Extended Time: The Antibiotic of Accommodations

The most predictable content in a psychoeducational evaluation written to support an exam accommodation typically comes in the form of a single recommendation: 50% extended time.

I read a lot of reports written by other psychologists. I write them myself, and I review them for credentialing bodies, and somewhere along the way I started keeping an informal tally of what exam-related report recommendations usually include. The referral questions vary; the diagnoses do, too, and especially the actual objective test data. Yet the reports often land, repeatedly, in the same place: extended time, 50%. Sometimes as much as 100%. The recommendation section occasionally reads like it could have been written before the client walked in the office.

As psychologists, and the clinicians most often called upon to make sound accommodation recommendations for applicants taking high-level exams, that pattern should bother us more than it does. When one intervention gets recommended across wildly different presentations – ADHD, SLD, anxiety, depression, a concussion from 10 years ago – one of two things is probably true. Either we’ve discovered a genuine panacea, or the recommendation has stopped being a clinical judgment and become a reflexive habit.

It’s the second one. And I think the clearest way to talk about it, for clinicians and applicants both, is this: extended time is the antibiotic of accommodations.

I mean that as a compliment, mostly. Antibiotics are real medicine. For the right infection, they are indispensable, and nobody serious wants to stop prescribing them. But an antibiotic is indicated for a specific, confirmed condition, not for the general experience of feeling unwell or stressed. It does hardly anything against the wrong target; you can take amoxicillin for a virus all week and not much changes other than your gut flora. The dose matters; more is not necessarily better. And over-prescription carries a cost that may eventually land on the applicant.

Each of those properties maps onto extended time, and the last one – resistance – is the one our field seems the least apt to discuss.

To be clear about what the metaphor is and isn’t saying: it isn’t saying extended time is fake, or that people who receive it are getting away with something. It’s saying extended time is a targeted intervention that we’ve been handing out like a general tonic – and that the people best served by getting this right are the ones whose limitations are unambiguous.

First, The Part That’s True

Any honest version of this conversation should start with the specific and real case for extended time.

Standardized exams are built to measure a target skill (e.g., your knowledge of contract law, your clinical reasoning, your command of organic chemistry, etc.). But demonstrating the target skill requires certain access skills, and reading speed is often a big one. If the MCAT intends to measure your science knowledge but you can't decode the passage fast enough to reach the questions, your score is partly a measure of your reading rate, which is not what anyone, including AAMC, really wants to measure. Extended time exists to remove that contamination. That's the whole theory, and it's a good theory.

This contamination is sometimes referred to in the literature as "construct-irrelevant variance," which is a mouthful for a simple idea: the test is accidentally measuring something it didn't set out to measure. An accommodation is supposed to strip out the accidental part without touching the intentional part.

And when the mismatch is real, extended time does what it's supposed to do. Lewandowski, Lovett, and Rogers (2008) are usually cited for their skeptical headline finding (we'll get there), but it contains a pro-accommodation result that doesn’t get mentioned much: with extra time, students with reading disabilities attempted as many items as their non-disabled peers had attempted under standard conditions. That's not an unfair boost. That's access, restored, by exactly the mechanism the accommodation was designed around.

Some newer data agree, with a small wrinkle. Wei and Zhang (2024) analyzed NAEP eighth-grade math data and found that students with learning disabilities who received and actually used extended time scored significantly higher than students with SLD who received nothing. Students who were granted the time but didn't use it scored significantly lower. And among the students who benefited, performance peaked at roughly 50% additional time. Not unlimited. Not double.

So: real medicine, and it helps the people it's meant for. Hold onto that. The problem isn't extended time. The problem is handing it to nearly everyone who asks, in whatever amount is customary, without first figuring out what's actually slowing the person down.

The Broad-Spectrum Problem

There are two general standards floating around for what makes a testing accommodation legitimate, and a lot of the confusion in this area comes from not noticing which one you're using when making recommendations for applicants.

The strict version is sometimes called the interaction hypothesis, or the maximum potential thesis: an accommodation is appropriate only if it helps examinees with disabilities and does essentially nothing for anyone else. On this view, non-disabled test-takers are already at their ceiling within the standard time limit, so extra time should be inconsequential.

The looser version – the differential boost hypothesis (Sireci, Scarpati, & Li, 2005) – concedes that extra time might help everybody a little, but says the accommodation is still justified so long as it helps the target group more.

