Back to History

Fields Changed

Registration

Field Before After
Trial Status in_development completed
Last Published March 23, 2026 07:19 AM March 31, 2026 05:40 AM
Primary Outcomes (Explanation) Photographs will be considered ethnically ambiguous when the mean perceived probability of minority origin is approximately equal to the midpoint (5) of the 0–10 scale. Candidate photos will be selected if they are statistically equivalent to this midpoint within a predefined equivalence margin of ±0.5 scale points. Equivalence tests will be conducted within each photo category to ensure that the selected photographs do not differ meaningfully in perceived ethnic origin. Photographs are selected based on proximity to the midpoint (5) on the 0–10 scale. Any remaining differences between photographs are accounted for in the main factorial survey experiment by including photograph fixed effects in the statistical analyses.
Intervention (Hidden) The photo pretest serves to validate the stimuli used in a factorial vignette experiment on hiring discrimination related to religious identity. A total of 48 candidate photographs are pretested. Respondents (HR professionals) each evaluate 12 randomly selected photographs. For each photo, they assess the perceived probability that the person is of a given ethnic origin (Maghrebi/Eastern European vs West European) on a 0–10 Likert scale, as well as perceived attractiveness. Photographs will be selected for the main experiment if they meet two criteria: - Ethnic ambiguity: the perceived probability of minority origin is approximately 5 on the 0–10 scale, indicating equal perceived likelihood of majority and minority origin. - Equivalence across selected photos: photographs within each category will be tested for equivalence using equivalence testing with a tolerance margin of 0.5 scale points. At least four photographs per category will be selected. The attractiveness scores will be retained and used as control variables in the main factorial survey experiment. The photo pretest serves to validate the stimuli used in a factorial vignette experiment on hiring discrimination related to religious identity. A total of 48 candidate photographs are pretested. Respondents (HR professionals) each evaluate 12 randomly selected photographs. For each photo, they assess the perceived probability that the person is of a given ethnic origin (Maghrebi/Eastern European vs West European) on a 0–10 Likert scale, as well as perceived attractiveness. Because it proved infeasible to identify a sufficient number of photographs within each category that were statistically equivalent within a ±0.5 margin around the midpoint of the ethnic ambiguity scale, the final selection procedure prioritizes photographs whose mean perceived ethnic origin is as close as possible to the midpoint of 5. The number of photographs selected per category (female Eastern European ambiguous, male Eastern European ambiguous, female Maghrebi ambiguous, male Maghrebi ambiguous) is determined by the requirement that each experimental deck in the main factorial survey contains only unique photographs, thereby increasing realism for respondents. In the main experiment, photograph fixed effects will be included in all regression analyses to control for all photograph characteristics, including perceived ethnicity and physical attractiveness.
Secondary Outcomes (Explanation) For each photograph, the mean attractiveness score will be calculated across respondents. These scores will not determine photo inclusion, but will be retained for use as a control variable in the main factorial survey experiment to account for potential attractiveness effects on hiring evaluations. For each photograph, the mean attractiveness score will be calculated across respondents.
Pi as first author No Yes
Back to top