The goal of this analysis is to evaluate the Levante Hearts and Flowers (H&F) IRT model's performance on two external datasets collected by Spark Lab, as a test of the model's generalizability.

Analysis of raw data, IRT estimates and reaction time data.

This analysis examined predictors of executive function (EF) development in early childhood using both long-format mixed-effects models and wide-format regressions.

Analysis of cross-site vocabulary item performance using raw accuracy and IRT estimates from the LEVANTE pilot data.

This analysis explores developmental patterns in children’s sentence comprehension using the TROG task, combining item-level accuracy and IRT-derived difficulty estimates to examine variation across grammar structures, ages, and sites.