This is like a hearing aid versus a megaphone. A hearing aid is targeted: it transforms things for someone with hearing loss and does nothing for someone without it. A megaphone makes everything louder for everyone in the room. Both "help." Only one tells you anything about who needed it. The entire empirical question is which of these extended time actually is.

Extended time has now been tested against both standards for a couple decades, and it fails the strict one outright and fails the loose one more often than not.

Lewandowski, Lovett, and Rogers (2008) gave high school students with and without reading disabilities the Nelson-Denny Reading Test (NDRT) under varying time limits. The non-disabled students benefited more from the extra time than the students with reading disabilities. The title of their article – Does a rising tide lift all ships? – speaks for itself.

Lewandowski, Cohen, and Lovett (2013) ran the college version, with 50% and 100% extensions on reading comprehension. Same pattern: non-disabled students gained more than the LD group. And an important additional finding: in the condition where only the LD group received double time, they outperformed their non-disabled peers. The accommodation didn't level the field; instead, it tilted it the other way.

The ADHD version is, if anything, more damaging to our field’s reflexive recommendation of 50% extended time. Lewandowski et al. (2007) gave students with and without ADHD a math test at standard time and time-and-a-half. The ADHD group had lower processing speed, lower math fluency, and lower achievement; so, by the numbers, they were genuinely slower. They still did not benefit more from the extra time than controls did. That means that slower did not translate into helped more by time.

Lovett's (2010) review in Review of Educational Research pulled this literature together and reached the conclusion that has mostly held since: the empirical case for extended time as a disability-specific accommodation is far weaker than the frequency of its recommendation implies.

Then there's arguably the most important statistic in this entire literature that speaks directly to the "but it helped me" argument. When Lewandowski and colleagues (2014) surveyed college students about whether 50% additional time would improve their performance on high-stakes tests, roughly 87% of non-disabled students said yes. Among students with disabilities? About 88%.

Those numbers are essentially the same. Nearly everyone believes extra time would help them, and the experimental data say nearly everyone is right.

This is why "the extra time helped me" isn't the argument people think it is. A runner is faster with a tailwind; that doesn't mean they have a limp - it demonstrates that tailwinds are nice and helpful. The question a testing agency is legally required to ask isn’t whether the accommodation improved your score. It’s whether, without it, your score would have understated your ability because of a documented disability and resultant impairment that separates you from the general population. “It helped” is evidence of almost nothing, because it’s true of almost everyone.

To balance this argument, the state of this literature is somewhat mixed. Some score-comparability studies find that accommodated administrations measure the same construct the same way; others find distortion. Witmer and Marinho (2024), working with NAEP process data, found that score comparability held whether or not eligible students actually used their extended time - which is an odd result if the accommodation is doing heavy lifting, and a perfectly sensible one if, for many of the students granted it, it wasn't doing much of anything.

But no version of the mixed evidence supports our field’s reflexive recommendation for extended time regardless of disability or impairment. Extended time is broad-spectrum. It is not specific to disability. It lifts a very large share of everyone who receives it, which means the fact that it lifts you is close to diagnostically worthless.

Wrong Bug, Right Drug

An antibiotic prescribed for a viral infection isn't neutral. It's bad practice and exposes the patient to side effects, delays the treatment that would have worked, and feeds resistance. The question was never whether the patient is sick. It's whether they're sick with the thing the drug treats.

I think of extended time the same way. It treats exactly one thing: an access barrier caused by insufficient speed on a skill the test doesn't intend to measure. That's it. The data demonstrates that it doesn't treat distractibility, dread, fatigue, working memory issues, or the feeling of walking into a testing center underprepared. It treats rate. So, the clinical question is never "does this person have a diagnosis?" It's "does this person's impairment express itself as a rate problem on the tasks this exam actually requires?"

These two questions come apart constantly. This is especially true for ADHD, which matters because ADHD is now the modal referral in adult accommodations work. Lovett and Leja (2015) looked directly at the relationship between ADHD symptoms and benefit from extended time in college students, and found something I rarely see being discussed among psychologists doing this work: students reporting more ADHD symptoms and more executive dysfunction benefited less from extra time. Not equally, but less. Their proposed explanation is almost embarrassing in its obviousness once you hear it. That is, using extra time productively requires sustained attention, self-monitoring, and the ability to stay in the chair, which are usually the capacities in question for individuals with ADHD. The students' own perceptions of how much they needed extra time, incidentally, did not predict how much they gained from it.

