The predictive validity of behavioral assessments in hiring
Structured, job-specific methods predict performance about twice as well as unstructured interviews.
Research, practical insight and best practice for leaders who want to hire with more certainty and fewer regrets.
Résumés and gut-feel interviews feel like evidence, but a century of research ranks them among the weakest predictors of performance we have. Behavior hiring is a structured alternative that compares candidates to the role instead of to each other, and it predicts performance far better than résumé screening and conversational interviews.
Behavior Hiring brings together what works and turns it into tools your team can use on the next hire.
Behavior hiring is a structured approach to talent selection that evaluates candidates based on observable, job-relevant behaviors — what people actually do at work — rather than credentials, gut instinct, or unstructured interviews. It draws on decades of organizational psychology research showing that how people behave at work predicts performance far better than what is written about them.
Traditional hiring methods — résumé screening, credential checks, and unstructured interviews — have a well-documented problem: they are poor predictors of actual job performance. Research consistently shows that what candidates say about themselves in interviews correlates weakly with how they perform once hired. The result is costly mis-hires, high early attrition, and teams that underperform their potential.
Behavior hiring replaces subjective impressions with structured data. Instead of asking candidates what they would do in a hypothetical situation, it asks what they have actually done, and uses validated assessment tools to measure the behaviors most predictive of success in a specific role. The process is systematic, repeatable, and auditable.
Résumés were never a good proxy for performance. In the research, years of job experience barely predicts how well someone will do (a validity of .07). Now the résumé is getting even weaker. AI tools can write a polished, keyword-matched résumé and cover letter for every posting in seconds, so every application looks strong and the differences between candidates disappear.
The same is happening in interviews. Candidates can rehearse answers with AI coaches, which makes polished responses to familiar questions cheap and common.
What AI can't easily fake is a consistent pattern of how someone works, verified against the specific behaviors a role needs and tested with real examples in a structured interview. When the paper signal gets noisier, behavior becomes the signal worth trusting.
The evidence base for behavior hiring spans decades of industrial-organizational psychology research. Meta-analyses consistently find that structured, behavior-based selection methods outperform unstructured alternatives in predictive validity, meaning they are better at identifying candidates who will actually succeed in the role. This site exists to make that research accessible to practitioners.
Structured, job-specific methods predict performance about twice as well as unstructured interviews.
Structured interviews average .42 validity; unstructured interviews average .19.
Structured interviews have the highest validity of any method studied and among the smallest subgroup differences.
Methods that ask candidates to describe themselves degrade under stakes; methods that ask for verifiable accounts degrade much less.
Perceived fit and social acceptance relate to early attrition about twice as strongly as task mastery.
Acquired companies lose about 21% of their executives a year, and the elevated rate persists for at least a decade.
Inside the PATH assessment and GoodJob's research-based behavior hiring process: how it measures behavior from role definition to onboarding, and how to apply each step yourself.
One management style fits few people. But what does the research say about tailoring your approach, and how to use behavior profiles without boxing people in?
Most onboarding treats new hires as interchangeable. The research points to a better design: a consistent backbone, tailored to how each person actually works.
The research only helps if it changes what happens in the room. These tools are built to make that easy.
A template for defining what a role must achieve and how someone will need to work to achieve it, before anyone sees a résumé.
How to design and run a structured behavioral interview, calibrate interviewers, and avoid the errors that pull a good process back toward a conversation.
A scorecard built so that in the room you only pick a number and write one line, because the criteria were agreed before anyone interviewed.
Twelve questions to score your current hiring process honestly, what each range of answers means, and the three changes worth making first.
Behavioral hiring is a structured approach to selection that defines the work behaviors a role needs, assesses candidates against them, and verifies the results in structured interviews. It compares candidates to the role instead of to each other.
In the 2022 Sackett et al. re-analysis of selection research, structured interviews had the highest average validity (.42), followed by job knowledge tests (.40) and empirically keyed biodata (.38).
They are among the weaker predictors, with an average validity of .19. Adding structure, meaning the same job-related questions and a shared rating scale, roughly doubles predictive value.
A personality test describes general traits. A behavioral hiring assessment measures how someone prefers to work and compares it to a defined profile of the role, then gets verified in an interview.
They are useful for self-awareness and team development, but they were not designed to predict performance in a specific role. MBTI's publisher advises against using it for selection.
PATH is GoodJob's behavioral self-assessment. It measures work-context behaviors such as motivation, problem solving, collaboration and work organization, and is used alongside a RoleDNA profile and structured interviews.
Only weakly. Sackett et al. (2022) estimate the validity of years of job experience at .07, among the lowest of any factor studied.
Start with one role. Define the behaviors that separate strong performers, replace open conversation with structured behavioral questions, and rate every answer against the same scale.
One email a month. New research, a practical tool, and what changed in the field.