Introduction
Ask a court whether a defendant is competent to stand trial and you’ll get a yes-or-no question back. It’s an understandable simplification — the case either proceeds or it doesn’t. But the clinical reality behind that switch is almost never binary, and pretending otherwise quietly distorts both the finding and the restoration that may follow.
This is the case for scoring competency on a five-level model instead of a single competent/incompetent verdict — the approach built into the MCES and J-MCES.
The problem with a binary finding
Competency exists on a continuum. A defendant who cannot name a single courtroom role is incompetent in a very different way than one who understands the proceedings but cannot, because of acute anxiety, sit through a hearing — who in turn differs from one whose grasp is shaky but workable with accommodation. A binary finding flattens all of that into one of two boxes.
That flattening has costs. It hides the marginal cases — the defendants near the threshold, where reasonable evaluators most often disagree and where a defensible, granular finding matters most. It gives a restoration program no starting point and no target. And it forces an examiner to round a nuanced clinical picture up or down, then defend that rounding on cross-examination.
Five levels, defined
The MCES and J-MCES replace the switch with a graded scale — what the technical manual describes as a move “beyond the simplistic ‘competent’ or ‘not competent’ dichotomy.” A defendant is placed at one of five levels:
- Level 1 — Incompetent (Very Low)
- Level 2 — Incompetent (Low)
- Level 3 — Borderline Incompetent / Minimally Competent (Marginal)
- Level 4 — Minimally Competent (Fair)
- Level 5 — Competent (Adequate)
The legal threshold still exists — it simply sits at a defined point on the scale: Level 4 or above corresponds to a competent finding, Level 3 or below to incompetent. The difference is that the model tells you how far from that line a defendant sits, and in which direction they’re moving.
What a graded scale gives the court
A level isn’t just a softer label — it carries information a binary finding cannot:
- Nuance for the judge. “Level 3, marginal” communicates something a flat “incompetent” doesn’t: this is a near-threshold case, not a profound one.
- A defensible position. A placement anchored to a standardized scale is far easier to support than a category assigned by impression alone — especially in the contested middle.
- A way to track change. Because the same scale is re-administered over time, “Level 2 at intake, Level 4 at exit” is a measured trajectory, not an assertion that the defendant “seems better.”
And the threshold itself is empirically grounded: when the instrument’s level-based classifications were compared against courts’ independent competency findings, agreement was strong (Cohen’s κ = .76; area under the curve = .95; roughly 92% accuracy). The graded model doesn’t trade rigor for nuance — it adds nuance on top of rigor.
From label to roadmap
The five-level finding pairs with a profile across the four content areas — General Legal Knowledge, Tolerating Stress and Behavior/Conduct, Working with an Attorney, and Ability to Consult with Counsel. So a result isn’t just “Level 2”; it’s “Level 2, with the deficit concentrated in legal knowledge and attorney consultation.” That’s a roadmap. It tells a restoration program exactly where to begin and which of the four MCAT phases to emphasize — turning an evaluation from a verdict into the first step of a plan.
That hand-off — from a graded evaluation to a targeted restoration — is the whole point of pairing the scales with a structured program. We follow the measurement across the phases in Measuring Progress, Not Just Status.
Want findings with texture, not a coin flip? The MCES and J-MCES score competency on the five-level model, with normed results you can defend. Evaluator subscribers also get APA-approved CE at a discount.