Their conclusion, stated about as plainly as peer-reviewed journal prose allows, was that there may be little justification for consulting a student's ADHD symptom count when making extended time decisions. Related work by Lewandowski, Gathje, Lovett, and Gordon (2013) on actual test-taking behavior points the same direction. Specifically, college students with ADHD frequently read at comparable rates and attempt comparable numbers of items under standard conditions.

None of this says ADHD isn't real, or that its impairments don't matter. It's a claim about mechanism. A person can be genuinely, substantially impaired by ADHD - in their studying, their sleep, their follow-through, their daily functioning - and still not be slow on a timed multiple-choice item. Impairment is not one undifferentiated substance that extra minutes dissolve.

Anxiety and depression are the second common mismatch. Being clear here is important, because I evaluate a lot of anxious professionals and their suffering is not in dispute. The intuition that anxiety slows people down is understandable, and there are presentations where it genuinely does. But the evidence that anxiety or depression produces the kind of rate impairment extended time was built to correct is thin, and there's a subtler problem underneath: when extra time helps an anxious examinee, it often works as a psychological safety net - the relief of knowing the clock isn't a threat - rather than as a correction for an actual access barrier. That relief is real and I don't dismiss it. But a safety net is not an accommodation, and the interventions that actually treat test anxiety (i.e., exposure, cognitive work, preparation, sometimes medication) are treatments. Prescribing extra time for anxiety is, at best, treating a symptom with a tool built for a different pathology. At worst it substitutes an accommodation request for care that would have actually helped the person.

And then there’s the measurement error I see more than any other and wrote more about here: treating a lower than average processing speed index score, by itself, as proof of a rate problem. Lovett et al. (2022) administered measures of cognitive processing speed and timed academic skills to 447 students with prior LD diagnoses. The correlation between the two was modest, specifically roughly six percent shared variance.

Six percent shared variance means the two things mostly travel separately. Knowing someone's score on a symbol-search or coding task tells you very little about how fast they actually read a passage, work a problem, or produce writing under pressure. If the claim is that someone is slow at what the exam requires, the only defensible move is to measure what the exam requires.

A WAIS processing speed index isn’t necessarily indicative of a test-taker’s reading rate. It’s also not necessarily reflective of how fast they compute math problems. And it’s not evidence of a substantial limitation in test-taking speed relative to the general population. But it keeps getting deployed as all three, and any reviewer who knows this literature — especially a fellow psychologist well versed in this work — knows that index score can't do that work alone. It needs corroboration from measures of the actual skill in question.

Where Did 50% Come From?

I’ve asked this question to more senior colleagues and experts working in this space, my peers, and in consultation groups: why time and a half?

What I hear mostly is "that's the standard," or "that's what's always been done." And when you go into the literature, you find researchers in this field saying the same thing, but not approvingly. Their read is that the number came from convention: a rule of thumb that hardened into policy through repetition, without any process that asked what amount of additional time actually equalizes access for a given impairment on a given task. They've been making that point for well over a decade, and the standard hasn't moved. If we dosed medication this way, it would be malpractice. In accommodations work, it's just another day.

The dose question turns out to have an evidence base, and the evidence doesn’t support the status quo. In their 2022 review of extended assessment time in post-secondary settings, Harrison, Pollock, and Holmes (2022) concluded that the research supports capping standardized time extensions at roughly 25% for most students with learning disabilities, with anything beyond that tied to documented, task-specific impairment. The same review cites work by Golan and colleagues (2020) making the blunter claim: no empirical or theoretical justification exists for the reflexive 50% extended time recommendation. And recall what happened in the Lewandowski, Cohen, and Lovett (2013) study at double time - the LD group didn't just reach their non-disabled peers, they passed them. An accommodation that produces scores above what comparable non-disabled examinees achieve isn't leveling anything. It’s become an advantage, which is exactly what testing entities and other exam-administering institutions are obligated to prevent, and exactly what makes the next applicant's request harder.

