Bar Exam Candidates with Anxiety: When Are Accommodations Reasonable?

This article originally appeared in The Bar Examiner print edition, Summer 2025 (Vol. 94, No. 2), pp. 19–24. By Benjamin J. Lovett, PhDLaw school has long been associated with stress and anxiety. In 1968, a second-­year law student at the University of Wisconsin published the results of his interviews of his first-year successors, detailing the many sources of their anxiety. Unsurprisingly, “the enormity of the exams” weighed heavily on these students.1 It is unlikely that the sponsors of the disability discrimination laws passed in the ensuing years had anxious law students in mind, but, today, students with anxiety disorders often receive disability accommodations, both in their classes and later on standardized tests such as the bar examination and the Multistate Professional Responsibility Examination.2 The most common accommodations are additional testing time, rest breaks, and a separate testing location with fewer people.3

Testing accommodations have generated much controversy. Critics charge that they provide an unfair advantage, incentivize students to malinger disability, and more generally lower academic standards.4 Providing accommodations for anxiety risks stoking such critics’ fires even more. After all, aren’t most bar candidates anxious? And if they can’t perform due to anxiety, how will they ever be good attorneys? As a practicing psychologist and a testing accommodations researcher, I try to assess concerns such as these with empirical evidence. In this article, I discuss the rise of anxiety-based accommodation requests, the logical basis of accommodations, and factors to consider when making accommodations decisions. I conclude by considering alternative approaches to helping anxious bar candidates.

The Rise of Anxiety-Based Accommodations

I often get asked: Where did all these accommodation requests for anxiety come from? Several factors have combined to fuel this trend.

Increase in Anxiety Among Young People

One factor is a rise in anxiety more generally in young people, and evidence from several sources bears this out. For example, the National College Health Assessment tracks higher education students’ mental and physical health. In the fall 2024 NCHA, over one third of students (34.3%) reported having, at some point, been professionally diagnosed with an anxiety disorder.5 Two decades earlier, that statistic was 7.8%, suggesting a more than fourfold increase in the condition.6 (This likely understates the increase, since by 2024, post-­traumatic stress and obsessive-­compulsive disorders were considered separately from anxiety disorders.) I know of no similar data specifically on law students or bar candidates, but it would be unlikely that these groups have lower anxiety than the general postsecondary population. Indeed, being in a high-achieving educational setting is a risk factor for mental health problems, rather than a protective factor.7

The general rise in anxiety in young people has spurred considerable social commentary and policy proposals, with some of them addressed in psychologist Jonathan Haidt’s recent best-selling book The Anxious Generation.8 Haidt sees the roots of anxiety in young people’s time spent on social media and their lack of early exposure to free, exploratory, physical play. He calls both for increasing children’s independence in the real world and for limiting their time online—reforms that are already prompting legal changes.9 It would be difficult to conduct controlled experiments to test Haidt’s hypotheses; regardless of the causes, however, anxiety has increased considerably.

Shifts in Attitudes toward Accommodations

A second factor is an apparent change in attitudes toward disability accommodations. Campaigns against disability stigma and mental health stigma have had some success, and more students identify—both privately and publicly—as having anxiety disorders. Therefore, the proportion of college students who register at their campus’s disability services offices for accommodations for all kinds of disability conditions has risen substantially in the past decade.10 Many such offices have lowered their standards of evidence for certifying a student as disabled, granting increasing weight to the students’ subjective reporting of their experiences.11 Even when self-reports are deemed insufficient, anxiety disorder diagnoses do not require extensive neuropsychological testing and are often documented with a physician or therapist’s brief note.

Accommodation requests for standardized tests have similarly increased, particularly among students seeking accommodations for mental health problems. In 2011, the US Government Accountability Office (GAO) published results of an audit of admissions testing companies, finding that only 4% of accommodation requests were for psychiatric disorders (most requests were instead for learning and attention disorders). In 2022, the GAO published an updated audit, finding that now 13% of requests were for psychiatric disorders, more than a threefold increase in the proportion.12

The Test Anxiety Industry

My candidate for a final factor in the rise of anxiety-based requests is more speculative. I call it the test anxiety industry. Simply put, there are people who benefit from raising examinees’ anxiety about tests and their consequences. These include test prep specialists and writers of test prep materials; anxiety is a powerful motivator, and it stimulates the purchase of goods and services. Other interested parties include testing critics and activists who campaign against the use of standardized tests; there is even an organization (FairTest) devoted to such work. Still other individuals are closer to the accommodations side of the issue; they include diagnosticians (even some of my fellow psychologists) who frame their services in terms of helping students get accommodations approved for tests. Consider one psychology practice in Manhattan that warns prospective clients, “Your SAT/ACT score can determine your future . . . Don’t you want to get the best score you can?”13 The practice admits that required diagnostic evaluations are expensive, but asks: “Isn’t your future worth it?”

The test anxiety industry builds on the other trends mentioned here—the general rise in anxiety and the increase in young people’s willingness to identify as having disabilities and to seek accommodations. Considering these factors, it should come as no surprise that anxiety-­based accommodation requests have been increasing.

