Adverse Impact & EEOC Compliance
What adverse impact means for an employer using our assessments, where the legal obligation sits, and what we do and do not measure.
Understanding Adverse Impact in Pre-Employment Testing
Adverse impact occurs when a selection procedure disproportionately screens out members of a protected group (race, sex, national origin, disability, or age) without being job-related and consistent with business necessity.
The EEOC's Uniform Guidelines on Employee Selection Procedures (UGESP) establish the four-fifths (80%) rule as the primary standard for assessing adverse impact. Our assessments are built to measure editorial skill directly rather than general aptitude, which is the design choice most relevant to that risk. We have not carried out a predictive-validity study, and we make no claim to one.
How Our Assessments Are Administered
These are the design and administration choices we actually make. They are stated as practices, not as evidence that the assessments are free of adverse impact — that would require a study we have not carried out:
Job-Related Content Only
Every test measures skills directly required for editorial, proofreading, and content roles. Questions are derived from real editorial tasks, not abstract cognitive measures that may create unnecessary barriers.
Job-Relevant Content
Test content is written to reflect the knowledge and skills editorial roles actually call for. We have not commissioned a formal content-validity study, and we do not claim one.
Standardised Administration
All candidates receive identical instructions, time limits, and question sets, so there is no interviewer variability between them.
Consistent, Rule-Based Scoring
Scores are computed against fixed answer keys, so the same submission always scores the same and no individual marker's judgement varies between candidates. That removes marker-to-marker variation; it is not the same as a demonstration that the test itself is free of bias, which would need a study we have not carried out.
Transparent Benchmarking
All scores are reported as percentiles against our 130,000+ candidate benchmark pool, so a pass mark can be set against the observed distribution of real candidate scores rather than an arbitrary number.
Enterprise Adverse Impact Monitoring
Proactive Compliance Monitoring
We do not collect demographic data from candidates, so we cannot break score distributions down by protected class and neither can anyone else from our records. What we can give you is the per-candidate outcome data your own analysis needs: the score, the percentile, the pass or fail against the mark you set, and the date. Run against the demographic data you already hold as the employer, that supports a four-fifths analysis of your own hiring — which is where the EEOC obligation sits.
Contact us to discuss what outcome data we can export for your account. Our position on bias auditing, and what it means for employers subject to NYC Local Law 144, is set out on our bias auditing page.
Validation Studies & Technical Documentation
If a regulator, internal audit or procurement review asks you for documentation, this is what we can and cannot supply. We hold no validation study; what we have is a description of how the assessments are built and scored, and your own candidates' data.
Available Documentation
- Assessment development methodology
- Item-level score data for your own candidates
- Your candidates' scores, exportable for your own analysis
- Data protection agreements
Validation Study Support
- What we can and cannot supply towards a validation study
- Role-specific analysis
- Expert consultation
- Answers to auditor questions about how scoring works
Ready to Strengthen Your Compliance Strategy?
Get expert guidance on adverse impact monitoring, validation studies, and compliance documentation tailored to your organisation's needs.