Appearance is potentially important in academic settings: in-person/virtual job-talks for academic hiring are common across many fields; personal websites and videos summarizing research articles are becoming standard; speeches and seminars are a routine part of publicizing papers. Are there definitive norms of appearance in academia, or are they more like unspoken rules and are there differences across disciplines? How does the public’s understanding of norms compare against the views of academics? We plan to answer these questions by surveying three groups of respondents—faculty, university students, and lay adults—about their perception of academic job candidates they know to be hypothetical. In particular, we will ask them to rate pictured headshots on perceived competence, professionalism, and agreeability. This study will generate novel evidence on norms of appearance in academic settings.
This Pre-Analysis Plan is being filed concurrently with the launch of the lay adult and student sample surveys.
Intervention
Two different information intervention experiments will be conducted on respondents to see whether the interventions will affect how respondents' rating of subsequent headshots.
Experimental design
Not available.
Design details
Randomization method
Randomization will be done by a computer.
Randomization unit
Randomization will be done at the individual respondent level.
The visual stimuli shown to each survey respondents will be randomly selected from a set of photos by a computer as well.
Clusters
Approximately 500 university students
Approximately 2500 lay adults
Observations
Each respondent will rate 32 photos. A subset of the photos are 'extras' that do not involve experimental manipulation; these will be excluded from the analysis. On average, 14% of the photos a respondent sees will be these extra photos, and the effective sample size will be the number of experimentally manipulated photos seen by all respondents.
Arms
List experiment: control (50 percent) and treatment (50 percent)
Information intervention: control (1/3), generic treatment (1/3), main treatment (1/3)
Experimenter demand induction: no demand (50 percent), weak demand (25 percent), and strong demand (25 percent)
Primary outcomes
- Ratings of headshots on three main metrics: based on perceived competence, professionalism, and agreeableness.
- The list experiment in our study with a control and treatment group (randomized with even 50% chance) will also produce a set of primary outcomes that will help estimate the share of the population which misreports their opinion on the sensitive statement.
List experiment will also produce measures of the shares of respondents who agree with veiled and/or direct versions of a sensitive statement
Cited by
No citation data available for this registration yet.