How to choose between scholarship offers: A new method

(Rankings, Profiles, Tuition, Student Life, . . . )
thep0ster
Posts: 2
Joined: Fri Nov 19, 2010 2:26 pm

How to choose between scholarship offers: A new method

Postby thep0ster » Sat Nov 20, 2010 10:24 am

Choosing between law schools is difficult enough with all other things being equal. But scholarship offers horribly complicate the matter. For instance, is a full ride at a T50 school better than paying sticker at a T20? What about a half ride?

I've therefore taken the liberty of creating a law school scholarship matrix that can be programmed to give an empirical score to each law school, taking scholarship into account, to help make a decision.

The matrix can be accessed through the google doc https://spreadsheets.google.com/ccc?key ... y=COm1rbgH.

I recommend exporting to excel for maximum functionality.

--ImageRemoved--

Here's how this works:

1. Enter all the schools you are applying to. For additional information, tier, include the USNWR rank, cost of tuition, total cost of attendance.

2. Assume you got a full ride to every school you're applying to. Rank the schools according the order you would attend them, all else equal, in the "No Cost Preference" Column. Obviously, this isn't quite realistic, but it gives you an idea of where you really want to go to school. Ties are okay.

3. This is where it gets tricky. The base preference is basically a weighted score that shows your actual interest in a school without any scholarship. Give your lowest ranked school a score of 100 (ties are ok).

Move up from there on your list of schools. To determine a school's score, think about it relatively. For instance, getting a half ride to vanderbilt for me would be a toss up between a full at Alabama, so I gave Vandy a score that was 50% higher. You can play around with this, especially once you start putting in scholarship values. Feel free to use mine as an example.

4. Enter the percent scholarship you receive into the "Scholarship" column. For instance, 50% = half ride, 100% = full ride.

5. Play around with it and figure out where you want to go to school! At the end of the day, you should choose the school that has the highest score in the "Final Preference Value" column.

Let me know what you all think, and if this is useful at all to anyone.
Last edited by thep0ster on Sat Nov 20, 2010 10:28 am, edited 1 time in total.

thep0ster
Posts: 2
Joined: Fri Nov 19, 2010 2:26 pm

Re: How to choose between scholarship offers: An new method

Postby thep0ster » Sat Nov 20, 2010 10:25 am

Also, TL;DR: I came up with a neat / fun way of weighting and comparing schools.

rundoxierun
Posts: 1893
Joined: Wed Dec 03, 2008 1:46 am

Re: How to choose between scholarship offers: A new method

Postby rundoxierun » Sat Nov 20, 2010 11:45 am

Lots of issues with this. For one, not useful because assumes too much linearity/equality. For instance, full scholly at Florida=/= sticker at Stanford by any means. A dollar increase in scholarship at Stanford =/= a dollar increase in scholarship at Florida. Because of different job prospects, COA is much less important at Stanford. A few ways to fix this would be two either work out a decent equation that applies a more curved equation to tier ranking (i.e. initial preference-ranking^1.2 or something) or maybe assign different equations to different schools taking into account relative importance of COA (i.e. initial pref-.02(cost of attendance) at harvard while equation for alabama is more like initial pref-.25(cost of attendance).

This is just off the top of my head so this is in no way comprehensive and those equations are just completely made examples with no thought given to the numbers or anything.

r6_philly
Posts: 10707
Joined: Sat Dec 19, 2009 4:32 pm

Re: How to choose between scholarship offers: A new method

Postby r6_philly » Sat Nov 20, 2010 12:03 pm

You can't quantify the value of the scholarship for comparison purposes because you can't quantify the value of schools. I think you can assign quantitative values to intangibles and subjective measures, sure, but you run the risk of one bad estimation ruining your whole analysis.

