myLSN.info Forum

[Grazzhoppa]

thanks man

[facile princeps]

This thing rocks. Thanks for sharing. Did you code it yourself?

[LSATSCORES2012]

This thing rocks. Thanks for sharing. Did you code it yourself?

Yes, but the data is from http://www.lawschoolnumbers.com I'm glad that you like it.

[facile princeps]

This thing rocks. Thanks for sharing. Did you code it yourself?

Yes, but the data is from http://www.lawschoolnumbers.com I'm glad that you like it.

It's very useful, man. I'm a self taught hobby coder. I could code LSN from scratch (except the flash graph; i'd need to dig up my actionscript books for that), but i don't know how you fetched and parsed their data. I know PHP is awesome and can do almost anything you can dream up.

Oh, and as for that colored graph with the mouse coordinates returning the percent of accepted applicants for the selected school, genius! I'd seriously have to reinvest some learning time to code that one lol. Great job.

[LSATSCORES2012]

This thing rocks. Thanks for sharing. Did you code it yourself?

Yes, but the data is from http://www.lawschoolnumbers.com I'm glad that you like it.

It's very useful, man. I'm a self taught hobby coder. I could code LSN from scratch (except the flash graph; i'd need to dig up my actionscript books for that), but i don't know how you fetched and parsed their data. I know PHP is awesome and can do almost anything you can dream up.

Oh, and as for that colored graph with the mouse coordinates returning the percent of accepted applicants for the selected school, genius! I'd seriously have to reinvest some learning time to code that one lol. Great job.

The fetch application is actually coded in python. Here's the code, if you're interested: --LinkRemoved--

The graph is actually interesting, because it's not an image or a series of images. It's just a table with a bunch of td's with colored backgrounds. So, probably not as complicated as you would think.

[facile princeps]

This thing rocks. Thanks for sharing. Did you code it yourself?

Yes, but the data is from http://www.lawschoolnumbers.com I'm glad that you like it.

It's very useful, man. I'm a self taught hobby coder. I could code LSN from scratch (except the flash graph; i'd need to dig up my actionscript books for that), but i don't know how you fetched and parsed their data. I know PHP is awesome and can do almost anything you can dream up.

Oh, and as for that colored graph with the mouse coordinates returning the percent of accepted applicants for the selected school, genius! I'd seriously have to reinvest some learning time to code that one lol. Great job.

The fetch application is actually coded in python. Here's the code, if you're interested: --LinkRemoved--

The graph is actually interesting, because it's not an image or a series of images. It's just a table with a bunch of td's with colored backgrounds. So, probably not as complicated as you would think.

Take some credit, bro. You got skills . (changes tab to google python)

[LSATSCORES2012]

For anyone who's interested, in celebration of nearing 10,000 searches I added a new column to the results page that lists the percentage of people in your score range who received fee waivers.

The data is still updating, so it's not available for every school/every year yet, but more and more will be added throughout the evening. If you're running a search now, just run it on the 2011-2012 cycle, because the cycles are being updated in reverse chronological order (meaning that as of this posting the 2011-2012 cycle is the only one for which I have data).

Hope you find it helpful!

[North]

About 100 of those searches are mine. You made a great tool, bro, thank you.

[LSATSCORES2012]

So I made a new addition, again themed after an idea from hourumd.

For schools that report their index formula to LSAC (among the top 30 schools), the website will now report the percentage of accepted applicants who had an index less than or equal to your own.

In some ways this is more useful than the percentage of applicants with your numbers that were accepted. There could, for example, be a random blip around your numbers, but if it reports that your index is greater than that of 70% of accepted applicants, your numbers likely went down either because of random chance or because of YP.

I think it's hard to say, but haphazardly guessing I'd say that if this number is greater than 15% you probably have a decent chance, 25% you probably have good chance, and if this is greater than 50% chance you're surely a lock. Just throwing random numbers out here, so I'd be interested to have your input.

I hope you find this to be useful - and thank you to all the support everyone has provided for this program so far

On a side note, I wonder if I could write a PS about making this program...? Probably not the best idea.

[Yukos]

So I made a new addition, again themed after an idea from hourumd.

For schools that report their index formula to LSAC (among the top 30 schools), the website will now report the percentage of accepted applicants who had an index less than or equal to your own.