The naturalistic data add some irony. We've been prescribing far more time than most recipients use. Multiple studies examining actual usage find that many students granted extended time finish within standard limits, and a large majority use well under their full extension. More recent work makes this visible in a way self-report never could, because computer-based testing now produces process data, specifically timestamped logs showing exactly who used how much time on which item. Witmer, Lovett, and Buzick (2023) used NAEP process data to pull apart three things our field routinely collapses into one: eligibility, use, and benefit. They are three different facts about a student, and knowing the first tells you surprisingly little about the third.

A student can be eligible for extended time, never use it, and score exactly as they would have anyway. Our documentation almost never distinguishes these cases, because our analysis stops at eligibility. That's writing the prescription and never checking whether the patient filled it, took it, or got better.

And like any drug, extended time has side effects we rarely name. The most obvious is the length of the day: time-and-a-half or double time turns an already-brutal eight-hour licensure exam into a marathon, and fatigue is not a neutral variable for anyone, least of all for the conditions that usually prompt the request. For an anxious examinee, more time in the room can mean more hours spent in the exact situation that produces the symptoms. And there’s a subtler cost clinicians who work with students know well: a person who succeeds with an accommodation they never needed learns to credit the accommodation instead of themselves. None of these costs shows up in the report, yet all of them are real.

The Careful Approach

So, what does the non-reflexive version look like? Not fewer recommendations for people who need them, just better-built, evidence-aligned ones. Recommend an accommodation the way a good doctor prescribes an antibiotic: be clear and specific on what you're treating, make sure the treatment fits, use no more than you need, and say why.

In practice that means the diagnosis gets confirmed with more than a screener - full diagnostic criteria, developmental history, collateral data, multiple sources. It means impairment gets documented against the general population, on age-based norms, because the comparison group the law cares about is most people, not other graduate, law, or medical students. It means that if the claim is a speed-based barrier, you measure speed on the relevant task - timed reading fluency, timed comprehension, math fluency, written production under pressure - rather than substituting a processing speed index that shares only six percent of its variance with the thing you're claiming.

It means checking that the documented impairment interacts with this specific exam. A math fluency deficit is not a rationale for extended time on a reading-heavy licensure exam; specify the demand, specify the barrier, and show the two touching. That interaction analysis is the heart of a defensible recommendation, and it's the piece most often missing.

Part of that interaction analysis is asking how speeded the exam actually is. Extra time only changes outcomes when time is scarce, and exams differ enormously on this: some are deliberately speeded, with more items than most examinees can comfortably finish, while others are built so nearly everyone reaches the last item with minutes to spare. The same documented impairment might warrant extended time on the first kind of exam and nothing at all on the second. A recommendation that never engages with the actual time demands of the actual test is half an analysis.

It also means ruling out the imposters. Is the slowness attributable to anxiety, to inefficient strategy, to inadequate preparation, to a medication effect, to sleep? Each of those has a different - and usually better - intervention than just more time, and mistaking one of them for a rate impairment is exactly how the wrong drug gets prescribed.

And, lastly, it means asking, every time, whether something narrower or more specific fits better. Stop-the-clock breaks for a medical or attentional condition. A less-busy environment. Text-to-speech for a decoding limitation (which very likely would also necessitate extra time). These are sometimes better matched to the actual documented barrier and considerably easier to defend on review, because the mechanism connecting the limitation to the remedy is legible. As a reviewer, some of the strongest recommendations I've read centered on something other than extended time entirely, specifically a narrower barrier and an accommodation matched to it.

Running underneath all of it: performance and symptom validity testing, reported transparently. The incentive structure in high-stakes accommodations work is what it is, and many reviewers have stopped extending the benefit of the doubt to reports that omit these measures. Including them isn't an accusation of anyone. It's the methodological floor for an evaluation whose conclusions will be relied on in a legal context.

The self-test I'd offer any evaluating psychologist, and the one I try to apply to my own work: if your recommendation section could have been written before the evaluation began, your evaluation didn't inform the recommendation, and your client may have paid you for something other than an objective opinion.

Resistance

Now the part of the metaphor that complicates things for everyone.

Antibiotic resistance isn't caused by the patients who needed antibiotics. It's caused by the accumulation of unnecessary prescriptions - each one individually defensible, well-intentioned, and easier than the conversation that would have avoided it. The harm is diffuse and delayed, and may only manifest later on.

