Data Analysis Strategy: Gender stereotyped beliefs and violent behavior scale (GSBVS)
Description
This dataset contains the raw responses from the application of the Gender stereotyped beliefs and violent behavior scale (GSBVS). The data were collected to examine the underlying dimensional structure of gender stereotypes and their association with justifications of violent behaviors in intimate partner relationships. The database includes 32 Likert-type items (ranging from 1 to 5). A total of 23 items were retained after psychometric analysis. The dataset also includes sociodemographic variables used to characterize the sample. Data availability: The full dataset (anonymized) is provided in CSV/SAV format. For further details, see the associated study (Deleon Villagran & González Gómez, 2026).
Files
Steps to reproduce
Data Analysis Strategy Step 1: Software and Data Preparation • Software used: o IBM SPSS Statistics 28.0.0.0 (190) with AMOS 24 extension o JASP 0.96.0 Step 2: Descriptive Analysis of Sociodemographic Variables • Objective: To describe the sample profile. • Techniques: o Measures of central tendency o Measures of dispersion o Measures of distribution o Percentages and frequencies Step 3: Processing of the Gender stereotyped beliefs and violent behavior scale (GSBVS) Divided into three stages: 3.1. Exploratory Factor Analysis (EFA) • Extraction method: Unweighted Least Squares (ULS) – suitable for ordinal Likert-type data without assuming multivariate normality. • Rotation method: Promax (allows correlation among latent variables). • Evaluation criteria: o Bartlett's test of sphericity (intercorrelations) o Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy o Kaiser's criterion (eigenvalues > 1) o Factor loadings > 0.400 for item retention • Expected outcome: Underlying dimensional structure. 3.2. Confirmatory Factor Analysis (CFA) • Estimation method: Maximum Likelihood with Robust Standard Errors (MLR) – corrects for lack of multivariate normality. • Goodness-of-fit indices: o CMIN/DF (Minimum Chi-square divided by Degrees of Freedom) o CFI (Comparative Fit Index) o RMSEA (Root Mean Square Error of Approximation) o SRMR (Standardized Root Mean Square Residual) • Procedure: o Model re-specifications (item deletion based on low standardized beta weights; correlation of residuals between theoretically justified item pairs). 3.3. Reliability Analyses • Coefficients used: o Cronbach's Alpha (α) o McDonald's Omega (ω) • Acceptance criterion: Values > .700 (for exploratory studies) • Computation: For each dimension and for the total scale. Step 4: Association Among Latent Variables • Method: Spearman's Rank Correlation (rho) • Objective: To evaluate associations among the resulting dimensions. Step 5: Final Univariate Descriptive Analysis • Descriptive results are presented for the 11 items taken from the Ambivalent Sexism Scale (Cárdenas et al., 2010).
Institutions
- Universidad Dr. Andrés BelloSan Salvador Department, San Salvador