Model 1: Waitlists Included Model
In the table below, you will find a list of variables that theoretically play a role in determining whether a law school accepts you or not. The dataset used in this model includes users on Law School Numbers who reported being either accepted, rejected, or waitlisted at the school in question (those who were first waitlisted and then accepted or rejected are treated as acceptances or rejections, respectively). For this model, I used an Ordered Logistic Regression, which allows you to look at how each of the independent variables affects the chances that you will be:
- Accepted rather than (combined waitlisted/rejected)
- (Combined accepted/waitlisted) rather than rejected
I know, I know...it's a little confusing. By way of example: in this model for the University of Pennsylvania, for two otherwise identical candidates, an additional point on the LSAT increases an applicant's chances of being accepted rather than either waitlisted or rejected by 31.9% (in other words, a 171 is 31.9% more likely than a 170 to get accepted rather than waitlisted or rejected, all else being equal). A 171 is also 31.9% more likely to get either accepted or waitlisted than rejected than is a 170, all else equal. If you have any questions, just e-mail me, or check out this link to the awesome UCLA stats site.
In any case, here are the results from this first model, in which I test the impact of LSAT score, GPA, each earlier month the application is sent (and by earlier month, I don't mean month earlier...I mean September vs. October, or October vs. November), binding early decision, URM status, non-traditional status, and female applicant status. Everything is based on LSN data from the 2003/2004 cycle to the present.
The first thing that jumps out at me is the comparatively tiny URM boost. URM candidates are only 3.3 times more likely to be accepted than otherwise identical non-URM candidates, which is pretty phenomenally low compared to most other schools. The other atypical result is that the GPA bump is larger than the LSAT bump, and this rarely the case. For each earlier month an applicant applies, he or she increases chances of acceptance by 25.7%, and if that applicant is a she, there is another 37.5% increase in chances of acceptance based on gender. Non-traditional applicants kind of get the shaft in this model, as they are only 70% as likely to be accepted as identical traditional applicants. Something else to take careful note of: there is a 122.6% increase for applying binding Early Decision. Something to take even more careful note of (but that is not reflected in this chart): that ED boost disappears for both splitter and reverse-splitter applicants. Non-splitter applicants actually enjoy a 134.9% increase in the likelihood of being admitted if they apply ED vs. RD, but there is no statistically significant advantage for ED application for splitters or non-splitters. Definitely food for thought, depending on your situation.
Model 2: Waitlists Excluded Model
The next model excludes waitlists, including only applicants that reported either being accepted or rejected (whether that was directly, or after first being waitlisted). The results in the table should be interpreted in the same way, but the interpretation is a little easier. The number given for each variable is simply the increase in likelihood of being accepted rather than rejected.
Here the URM boost grows a little, to 596.2%, but it is still pretty darn small, especially for a top school. The boost for female applicants also increases, as do the boosts for LSAT and GPA (and the LSAT boost is also larger, which is more in line with other schools, but the numbers are still much closer than they usually are). The advantage for each month earlier one applies increases a bit to 27.4% (although there is more going on here than meets the eye, which we will get to shortly). The ED bump here grows quite a bit as well, as ED applicants are about 2.5 times as likely to get in as identical RD applicants. Again, though, that ED boost applies only to applicants who do not fall into the splitter or reverse-splitter categories. The actual ED boost for non-splitters is 159.1%, and is completely statistically insignificant for splitters and reverse-splitters. It doesn't seem to make a bit of sense for the typical splitter or reverse-splitter to apply ED to Penn, based on these numbers. But wait! All hope is not lost! If we break down that early application advantage, this is what it looks like:
- Non-splitters: 24.6% more likely to be accepted than identical RD applicants
- Splitters: 84.8% more likely to be accepted than identical RD applicants
- Reverse-splitters: 118.0% more likely to be accepted than identical RD applicants
So, while applying ED might not give you any advantage if you are a splitter or reverse-splitter, applying early sure as hell does. ED application and earlier application are the two factors that we have been looking at that the applicant clearly has the most control over. While it is very abstract to say "URM candidate and identical non-URM candidate", the difference between "ED candidate and identical RD candidate" is essentially just the difference between the same applicant choosing one over the other, and the early application variable is just an applicant applying in January vs. December. You don't have to delve into the theoretical here...if you're applying, the "otherwise identical applicant" you are comparing yourself to is actually, uh...you.
Non-Splitters, Splitters, and Reverse-Splitters: Acceptance/WLs/Rejections, and Means
Last, I'm including a table that breaks down how non-splitters, splitters, and reverse-splitters are represented in the data. This one you really have to be careful with, because the data on LSN does skew towards higher-caliber applicants, and so acceptances are more highly represented than they are in the applicant pool. Really, the value of this kind of thing will become more clear when we can compare schools, because that same "higher-caliber" applicant caveat will apply across the board, so we can probably draw somewhat valid conclusions by comparing schools. For now, I'll include it for interested parties, but please do not look at this and say to yourself, "Self, as a splitter I have an X% chance of getting into Penn!"
These numbers are atypical in that splitters are less likely to be accepted than non-splitters (reverse splitters, as usual, come in last). The difference isn't great, but it's worth noting. LSAT doesn't seem to matter as much for splitters as GPA (which is typical), but it's worth noting that there isn't much of a difference in the GPAs of accepted and waitlisted applicants, and the dropoff for rejected applicants is pretty sharp. For reverse-splitters, the LSAT seems to matter, and there are relatively fine gradations between accepted, waitlisted, and rejected candidates. Otherwise, not much to report here.
ED vs. RD tables (non-URM and URM applicants)
In the final two tables (which are pretty busy, I realize), I have broken down results and means for non-URM and URM ED and RD applicants. These are just raw numbers.
This table is just aggregated data, but it does seem to contradict what we found in our regression analysis regarding applying ED as a splitter - the % of acceptance was much higher among splitter ED applicants than it was among splitter RD applicants. The LSAT is slightly higher for accepted ED applicants, although the GPA was considerably lower for that group. This table doesn't take into account the other factors, though, so I'd put a whole lot more stock in the regression analysis, which controls for other factors. Also, there's a great chance that a lot of those splitter ED applicants would have gotten in if they'd applied RD. The same thing could be said for reverse-splitter ED applicants (although you'd swap LSAT for GPA), but the same caveat applies. With the non-splitter ED applicants, though, you can pretty clearly see that they were getting accepted with significantly lower numbers than the RD applicants were, which is quite in line with what our earlier models reflect.
With URM candidates, the low sample size makes this even more dicey to interpret, although because it's Penn we're talking about, there is a little more data than usual. Still, it's not much to go on. The only category we might say anything about is the non-splitter category, and even then, the differences between accepted ED and RD candidates is not enormous, and definitely smaller than it was for non-URM students.
If you're a URM throwing an application Penn's way, you can expect an edge, but not nearly as big as you'd get at most top schools. If you're absolutely committed to Penn, regardless of your race or ethnicity, an ED application is not a bad idea, but if you're a splitter or non-splitter, it certainly seems that there's no reason to commit in such a way. Those applicants would be far better served by keeping their options open, but applying as early as possible to Penn, because that's where the real traction is in gaining an edge. Non-trads are out of luck, and female applicants get a nice little boost.