I find that accommodations resistance works in a similar way. Every extended time recommendation built on a thin rationale - a low processing speed index alone, a positive screener, a discrepancy from estimated potential, a school accommodation granted years ago on a one-page note - teaches reviewers (and agencies) that documentation written by psychologists can't be trusted at face value. Reviewers have learned the lesson thoroughly. The result is the system every applicant now faces: heightened scrutiny, longer timelines, and heavier documentation demands. Including - especially - the applicants whose limitations are severe and unambiguous. They are sometimes the individuals paying for everyone else's easy prescriptions.

There’s also an equity dimension here that deserves more attention than it gets. A private evaluation of the kind these agencies expect costs real money, and the reflexive prescription flows disproportionately to the people who can afford the prescriber. When extended time functions as a general performance enhancer handed out on thin evidence, it erodes reviewer trust and it becomes one more advantage that tracks financial resources rather than documented need. The tighter, better-matched version of this work isn’t only more defensible. It’s fairer.

Our field's contribution to this is documented, and it is self-reported. In survey work summarized by Harrison and Sparks (2022), 14 percent of psychologists who had authored disability documentation acknowledged they would bend rules or ignore published diagnostic criteria to help a client secure a desired accommodation. And that’s a self-reported number. Almost half believed the primary purpose of the assessment was to help their client obtain the accommodations they had requested. That’s a description of advocacy, not assessment, and it came from the clinicians themselves.

I understand the impulse completely. The person in your office is struggling, they’re paying you money, the exam is brutal, extra time is easy to write, and nobody has ever been sued for being generous (I think). The cost of the reflexive prescription may be invisible to you and to your client in the moment. But it can surface three years later, in a letter from the state board of bar examiners telling your client that the accommodation they leaned on all through law school was likely never well supported — and that they won't have it on the Bar.

What This Means If You’re An Applicant

If you're preparing to request extended time on the Bar, the USMLE, the MCAT, or a Board exam, here is what I'd want you to walk in understanding.

Your struggle isn't in question, and nothing here says your difficulty isn't real. But extra time is built for one barrier: running out of clock. If what's hardest is recalling what you learned, or understanding it to begin with, or blanking when the stakes are high, extra minutes may not be the thing that helps — even though your difficulty is real.

"It helps me" isn't the evidence many applicant’s think it is. Roughly nine in ten people without any disability say the same thing, and the data suggest they're correct. A reviewer who accepted "it improved my performance" as proof of need would have to grant the accommodation to nearly everyone who applied.

Neither is "I've had it since high school" — at least, not by itself. To be fair to the full picture: the law does require testing agencies to give “substantial weight” to your history of receiving accommodations, and a documented track record genuinely matters. But substantial weight isn't controlling weight. A history of extended time tells a reviewer that someone once granted it — not that your impairment meets the standard now, on this exam. School-based accommodations are often granted under a lower bar than the one a licensing boards apply, which is why agencies often still ask the current question rather than inheriting an old answer. A past grant supported by documented clinical evidence is a strong case. A past grant instead of documented clinical evidence is the weakest one available. Why school accommodations don't automatically transfer is its own conversation, and one I've written about separately.

Ask what else might fit, because this is the question that actually serves you better. If your barrier is a medical condition that requires you to eat or check glucose, the accommodation that likely fits is stop-the-clock breaks, not a longer clock. If it's environmental, a different, less busy setting may be most indicated for you. A request that names a specific barrier and a specific matched remedy is dramatically more defensible than a request for more time stapled to a diagnosis - and it is often what you actually needed in the first place.

And if your evaluating clinician declines to recommend extended time, that isn't a betrayal. A well-built request for the right accommodation beats a poorly built request for the popular one, every time, and especially on appeal, where a denial already sits in your record.

Conclusion

I'll end by pointing at something bigger than any single report.

Nearly every problem in this article traces back to one design decision: high-stakes exams run on time limits that a meaningful fraction of perfectly competent examinees find constraining. We've then built an enormous clinical, legal, and administrative apparatus to adjudicate, one person at a time, who gets relief from a constraint that was somewhat arbitrary to begin with. There's a version of this world (i.e., the universal design version) where limits are set generously for everyone, or where speed is removed as a factor unless speed is the thing being measured, and where most individual extended time requests simply stop being necessary. That's a better long-term answer than anything psychologists can produce one evaluation at a time. Until testing bodies build it, we're stuck adjudicating.

