Guidelines for UCLA Astronomy Graduate Admissions

Goals

Although it may seem obvious to state the goals of our admissions process, it is worthwhile being explicit about the expectations from both the point of view of the department and the applicants, as this may influence how we rank applications from different candidates.

It is clear, from reading the personal statements of the applicants, that the vast majority enter into the process with the hope and expectation that graduate school will provide them with the skills and training to become professional scientists. Although only a fraction can reasonably expect to become tenure track academics, there are a wide range of related professions which require a similar skill set based on a general understanding of physics, mathematical and computational sophistication, and the experience of analyzing and solving open ended problems.

We have collected the data on the first permanent jobs our recent graduates accept in order to assess how often these desires are met. The results are shown here

The Astronomy department as a whole also has expectations. Recruiting and training the best students enables the research program of the faculty directly, and the success of our students after graduation has a positive feedback on the admissions process by providing concrete examples of success that enhances our recruiting of in demand applicants.

The departmental role (as a component of a public University) in serving the broader society must also be taken into account. Our admissions process must be mindful of the range of backgrounds and varying levels of preparation from which our applicants are drawn, and that every metric for evaluation of an applicants readiness for the program is incomplete and uncertain. As such, we must make an effort to evaluate everyone on an equal footing in regards of what we estimate their ceiling could be, rather than their starting point. Indeed, it is in our interest to find candidates whose potential may be undervalued by other programs, as we are not a standalone program and are usually competing with other schools for the best applicants.

Evaluation Process

The timeline for admissions is such that the committee starts to review applications in the first week of January. The usual timeline seeks to have a proposed list of candidates to present to the faculty in the first week of February. After receiving the approval of the Faculty, we contact the admitted students, with a visiting day in early March. The acceptance deadline for offers is April 15.

We receive of order 100 applications each year. As such, the initial review of applications is performed via a triage process, with each application being read and graded by three members of the five member committee. We provide no explicit requirements for how each committee member grades the applications. We recognize that all of the information in the application is relevant, but imperfect, and so we want to solicit a variety of individual perspectives in order to spur an informed discussion. However, we do offer some guidelines to try and ensure that everyone is working towards a common goal. We start the process off by asking each reviewer to refresh their memory/perspective regarding implicit bias. This is an unavoidable element of such a subjective process, since we all bring our individual perspectives to the review, but we can, at least, be mindful of the pitfalls.

We require all applicants to take the Physics GRE, but we accept unofficial reporting of scores until an offer of admission is made, in recognition of the expense incurred for official score reports. This has become a politically sensitive topic of late and has been discussed at some length amongst the faculty. The consensus position is that the PGRE is not an accurate predictor of success in graduate school, but is also not completely devoid of information about an applicants preparedness. As such, there is no threshold score required for admission, but it is considered as an element in the admission process.

The initial grading scheme asks the committee members to grade each application on a 5 point scale to assess the likelihood that the student can succeed in our program. This is only the first step, and is designed to select a manageable subset for a more in-depth discussion. This is done by selecting those applications that received an average grade of 4 or better. In addition, students from underrepresented groups who received a grade of 4 or better from at least one reviewer are also added to the mix if their other grades were 3. This is to be mindful of the possibility of biases in our reviews of applicants from nontraditional backgrounds. Finally, applications with a large dispersion in grades are added to the discussion so that additional readers can offer a normalizing perspective. This process usually yields a subset of 30-40 applications for further discussion. At this point, each reviewer reads the applications from this subset that were not included in their triage section. Thus, all applications in subsequent parts of the review are read by the entire committee.

We start the discussion with a ranked list based on a simple average of the above grading scheme, but now including the grade from the additional reviewers. The generic outcome of this is that we usually have of order 10 applications that received high grades (5 or almost 5) from all the committee members. This forms the core of the admissions group, since there is little dispute that these applications are outstanding. Our usual target number of admissions offers is between 15-20, so most of the subsequent discussion centers around which of the remaining subset should receive an offer.
At this point in the process, it becomes a matter of individual weighting of different elements of the application. The crux of this process is finding the reason or argument why a particular candidate should be preferred to the others in this cohort. Much of the discussion of graduate admissions procedures in the literature seems to focus on metrics used to rule candidates out of contention (and usually why such metrics are bad). However, in our experience, it is more relevant to find the elements that rule a candidate ``in'', rather than ``out''. This is one reason why we resist the currently fashionable movement to do away with the Physics GRE score.

The argument against the PGRE is that some students who perform poorly on this exam may nevertheless make excellent researchers for other reasons, such as the ability to overcome obstacles. There are indeed many cases of this, but it presupposes that such applicants have received the opportunity to demonstrate these skills through exposure to research programs. There are also many students, who attend smaller schools with limited research facilities, who simply do not receive these opportunities and whose only opportunity to stand out is through an excellent performance on a standardized test. As such, we believe our approach offers us the opportunity to find both kinds of students and maximizes our chances to make our graduate program the strongest it can be.

The kinds of factors that can elevate one application over another are varied, and at the discretion of the committee. Ideally this should be based on evidence from the transcript. The committee is asked to avoid such arguments as ``having a sense'' or ``a good feeling'' about a particular candidate. This is how implicit biases creep in. A common type of good argument is whether an application is particularly well suited to this department. One example is if the student already has instrumentation experience and skills and wants to work in the IR lab. Another is if the student is already aligned with a particular faculty member or research program, such as the Galactic Center group.

The intended outcome of this discussion phase is the definition of the subgroup of applicants to receive an offer of admission to the graduate program. An element of this is a projection of the desired number of incoming graduate students. This is dictated by the current state of the department. In the fall the admissions chair emails the faculty to enquire how many new students each faculty is likely to be able to advise in the near future. We evaluate this information in consultation with the Astronomy Vice chair regarding the availability of both office space and department funding to determine a target number. We then combine the number of offers made with the expected yield of acceptances to meet the target number. We base our projection on the past history of success.

Once the committee and the vice-chair have agreed on the admissions list, the grad admissions chair makes a presentation to the faculty, seeking approval of the procedures and outcome of the process, and clearance to make the admissions offers. Once we have received approval, the committee chair and designated substitutes contact the people admitted to inform them of their admission and encourage them to attend the visiting day. The official offer letter, containing the detailed information about finances and other official data, usually goes out within a week after the initial contact.