Graduation Year

2026

Document Type

Dissertation

Degree

Ph.D.

Degree Name

Doctor of Philosophy (Ph.D.)

Degree Granting Department

Educational and Psychological Studies

Major Professor

Nathaniel von der Embse, Ph.D.

Committee Member

Robert Dedrick, Ph.D.

Committee Member

Stacy-Ann January, Ph.D.

Committee Member

Lise Fox, Ph.D.

Committee Member

Qunshan Zheng, Ph.D.

Keywords

Early Childhood, Measurement, Parent, Universal Screening

Abstract

Multi-informant approaches provide a more comprehensive understanding of behavior patterns and are considered the gold standard for behavioral assessment. However, few screening tools for young children in school settings incorporate parental perspectives, and those that do remain under-researched with respect to their validity and reliability. To address these gaps, the Identifying Behavioral and Emotional Growth Indicators and Needs-Parent Form (I-BEGIN-P) was developed as an extension of the I-BEGIN-T (Teacher) scale for children aged 3 to 5 years. The purpose of this study was to develop and initially validate I-BEGIN-P. First, the content validity of I-BEGIN-P was assessed using expert panel ratings (n = 5) and feedback from cognitive interviews with parents (n = 3). The Scale-level Content Validity Index using Universal Agreement (S-CVI/UA) of I-BEGIN-P fell below the ideal threshold (≥ 0.80) but was typical for early-stage instruments. The Scale-Level Content Validity Index using the average method (S-CVI/Ave) demonstrated stronger support, with the relevance index (0.74) approaching acceptable standards and the clarity index (0.81) meeting the recommended criterion ≥ 0.80. Following these analyses, the final I‑BEGIN‑P item bank used for screening comprised 36 items. Second, confirmatory factor analyses (CFAs) were conducted using screening data provided by parents (n = 166) of children aged 3 to 5 who completed the final 36-item survey of the I-BEGIN-P items. CFAs supported the fit of a bifactor model, in which each item corresponds to a general factor (i.e., General Behavior) and one of three specific factors (i.e., Social, Emotional, and Early Learning Behavior). Third, the internal reliability of each factor was assessed. Omega total (ω) and hierarchical omega (ωₕ) coefficients indicated overall total score reliability was high (ω = .91), with a large proportion of variance coming from the general factor (ωₕ = .85). However, although domain‑specific scales displayed acceptable levels of total reliability (ω ≈ .68–.72), hierarchical omega values were much lower (ωₕ = .08–.18), suggesting that there was little variance after accounting for the general factor. Further implications for practice and future directions are discussed.

Download

Included in

Psychology Commons

Share

COinS