What Candidates Actually Think About Pre-Employment Tests

Most of what gets written about pre-employment testing is aimed at employers: which assessments to use, how to score them, how to defend them legally. Almost none of it asks the other person in the process what they think. That’s a gap worth closing, because the data that does exist paints a more specific picture than “candidates hate being tested.” Most have already been tested before. A majority actually prefer it to a resume-only process. And the frustration that does exist has a short, identifiable list of causes, most of which have nothing to do with the test itself.

This piece pulls together what recent, dedicated candidate-side research actually shows: where trust holds up, where it breaks down, and what that means for how an assessment should be built and communicated.

The Short Answer

Candidates are more receptive to skills testing than most hiring teams assume. Seventy-three percent of job seekers have already completed a skills assessment, and a majority say they prefer a hiring process that includes one over a resume-only process. The friction that exists concentrates in three specific places: undisclosed AI use, tests that run long without warning, and silence afterward, not in the basic idea of being tested.

What the Data Actually Shows

The research below comes from two dedicated 2026 surveys of job seekers themselves, not employers, plus platform-level data drawn from actual assessment completions. Read together, they show a candidate population that has normalized testing far more than most hiring teams assume.

Most Candidates Have Already Done This

According to Employ’s 2026 Job Seeker Nation report, a survey of more than 1,500 U.S. job seekers, 73% have completed a skills assessment as part of a hiring process. That figure covers a wide range of formats: timed skills tests, personality inventories, situational judgment scenarios, and video-based assessments all count toward it. Testing isn’t a novel or unusual ask anymore. For most candidates, it’s an expected part of applying, closer to a standard application field than a special hurdle.

That normalization matters strategically. A candidate encountering their first-ever assessment reacts differently than one who has done a dozen across different job searches, and the research suggests most candidates now fall into the second group. That shifts the real friction point away from “why am I being tested at all” and toward “is this particular test worth my time.” NACE’s own employer-side data reinforces the same trend from the other direction: 70% of employers now use some form of skills-based hiring for entry-level roles, up from 65% the year before, so the rise in tested candidates tracks closely with the rise in employers doing the testing.

A Majority Actually Prefer It

The same Job Seeker Nation report found that 79% of job seekers say assessments help employers identify qualified candidates, a notably positive number for a step that’s often assumed to be an annoyance at best. Separately, Test Gorilla’s State of Skills-Based Hiring research found that 54.3% of candidates prefer a hiring process that includes skills testing over one that doesn’t.

The reasoning behind that preference tends to center on fairness rather than convenience. A candidate who has watched a less-qualified but better-connected applicant get an offer has a specific reason to want a process that measures ability directly, rather than resume polish or interview charisma. A test doesn’t care who someone knows. For a candidate without an insider advantage, that isn’t a hurdle. It’s the first fair shot they’ve gotten in a while.

Realistic Assessments Build Real Confidence

Personnel psychologists have a specific term for what a well-designed assessment does for a candidate: a realistic job preview. The U.S. Office of Personnel Management, the federal government’s own HR agency, has long recognized realistic job previews as a way to give candidates an honest, balanced look at a role before they accept it, on the reasoning that a candidate who understands what the job actually involves makes a better-informed decision and is less likely to leave once the gap between expectation and reality shows up.

A structured work sample or job simulation functions as exactly this kind of preview. A generic aptitude test, abstract logic puzzles, spatial reasoning, tells a candidate nothing about the actual day-to-day work. A simulation built around real tasks, drafting a support ticket response, working through an actual dataset, writing a short piece of code, does double duty: it gives the employer a genuine performance sample, and it gives the candidate the same honest preview OPM’s own guidance calls for, before they’ve committed months of their life to the role. A candidate who completes a realistic simulation and still wants the job has, in effect, test-driven it first.

Where Candidate Trust Breaks Down

The same research that shows broad acceptance of testing also shows exactly where it curdles into distrust. Three issues account for most of it, and none of them are about the existence of testing itself.

AI Is the Biggest Trust Gap

The 2026 Job Seeker Insights Report from Resume Genius, based on 1,000 active U.S. job seekers, found that 22% of candidates are already using AI in real time during interviews, not just to polish a resume beforehand. Employ’s Job Seeker Nation research adds the other half of the picture: 34% of U.S. job seekers believe they’ve been automatically rejected by AI at some point, whether or not that’s actually what happened.

Put those two numbers next to each other and the picture gets more interesting than simple AI aversion. A meaningful share of candidates use AI to help them get through the process, while a larger share distrust AI being used to evaluate them. That’s less a contradiction than an asymmetry: candidates seem comfortable with AI as a tool they control and uncomfortable with it as a gatekeeper they can’t see inside. What consistently shows up as the actual complaint isn’t the technology, it’s the silence around it, not knowing whether a human ever reviews the application, what an algorithm scored, or why a candidate was screened out. Disclosure, not avoidance, appears to be what actually rebuilds trust here. A short, plain statement of what’s automated and what isn’t changes the emotional register of the entire process, even when the underlying screening logic doesn’t change at all.

Length Is the Silent Dealbreaker

Employ’s research found that 36% of candidates say the length of the hiring process has the single biggest impact on their overall experience, more than any other factor measured. Platform research covering roughly half a million tests found completion holding around 80% up to 40 minutes before dropping toward 60% beyond an hour, a threshold that holds up consistently across independent assessment platforms.

There’s a wrinkle worth understanding rather than just repeating the headline number. Separate research from Modern Hire, covering close to 30 million assessments, found that test length alone isn’t actually the strongest predictor of who abandons a test. Most candidates who quit do so in the first several minutes, often before they’ve gotten far enough in to judge whether the test is even long. That points to the real lever being less about trimming minutes off a test and more about what happens at the very start: does the candidate immediately understand what’s being asked of them and why, or are they staring at an unfamiliar interface with no sense of how much time this is going to take. A well-explained 40-minute test will often outperform a confusing, unexplained 15-minute one.

