Admissions Practices
Every physics graduate program must decide which applicants to let in and which to turn down. Unfortunately, the traditional admissions criteria that many departments use can unintentionally select against women and underrepresented minorities.
The good news is that it’s possible to select, in an unbiased way, students who will succeed and become productive scientists. To do so, faculty in charge of graduate physics admissions may need to rethink how these important decisions are made in their departments. Attributes such as a student’s ability to persevere when faced with adversity and a student’s ability to accurately self-assess strengths and weaknesses are important to success in a graduate physics program and should be considered alongside knowledge of physics and mathematics in admissions decisions.
Standardized measures such as the GRE do not attempt to measure these “non-cognitive” skills; moreover, evidence shows that the GRE and other standardized tests do not provide unbiased measures of student knowledge. And knowledge of physics is not synonymous with an ability to learn or to excel in research.
To help physics faculty efficiently evaluate applications and select students who can be successful and thrive, we present here some techniques used at APS Bridge Programs, whose retention rates are higher than the national average for physics graduate programs.
Effective Practices:
Develop processes that will allow the admissions committee to build internal consistency.
- Begin by having every committee member review a few common applications. Then meet to discuss and understand each other’s viewpoints. This allows committee members to decide on common admissions goals and measures.
- Prepare a rubric ahead of time for evaluating applicants, so that all evaluators rate applicants similarly.
- Allow sufficient time in the admissions timeline for multiple reviews of applications. This can help build departmental confidence in the selection process and catalyze new ideas for evaluating applicants.
- Create a written admissions policy, and make sure it is known to all committee members. Transparency promotes consistency and equitable treatment of applicants.
Ensure that your admissions process does not rely too heavily on GRE scores.
- Remove GRE scores during the initial review of applications, to avoid anchoring bias. Anchoring bias is the tendency to place too much weight on an initial piece of information (the anchor) when making subsequent decisions. A solution is to remove this information from the initial review and make it available only after short lists have been constructed.
- Do not use GRE scores to eliminate any students from consideration.
- Using a cutoff is specifically discouraged by the Educational Testing Service (ETS), the designer of this test.
- GRE cutoff scores can cause admissions committees to overlook many qualified applicants.
- Consider dropping the GRE as a requirement for application to your program.
- The APS Bridge Program has conducted interviews with many students who did not take the GRE. Reasons that these students cited for not taking the test include the cost of taking the test and sending scores to graduate programs, lack of time to study because of work obligations, and fear of poor performance due to insufficient undergraduate coursework. These students do not consider graduate schools that require the GRE. As a result, those schools receive fewer applicants, especially of lower socioeconomic status.
- Students also indicated that listing the GRE as “optional” is a sub-optimal solution, because this is often interpreted as, “We don’t require the GRE, but we won’t consider those who don’t provide this information.”
- If you decide to use GRE scores, read the ETS’s guidelines to ensure that you and other committee members understand how scores are designed to be used.
- Review literature on the limits of the GRE’s predictive power:
Screen applications using methods that go beyond traditional metrics.
- Use evaluative measures that consider non-cognitive aspects of applicant quality.
- Non-cognitive skills are non-academic attributes that are essential for student success. They can include perseverance in the face of adversity, accurate self-assessment of strengths and weaknesses, possession of long-range goals, creativity in problem solving, time management skills, and leadership abilities.
- Probe letters of reference for evaluations of non-academic performance. These indicators can help evaluate applicants’ potential for success in the laboratory and the classroom.
- Review applications for trends rather than averages.
- Search multiple components of each application for evidence of improvement over time. This can often provide information about a student’s non-cognitive skills, which are critical for future scientists and should play a prominent role in admissions decisions.
- When reviewing a transcript or GRE record, look for evidence that the student’s grades or scores improved over time.
- Examine letters of recommendation and personal statements for examples of dedication to improvement in the applicant’s professional life.
- Look for events in a student’s life that might have dealt temporary setbacks and evidence that the student overcame these obstacles.
Incorporate video interviews into the selection process.
- Keep logistics in mind:
- Use free video conferencing software such as Skype or Google Hangouts.
- Allot 20 to 30 minutes per applicant.
- Ensure that at least two admissions committee members are present for each interview. These faculty should compare their evaluations before presenting recommendations to the full committee.
- Limit interviews to a “short list” of candidates, to reduce the load on admissions committee members’ time. APS Bridge Program sites typically interview about two students for every available position.
- Use interviews to probe non-cognitive skills.
- Evaluation of non-cognitive skills allows for a broader understanding of an applicant’s path toward graduate education, including hurdles the student may have encountered. Suggested questions include:
- If you find that you are unable to solve a problem on the first few tries, what do you do?
- Tell me about a problem that you are really proud to have solved, and explain how you solved it.
- Evaluation of non-cognitive skills allows for a broader understanding of an applicant’s path toward graduate education, including hurdles the student may have encountered. Suggested questions include:
Make offers to students to fill specific departmental or research area needs.
- Mentoring students requires a sizable investment of departmental resources and time. If a student arrives only to find that they will not be able to pursue research in their field of interest, you risk losing that student.
- Students who are engaged with their research area early in their graduate studies and who are made to feel like valued members of the department are more likely to be retained.
Regularly discuss your department’s overall admissions strategy.
- Have regular conversations among faculty about your department’s admissions goals and how your strategies for evaluating applicants help meet those goals. This helps generate buy-in to the broader mission of educating students.
- Questions to discuss might include:
- Can your admissions process identify talented individuals who can succeed with the support you can offer?
- What does it mean to be talented?
- How are GRE scores, undergraduate GPA, and other measures used in admissions?
- Are faculty aware of issues associated with these measures that might bias selection?
- Can you accept students who are not yet ready for all of your graduate-level courses and must take a few undergraduate courses first?
- How will you accommodate these students?
- Are your admissions efforts linked to induction and matriculation programs that can integrate students into your program?
References:
Willingham, W. 1976. Validity and the Graduate Record Examinations Program. ETS GRE Board Report. 
Miller, C. and Stassun, K. 2014. A Test that Fails. Nature, 510, pp 303-304.
