Publication Date

2025

Document Type

Dissertation/Thesis

First Advisor

Styck, Kara M.

Degree Name

M.A. (Master of Arts)

Legacy Department

Department of Psychology

Abstract

As SEL assessments gain traction globally, questions concerning their cross-cultural validity have become increasingly salient, particularly in linguistically diverse contexts where language and cultural schemas may differentially shape students' interpretation of socioemotional constructs. This study employed Rasch modeling and differential item functioning (DIF) analyses to examine the psychometric performance of the Social Skills Improvement System - Social Emotional Learning Brief Scales (SSIS-SELb) among 192 middle school students attending an international school in Lagos, Nigeria. All participants were proficient in English; however, they differed in native language (English: 43.1%, Yoruba: 23.1%, Igbo: 16.5%) and country of birth (Nigeria: 22.0%, United States: 36.5%, United Kingdom: 23.1%), providing a unique opportunity to isolate linguistic and cultural factors while controlling for comprehension. Analyses investigated whether item functioning varied systematically across language and national origin groups and whether Rasch-modeled item difficulty aligned with students' perceived difficulty ratings. Results revealed selective DIF patterns concentrated in communication and interpersonal domains. Four items showed practical DIF (≥0.50 logits) across native language groups, all involving comparisons with Igbo speakers. Item 9 (eye contact during communication) demonstrated significant DIF (t = 2.10, p = .040), with Igbo speakers finding this behavior easier than English speakers. Country of birth analyses revealed non-uniform DIF across six items, with Item 11 (communication about problems) showing significant differences between United States and United Kingdom students (t = -2.08, p = .040). UK students consistently found help-seeking behaviors more difficult across all ability levels. The comparison between Rasch-estimated and perceived item difficulty revealed a systematic mismatch (ρ = -.43), consistent across all cultural subgroups, suggesting universal developmental factors rather than cultural measurement bias. Findings demonstrate that cultural influences operate selectively on communication-related competencies rather than creating pervasive measurement bias, supporting cultural schema theory while challenging assumptions about uniform cultural effects. The concentration of DIF in specific domains suggests overall screening utility may be preserved while domain-specific interpretations require cultural sensitivity. Results inform future adaptation and validation of SEL measures for culturally and linguistically diverse populations and contribute to understanding the intersection of language, culture, and educational assessment in global context.

Extent

124 pages

Language

en

Publisher

Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type

Text

Download

Share

COinS