Four years of longitudinal social-emotional survey scores analyzed under competing IRT scoring models¶

Element · All elements
Evidence · 1 claim (1 for) · 1 study (1 associational), q2 · 0 of 1 report an effect size · 1 claim rests on one study

Description¶

The study's data and analysis setup: four years of longitudinal survey scores on social-emotional constructs, scored using item response theory (IRT) models that do and do not account for response styles. The resulting scores were then used in growth models and results compared. The article describes conducting "empirical and simulation analyses" with these scores to test effects on growth parameters.

Design Implications¶

Context¶

Requirements¶

• Longitudinal survey administration across four years
• IRT scoring models both with and without response-style parameters

Constraints¶

• Analyses concern social-emotional constructs measured via Likert-style surveys

Target Learners¶

• Survey respondents providing social-emotional ratings across four years

Target Learning Goals¶

• Estimating student growth on social-emotional constructs

Claims¶

• The effect of response styles on growth estimates varies by psychological construct, response style, and IRT model used [+W]

Related Elements¶

• Simulation and empirical comparison of scoring approaches for longitudinal survey growth estimation
• Longitudinal SEL dataset from a large urban district with about 3,000 students per timepoint

Examples¶

• Compare growth estimates from IRT scoring models that do and do not account for response styles when evaluating longitudinal survey scores

Key Sources¶

• James Soland, Megan Kuhfeld. (2020). Do response styles affect estimates of growth on social-emotional constructs? Evidence from four years of longitudinal survey scores. Multivariate Behavioral Research. https://www.nwea.org/research/publication/do-response-styles-affect-estimates-of-growth-on-social-emotional-constructs-evidence-from-four-years-of-longitudinal-survey-scores-journal/