Does bias affect all minority groups the same way?
No, the type and severity of bias differ across minority groups. A 2025 study on psychopathy testing found that self-report measures 'overpredict personality psychopathology and related impairment in sexual minorities,' especially traits like detachment and disinhibition [1]. This means the tests systematically scored sexual minority individuals as more pathological than they actually are. However, the same study noted that for antisocial behaviors, the overprediction did not reach a 'clinically meaningful' threshold, so the bias was real but not always large enough to change a diagnosis [1]. This shows that bias can be subtle and measure-specific.
Is the bias only in the test questions, or is it also in how scores are used?
The bias is both in the tests themselves and in the broader system around them. A 2024 review of educational testing concluded that standardized tests are 'full of biases against economically disadvantaged students and students of color' and that these biases affect teacher attrition and long-term educational opportunities [3]. Additionally, a 2021 study of Black mathematics teachers found that experiences of 'anti-Black, racist microaggressions' accounted for most of the variance in teachers' thoughts of leaving the profession [5]. This means that even when test scores are 'objective,' the way they are interpreted and the environment in which they are used can perpetuate bias. A separate 2024 study on the 'minority extremity bias' found that people tend to rate minority groups as more extreme in their characteristics than majority groups, which can influence how test results are perceived [2].
What does this mean for fairness in education and hiring?
The evidence shows that standardized tests can systematically disadvantage minority groups, but the solution is not simply to discard all tests. Rather, the findings suggest that tests should be validated for each specific population and that scores should be interpreted with caution. For example, the psychopathy study [1] recommends that 'the risk of both testing and clinician bias should continue to be assessed in sexual minority persons and other diverse groups to ensure equitable evaluations.' Similarly, the review on educational testing [3] argues that teachers need training to help students 'overcome these biases.' In short, the tests themselves are not neutral—they reflect and can amplify existing inequalities, and addressing this requires both better test design and more equitable systems of interpretation.
Sources used in this answer
Psychopathy testing bias in sexual orientation minorities.
Psychopathy measures overpredict personality pathology in sexual minorities, especially detachment and disinhibition, but the overprediction of antisocial behaviors was not clinically meaningful.
The minority extremity bias
A series of studies (N=957) found a 'minority extremity bias' where people rate minority groups as more extreme in their characteristics than majority groups.
Overcoming Bias in Standardized Testing
A 2024 review concluded that standardized tests are 'inherently biased against students of color' and economically disadvantaged students, with dire long-term consequences for education.
We Won’t Go Back: Asian Americans and Racial Justice After Affirmative Action
The paper discusses how the 'model minority' myth about Asian Americans is used to argue against affirmative action, but notes that standardized tests like the SAT reflect systemic inequalities.
Exploring Racialized Factors to Understand Why Black Mathematics Teachers Consider Leaving the Profession
A survey of Black mathematics teachers found that experiences of microaggressions accounted for most of the variance in their thoughts of leaving the profession.