Test Anxiety Is Real, Even Where It Isn’t the Main Driver

Some candidates carry real, specific anxiety into any testing situation, often shaped by earlier experiences with school exams or irrelevant, generic aptitude tests that had little to do with an actual job. That anxiety is genuine and worth designing around, but the research above suggests it isn’t the dominant factor in how candidates experience testing overall. Relevance, transparency, and speed of response carry more of the weight than the format of the test itself.

Anxiety concentrates unevenly, though, and neurodivergent candidates, people with ADHD, autism, dyslexia, and related conditions, face a distinct set of barriers that a general anxiety framing understates. Practitioner guidance from organizations working directly on neuroinclusive hiring points to a consistent set of low-cost accommodations that make the biggest difference: explaining the format and duration of an assessment in advance rather than letting a candidate walk in blind, offering extra time on request, providing alternative formats such as written responses in place of verbal ones, and using plain, jargon-free instructions throughout. None of these change what the assessment measures. They change whether a candidate whose actual job-relevant skills are strong gets a fair opportunity to demonstrate that, rather than losing points to an unfamiliar interface or an unexpected time crunch. A visible, easy path to request an accommodation, rather than one buried in fine print, is itself part of what signals a fair process before the test even starts.

The Generational Gap Employers Underestimate

Candidate patience with a rough hiring process isn’t evenly distributed. Employ’s research found that 25% of Gen Z candidates have declined a job offer specifically because of a poor hiring experience, compared with just 6% of Baby Boomers, a four-times difference on the single most consequential outcome a bad process can produce: a qualified candidate walking away after the offer is already on the table.

Part of this gap is generational expectation-setting. Gen Z candidates have spent their entire adult lives interacting with consumer technology that’s fast, transparent, and responsive by default, so a hiring process that’s slow, opaque, and unresponsive reads to them as a signal about the employer, not just an inconvenient step. Part of it is structural too: younger candidates are more likely to be applying to several roles at once and have less financial cushion pushing them to tolerate a bad process out of necessity. For a hiring team building an assessment process meant to run for years, this isn’t a minor demographic footnote. It’s a preview of what every candidate pool will expect as the workforce continues to skew younger. A clunky, opaque process that an older candidate might tolerate is a live, near-term reason for a younger one to walk, and the generational mix of most applicant pools is only moving in one direction.

What This Means for How You Run Assessments

None of this argues against testing. It argues for running it in a way that matches what the data shows candidates actually respond to.

Say when AI is involved. A short line explaining that scoring is automated, a human reviews borderline cases, and whether there’s a path to request manual review, addresses the single biggest trust gap in the research directly. It costs nothing to implement and changes how the process is perceived, even when the underlying scoring logic doesn’t change at all.

State the time commitment upfront. “This takes about 20 minutes” before the test starts does more for completion rates than shortening the test itself, given that most abandonment happens in the first few minutes rather than scaling with total length. Pair the estimate with a one-line description of what the test covers so candidates aren’t guessing at both the clock and the content at once.

Make the test look like the job. A work sample or simulation tied to real tasks builds candidate confidence, not just employer confidence, in a way a generic aptitude test doesn’t. If the role involves handling a difficult customer, simulate that. If it involves writing code, have them write code. The closer the test mirrors the actual work, the more it functions as a preview both sides can use to judge fit.

Respond, even briefly. Silence after a completed assessment is one of the most consistently cited frustrations in candidate research, alongside application ghosting more broadly. A short, even automated, acknowledgment that confirms the assessment was received and gives a rough timeline for next steps closes a gap that costs very little to close, and measurably protects how the employer is remembered, win or lose.

Build in accommodation paths. A visible, easy way to request an adjustment, extended time, an alternative format, a different interface, reduces the anxiety and access barriers that keep some qualified candidates from finishing, or starting, an assessment at all. Making this an explicit, findable option rather than something a candidate has to ask about defensively signals fairness before the test even begins.

Frequently Asked Questions

Do candidates actually like pre-employment tests?

Most don’t dislike them outright. Research from Employ’s Job Seeker Nation 2026 report found 79% of candidates say assessments help employers identify qualified candidates, and TestGorilla found 54.3% actively prefer a hiring process that includes skills testing over one that doesn’t.

What’s the biggest complaint candidates have about assessments?

Undisclosed AI use and process length top the list. Roughly half of candidates in recent research see AI-led screening as unfair when it isn’t explained, and over a third say the length of the hiring process has the biggest impact on how they experience it.

Does using AI in screening make candidates trust the process less?

Undisclosed AI use does. Candidates are not uniformly opposed to AI in hiring, since a meaningful share already use AI tools themselves during assessments, but a large share view screening as unfair specifically when they aren’t told it’s happening.

How long should an assessment be before candidates start abandoning it?

Multiple independent studies point to a similar window: completion holds up well through roughly 30 to 40 minutes and drops off meaningfully beyond that. Stating the expected time upfront matters as much as the length itself.

Do candidates prefer skills tests over traditional interviews?

Preference splits by format rather than favoring one over the other outright. Realistic, Job-relevant assessments tend to build candidate confidence, while generic or unrelated tests are the ones candidates react against, regardless of whether an interview is also part of the process.

Conclusion

The candidate on the other side of an assessment isn’t the skeptic hiring teams often assume. Most have done this before, and a majority see real testing as fairer than a resume alone. The trust that does break down breaks down for specific, fixable reasons: not knowing when AI is involved, not knowing how long something will take, and not hearing back afterward. None of those require abandoning testing. They require running it the way the candidates being tested already say they’d respond to.