User avatar
northwood
Posts: 4872
Joined: Fri May 14, 2010 7:29 pm

Re: How to choose between scholarship offers: A new method

Postby northwood » Sat Nov 20, 2010 12:21 pm

you are making this more difficult and complex than necessary:
( you need to know the reasoning behind your ranking before you do this- i am assuming you did this before you applied)
Make a ranking of the schools before you hear scholarships/ or decisions
Make a ranking of the regions you want to work ( can do this before the schools)
Once you are in, determine the final cost of attendance
Rank the schools based on these factors ( can make a chart and average the rankings)

Choose your best ranked school, and start finding a decent place to live.

rundoxierun
Posts: 1893
Joined: Wed Dec 03, 2008 1:46 am

Re: How to choose between scholarship offers: A new method

Postby rundoxierun » Sat Nov 20, 2010 12:35 pm

r6_philly wrote:You can't quantify the value of the scholarship for comparison purposes because you can't quantify the value of schools. I think you can assign quantitative values to intangibles and subjective measures, sure, but you run the risk of one bad estimation ruining your whole analysis.


Yeah you can in both cases. It doesnt have to be super exact like "florida is worth .02657672 times as much as harvard" but you can easily say Harvard is my number so that school is 1. I like Yale 80% as much as harvard so .8, so forth and so on. Its done plenty of times in plenty of things. You just have to set out your assumptions beforehand and have an idea of what measurements you are going to quantify and how..

God Im going to miss analytical style math/science in law school..

r6_philly
Posts: 10707
Joined: Sat Dec 19, 2009 4:32 pm

Re: How to choose between scholarship offers: A new method

Postby r6_philly » Sat Nov 20, 2010 1:40 pm

tkgrrett wrote:
r6_philly wrote:You can't quantify the value of the scholarship for comparison purposes because you can't quantify the value of schools. I think you can assign quantitative values to intangibles and subjective measures, sure, but you run the risk of one bad estimation ruining your whole analysis.


Yeah you can in both cases. It doesnt have to be super exact like "florida is worth .02657672 times as much as harvard" but you can easily say Harvard is my number so that school is 1. I like Yale 80% as much as harvard so .8, so forth and so on. Its done plenty of times in plenty of things. You just have to set out your assumptions beforehand and have an idea of what measurements you are going to quantify and how..

God Im going to miss analytical style math/science in law school..


It is very hard to do it in the right scale. Choosing the wrong proportions for coefficients is going to skew your result. God how many times I have fixed systems that relied on bad coefficients.

Using your example, you can't just arbitrarily assign .8 to Y as compared to H. Yes you can apply that .8 to one factor, but how much does that factor weigh in the actual formula. We do cross section regression like that yes but usually not with made up formulas, they are based form existing data. Like I said, if you really like Y .7 instead of .8, your results will be very skewed. How do you quantify the difference between 70% like and 80% like? Assigning quantitative values to subjective criteria is futile :D Don't forget what I do, I wish people would stop doing them. :P

User avatar
Borhas
Posts: 4852
Joined: Sun Sep 27, 2009 6:09 pm

Re: How to choose between scholarship offers: A new method

Postby Borhas » Sat Nov 20, 2010 1:43 pm

sigh... economics majors....

r6_philly
Posts: 10707
Joined: Sat Dec 19, 2009 4:32 pm

Re: How to choose between scholarship offers: A new method

Postby r6_philly » Sat Nov 20, 2010 1:52 pm

Borhas wrote:sigh... economics majors....


How's your Saturday treating you?

Just by chance I am in a finace/economics class right now :lol:

User avatar
Borhas
Posts: 4852
Joined: Sun Sep 27, 2009 6:09 pm

Re: How to choose between scholarship offers: A new method

Postby Borhas » Sat Nov 20, 2010 1:58 pm

r6_philly wrote:
Borhas wrote:sigh... economics majors....


How's your Saturday treating you?


too early to tell... at some point I'll do my morning run, and then start being productive. I'm thinking 2 more hours of preparing to start work before I actually start.

rundoxierun
Posts: 1893
Joined: Wed Dec 03, 2008 1:46 am

Re: How to choose between scholarship offers: A new method

Postby rundoxierun » Sat Nov 20, 2010 2:00 pm

r6_philly wrote:
tkgrrett wrote:
r6_philly wrote:You can't quantify the value of the scholarship for comparison purposes because you can't quantify the value of schools. I think you can assign quantitative values to intangibles and subjective measures, sure, but you run the risk of one bad estimation ruining your whole analysis.


