Definition:
A cut score (also called a cutoff score or passing score) is the predetermined score threshold on a test that separates categories — typically “pass” from “fail,” or one proficiency level from another. On the JLPT, for example, each level (N5–N1) has a cut score that determines whether a test-taker passes.
In-Depth Explanation
How cut scores are set:
Cut scores are not arbitrary — they’re established through a formal standard-setting process involving expert judgment and statistical analysis. Common methods include:
| Method | How it works |
|---|---|
| Angoff method | Experts estimate the probability that a “minimally competent” person would answer each item correctly. The sum of these probabilities becomes the cut score. |
| Bookmark method | Items are ordered by difficulty. Experts place a “bookmark” at the point where a minimally competent person would begin struggling. |
| Borderline group method | Identify test-takers who are judged “borderline” by teachers/raters. Their median score becomes the cut score. |
| Contrasting groups | Compare known-pass and known-fail groups. The cut score is set where the two distributions intersect. |
JLPT cut scores:
| Level | Total Possible | Passing Score | Section Minimums |
|---|---|---|---|
| N1 | 180 | 100 | 19 per section |
| N2 | 180 | 90 | 19 per section |
| N3 | 180 | 95 | 19 per section |
| N4 | 180 | 90 | 38 per two-section composite |
| N5 | 180 | 80 | 38 per two-section composite |
Note: JLPT uses scaled scores (not raw scores), and you must meet both the total cut score and section-level minimums to pass.
Consequences of cut scores:
Cut scores create binary outcomes from continuous data. A person scoring 89 on JLPT N2 has essentially the same ability as someone scoring 90, but only the latter “passes.” This is an inherent limitation of cut-score-based systems.
Good testing programs address this by:
- Reporting confidence intervals around the cut score
- Using multiple measures for high-stakes decisions
- Setting cut scores through rigorous, documented standard-setting processes
- Accounting for measurement error in the cut score region
Related Terms
See Also
Research
- Cizek, G. J., & Bunch, M. B. (2007). Standard Setting: A Guide to Establishing and Evaluating Performance Standards on Tests. Sage. — Comprehensive guide to cut score setting methods.
- Tannenbaum, R. J., & Wylie, E. C. (2008). Linking English-language test scores onto the Common European Framework of Reference: An application of standard-setting methodology. ETS Research Report, RR-08-34. — Demonstrates cut score setting for language proficiency levels.