Harvard, C/O 2017, Applicants Thread (2013-2014) Forum

[lawschool22]

But in all seriousness, the problem with this is it (I assume) ignores 25/50/75 percentiles, which we know are important. Also 5 years of data might actually be a bad thing. We are in a vastly different world with admissions than we were in the 2008-2009 cycle.

Finally, I'm curious about your regression. What confidence level are you at, and what was your correlation coefficient?

Here's the STATA printout. Again this was just a straight up, no tweaks, no data scrubbing "logit admit lsat gpa":

Model For Above Median LSAT Data Points

Logistic regression Number of obs = 690
LR chi2(2) = 369.27
Prob > chi2 = 0.0000
Log likelihood = -287.48419 Pseudo R2 = 0.3911

------------------------------------------------------------------------------
admit | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
gpa | 11.05645 .8674196 12.75 0.000 9.356334 12.75656
lsat | .3172556 .0563768 5.63 0.000 .206759 .4277521
_cons | -97.13444 11.28062 -8.61 0.000 -119.244 -75.02483
------------------------------------------------------------------------------


Model for Below Median LSAT Data Points

Logistic regression Number of obs = 817
LR chi2(2) = 236.28
Prob > chi2 = 0.0000
Log likelihood = -257.6438 Pseudo R2 = 0.3144

------------------------------------------------------------------------------
admit | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
gpa | 10.25789 1.111142 9.23 0.000 8.080089 12.43568
lsat | .54094 .0758397 7.13 0.000 .3922969 .6895831
_cons | -133.1748 14.55945 -9.15 0.000 -161.7108 -104.6388
------------------------------------------------------------------------------

And in case you're not familiar with that format, p-value for everything is 0.

So you're saying there's a chance...

That quote is all I ever hear in my head when we start talking chances...

[esrom55]

But in all seriousness, the problem with this is it (I assume) ignores 25/50/75 percentiles, which we know are important. Also 5 years of data might actually be a bad thing. We are in a vastly different world with admissions than we were in the 2008-2009 cycle.



And in case you're not familiar with that format, p-value for everything is 0.

So you're saying there's a chance...

That quote is all I ever hear in my head when we start talking chances...

You rang??

[chneyo]

...

[Mort_the_Tort]

1% chance, as well.

I'll take it !

will ¿tomorrow? be a blood bath or a victory lap ?

[chneyo]

...

[jttoplawschools93]

1% chance, as well.

I'll take it !

will ¿tomorrow? be a blood bath or a victory lap ?

can someone explain for the mathematically challenged how the percents are being calculated

[midwest17]

1% chance, as well.

I'll take it !

will ¿tomorrow? be a blood bath or a victory lap ?

can someone explain for the mathematically challenged how the percents are being calculated

Basically, it assumes that the graph of admission chances is shaped like an S (so pretty flat for a while, then rapidly increasing, then leveling off near 100%) for both LSAT and GPA. It then figures out which S "fits the data best" (i.e., predicts as close as possible to 100% for everyone who got in, and as close to 0% as possible for everyone who didn't get in). Once it has that S, you just plug your numbers in and it tells you where you fall.

But in all seriousness, the problem with this is it (I assume) ignores 25/50/75 percentiles, which we know are important. Also 5 years of data might actually be a bad thing. We are in a vastly different world with admissions than we were in the 2008-2009 cycle.

Finally, I'm curious about your regression. What confidence level are you at, and what was your correlation coefficient?

Here's the STATA printout. Again this was just a straight up, no tweaks, no data scrubbing "logit admit lsat gpa":

Model For Above Median LSAT Data Points

Logistic regression Number of obs = 690
LR chi2(2) = 369.27
Prob > chi2 = 0.0000
Log likelihood = -287.48419 Pseudo R2 = 0.3911

------------------------------------------------------------------------------
admit | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
gpa | 11.05645 .8674196 12.75 0.000 9.356334 12.75656
lsat | .3172556 .0563768 5.63 0.000 .206759 .4277521
_cons | -97.13444 11.28062 -8.61 0.000 -119.244 -75.02483
------------------------------------------------------------------------------


Model for Below Median LSAT Data Points

Logistic regression Number of obs = 817
LR chi2(2) = 236.28
Prob > chi2 = 0.0000
Log likelihood = -257.6438 Pseudo R2 = 0.3144

------------------------------------------------------------------------------
admit | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
gpa | 10.25789 1.111142 9.23 0.000 8.080089 12.43568
lsat | .54094 .0758397 7.13 0.000 .3922969 .6895831
_cons | -133.1748 14.55945 -9.15 0.000 -161.7108 -104.6388
------------------------------------------------------------------------------

And in case you're not familiar with that format, p-value for everything is 0 or in other words 99.999% confidence.

So you did separate logistic regressions for above median and below median?

Wouldn't it make more sense to do a regression discontinuity? My stats is rusty/never that developed, but that seems like it would be a better model than 2 logistic curves.

ETA: Actually, maybe that would do the same thing. Like I said, my stats is rusty.

[Gary]

1% chance, as well.

I'll take it !

will ¿tomorrow? be a blood bath or a victory lap ?

second Mort's question. What's everyone thinking tomorrow will bring? Dings/ WLs or Acceptances?

[wtrc]

1% chance, as well.

I'll take it !

will ¿tomorrow? be a blood bath or a victory lap ?

second Mort's question. What's everyone thinking tomorrow will bring? Dings/ WLs or Acceptances?

Totally a guess, but my guess is tomorrow will be a sad day and Monday will be a happier day.

[phillywc]

I also believe tomorrow will be a ding day...

