Replication Data for: Cross-Context Behavioral Adjustment Among Adolescents With and Without Specific Learning Difficulties: A Multi-Informant Analysis of Home and School Functioning

Description

Dataset Overview & Research FrameworkThis repository provides the primary empirical dataset ($N = 500$ male adolescents aged 14–16 years, equally split between specific learning difficulties [SLD; $n = 250$] and typically developing [TD; $n = 250$] groups) examining cross-context behavioral adjustment. To overcome the informant variance problem, the study integrates father-rated Home and Community Social Behavior Scales (HCSBS) and teacher-rated School Social Behavior Scales (SSBS) evaluating Social Competence and Antisocial/Problem Behavior.Research HypothesesH1 (Group Differentiation): Adolescents with SLD exhibit significantly lower social competence and higher antisocial/problem behaviors than TD peers across settings.H2 (Cross-Context Stability): Behavioral differences remain stable across ecological contexts, yielding non-significant Group $\times$ Context interactions in multilevel models.H3 (Full-Sample Convergence): Father and teacher assessments demonstrate strong overall bivariate correlation.H4 (Within-Group Variability): Subgroup analyses reveal contextual variations reflecting unique environmental influences.Empirical FindingsWelch's independent-samples t-tests demonstrate massive group differences across all domains:Social Competence (Home - Father): SLD Mean = $67.10$ ($\text{SD} = 8.58$), TD Mean = $141.31$ ($\text{SD} = 5.14$), $t = -117.28$, $p < 10^{-300}$, Cohen's $d = -10.49$.Social Competence (School - Teacher): SLD Mean = $68.20$ ($\text{SD} = 9.76$), TD Mean = $141.80$ ($\text{SD} = 4.61$), $t = -107.83$, $p < 10^{-270}$, Cohen's $d = -9.64$.Antisocial Behavior (Home - Father): SLD Mean = $108.30$ ($\text{SD} = 11.47$), TD Mean = $45.83$ ($\text{SD} = 4.32$), $t = 80.60$, $p < 10^{-200}$, Cohen's $d = 7.21$.Antisocial Behavior (School - Teacher): SLD Mean = $106.79$ ($\text{SD} = 11.55$), TD Mean = $46.31$ ($\text{SD} = 4.94$), $t = 76.14$, $p < 10^{-200}$, Cohen's $d = 6.81$.Full-sample inter-informant correlations are exceptionally strong ($r = 0.964$, $p < 10^{-288}$ for Social Competence; $r = 0.921$, $p < 10^{-206}$ for Antisocial Behavior).Interpretation & UsageThe data confirm that learning difficulties are associated with pervasive cross-context psychosocial vulnerabilities (H1 & H2 supported) while maintaining high inter-informant convergence (H3 supported). Researchers can use the dataset (.xlsx), Python script (.py), and documentation (README.txt) to replicate multilevel mixed models accounting for institutional clustering ($u_j$) and Bland–Altman sensitivity analyses.

Steps to reproduce

To reproduce the analyses and statistical models contained within this dataset, execute the provided Python script (Statistical_Analysis.py) alongside the main data workbook (Primary_Empirical_Dataset_SLD_Dual_Lens_N500.xlsx) using the following workflow:Environment Setup: Install Python 3.9 or a newer version, along with the required scientific computing libraries (pandas, numpy, scipy, and statsmodels).Data Ingestion: Load the dataset into your workspace, ensuring that Sheet 1 (Raw_Data) and Sheet 2 (Codebook) are accessible.Descriptive & Reliability Verification: Run descriptive statistics and verify internal consistency scores (Cronbach’s $\alpha = 0.88$ to $0.94$) across the Home and Community Social Behavior Scale (HCSBS) and School Social Behavior Scale (SSBS) measures.Group Comparisons (H1): Execute Welch’s independent-samples t-tests to evaluate group differences in social competence and antisocial behaviors between adolescents with specific learning difficulties (SLD) and typically developing (TD) peers.Multilevel Mixed-Effects Modeling (H2): Fit multilevel mixed-effects models (HLM) using school-level random intercepts ($u_j$) to account for institutional clustering across the three public secondary schools, evaluating the Learning Status $\times$ Context interaction.Agreement & Sensitivity Analyses (H3 & H4): Calculate Pearson correlation coefficients, two-way mixed-effects absolute agreement intraclass correlation coefficients ($\text{ICC}(2,1)$), and Bland–Altman procedures across the full sample and subgroup splits.

Institutions

• Central China Normal University

  Hubei, Wuhan

Categories

Licence