gender identity (n = 437 female or n = 391 male) and ethnoracial identity (n = 489 White or n = 334 non-White). The levels of the help-seeking history (n = 565 with a helpseeking history or n = 261, without a help-seeking history), education (n = 315 high school diploma or n = 501 undergraduate degree or beyond), and income (n = 533 above the poverty line or n = 233 below the poverty line) variables met the minimum sample size requirement for MCFA. Pending at least acceptable internal structure validity evidence of IWI scores (RQs 1 and 2), we will test convergent validity evidence by computing Pearson product moment correlations (r) between the IWI and the following well-established screening tools: the PHQ-9, GAD-7, and MHI-5. Strong negative correlations between the IWI and both the GAD-7 and PHQ-9 would prove convergent validity evidence of scores. A strong positive correlation between the IWI and MHI-5 would support convergent validity. A threshold of r > +/− .50 was used to evidence convergent validity of scores (Drummond et al., 2016). Results The unidimensional and two-dimensional IWI models were entered into two separate CFAs to test the dimensionality of scores on each model with adults in the United States (RQ 1). Pending acceptable model fit, the IWI items will be entered into an MCFA for invariance testing (RQ 2). The CFAs and MCFA were computed in IBM SPSS Amos version 26 with a maximum likelihood estimation method. Single-Order Confirmatory Factor Analysis The unidimensional IWI items were entered into the first CFA, and all the incremental fit indices displayed an acceptable model fit: CFI = .94, NFI = .93, IFI = .94. For the absolute fit indices, the SRMR indicated a strong model fit (SRMR = .04); however, the CMIN (χ2 [65] = 498.61, p < .001, X2/df = 7.67) and
RkJQdWJsaXNoZXIy NDU5MTM1