Hybrid Measurement Models for Technology-Enhanced Assessments Through mIRT-bayes
Bibliographic record
Abstract
Technology-enhanced assessments (TEAs) are rapidly emerging in educational measurement. In contexts such as simulation and gaming, a common challenge is handling complex streams of information, for which new statistical innovations are needed that can provide high quality proficiency estimates for the psychometrics of complex TEAs. Often in educational assessments with formal measurement models, latent variable models such as item response theory (IRT) are used to generate proficiency estimates from evidence elicited. Such robust techniques have become a foundation of educational assessment, when models fit. Another less common approach to compile evidence is through Bayesian networks, which represent a set of random variables and their conditional dependencies via a directed acyclic graph. Network approaches can be much more flexibly designed for complex assessment tasks and are often preferred by task developers, for technology-enhanced settings. However, the Bayesian network-based statistical models often are difficult to validate and to gauge the stability and accuracy, since the models make assumptions regarding conditional dependencies that are difficult to test. Here a new measurement model family, mIRT-bayes, is proposed to gain advantages of both latent variable models and network techniques combined through hybridization. Specifically, the technique described here embeds small Bayesian networks within an overarching multidimensional IRT model (mIRT), preserving the flexibility for task design while retaining the robust statistical properties of latent variable methods. Applied to simulation-based data from Harvard's Virtual Performance Assessments (VPA), the results of the new model show acceptable fit for the overarching mIRT model, along with reduction of the standard error of measurement through the embedded Bayesian networks, compared to use of mIRT alone. Overall for respondents, a finer grain-size of inference is made possible without additiona testing time or scoring resources, showing potentially promise for this family of new hybrid models.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".