NEW UPDATE: Completed trials may now upload and register supplementary documents (e.g. null results reports, populated pre-analysis plans, or post-trial results reports) in the Post Trial section under Reports, Papers, & Other Materials.
Social identity and career choices
Initial registration date
September 01, 2017
May 17, 2020 4:29 AM EDT
This section is unavailable to the public. Use the button below
to request access to this information.
London School of Economics
Other Primary Investigator(s)
Additional Trial Information
How does social identity affect career choices? Why do men not apply to jobs traditionally dominated by women? While there is a rich and growing literature on stereotypes and aspirations, there are a few studies investigating how counter-stereotypical messages in crucial moments of choice can actually induce more people to consider and even pursue a career which is uncommon for their group of origin.
I design an experiment in collaboration with a UK organization operating in a sector traditionally made of more than 75% of women. The experiment randomly assigns people interested in applying to the organization in different treatment groups, which manipulate the content of an invitation-to-apply email. The different treatments are meant to investigate the extent to which social identity matters for this career choice and disentangle some of the channels through which it affects decisions at different stages of the recruitment and selection process, for different demographic groups. The main outcome variables will be information gathering about the job, application submission and performance in the selection process, offer acceptance, performance on the job. I will also validate the mechanisms behind the intervention in an additional online experiment with a different sample. Registration Citation
Intervention Start Date
Intervention End Date
Primary Outcomes (end points)
Application submission, access to the application portal, information gathering about the job (event participation, website use, brochure collection), performance in the selection process (duration, scores), offer acceptance/rejection, performance in the program, beliefs about job effectiveness (survey measures).
Primary Outcomes (explanation)
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
The experiment is going to be conducted in two different stages of the recruitment process of the partner organization. The first stage will be across universities and will investigate what types of messages the employer can send to students to attract a more diverse pool of applicants and increase exposure to the job across demographic groups. Two different types of messages will be tested, which differ to the extent to which they represent stereotypical or counter-stereotypical examples of previously successful applicants. The second stage will target applicants and will manipulate perceived similarity with previously successful applicants and/or effectiveness on the job by providing information of the performance of previous cohorts (4 treatments).
An additional online survey will be implemented to validate and understand better what is the interpretation and emotional response to the different treatments used in these two stages of the recruitment process.
Experimental Design Details
Randomization programmed in the application software and done by the computer throughout the selection process period.
Two levels of randomization: university level and individual level randomization. The latter will be done at two different stages: participants to the organization's events and applicants.
Was the treatment clustered?
Sample size: planned number of clusters
Sample size: planned number of observations
The sample of applicants will comprise between 3000 and 5000 people.
The total number of universities in the treatment and control groups is 28 and 14 respectively. The sample of participants to events will be collected across both treatment and control universities and will comprise between 1500 and 2000 people (according to historical estimates).
Sample size (or number of clusters) by treatment arms
The sample of event participants is going to be assigned to three treatment arms: this means approximately 500 people for each arm, 500 in control group and 500 in each treatment.
Applicants are going to be assigned to five treatment arms. Assuming that the number of applicants will be approximately 3000, this means 600 people in each treatment arm: 600 control and 600 in each of the four treatment groups.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
INSTITUTIONAL REVIEW BOARDS (IRBs)
LSE Research Ethics
IRB Approval Date
IRB Approval Number
Ref # 000611