But we can adjudicate well, and better. Extended time is real medicine, and like real medicine it deserves to be prescribed for a reason. Not because a diagnosis appeared. Not because a discrepancy showed up somewhere in a broad battery. Not because it's what the last twenty reports said. Because a documented, general-population-referenced impairment in a rate function collides with the specific demands of a specific exam - and because extra time is the narrowest intervention that levels the playing field for someone with documented functional impairment.

The applicants who genuinely need this accommodation to level the playing field are counting on the rest of us to be able to tell the difference. 

If you have questions about navigating an accommodations evaluation, documentation for a high-stakes exam, or a credentialing review, you can reach me at alex@lightsidepsych.com.

References & Additional Reading

Golan, S., et al. (2020). [As cited in the 2022 post-secondary extended assessment time review — verify primary source before publishing.]

Harrison, A. G., Pollock, B., & Holmes, A. (2022). Provision of extended assessment time in post-secondary settings: A review of the literature and proposed guidelines for practice. Psychological Injury and Law, 15(3), 295-306.

Harrison, A. G., & Sparks, R. (2022). Disability diagnoses: Seven sins of clinicians. Psychological Injury and Law.

Lewandowski, L. J., Cohen, J., & Lovett, B. J. (2013). Effects of extended time allotments on reading comprehension performance of college students with and without learning disabilities. Journal of Psychoeducational Assessment, 31(3), 326–336.

Lewandowski, L. J., Gathje, R. A., Lovett, B. J., & Gordon, M. (2013). Test-taking skills in college students with and without ADHD. Journal of Psychoeducational Assessment, 31(1), 41–52.

Lewandowski, L., Lambert, T. L., Lovett, B. J., Panahon, C. J., & Sytsma, M. R. (2014). College students' preferences for test accommodations. Canadian Journal of School Psychology, 29(2), 116-126.

Lewandowski, L. J., Lovett, B. J., Parolin, R., Gordon, M., & Codding, R. S. (2007). Extended time accommodations and the mathematics performance of students with and without ADHD. Journal of Psychoeducational Assessment, 25(1), 17–28.

Lewandowski, L. J., Lovett, B. J., & Rogers, C. L. (2008). Extended time as a testing accommodation for students with reading disabilities: Does a rising tide lift all ships? Journal of Psychoeducational Assessment, 26(4), 315–324.

Lovett, B. J. (2010). Extended time testing accommodations for students with disabilities: Answers to five fundamental questions. Review of Educational Research, 80(4), 611–638.

Lovett, B. J., Harrison, A. G., & Armstrong, I. T. (2022). Processing speed and timed academic skills in children with learning problems. Applied Neuropsychology: Child, 11(3), 320–327.

Lovett, B. J., & Leja, A. M. (2015). ADHD symptoms and benefit from extended time testing accommodations. Journal of Attention Disorders.

Lovett, B. J., Lewandowski, L. J., & Potts, H. E. (2017). Test-taking speed: Predictors and implications. Journal of Psychoeducational Assessment.

Ofiesh, N. S., Hughes, C., & Scott, S. S. (2004). Extended test time and postsecondary students with learning disabilities: A model for decision making. Learning Disabilities: Research and Practice, 19, 57–70.

Ofiesh, N. S., Mather, N., & Russell, A. (2005). Using speeded cognitive, reading, and academic measures to determine the need for extended test time among university students with learning disabilities. Journal of Psychoeducational Assessment, 23(1), 35–52.

Sireci, S. G., Scarpati, S. E., & Li, S. (2005). Test accommodations for students with disabilities: An analysis of the interaction hypothesis. Review of Educational Research, 75(4), 457–490.

Spenceley, L. M., & Wheeler, S. (2016). The use of extended time by college students with disabilities. Journal of Postsecondary Education and Disability, 29(2), 141–150.

Wei, X., & Zhang, S. (2024). Extended time accommodation and the academic, behavioral, and psychological outcomes of students with learning disabilities. Journal of Learning Disabilities, 57(4), 242-254.

Witmer, S. E., Lovett, B. J., & Buzick, H. M. (2023). Extended time accommodations on the 2017 NAEP Grade 8 mathematics test: Eligibility, use, and benefit. Journal of Psychoeducational Assessment, 41(2), 123–135.

Witmer, S. E., & Marinho, N. (2024). Extended time test accommodations: Does use correspond to score comparability for students with disabilities deemed in need? Psychology in the Schools, 61(11), 4175–4188.