Writing Reports That Survive Scrutiny: What High-Stakes Reviewers Actually Look For

Every psychologist who conducts accommodations evaluations believes they're writing a solid report. Many of them are wrong – not because they lack clinical skill, but because writing for a reviewer is a fundamentally different task than writing for a treatment team, a school, or a referring physician

I've been on both sides of this process. I conduct evaluations for professionals seeking accommodations on high-stakes exams, and I review documentation submitted to credentialing bodies. That dual vantage point has taught me something that doesn't get discussed enough amongst evaluating psychologists: the gap between a clinically accurate report and a reviewable report is wide, and it costs applicants every day. 

This isn't about gaming the system. A well-written report doesn't manufacture need. Rather, it makes genuine need legible to someone who will spend fifteen to twenty minutes with the report you wrote for your client before making a consequential decision. That's the actual task.

The information below is what I look for when I'm on the reviewing end, and what I try to build into every report I write.

A Clearly Established Developmental History, Not Just a Current Snapshot

One of the most common reasons a report loses credibility with a reviewer is because it’s missing a story. I’m looking for evidence that the condition being documented didn't appear out of nowhere at age 31 (i.e., statistically, the average age of Bar examinees). 

This doesn't mean every client will have a pristine paper trail. Many won't. What it means is that your report needs to actively grapple with the history (or lack thereof). If a client was never diagnosed as a child but describes years of compensatory strategies, academic struggle relative to ability, or a family or cultural history that adds context, tell me that in plain language. Don't leave me to connect those dots alone, because I don’t take those liberties as a reviewer. 

Functional Impairment That Is Specific, Documented, and Tied to the Exam Context

A diagnosis is not an accommodation. This sounds obvious, but a surprising number of reports I review stop at establishing that a condition exists without building the bridge to why that condition, in this person, meaningfully limits this specific functional demand in this specific test setting.

I’m not asking whether your client has ADHD; though, I read some reports where I genuinely wonder how certain conclusions were reached. That aside, I’m asking whether sustained attention under timed, high-pressure testing conditions is genuinely compromised for this individual — and whether your data shows it. Name the functions. Tie them to specific test performances. Then connect those performances explicitly to what a Bar exam, licensing exam, or board exam actually requires.

Internal Consistency Across the Entire Report

This is where I see many reports quietly fall apart. As a reviewer, I’m running a continuous coherence check the entire time I spend reading your report. Does the history match the presenting concerns? Do the test scores align with the clinical narrative? Does the severity described in your summary match what the scores actually show?

When things don't line up, good reviewers notice. A client described as severely impaired whose scores fall largely in the average range isn't automatically disqualified; but, that discrepancy needs to be explained in the report, not left as a question mark. Unresolved inconsistencies between the history and data obtained read as either clinical carelessness or, worse, advocacy without foundation. 

Performance Validity Testing (PVT) Is Not Optional

This may be the most consistent gap I see in reports submitted for high-stakes review: the absence of PVT data altogether. In a standard clinical evaluation, this omission might go unremarked, but in evaluations where one of the primary functions is to determine appropriateness for accommodations on a high-stakes exam, it’s a big problem.

As a reviewer, what's at stake for applicants isn’t lost on me. I get it. That said, accommodation-seeking creates an incentive structure that demands we, as evaluating psychologists, rule out suboptimal effort before we interpret any cognitive or achievement data as meaningful. Trust but verify. When a report arrives without any embedded or freestanding validity indicators, I immediately question the integrity of the data that follows, regardless of how well the rest of the report is written – and especially when it’s clearly stated that your client sought you out specifically for documentation necessary to submit a request for nonstandard testing accommodations (NTA). 

Including PVTs, and reporting the results explicitly, doesn't signal distrust of your client; it communicates that you understand the evidentiary standard the context requires, and that your data can withstand it. 

Collateral Information Is Evidence, So Treat It Like One

Self-report is inherently limited, and in high-stakes evaluations, reviewers know it. A client's account of their own struggles is clinically relevant, but it’s rarely sufficient on its own to establish the kind of longitudinal, cross-situational impairment that credentialing bodies are looking for.

Collateral information, including records from past treating providers, prior neuropsychological or psychoeducational testing, academic transcripts, employer or teacher observations, or targeted collateral interviews with family members, serves a function that no battery of tests fully replaces. It corroborates and contextualizes. It also shows me that the picture you're presenting wasn't assembled entirely from one morning with one motivated client.

