This (unidimensional) dataset includes item parameters from a large-scale reading assessment. The parameters were estimated using a combination of the three parameter logistic model (3PL) and the generalized partial credit model (GPCM). There are six sets of parameters which are based on tests administered in four grades over three years. In particular, the data include a grade 3 and grade 4 test in year 0, a grade 4 and grade 5 test in year 1, and a grade 5 and grade 6 test in year 2. The label for the information related to the grade 3 test is grade3.0 where the value after the decimal indicates the year. Similar labels are used for the other grade/year combinations. This dataset is used for illustrative purposes to show in a multi-group scenario how items from different response models can be mixed together and common items can be in different positions across groups.
A list of length four. The first element is a list of length six with item parameter estimates for each grade/year. There is no location parameter for the GPCM items. The second list element is a list identifying the number of response categories for the six grade/year combinations. The third element specifies which items correspond to the different item response models for each grade/year respectively. The last element is a list of common item matrices for each adjacent grade/year group.
Briggs, D. C. & Weeks, J. P. (In Press) The Impact of Vertical Scaling Decisions on Growth Interpretations. Educational Measurement: Issues and Practices.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.