In some ways this is more useful than the percentage of applicants with your numbers that were accepted. There could, for example, be a random blip around your numbers, but if it reports that your index is greater than that of 70% of accepted applicants, your numbers likely went down either because of random chance or because of YP.

I think it's hard to say, but haphazardly guessing I'd say that if this number is greater than 15% you probably have a decent chance, 25% you probably have good chance, and if this is greater than 50% chance you're surely a lock. Just throwing random numbers out here, so I'd be interested to have your input.

I hope you find this to be useful - and thank you to all the support everyone has provided for this program so far

On a side note, I wonder if I could write a PS about making this program...? Probably not the best idea.

The index thing is cool, but I actually believe %accepted more. According to % of accepts w/worse I'm a lock at Boalt, but %accepted shows it's more a 50/50 proposition -- which, considering how black boxed Boalt is, is more accurate. More data is never a bad thing though!

[LSATSCORES2012]

The index thing is cool, but I actually believe %accepted more. According to % of accepts w/worse I'm a lock at Boalt, but %accepted shows it's more a 50/50 proposition -- which, considering how black boxed Boalt is, is more accurate. More data is never a bad thing though!

Yea, I think it definitely depends on the school. For Berkeley it probably will not mean much because Berkeley seems to be one of those black-box schools.

Speaking of which, I was again curious, so I thought I'd see which schools care the most about your numbers. Here's a ranking of the T14 schools, from those that care most about your numbers to those that care least.

1. Columbia (.78)
2. NYU (.72)
3. Georgetown (.69)
4. Duke (.66)
5. Northwestern (.66)
6. UVA (.61)
6. Harvard (.61)
8. Penn (.58)
8. Chicago (.58)
10. Michigan (.57)
11. Cornell (.54)
12. Berkeley (.49)
13. Stanford (.47)
14. Yale (.35)

The values in parentheses represent the Pearson correlation coefficient for the school (that is, how well one's numbers correlate to their acceptance). I also excluded waitlists in order to try to avoid pollution from yield protection. Interestingly, the schools for which I made an artificial index for comparison correlate better than the schools that provided their index (meaning the indexes that I made are pretty good! )

I also thought I'd check how well the correlations fare for URM's. Here's what we get:

1. Columbia (.68)
1. Harvard (.68)
1. Northwestern (.68)
4. NYU (.64)
5. Duke (.64)
6. Georgetown (.63)
7. Cornell (.62)
7. Berkeley (.62)
9. Yale (.59)
10. Chicago (.58)
10. Stanford (.58)
12. Michigan (.54)
13. UVA (.37)
14. U Penn (.32)

On average, they're about the same (.58 URM vs .59 non-URM), but it seems like URM cycles are a bit less consistent.

[bernaldiaz]

The index thing is cool, but I actually believe %accepted more. According to % of accepts w/worse I'm a lock at Boalt, but %accepted shows it's more a 50/50 proposition -- which, considering how black boxed Boalt is, is more accurate. More data is never a bad thing though!

Yea, I think it definitely depends on the school. For Berkeley it probably will not mean much because Berkeley seems to be one of those black-box schools.

Speaking of which, I was again curious, so I thought I'd see which schools care the most about your numbers. Here's a ranking of the T14 schools, from those that care most about your numbers to those that care least.

1. Columbia (.78)
2. NYU (.72)
3. Georgetown (.69)
4. Duke (.66)
5. Northwestern (.66)
6. UVA (.61)
6. Harvard (.61)
8. Penn (.58)
8. Chicago (.58)
10. Michigan (.57)
11. Cornell (.54)
12. Berkeley (.49)
13. Stanford (.47)
14. Yale (.35)

The values in parentheses represent the Pearson correlation coefficient for the school (that is, how well one's numbers correlate to their acceptance). I also excluded waitlists in order to try to avoid pollution from yield protection. Interestingly, the schools for which I made an artificial index for comparison correlate better than the schools that provided their index (meaning the indexes that I made are pretty good! )

I also thought I'd check how well the correlations fare for URM's. Here's what we get:

1. Columbia (.68)
1. Harvard (.68)
1. Northwestern (.68)
4. NYU (.64)
5. Duke (.64)
6. Georgetown (.63)
7. Cornell (.62)
7. Berkeley (.62)
9. Yale (.59)
10. Chicago (.58)
10. Stanford (.58)
12. Michigan (.54)
13. UVA (.37)
14. U Penn (.32)

On average, they're about the same (.58 URM vs .59 non-URM), but it seems like URM cycles are a bit less consistent.