[rui]

I doubt anyone will listen to this, but, a word of advice from someone who went through this process (and is now watching a good friend do it, which is what brought me back to TLS): submit your app, and then forget you applied to law school for as long as you can before you need to start considering scholarship deadlines. Until that point just log out of here and go outside or read a book or do anything but compulsively refresh these threads. Once you start 1L, at HLS or elsewhere, your life will be consumed with all things law school. No need to get a jump on it. I promise! Enjoy these last few months.

[Mort_the_Tort]

tomorrow will be a great day, and so will the day after tomorrow, and so forth -- whatever may come.

we shall will it and make it so!

[Balthy]

I doubt anyone will listen to this, but, a word of advice from someone who went through this process (and is now watching a good friend do it, which is what brought me back to TLS): submit your app, and then forget you applied to law school for as long as you can before you need to start considering scholarship deadlines. Until that point just log out of here and go outside or read a book or do anything but compulsively refresh these threads. Once you start 1L, at HLS or elsewhere, your life will be consumed with all things law school. No need to get a jump on it. I promise! Enjoy these last few months.

One does not simply ignore status checkers.

But this does sound like solid advice. Just wish i could follow it.

[drawstring]

Ya, I would totally follow that advice if I could. It's like some force always draws me back here no matter how hard I try to avoid it though

[FuriousDuck]

...

[FuriousDuck]

...

[barrelofmonkeys]

I actually completely ignore the status checkers... I'm not even totally sure why they exist. Don't the schools usually e-mail or call you whenever anything important happens?

I have basically found out about 3 acceptances on my status checkers before anything else (or at least, known to expect a decision because of the status checker).

[FuriousDuck]

...

[TripTrip]

I doubt anyone will listen to this, but, a word of advice from someone who went through this process (and is now watching a good friend do it, which is what brought me back to TLS): submit your app, and then forget you applied to law school for as long as you can before you need to start considering scholarship deadlines. Until that point just log out of here and go outside or read a book or do anything but compulsively refresh these threads. Once you start 1L, at HLS or elsewhere, your life will be consumed with all things law school. No need to get a jump on it. I promise! Enjoy these last few months.

One does not simply ignore status checkers.

But this does sound like solid advice. Just wish i could follow it.

Short of following this advice, as long as you haven't been rejected from Harvard, don't tell anyone where you plan on going. People start to think you're weird if you keep changing the school you say you'll be attending. If they say "Which schools are you considering?" to try to draw it out of you, just say it's between quite a few. It would have mentally helped me a lot if friends and family weren't trying to solidify a school in my head for the sake of polite conversation.

[IrishJew]

I developed an admissions calculator specifically for Harvard. The regression model used is based on about 5 years of non-URM LSN data.

--LinkRemoved--

If I get much response, I'll put more work into the regression model and test more for accuracy, and maybe do the rest of the T14.

Cool. I like that it will display the "what if curve" for LSAT or GPA, it's a nice feature, especially if someone is thinking of retaking the LSAT (or...college?). But I did find a bug: it gave me lower odds than mylsn did, so yours must be broken :p

Thanks for the cool app!

[IrishJew]

But in all seriousness, the problem with this is it (I assume) ignores 25/50/75 percentiles, which we know are important. Also 5 years of data might actually be a bad thing. We are in a vastly different world with admissions than we were in the 2008-2009 cycle.

Finally, I'm curious about your regression. What confidence level are you at, and what was your correlation coefficient?

Here's the STATA printout. Again this was just a straight up, no tweaks, no data scrubbing "logit admit lsat gpa":

Model For Above Median LSAT Data Points

Logistic regression Number of obs = 690
LR chi2(2) = 369.27
Prob > chi2 = 0.0000
Log likelihood = -287.48419 Pseudo R2 = 0.3911

------------------------------------------------------------------------------
admit | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
gpa | 11.05645 .8674196 12.75 0.000 9.356334 12.75656
lsat | .3172556 .0563768 5.63 0.000 .206759 .4277521
_cons | -97.13444 11.28062 -8.61 0.000 -119.244 -75.02483
------------------------------------------------------------------------------


Model for Below Median LSAT Data Points

Logistic regression Number of obs = 817
LR chi2(2) = 236.28
Prob > chi2 = 0.0000
Log likelihood = -257.6438 Pseudo R2 = 0.3144

------------------------------------------------------------------------------
admit | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
gpa | 10.25789 1.111142 9.23 0.000 8.080089 12.43568
lsat | .54094 .0758397 7.13 0.000 .3922969 .6895831
_cons | -133.1748 14.55945 -9.15 0.000 -161.7108 -104.6388
------------------------------------------------------------------------------

And in case you're not familiar with that format, p-value for everything is 0.

So you're saying there's a chance...

+!

[Mort_the_Tort]

¿ today ?

[IrishJew]

I actually completely ignore the status checkers... I'm not even totally sure why they exist. Don't the schools usually e-mail or call you whenever anything important happens?

I took the EXACT same attitude as you. Couldn't agree more.

And then I found out I was in at a school via status checker with no e-mail or phone call. Commence my compulsive freakout.

But I think it depends on the school. Said school apparently does that as the norm; we know Harvard does otherwise, and even said school is sending physical mail soooo.....

[sah]

At which T14s does the status checker matter? So I can stop wasting my time.... For instance, I know NYU and Harvard just called before any updates, and Columbia states explicitly that they will not announce decisions online.

[neprep]

At which T14s does the status checker matter? So I can stop wasting my time.... For instance, I know NYU and Harvard just called before any updates, and Columbia states explicitly that they will not announce decisions online.

Actually, NYU indicates "Decision letter sent" before you get a call or an email.