The Logic of Testing Accommodations

In considering when—if ever—accommodations should be provided for anxiety, it is helpful to step back and ask why any accommodations are appropriate on tests. All tests are standardized to one degree or another; they are designed to be administered and scored in the same way for all examinees, which allows fair comparisons of performance across the examinees. Given that, why do we ever break standardization and provide accommodations to students with disabilities?

Doing well on tests requires two sets of skills.14 Target skills are what a test is designed to measure. On the bar exam, this would include knowledge of the different areas of law, the tasks that newly licensed attorneys are expected to be able to do, etc. In contrast, access skills are what allow an examinee to participate fairly in a test under standard administration conditions. For instance, the standard bar exam format assumes that candidates have the visual acuity needed to read printed text. The bar exam isn’t designed to test vision, but participating in the standard test format requires vision skills.

Accommodations are designed to address deficits in access skills; if a visually impaired examinee can’t fairly participate in the bar exam in its typical format, an accommodation such as a Braille test format, use of text-to-speech software, or even enlarged print might allow that participation and let the student demonstrate whatever target skills they have. If the student fails the exam, it will be due to deficits in target skills, and those deficits are exactly what the test is designed to catch.

The logic of accommodations is easiest to see with these kinds of sensory disabilities, but today they are far rarer than accommodations for learning disabilities and attention-­deficit hyperactivity disorder (ADHD). In those cases, the basic logic is still present but less certain. A candidate with a learning disability may read slowly and need additional time to fairly participate in a test. It’s unlikely that bar exam developers intended to measure reading speed, and so reading speed would appear to be an access skill, not a target skill. Extended time would account for the deficit in reading speed and allow the candidate to show whatever target skills they have. But critics might argue that it’s important for attorneys to be able to read and understand documents at a reasonable rate of speed (working under time pressure has been cited as a core competency of attorneys).15 Moreover, extended-time accommodations have been shown to help students regardless of disability status,16 and so giving extra time has the potential to provide an unfair advantage if it’s not calibrated well to each student’s needs. Students seem to know this; most nondisabled students report that they would benefit from extended time on a standardized test, according to survey research.17 To be clear, many students with learning and attention problems are entitled to extended time (and sometimes other accommodations), but determining need and appropriateness is a more complex task.

Finally, we come to the case of anxiety disorders. The logical argument here is that anxiety prevents a student from fairly participating in a test under typical administration conditions. Anxiety is claimed to impair an examinee’s access skills, perhaps because worries take up testing time that extended time replaces, or because anxiety causes fatigue and necessitates rest breaks, or because a separate room would reduce anxiety directly.

Unfortunately, there is scant objective evidence suggesting that anxiety has direct effects on test access skills.18 Although anxious examinees tend to do somewhat worse on tests, this may be because examinees who have lower levels of target skills feel (rationally) more anxious about their likely test performance. When researchers have tried to look for direct effects of anxiety, there generally haven’t been any reliable or substantial ones. For instance, a classic study found that high school students performed at about the same level on real SAT sections (which induced substantial anxiety) and practice SAT sections (which did not).19 In recent research, more-anxious German medical students taking a licensure examination showed lower performance levels, but the relationship vanished when the researchers controlled for the students’ performance on practice tests, suggesting that the more-anxious students simply had a lower skill level.20 And a team of clinical psychology researchers found that higher education students with clinical levels of anxiety did not generally have any deficits in their timed reading comprehension skills or in the speed with which they could retrieve information from memory.21 The available research, then, suggests that, in many cases of anxiety, students may believe that accommodations are being given for access, but in fact they are being given for comfort. After all, anxiety is not inconsistent with high performance, and a fair number of high-­performing students are also very anxious.22 All this undermines the logic for accommodations in cases of anxiety, at least as a general rule.

Factors to Consider in Anxiety-Based Accommodations Requests

Even when the logic of accommodations is clearest (e.g., in the case of sensory disabilities), accommodation determinations should always be made on an individualized basis, reviewing a candidate’s request and assessing it against the available supportive documentation. Each testing agency has its own process, guided by legal regulations and logistical realities, but here I discuss factors that I consider in making a clinical-forensic determination of whether an accommodation is needed in the case of anxiety disorders.

First, an anxiety disorder should be present. Anxiety is not itself a disorder; it’s a part of the human condition. Many good things come from our ability to anticipate future events, but one unpleasant side effect is anxiety. Even if a therapist reports that a candidate is anxious, this does not necessarily indicate a disorder. Therefore, a clear, explicit anxiety-disorder diagnosis should be present, and there should generally be a description of symptoms that is consistent with the official diagnostic criteria for disorder. (Note that “test anxiety” is not a recognized disorder; it’s in fact present in almost all students to one degree or another).23

If a recognized anxiety disorder is well-documented, I then look for evidence that the student is unable to access tests (under standard administration conditions) when they are anxious. As the research evidence reviewed earlier shows, our default assumption should be that anxiety will not prevent access to tests; therefore, specific evidence of impaired test access is needed to overcome that assumption. One type of such evidence comes from testing of academic skills performed in diagnostic evaluations; if a candidate consistently does poorly on these tests when exhibiting signs of anxiety, this is very helpful to know. Another type of specific evidence comes from prior performance on real-world tests taken without accommodations; if a candidate is unable to access tests while anxious, prior unaccommodated tests should usually show poor scores. If, instead, a candidate reports anxiety that dates back many years, but has also consistently performed well without accommodations, it would appear the candidate can access tests while anxious. Remember that doing well on a test requires both target and access skills; a high test score necessarily shows access, whereas a low test score may be due to deficits in target skills, access skills, or both.