You seem to be stats whiz; it might be interesting if you could quantify different URM bumps. I know they are unpredictable it might be helpful for URMs to see where they might fall and fascinating for the rest of us.

[LSATSCORES2012]

You seem to be stats whiz; it might be interesting if you could quantify different URM bumps. I know they are unpredictable it might be helpful for URMs to see where they might fall and fascinating for the rest of us.

A few weeks ago I actually downloaded this information but it was so much data that it crashed the database. I suppose just doing the URM type might work (before I had downloaded all sorts of stuff like the kind of EC's they participated in, etc.) From what I remember, AA's got the biggest boost, and NA's actually seemed to get hurt in the application process (that's what I care about because I'm NA.) Not sure what's up with that, but I'm hoping it's just a symptom insufficient data points. I also had information like majors, undergrad types, etc. in there.

The other disadvantage is that it uses up massive amounts of bandwidth to load each user's page, but I suppose that I have to do that anyway now to track fee waivers.

I'll see what I can do As Yukos said, more information is better

[LSATSCORES2012]

I know that TLS has this chart of decision times for the 2009-2010 cycle, but that's a little bit out of date.

So, on mylsn you can now find the more updated data. Enjoy! Let me know if you have any suggestions/comments/problems. Note that the page is INCREDIBLY large, as it will load a chart for every T1 school. So it may take a while to load. But don't worry about overloading my server... it uses google's graphing server!

[Yukos]

I know that TLS has this chart of decision times for the 2009-2010 cycle, but that's a little bit out of date.

So, on mylsn you can now find the more updated data. Enjoy! Let me know if you have any suggestions/comments/problems. Note that the page is INCREDIBLY large, as it will load a chart for every T1 school. So it may take a while to load. But don't worry about overloading my server... it uses google's graphing server!

You are the man.

It's really amazing Yale manages to have the least numbers-based acceptance method, and yet still has the highest medians. I guess that's what they get for being the best.

[LSATSCORES2012]

I know that TLS has this chart of decision times for the 2009-2010 cycle, but that's a little bit out of date.

So, on mylsn you can now find the more updated data. Enjoy! Let me know if you have any suggestions/comments/problems. Note that the page is INCREDIBLY large, as it will load a chart for every T1 school. So it may take a while to load. But don't worry about overloading my server... it uses google's graphing server!

You are the man.

It's really amazing Yale manages to have the least numbers-based acceptance method, and yet still has the highest medians. I guess that's what they get for being the best.

I know, it's crazy. I kind of think of them as Harvard and Stanford combined. Stanford, you gotta be awesome outside of the classroom. Harvard, you gotta be awesome inside of the classroom. For Yale, you gotta be both.

[LSATSCORES2012]

This is probably the last big update for a while, because I'm fresh out of ideas (but if you have any, let me know!) and I should really be working on my own applications.

I've put up a new grapher that you can access at --LinkRemoved--

I'm not listing it on the website yet because I want to get people's input/see if they notice any problems. So please let me know if you do!

With this one, you can graph by major, race, sex, etc. These fields are all optional, but if you select one of the checkboxes that means that it will only include results that match at least one of the criteria (they need not match all of the criteria).

Edit: Also, I always thought that in the LSAC rankings, philosophy being grouped with theology hurt philosophy majors' average LSAT. So I thought I'd check based on LSN data. Here's an LSN ranking of majors:

1. Physics - 166.9
2. Engineering - 165.5
3. Chemistry - 164.8
4. Philosophy - 164.4
5. Religion/Theology - 164
6. English/Literature - 163.9
7. Biology - 163.5
8. Business/Economics/Finance/Accounting - 162.6
9. Psychology/Sociology/Anthropology - 162.5
10. Double - 162.1
11. Political Science/History/Public Policy - 160.6
12. Communications/Journalism/Marketing - 158
13. Criminal Justice/Government/Legal Studies - 155.6

And here are majors ranked by difficulty (LSAT divided by GPA):

