Latent State Trait (LST) decomposition of test disengagement over repeated administrations¶

Theory · All theories
Evidence · 2 claims (2 for) · 1 study (1 associational), q2 · 0 of 1 report an effect size · 2 claims rest on one study

Description¶

The study applies a Latent State Trait (LST) model to decompose an individual's relative standing at each test occasion into two components. As the abstract explains, standing "is decomposed into a trait factor that is free of situational or interactional effects and an occasion factor capturing the situational or interactional component." If rank ordering is entirely consistent over time, observed variance is fully attributable to the trait factor. The authors use this framework to judge whether rapid guessing is trait-like or state-like across six administrations over three years.

Design Implications¶

Context¶

Requirements¶

• Requires achievement test data from multiple administrations over time so that trait and occasion variance components can be separated

Constraints¶

Target Learners¶

• K-12 students taking interim achievement tests

Target Learning Objectives¶

• Understanding and detecting test-taking disengagement (rapid guessing)

Claims¶

• Rapid Guessing State Like Not Trait [+M]
• Rapid guessing can be examined over three years using six test administrations to distinguish trait-like from situational disengagement [+W]

Related Theories¶

Examples¶

-

Key Sources¶

• James Soland, Steven Wise. (2020). Do students rapidly guess repeatedly over time? A longitudinal analysis of student test disengagement, background, and attitudes. NWEA Research presentation, 2020 National Council on Measurement in Education virtual conference. https://www.nwea.org/research/publication/do-students-rapidly-guess-repeatedly-over-time-a-longitudinal-analysis-of-student-test-disengagement-background-and-attitudes/