Finally, the role of self-­reported information in anxiety-based accommodation requests should be clarified. Anxiety is a private experience, and although it sometimes generates observable signs, some people conceal their anxiety well. Therefore, self-report is the primary form of evidence of symptoms that could lead to diagnosing an anxiety disorder. But even if we implicitly trust candidates’ reports of their anxiety symptoms, we cannot trust candidates’ reports of their ability to access tests. Indeed, if anything, genuinely high anxiety makes candidates’ reports concerning their access skills less accurate, since anxiety tends to make people underestimate their abilities.24 Therefore, objective data are needed. Thankfully, such data are often available.

I have only touched on some of the basic elements of reviewing anxiety-based requests; unfortunately, there is no simple flow chart that works in all decision-making situations. In practice, examinees who report anxiety disorders often report other disability conditions (e.g., ADHD) as well. In such cases, the evidence for each condition should be evaluated separately. Moreover, some examinees report highly specific reasons for requesting accommodations (e.g., needing additional time to compulsively check work, or being distracted when seated near a door through which other examinees are leaving), and the evidence must be examined to evaluate these claims. In addition, other documentation, such as those relating to a history of accommodations, must be considered when testing agencies make decisions. Finally, what counts as “poor” performance on diagnostic tests and prior real-world tests varies from one measure to another. The 25th percentile is a useful cutoff, because on diagnostic tests of cognitive and academic skills, it is generally where the average range starts. However, on real-world tests, the comparison sample may not be the general population, and may instead be other law school applicants or law students, meaning that even low scores on such tests may represent average skills compared to a general population standard.  Space limitations prevent me from covering these and other issues in detail, but I hope that this broad outline of decision analysis is a helpful start.

Beyond Accommodations: Helping Bar Candidates with Anxiety

In many cases, anxious bar candidates do not need accommodations, even if the candidates have diagnoses of anxiety disorders. But this is not a reason to simply give up on helping such candidates. Even though anxiety does not generally prevent access to tests, it can still be very uncomfortable for those who experience it, and that is reason enough to address it. Thankfully, there are evidence-based therapy techniques that can do just that.

My own preferred intervention approach falls within acceptance-­based behavior therapy. The philosophy behind this approach is simple. Emotions such as anxiety are unpleasant but unavoidable. The key lies in how we respond to those emotions. It is natural for many people to respond with avoidance, trying to stay away from any situations that might trigger the anxiety, but this tends to make the issue worse in the long run, and it tends to have collateral negative effects on the person’s life. In acceptance-­based behavior therapy, people are exposed to thoughts and feelings related to anxiety but learn that those thoughts and feelings are not dangerous.25

This learning process starts with psychoeducation—in short, educating participants about the psychology of anxiety. They learn about how anxiety is common and in fact central to our lives and even our evolutionary histories. Anxiety about tests is particularly common and certainly not suggestive of any disorder or dysfunction. Participants also learn about the relationship between anxiety and performance; they learn that anxiety does not directly impair performance, but that it can influence one’s approaches to studying and taking tests. Many examinees who are anxious about tests procrastinate on studying, for instance, or they use less effective study techniques. This kind of psychoeducation can be done by a mental health professional, but law schools and even testing entities (such as boards of law examiners) can also provide informational resources about anxiety. Many universities have webpages with such resources and those pages can serve as a useful model.26

Examinees who feel physiological symptoms of anxiety before, during, or after tests can also benefit from behavioral relaxation techniques. Slow breathing and muscle relaxation exercises reliably reduce feelings of anxiety for many. Some examinees who feel more intense panic benefit from interoceptive exposure exercises, where one deliberately brings on sensations similar to panic symptoms (e.g., by hyperventilating), and practices these exercises until the resulting sensations are no longer distracting or concerning. Finally, exposure to worries can also be helpful; use of a “worry journal” in which examinees take several minutes each day to write down worries can help to identify action steps or even simply expose worries as irrational. These more structured psychotherapy exercises are best completed with a trained mental health professional, although some participants can benefit from detailed written instructions or videos explaining how to implement these techniques.

Relative to many other psychological problems, anxiety tends to be responsive to intervention, and so even if many anxious bar candidates do not qualify for accommodations, they should not be forever stuck with severe anxiety. Reserving accommodations for those who truly need them for access—rather than comfort—maintains the sound logic of carefully altering standardized tests, puts all examinees on a fair and level playing field, and upholds standards for competence in a profession.

Contact us to request a pdf file of the original article as it appeared in the print edition.