1. Chemistry - 49.4
2. Biology - 49
3. Engineering - 48.8
4. Physics - 48.4
5. Communications/Journalism/Marketing - 47.9
6. Psychology/Sociology/Anthropology - 47.1
7. English/Literature - 46.9
7. Philosophy - 46.9
9. Business/Economics/Finance/Accounting - 46.8
10. Political Science/History/Public Policy - 46.7
11. Criminal Justice/Government/Legal Studies - 46.5
12. Double - 46.1
13. Religion/Theology - 45.9

[Yukos]

And here are majors ranked by difficulty (LSAT divided by GPA):

1. Chemistry - 49.4
2. Biology - 49
3. Engineering - 48.8
4. Physics - 48.4
5. Communications/Journalism/Marketing - 47.9
6. Psychology/Sociology/Anthropology - 47.1
7. English/Literature - 46.9
8. Philosophy - 46.9
9. Business/Economics/Finance/Accounting - 46.8
10. Political Science/History/Public Policy - 46.7
11. Criminal Justice/Government/Legal Studies - 46.5
12. Double - 46.1
13. Religion/Theology - 45.9

lolwut.

[LSATSCORES2012]

And here are majors ranked by difficulty (LSAT divided by GPA):

1. Chemistry - 49.4
2. Biology - 49
3. Engineering - 48.8
4. Physics - 48.4
5. Communications/Journalism/Marketing - 47.9
6. Psychology/Sociology/Anthropology - 47.1
7. English/Literature - 46.9
8. Philosophy - 46.9
9. Business/Economics/Finance/Accounting - 46.8
10. Political Science/History/Public Policy - 46.7
11. Criminal Justice/Government/Legal Studies - 46.5
12. Double - 46.1
13. Religion/Theology - 45.9

lolwut.

I know! Could it be marketing? I wasn't sure if marketing was communications or business, so I grouped it in with communications.

Edit: communications gets dropped to #7, tied w/ english and philosophy, if I put marketing on business

[BallHog]

I find it hilarious law concentrated majors in UG do the worst on the LSAT.

Good work on the website! This is going to help me immensely during my cycle!

[Yukos]

And here are majors ranked by difficulty (LSAT divided by GPA):

1. Chemistry - 49.4
2. Biology - 49
3. Engineering - 48.8
4. Physics - 48.4
5. Communications/Journalism/Marketing - 47.9
6. Psychology/Sociology/Anthropology - 47.1
7. English/Literature - 46.9
8. Philosophy - 46.9
9. Business/Economics/Finance/Accounting - 46.8
10. Political Science/History/Public Policy - 46.7
11. Criminal Justice/Government/Legal Studies - 46.5
12. Double - 46.1
13. Religion/Theology - 45.9

lolwut.

I know! Could it be marketing? I wasn't sure if marketing was communications or business, so I grouped it in with communications.

Edit: communications gets dropped to #7, tied w/ english and philosophy, if I put marketing on business

My friend went to J-school and it was super grade-uninflated (not sure if that's the norm or just Medill) so it might just be communications was put with two difficult majors.

[bernaldiaz]

I'm bumping this thread. There seems to be a lot of new users with real basic questions about where they can get in or should apply. I encourage you to peruse this website. It's a great resource and can give much better advice than any of us are capable of.

[LSATSCORES2012]

Just put together a graph of GPA-LSAT correlation based on LSN... though I'd post it, it's a bit interesting. It clearly shows there's not much of a correlation between UG GPA and LSAT scores.

I was also planning to label points by major and undergraduate, but I ended up running out of time/getting distracted, so maybe I'll do it later.



Also, thanks for the bump above, bernaldiaz!

[LSATSCORES2012]

... I was looking at the demographics for the website, there are a ton of hits from New Haven, Connecticut. Are all the Yale peeps checking out my website? Yale, I mentioned this in my application! You should take me despite my mediocre numbers!

Oddly enough, it spiked the day I submitted my application. They couldn't have reviewed my application already, could they? although I suppose that also coincides with the day october testers got their scores back, so there could have just been more people on the site in general (which is true.)

[North]

Popping in to say that I like what you're doing with the links on top of the myLSN page. It provides easy access to some of TLS's most useful information to people who might not otherwise be able to find it. As usual, great work bro.