Assessment Pass-Mark Approval and Governance: Employer Decision Records

A company invites 100 applicants to complete a skills test. Before the test begins, the hiring manager announces that everyone must score at least 70% to proceed. The figure sounds reasonable. But why 70%? Would someone who scores 68% be unable to do the job? Would a person who scores 72% necessarily be ready?

Without a clear answer, the pass mark is only a number. A fairer approach starts with the work candidates must be able to perform, then uses evidence and informed judgement to decide what level of performance is good enough for the intended purpose.

This guide explains how employers can set and review pass marks for recruitment, promotion and employee assessments. It uses plain language and practical examples. It also explains when an assessment should inform a wider decision rather than act as an automatic pass-or-fail barrier.

What is an assessment pass mark?

A pass mark is the minimum score someone must achieve to meet a defined standard. It is sometimes called a cutoff score or passing standard.

For example, a test has 40 equally weighted questions. If an employer requires 28 correct answers, the pass mark is:

28 ÷ 40 × 100 = 70%.

The calculation is straightforward. The difficult question is whether 28 correct answers represent the knowledge or skills required for the role. A mathematically correct pass mark is not automatically a suitable one.

A passing standard may be used to decide who proceeds to an interview, who qualifies for a training programme, or who has demonstrated a particular level of skill. Those purposes are different, and each needs its own justification.

Start with the job, not the percentage

The employer should retain a signed job-relevance statement identifying essential tasks, risks and the evidence used to propose a minimum standard. The purpose of this guide is to govern the approval and subsequent use of that standard; it does not substitute for the technical standard-setting methods explained in the existing Talentrenza recruitment-assessment guidance. Record who proposed the threshold, who reviewed its consequences, who approved it and when it becomes effective.

1. What must the person be able to do? Identify the tasks that matter in the job.
2. What does acceptable performance look like? Describe what a person who is ready for the role can do without unreasonable supervision.
3. What does the assessment actually measure? Check that its questions or tasks represent the important requirements.

Imagine an employer recruiting a payroll assistant. The role involves reading payroll records, checking calculations, spotting unusual deductions and handling confidential information. A test containing only general arithmetic questions would not cover the whole job, regardless of the pass mark.

A sensible starting point is a simple job-skills table:

This table does not establish that a particular test is valid. It helps the employer see what evidence is needed before setting a standard.

Record the basis for a proposed passing standard

This article addresses governance of an approved pass mark, not a substitute tutorial on every standard-setting method. The designated subject-matter reviewers should document the job tasks, evidence supporting the proposed threshold, assumptions about test content, and uncertainty in borderline decisions. Where a formal standard-setting method is appropriate, obtain qualified technical review and preserve the underlying calculations. The separate Talentrenza guide on pre-employment pass marks explains the principal methods and worked examples; neither article establishes a universally valid cutoff.

Worked example: Setting a standard for a 40-question test

Talentrenza's commonly used assessment session format provides a useful example: 40 questions completed in 45 minutes. This format alone does not establish an appropriate pass mark, and the following numbers are fictional.

Assume an employer has developed a 40-question test for an accounts assistant role. After job review, the employer divides the questions into four areas:

A trained review panel examines all 40 questions and estimates the expected performance of someone who just meets the minimum job standard. Suppose its properly documented estimates add up to 27.2 expected correct answers.

The estimated percentage is:

27.2 ÷ 40 × 100 = 68%.

Because candidates cannot answer 0.2 of a question correctly, the employer needs a documented rule for turning the estimate into a usable whole-number cutoff. Choosing 28 correct answers would produce a 70% requirement. Choosing 27 would produce 67.5%. Neither choice should be automatic: the panel should consider the method, uncertainty, importance of the decision and any other essential skill requirements.

A pilot may also show that the 45-minute limit prevents capable candidates from finishing. In that case, the employer should investigate the timing and the questions before finalising the passing standard. Raising or lowering the cutoff cannot fix a poorly designed test.

Why a total score may not be enough

Suppose Candidate A answers 30 questions correctly, but gets only 2 of 8 workplace-judgement questions right. Candidate B answers 28 correctly, including 7 of 8 workplace-judgement questions.