When collateral is genuinely unavailable – and sometimes, this is the case – say so and tell me why. A report that acknowledges its own evidentiary limitations is far more credible than one that simply doesn't address them.

Relative Weaknesses Are Not the Same as Impairment — And Reviewers Know the Difference (And So Should You)

This may be the most clinically nuanced issue on this list, and also one of the most consequential. Intra-individual discrepancy analysis, specifically identifying areas where a person performs significantly below their own estimated ability, has legitimate diagnostic utility. But it’s frequently misapplied in accommodations reports as a primary rationale for functional impairment, and good reviewers are increasingly sophisticated about this distinction.

For example, a processing speed index score at the 30th percentile is not, by itself, evidence of impairment. If that same score falls in a range that the general population navigates without accommodation, citing it as the basis for extended time requires substantially more justification than the discrepancy alone provides. I’m asking whether the person is impaired relative to the demands of the exam and the broader population, not simply whether one of their scores is lower than another.

When discrepancy data is part of your rationale, anchor it. Show and tell me how the relative weakness translates to a specific, documented functional limitation. Without that bridge, the argument doesn't hold up under scrutiny, and your client’s denial letter will specifically say so. 

Where Reports Tend to Fall Short

After reviewing a significant number of accommodation requests, certain patterns have become immediately recognizable. These aren’t reasons to deny a request outright, but they signal that the evaluation behind the report may not meet the evidentiary standard high-stakes review requires. These are the things that make me pause before I’ve finished reading your report:

No performance validity data. When a report contains no PVTs — embedded or freestanding — the integrity of every cognitive and achievement score that follows is open to question. In a high-stakes accommodations context, this is a critical and foundational omission.

Accommodation recommendations that aren't anchored in data. A recommendation for double time that isn't traceable to any specific finding in the report is difficult to defend. I’m specifically looking for the data that justifies the specific accommodation requested. When it isn't there, your recommendation reads as predetermined rather than derived.

A thin or absent developmental history. A two-paragraph background section covering a client's entire developmental, academic, and occupational history is not sufficient. It signals either that the information wasn't gathered or that it wasn't considered important, and neither is great. 

No collateral data, and no explanation for its absence. Failing to obtain collateral information is a problem. Failing to even acknowledge that it wasn't obtained is even worse. It’s clear when a report is built entirely on self-report and a single testing session with no outside corroboration and no discussion of why that corroboration couldn't be secured.

Misusing timed/untimed score comparisons as rationale for extended time. The Nelson-Denny can be a legitimate data point in an accommodations-focused evaluation, but only when interpreted correctly. A finding that an applicant's score improves from average to above average when moving from timed to untimed conditions isn’t evidence that extended time is warranted. It’s simply evidence that additional time changes performance, which is true for virtually everyone regardless of disability status. What I need to see is that the applicant's timed performance reflects a genuine functional limitation, not merely that untimed performance is better. Using a timed-to-untimed gain as the primary or sole rationale for extended time conflates a universal testing phenomenon with documented impairment, and reviewers familiar with the measure will recognize the error immediately.

Reports that read as advocacy from the first page. There is a meaningful difference between a report that builds a thorough, well-supported case and one that feels, from the opening paragraph, like it was written to reach a predetermined conclusion. Reviewers feel that distinction quickly. When clinical language is consistently superlative, when contradictory data is absent rather than addressed, when every finding is framed to support the request and nothing is examined critically, the report stops reading like an objective evaluation and starts reading like a brief for the defense. That shift undermines everything that follows — including findings that might otherwise be perfectly legitimate. A reviewer who feels they are being sold something will scrutinize the entire document more skeptically, not less. The most credible reports are the ones that acknowledge complexity, sit with ambiguity where it exists, and let a well-constructed evidentiary argument speak for itself. Objectivity isn't just an ethical obligation in this context — it's a strategic one. Ben Lovett writes more about this here. 

The Bottom Line

High-stakes accommodations evaluations are some of the most consequential documents we produce as psychologists – not just for our clients, but also for the integrity of the credentialing systems that rely on our objectivity. Getting them right is both good clinical practice and a professional obligation.

If you have questions about any of the points I discuss here, navigating a particularly complex case, or interested in consultation for evaluations or documentation for high-stakes exams, reach out to me: alex@lightsidepsych.com.

Up next: what happens when you’ve done everything right, but the data still says no.