Student Growth Percentiles Based on MIRT: Implications of Calibrated Projection

This research concerns a new proposal for calculating student growth percentiles (SGP, Betebenner, 2009). In Betebenner (2009), quantile regression (QR) is used to estimate the SGPs. However, measurement error in the score estimates, which always exists in practice, leads to bias in the QR-based estimates (Shang, 2012). One way to address this issue is to estimate the SGPs using a modeling framework that can directly account for the measurement error, such as Multidimensional IRT (MIRT) which is the one utilized here. To maximize the generality of the approach, the SNP-MIRT model (Monroe, 2014), which estimates the shape of the latent variable density, is used to obtain model parameter estimates. These estimates are then used with the calibrated projection linking methodology (Thissen, Varni, et al., 2011, Thissen, Liu, Magnus, & Quinn, 2014, Cai, in press-a, Cai, in-press-b) to produce SGP estimates. The methods are compared using simulated and empirical data.