New German Scale Reveals Hidden Prejudice Against Disabled People in Medical Training

Prejudice against disabled people often hides in plain sight. It rarely announces itself through open hostility; instead, it surfaces in polite refusals, well-meaning paternalism, and quiet discomfort. Researchers call this phenomenon symbolic ableism, and a team led by Anna-Lena Hillebrecht of the University of Freiburg has now built the first German-language instrument designed to measure it specifically in health professions education. The work, published in BMC Medical Education, describes the development and initial psychometric evaluation of the German Symbolic Ableism Scale, or SAS-G, a brief questionnaire that could eventually help medical and dental schools identify and address attitudes that quietly undermine equitable care.

The stakes are far from abstract. Ableist attitudes among future physicians and dentists can shape clinical decisions, communication, and even whether disabled patients feel safe seeking treatment at all. They also influence how learners and professionals with disabilities are treated within the health system itself, affecting who gets admitted, who gets supported, and who quietly leaves the profession. The United Nations Convention on the Rights of Persons with Disabilities obliges signatory states, including Germany, to dismantle such barriers, yet until now there has been no brief, context-sensitive German tool to quantify these subtle attitudes in the very institutions that train the country’s clinicians.

What makes the new study distinctive is its methodology. Conducted between 2022 and 2024, the three-phase instrument-development project placed people with disabilities at the center of the process from the outset. Rather than treating disabled people as subjects to be measured, the researchers involved them directly as experts by experience in interpreting the construct, generating new items, and reviewing content alongside specialists in disability studies and clinical practice. This participatory approach reflects a growing consensus in disability studies that instruments about disabled people should not be built without them, and it likely explains why the final scale captures dimensions of ableism that purely academic item-writing might have missed.

The technical groundwork began with the original English Symbolic Ableism Scale, which the team subjected to forward-backward translation into German. But translation alone was not enough. The researchers complemented the translated items with de novo items generated through the participatory process, then used qualitative content analysis and content validity ratios to refine the pool. Content validity ratios quantify how many experts judge each item as essential, giving the team a numerical handle on which questions genuinely belonged. By the end of Phase 1, a 25-item pool was ready for field testing, blending classic items on denial of rights and interpersonal comfort with new material grounded in lived experience.

Phase 2 put that pool to the test in a general-population pilot sample of 210 participants. Item-level statistics from this pilot guided both conceptual and linguistic redevelopment, trimming and reworking the instrument into a 21-item pre-final version tailored for students. This iterative winnowing is standard practice in scale construction, but the Freiburg team added an unusually rigorous statistical apparatus in the final phase, in which 173 medical and dental students completed the pre-final version. Deciding how many factors underlie a set of questionnaire items is one of the most contested problems in psychometrics, and the researchers attacked it from several directions at once.

Factor retention was guided by parallel analysis, a technique that compares observed eigenvalues against those generated from random data, using both Pearson and polychoric correlations. The team also applied Velicer’s minimum average partial criterion, which identifies the point at which removing further components no longer meaningfully reduces the correlations among remaining items. A primary exploratory factor analysis using maximum-likelihood extraction with oblimin rotation, which allows the resulting factors to correlate, was then supplemented by a sensitivity analysis designed for ordinal data, employing polychoric correlations and minimum residual extraction. The convergence of these methods on the same solution strengthens confidence that the two-factor structure is not a statistical artifact.

The final scale retained just ten items organized into two complementary factors. The first, Denial of Rights and Structural Barriers, comprises six items and showed strong internal consistency, with Cronbach’s alpha of 0.81 and McDonald’s omega of 0.83. The second, Interpersonal Comfort, comprises four items with alpha of 0.70 and omega of 0.74, values acceptable for a brief subscale. The unrotated maximum-likelihood extraction accounted for 44.3 percent of total item variance, and critically, the ordinal-data sensitivity analysis reproduced every primary item-to-factor assignment, suggesting the structure is robust to the choice of statistical treatment. The two dimensions echo a classic insight from prejudice research: someone can endorse abstract principles of equality while still feeling uneasy in personal interactions with members of a stigmatized group.

Exploratory regression analyses offered intriguing, if preliminary, correlates. Lower scores on Denial of Rights and Structural Barriers were associated with female gender, enrollment in medicine rather than dentistry, and friendship with a person with disabilities. Lower Interpersonal Comfort scores were associated with medical study and with such friendship. The friendship finding is particularly interesting, because sustained personal contact has long been identified in social psychology as one of the most reliable reducers of prejudice, and the pattern here is consistent with that literature. The authors are careful, however, to frame these regressions as hypothesis-generating rather than confirmatory, given the modest sample size and single-site design.

Indeed, the team is unusually candid about the limits of the evidence. The findings provide initial exploratory support for the scale’s structure and internal consistency in a single-site student sample, but independent multisite confirmation is required before the SAS-G can be deployed to evaluate interventions or compare groups or countries. Evidence of temporal stability, meaning that scores remain consistent over time in the absence of real change, is still lacking, as are data on responsiveness to change and measurement invariance, the property that a scale measures the same construct across different populations. These are exactly the psychometric milestones that separate a promising research instrument from one suitable for high-stakes program evaluation.

Even so, the arrival of the SAS-G marks a meaningful step for German-speaking health professions education. A ten-item questionnaire that can be completed in minutes gives educators, researchers, and curriculum designers a practical tool to benchmark attitudes, track the effects of disability-awareness training, and open honest conversations about how ableism operates in clinical culture. Its participatory pedigree may prove as important as its statistics: by building the scale with disabled people rather than merely about them, the Freiburg team has modeled the kind of inclusion the instrument is designed to promote. As medical and dental faculties across Germany and beyond grapple with their obligations under the CRPD, the SAS-G offers a way to measure how far there is still to go, and whether the steps being taken are actually working.

Subject of Research: Development and psychometric evaluation of a German-language scale measuring symbolic ableism in health professions education

Article Title: Symbolic ableism in health professions education: development and initial psychometric evaluation of the German symbolic ableism scale

Article References: Hillebrecht, A.-L., Hausdorf, D., Warth, R. V. D., Jeung-Maarse, H., Fabry, G., Linde, P., Auerbacher, M., Spies, B. C., & Reissmann, D. R. (2026). Symbolic ableism in health professions education: development and initial psychometric evaluation of the German symbolic ableism scale. BMC Medical Education. https://doi.org/10.1186/s12909-026-10636-6

Image Credits: AI Generated

DOI: 10.1186/s12909-026-10636-6

Keywords: symbolic ableism, health professions education, psychometrics, scale development, disability studies, medical students, dental students, exploratory factor analysis, participatory research, prejudice measurement, Germany, questionnaire validation