The employer should not assume that Candidate A is necessarily the better choice. It depends on what the job requires and whether each part of the assessment has sufficient evidence to support a separate decision.

If a minimum score is required on an essential skill, it must be justified and approved before candidate results are reviewed. Small sections can produce unstable results, so separate section cutoffs need particular care.

Why a 70% pass mark is not automatically fair

Many employers choose 60%, 70% or 80% because those figures feel familiar from school examinations. Recruitment assessments are different: the question is not what percentage looks respectable, but whether the score distinguishes acceptable from unacceptable performance for a defined job purpose.

A 70% cutoff on an easy test may be too low. The same cutoff on a difficult test may exclude many capable candidates. A test containing ambiguous questions may not support any meaningful cutoff until those questions are fixed.

The difficulty and quality of the questions matter as much as the percentage.

Avoid setting the pass mark after seeing candidate results

A common mistake is to administer a test, examine the scores and then choose a cutoff that produces the desired number of candidates.

For example, a company needs to shortlist ten people. It may be tempting to call the tenth-highest score the pass mark. That is a ranking decision, not evidence that the tenth candidate meets a defined job standard.

Ranking candidates can sometimes form part of a selection process, but employers should be clear about the distinction. If the purpose is to identify people who meet a minimum standard, the standard should be established in advance.

Document any later change, explain why it was necessary, and assess whether affected candidates need consistent treatment or a new opportunity.

Pass marks for recruitment and promotion may differ

Recruitment and promotion involve different decisions. A new employee may receive training and supervision. A promotion candidate may be expected to make decisions independently from the first day.

That does not mean promotion tests should automatically have a higher percentage. Instead, the employer should review the responsibilities, evidence and acceptable level of performance for the new role.

For example:

A pass/fail result is not always the most useful outcome. Development assessments may be more valuable when they identify strengths and gaps without imposing a cutoff.

Review borderline outcomes without changing the rule

A result close to a cutoff should prompt careful interpretation, not an improvised change to the threshold after scores are known. Check the approved decision policy, the quality of the evidence and any relevant accommodations or incidents. For how measurement uncertainty affects borderline decisions, see Understanding Score Accuracy and Uncertainty; for combining a minimum standard with comparative ranking, see Job Standards Versus Candidate Comparisons.

Fairness, access and reasonable adjustments

A passing standard should measure job-relevant ability rather than avoidable barriers. Employers should review whether language complexity, inaccessible interfaces, poor internet connections or unsuitable timing affect candidates in ways unrelated to the intended skill.

A candidate who uses an approved accessibility adjustment should not automatically be judged against a different or lower job standard. Instead, consider whether the adjustment changes how the skill is demonstrated without changing the skill itself. This requires individual review and attention to applicable local requirements.

Where group outcomes differ unexpectedly, investigate the test design, administration and job relevance. A difference alone does not establish discrimination, but it should not be ignored.

How many reviewers should set a pass mark?

There is no universal number suitable for every assessment. The review group should include people who understand the job, the assessment and the consequences of the decision. A larger panel can offer more perspectives, but adding people without relevant expertise does not automatically improve the result.

For a high-stakes programme, obtain appropriately qualified assessment expertise. Ask reviewers to make initial judgements independently before discussing disagreements. Record the reasoning, not just the final number.

Reviewers should also understand the description of the minimally ready candidate. Otherwise, one reviewer may imagine a beginner while another imagines an experienced employee, producing incompatible estimates.

A simple employer process for approving a pass mark

Use the following sequence before a test goes live.

Step 1: Confirm the purpose. Record whether the assessment is for hiring, promotion or development.

Step 2: Describe the minimum standard. Write a short, job-based description of acceptable performance.

Step 3: Check the test content. Make sure the questions reflect the important skills and are free from known errors.

Step 4: Choose a review method. Use job-expert judgement, practical evidence and other suitable information according to the stakes.

Step 5: Document the proposed cutoff. Record the score, how it was calculated and any uncertainty or limitations.

Step 6: Check likely consequences. Examine whether the rule could unfairly exclude capable candidates or overlook essential skills.

Step 7: Approve before testing. Name the person authorised to approve the standard and any separate minimum requirements.