Yeah you can in both cases. It doesnt have to be super exact like "florida is worth .02657672 times as much as harvard" but you can easily say Harvard is my number so that school is 1. I like Yale 80% as much as harvard so .8, so forth and so on. Its done plenty of times in plenty of things. You just have to set out your assumptions beforehand and have an idea of what measurements you are going to quantify and how..

God Im going to miss analytical style math/science in law school..


It is very hard to do it in the right scale. Choosing the wrong proportions for coefficients is going to skew your result. God how many times I have fixed systems that relied on bad coefficients.

Using your example, you can't just arbitrarily assign .8 to Y as compared to H. Yes you can apply that .8 to one factor, but how much does that factor weigh in the actual formula. We do cross section regression like that yes but usually not with made up formulas, they are based form existing data. Like I said, if you really like Y .7 instead of .8, your results will be very skewed. How do you quantify the difference between 70% like and 80% like? Assigning quantitative values to subjective criteria is futile :D Don't forget what I do, I wish people would stop doing them. :P


I understand exactly where are coming from but thats a bit exact for this. Well I mean you could apply .8 to specific factor then weigh the specific factor.. I.e. .8 for QOL and QOL is 25% of decision. Regardless, we are getting way to precise for this kind of thing if we are arguing down to this point. I think most people just want kind of a standardized view, not exact down to a few hundred dollars.

User avatar
NZA
Posts: 1285
Joined: Fri Nov 19, 2010 10:01 pm

Re: How to choose between scholarship offers: A new method

Postby NZA » Sat Nov 20, 2010 2:06 pm

Just like in chess or go, there are right moves and wrong moves. Similarly, I think, in law school apps. Like, why wouldn't you go to HYS if you got in? At the same time, there are moves and choices that need to be made in situ. And that's 90% of the game.

For instance, if someone really wanted to practice or live in Seattle, go to UW. If they wanted to practice in NYC, wouldn't Fordham be as great a choice as any (aside from the NYU/Columbia thing)?

In other words, numbers might give a person an idea of how things "objectively" shape up, but in the end, they can't tell you what you really want. That's something you have to figure out by introspection, I think.

Anyway, if this works for you, great! :) But I don't know if everyone will feel that an Excel spreadsheet is going to make a difference in their evaluation of law schools with or without scholarships. But then again, I'm just some random guy on the internet, so wtf do I know? :P

r6_philly
Posts: 10707
Joined: Sat Dec 19, 2009 4:32 pm

Re: How to choose between scholarship offers: A new method

Postby r6_philly » Sat Nov 20, 2010 2:07 pm

tkgrrett wrote:
I understand exactly where are coming from but thats a bit exact for this. Well I mean you could apply .8 to specific factor then weigh the specific factor.. I.e. .8 for QOL and QOL is 25% of decision. Regardless, we are getting way to precise for this kind of thing if we are arguing down to this point. I think most people just want kind of a standardized view, not exact down to a few hundred dollars.


Agreed. I just leery of people making $100k+ decisions based on an imprecise formula. It may actually leave this kind of financial analysis subjective. Because putting a hard number on it may be counter productive.

User avatar
AreJay711
Posts: 3406
Joined: Tue Jul 20, 2010 8:51 pm

Re: How to choose between scholarship offers: A new method

Postby AreJay711 » Sat Nov 20, 2010 2:37 pm

This is a cool attempt to model decision making behavior BUT it is pointless because people can obviously tell which one they prefer better without this. People almost never have a problem making ordinal decisions. Plus, if in their minds it is a close decision the fact that this formula says one is superior doesn't change the fact that the person making the decision doesn't think so.

What your model actually shows is what you could expect a person to choose if you didn't know his or her exact preferences but knew the other measures (and you didn't have data to run a regression). Individual people are just to quirky to have one model for exact decisions that can be applied to multiple people.




Return to “Choosing a Law School”

Who is online

Users browsing this forum: Yahoo [Bot] and 3 guests