Search
On-Site Program Calendar
Browse By Day
Browse By Time
Browse By Person
Browse By Room
Browse By Unit
Browse By Session Type
Search Tips
Change Preferences / Time Zone
Sign In
X (Twitter)
This presentation introduces a novel approach for measuring latent change using IRT without repeating anchor items. This is achieved by decomposing item parameters into parameters of item attributes. These parameters are then constrained to the same values across measurement occasions. The method enables vertical equating between any two tests that share the same item attributes, and not necessarily the items themselves. The process relies on the Linear Logistic Test Model and non-Linear Logistic Test Model for difficulty and discrimination decomposition. We also describe methods to examine longitudinal measurement invariance for this approach to vertical equating. The method’s main limitation is its sensitivity to mis-specification of item attributes, such as an incorrect Q-matrix or violated measurement invariance.