Law School Discussion

Nine Years of Discussion
;

Author Topic: Admissions calculators  (Read 280273 times)

RocketBot

  • Guest
Re: Admissions calculators
« Reply #40 on: October 27, 2005, 02:07:22 AM »
What about when they look like a 'star' does that mean $$$? more than one app with the same stats?

I think it means they were accepted and are attending.  A lot of people don't mark the ones they are attending tho, hence there not being many stars.

SanchoPanzo

  • Sr. Citizen
  • ****
  • Posts: 2098
  • My avatar is hotter than your avatar
    • View Profile
Re: Admissions calculators
« Reply #41 on: October 28, 2005, 10:35:05 PM »
I think the LSAC's indicator might be pretty good. It came out about right for all my friend's decisions when he applied last year. That's pretty anecdotal though.

Is you friend a white male with average scores and average soft factors?
1L
order in the midst of chaos surrounded by order and a sprinkling of mist with chaos flavor. bake 2 hours then serve

JCD

  • Full Member
  • ***
  • Posts: 42
    • View Profile
Re: Admissions calculators
« Reply #42 on: October 29, 2005, 05:22:42 PM »
I personally think the calculators are ok to use as a rough guide to decide where to apply, but they don't predict the your chances with any accuracy, particularly at the low end. For example, if it reports a 50% chance of admission, your chances of admission are not 50-50. Your chances of admission are probably more like 10% or 90%. Basically, if your soft factors put you in the top half of people with similar numbers, you're very likely to be admitted. Conversely, if your soft factors put you in the bottom half of people with similar numbers, your chances of admission are very slim.

Something like Chiashu is not predicting your chances of admission. It is telling you how many people(with similar numbers) you need to beat to be accepted. The problem is compounded by the fact that the number these calculators spit out is innaccurate due to old data and URMs.

As a rough guide, if it puts your probability at 60% or below, subtract 20% from the number. If you fall between 60%-75%, subtract 10%. I'd also suggest subtracting 15% from your Harvard and NYU probabilities. Once you start getting up above 80%, you will almost assuredly be admitted as long as you don't screw up your application.

Example: Chiashu says my chances at harvard are 65%. I make the suggested adjustments and the new figure is 40%. This means that if my soft factors/application quality fall within the top 40% of people with my numbers, I'll probably be admitted.

Obviously this method is very rough and arbitrary, but I think it's closer to the truth than chiashu numbers.

LawyersGunnersnMoney

  • Sr. Citizen
  • ****
  • Posts: 964
    • View Profile
Re: Admissions calculators
« Reply #43 on: October 29, 2005, 05:32:15 PM »
I don't get what your saying.  How can you reasonably compare your soft factors unless they are extremely good or bad?  Also, why would %40 and good soft factors mean 'probable admission?'

JCD

  • Full Member
  • ***
  • Posts: 42
    • View Profile
Re: Admissions calculators
« Reply #44 on: October 29, 2005, 10:06:10 PM »
You're attempting to attribute distinctions to me that I did not make. I will not be comparing my soft factors to other candidates, that is up to the admissions council. I am simply pointing out that an admissions calculator often does not provide your chances of admission...rather it simply tells you what percentage of people in your index range you need to be ahead of to get admitted.

The logic is quite simple. If I know that Harvard admits roughly 40% of the non-URM applicants with similar indexes, I know that I will probably be admitted as long as the nonquantative aspects of my application are superior to the 60% that don't get admitted. There is no way to know what will happen until I apply, but this method is a lot more realistic than looking at the .65 chiashu spits out and blindly assuming I have a 2 out of 3 chance of getting into Harvard.

Given that I'm well aware my nonquantative aspects probably fall into the bottom half, I know that I probably won't get in despite the predictions of overly optimistic admissions calculators.

eatdrinkbemerry

  • Sr. Citizen
  • ****
  • Posts: 164
    • View Profile
Re: Admissions calculators
« Reply #45 on: October 31, 2005, 08:00:40 PM »
www.lawschoolstats.com


how accurate is this tool? Broad question I know, but I just wanted to get some feeback of how people felt about this.

SkullTatt

  • Guest
Re: Admissions calculators
« Reply #46 on: October 31, 2005, 08:25:47 PM »
It basically takes each school's index formula and then plugs your numbers into it, and shows how you compare to the school's 25th and 75th percentiles. This is slightly more useful than just comparing your own numbers visually to a school's 25th and 75th percentile since your numbers are being weighted according to how the school values them.

