cdCAT 0.1.0
- Initial release.
- Session-based CD-CAT engine via the
CdcatSession R6
class.
- Support for DINA, DINO, and GDINA cognitive diagnosis models.
- Item selection criteria: KL, PWKL, MPWKL, SHE, sequential, and
random.
- Attribute estimation methods: MLE, MAP, and EAP.
- Informative priors over skill profiles via
cdcat_prior().
- Content balancing (Kingsbury & Zara, 1991).
- Exposure control (Sympson-Hetter and Randomesque).
- Shadow CAT via user-supplied constraint functions.
- Stopping rules: fixed length, attribute-level posterior threshold,
and dual posterior threshold.
- Per-step history tracking with
history_df().
- Built-in simulated datasets (
cdcat_sim).