Bayesian model averaging of (a)symmetric item response models in small samples.

Asymmetric IRT models present theoretically desirable features, but often require large sample sizes for stable estimation due to additional item parameters. When applying item response theory (IRT) to small samples, it is often the case that only mo…

Preview only. Read the full abstract at the source