You can basically do the same thing using LSAC's calculator, however many schools boycott that calculator, but they do publish their indexes.

moonpie

  • Sr. Citizen
  • ****
  • Posts: 391
  • No it's not mexico
    • View Profile
Re: Admissions calculators
« Reply #47 on: November 01, 2005, 04:11:03 AM »
It would be nice if schools disaggregated their URM and non-URM admissions indices, but that would send a catastrophically incorrect message about what aa is intended to do.

SCgrad

  • Guest
Re: Admissions calculators
« Reply #48 on: November 01, 2005, 05:00:33 AM »


The logic is quite simple. If I know that Harvard admits roughly 40% of the non-URM applicants with similar indexes, I know that I will probably be admitted as long as the nonquantative aspects of my application are superior to the 60% that don't get admitted. There is no way to know what will happen until I apply, but this method is a lot more realistic than looking at the .65 chiashu spits out and blindly assuming I have a 2 out of 3 chance of getting into Harvard.



You are assuming that schools pick a percentage of applicants with a certain index and admit that number without fail.  You are also assuming they hand this info out in the form of past results. ???  Your method is as good as any, but really doesn't cut it, nor does anyone elseís' haphazard attempt to judge their chances.  Here is how you determine if you will get in.  If there are fewer seats available then students who A) are desired more by the school than you; and B) who end up desiring to go to this school, your admissions chances are zero.  If not, they are 1.  Iím sure someone canít poke a hole straight through this, but save your time.  These calculators should be used as very general guides.  If every student with your numbers got into a school the year before and you apply to this school alone and get rejected, what are you going to do?  Sue them?  If you apply to several schools that are within your LSAT and GPA ranges, assuming you donít have some retched quality that permeates your app like a giant metaphoric turd, you will get into at least one law school and all these numbers wonít mean *&^%(pun intended). ;)


JCD

  • Full Member
  • ***
  • Posts: 42
    • View Profile
Re: Admissions calculators
« Reply #49 on: November 01, 2005, 06:04:29 AM »
www.lawschoolstats.com


how accurate is this tool? Broad question I know, but I just wanted to get some feeback of how people felt about this.

Good for figuring out where you may want to apply, terrible for predicting actual chances at admission. For example, my index falls right at 50% for Yale, yet looking at Yale's admission grid shows that people with my index usually get in 20-25% of the time. Use it to see if you'll be dinged on numbers alone, not to predict your chances.


Quote
You are assuming that schools pick a percentage of applicants with a certain index and admit that number without fail. You are also assuming they hand this info out in the form of past results.

Your first statement would be true if my goal was to create a 100% accurate predictor. Unfortunately that is impossible, so I did not attempt it. I concede that it is an accurate criticism, but one that is only relevant to a few applicants at the margin (ie: the choice may come down to a 175/3.8 vs 171/3.7). In that case, a fluctuation in the percentages of indexes admitted would be meaningful to someone obviously, yet most would not be affected. This information is 'handed out in the form of past results'. You can see it in an admissions grid, or in the data plots on LSN(the latter is obviously very informal). IE: If Yale admits roughly 50% from the highest index grid section and 25% from the second highest, a fluctuation to 51%/24% only affects a few people. I doubt the percentages go through vast fluctuations from year to year, although I concede the possibility.

Quote
Here is how you determine if you will get in.  If there are fewer seats available then students who A) are desired more by the school than you; and B) who end up desiring to go to this school, your admissions chances are zero.  If not, they are 1.  Iím sure someone canít poke a hole straight through this, but save your time.

The odd thing is, you seem to express disagreement with me when your idea is simply a condensed restatement of mine. I completely agree with this principle, although I'm not willing to ignore the huge factor an index number plays in determining the desireability of a student. If a student has a 25th percentile index, there is a much larger burden to impress the committee. My goal was to cause people to consider how large their index-dictated burden is, not predict admissions chances.

Quote
These calculators should be used as very general guides.  If every student with your numbers got into a school the year before and you apply to this school alone and get rejected, what are you going to do?  Sue them?  If you apply to several schools that are within your LSAT and GPA ranges, assuming you donít have some retched quality that permeates your app like a giant metaphoric turd, you will get into at least one law school and all these numbers wonít mean sh*t(pun intended).

Again, I agree with you.....this is basically what I was advocating. Perhaps the misunderstanting lies in my stat background. My statements were purely interpretative in nature, but I can see how they might be viewed as predictive by those in other fields.

Summary: Use an index calculator to figure out the highest ranked schools that won't ding you on numbers, apply to them, then assign yourself an admissions probability of 1 or 0 when you get the acceptance/rejection letter. The only time for further analysis is if you are poor and trying to weigh dollars spent on applications vs. chances of acceptance. I analyze not because it is required, but rather because I am freakish, twisted, and obsessive.