Step 8: Explain the rule clearly. Tell candidates what they need to know without exposing confidential questions or answer keys.

Step 9: Review results responsibly. Investigate unusual outcomes rather than automatically changing the cutoff.

Step 10: Revisit the standard after meaningful changes. A new job requirement, major question-bank revision or different test form may require a fresh review.

A practical pass-mark approval record

An employer can keep a short record with these fields:

The sample is a record-keeping template, not evidence that the example cutoff is suitable.

Common mistakes to avoid

Choosing a familiar percentage without evidence. A number that sounds reasonable may not match the difficulty of the test or the demands of the role.

Using the same cutoff for every job. Different roles and tests require different standards.

Confusing top scores with acceptable performance. Ranking candidates does not establish who meets a minimum standard.

Changing rules after viewing results. This can undermine consistency and confidence in the process.

Ignoring weak questions. A wrong answer key or ambiguous question can unfairly change who passes.

Assuming every test form is equally difficult. Different sets of questions may need evidence of comparability before scores or cutoffs are treated as interchangeable.

Using a single score as the whole hiring decision. Some important abilities may not be measured by the test.

Treating a borderline score as a precise measure of ability. Small score differences may not support strong conclusions.

How employers can use Talentrenza Assess responsibly

Talentrenza Assess can support structured skills assessment workflows where relevant features are available. Employers should choose assessments that match the role, communicate instructions clearly and decide how results will contribute to their selection or development process.

A standard 40-question session is a delivery format, not a universal passing standard. Employers should not assume that a 70% score is suitable across all skills, roles, candidate groups or test versions.

Where a role requires a high-stakes decision, employers may need specialist review of the assessment evidence and the proposed decision rule. Platform reporting should support informed judgement rather than replace it.

Frequently asked questions

Is 70% a good pass mark for a recruitment test?

It might be appropriate for a particular test and role, but there is no universal rule. The employer should explain why the chosen score reflects the minimum acceptable performance for the intended decision.

Can employers use 50% as a pass mark?

A 50% cutoff is not automatically too low or too high. Its suitability depends on the test's content, difficulty, scoring and purpose. It should be supported by evidence rather than convenience.

Should every candidate receive the same pass mark?

Candidates taking the same assessment for the same decision should generally be treated under consistent, pre-approved rules, subject to lawful and appropriate individual adjustments. Different test forms or purposes may require careful review of comparability.

Can we change a pass mark when too many people fail?

A high failure rate is a reason to investigate, not automatically lower the standard. Review question quality, instructions, timing, candidate population and the original justification. Any change should be documented and applied fairly.

What if a candidate misses the cutoff by one question?

A one-question difference should not be treated as perfect proof of a difference in job readiness. Consider the available evidence, uncertainty and any pre-approved review or retake procedure.

Should promotion assessments have higher pass marks than recruitment assessments?

Not automatically. The standard should reflect the duties and level of performance required for the specific role, not a blanket assumption that promotions require a higher percentage.

Can we set a pass mark using only the average candidate score?

An average describes how a group performed. It does not show the minimum level of skill needed for a job. It may help analyse results, but should not replace a job-based standard-setting process.

Do employers need an assessment specialist?

The need depends on the stakes and complexity. High-impact selection decisions, complex scoring, multiple test forms and claims about score precision often justify qualified specialist input.

Final employer checklist

Before using a pass mark, confirm that:

• The purpose and job requirements are documented.
• The assessment covers the important skills for that purpose.
• A minimally ready candidate has been described clearly.
• The proposed score has a recorded basis beyond a familiar percentage.
• Questions, answer keys and scoring rules have been reviewed.
• The consequences for borderline candidates have been considered.
• Accessibility and candidate conditions have been reviewed.
• The rule is approved before candidate scores are known.
• Reviewers know when a second source of evidence is needed.
• The standard will be revisited after significant changes.

Conclusion

A fair assessment pass mark begins with a simple question: What must someone be able to do to perform this job acceptably? Employers can then review the test, consult knowledgeable people, examine suitable evidence and approve a clear decision rule.

For the broader treatment, see the existing Talentrenza article: How to Set Pass Marks and Scoring Criteria for Pre-Employment Assessments.