2.6. Statistical Analysis

Stata 14.0 software was used for analyzing data. Sociodemographic and HIV-related characteristics of respondents were described using frequency, mean, and standard deviation (SD). Cronbach's alpha was employed to estimate the internal consistency reliability of the stigmatization measures. Factor analysis was utilized to explore the construct of the scale by determining factors and restructuring the items into appropriate factors to increase the interpretability of the measure. Polychoric correlation matrix was applied because each item included in the factor analysis was binary variables. We extracted the factors using the principal component analysis (PCA) with the eigenvalue of 1.0 as a threshold for flattening out the eigenvalue curve and used Orthogonal Varimax rotation with Kaisers' normalization to reorganize the items. We used a value of 0.4 as a cut-off point for factor loadings. Multiple Zero-inflated Poisson regression models were constructed to examine the correlates of the rate of numbers of types of stigma that patients experienced. Candidate independent variables included socioeconomic, drug injection, and HIV-related characteristics of respondents. Statistical significance was defined when the p-value was less than 0.05.