PHILOSOPHY ADMISSIONS SURVEY

[ianfaircloud]

We have 80 responses to the survey. I hope that a lot of people will fill it out after the season ends (in a week or so). Then I hope some of you will help me analyze the data and draw out some conclusions. Of course I'll publish the data, but I will be very interested to hear what people conclude (if anything) from it. Maybe we'll get 100 responses. For some reason, that seems like the 'magic number'.

[kant_get_in]

I think your funding survey is an underrated, useful tool. Perhaps we should propagate that too! The last submission (besides mine) was late March. 

[ianfaircloud]

87 responses as of today. I'll close the survey on May 1. I hope others will participate before then! Maybe we'll hit 100...

[primenumbers]

I filled it out and can't wait to see responses...  I did better than expected on admissions, and I'm curious about why!

[humean_skeptic]

I filled it out and can't wait to see responses...  I did better than expected on admissions, and I'm curious about why!

 

Isn't it almost certain that you did better than expected on admissions because your writing sample was better than you thought? I mean, sure, you had good stats and all, but so does almost everyone else. The quality of the writing sample is *absolutely* crucial, especially if all else is equal, and it's not measured by the admissions survey, as far as I know. 

[primenumbers]

Isn't it almost certain that you did better than expected on admissions because your writing sample was better than you thought? I mean, sure, you had good stats and all, but so does almost everyone else. The quality of the writing sample is *absolutely* crucial, especially if all else is equal, and it's not measured by the admissions survey, as far as I know. 

Well, maybe.  I am not sure that there isn't a bias, though, for people on sites like this to only post their good stats.  As such, I wonder if maybe I didn't do better than my stats would predict.  My writing sample was fun, but it never got a thorough professor read before I sent it, so I find it hard to believe that it is that much better than other people's heavily edited samples.  I thought it might have been my letters...

[humean_skeptic]

Well, maybe.  I am not sure that there isn't a bias, though, for people on sites like this to only post their good stats.  As such, I wonder if maybe I didn't do better than my stats would predict.  My writing sample was fun, but it never got a thorough professor read before I sent it, so I find it hard to believe that it is that much better than other people's heavily edited samples.  I thought it might have been my letters...

 

I think you said you have a lower than normal GPA, though? 

 

I double checked your posts and yes, you did say "I have a lower-than-average GPA". 

Edited April 17, 2014 by humean_skeptic

[primenumbers]

I think you said you have a lower than normal GPA, though? 

 

I double checked your posts and yes, you did say "I have a lower-than-average GPA". 

But without data, how am I to know that my perception of normal successful applicant GPA is accurate?

[humean_skeptic]

But without data, how am I to know that my perception of normal successful applicant GPA is accurate?

 

http://thegradcafe.com/survey/index.php?q=philosophy

[primenumbers]

http://thegradcafe.com/survey/index.php?q=philosophy

But again, I think there may be a reporting bias in that only people with good stats attach them there, whereas theoretically, people who take the survey have an interest in accurate data and are more likely to report less excellent scores.

[humean_skeptic]

But again, I think there may be a reporting bias in that only people with good stats attach them there, whereas theoretically, people who take the survey have an interest in accurate data and are more likely to report less excellent scores.

 

I see. So you think people are more likely to anonymously provide false information when submitting results to the grad cafe results page than anonymously provide false information when filling out the grad admissions survey? 

 

Seems like a very speculative suspicion to me. 

 

To be a little more precise: I see no reason to posit that the probability of applicants with mediocre stats filling out the grad admissions survey is higher than the probability of applicants with mediocre stats reporting those stats on the grad cafe results page.

 

But, of course, the main issue is this: 1) The quality of the writing sample can't really be quantitatively measured, and 2) The quality of the writing sample is, according to all the admissions advice and information I've read on every department page, an extremely important part -- if not the most important part -- of one's application. 

Edited April 17, 2014 by humean_skeptic

[primenumbers]

I see. So you think people are more likely to anonymously provide false information when submitting results to the grad cafe results page than anonymously provide false information when filling out the grad admissions survey? 

 

Seems like a very speculative suspicion to me. 

 

To be a little more precise: I see no reason to posit that the probability of applicants with mediocre stats filling out the grad admissions survey is higher than the probability of applicants with mediocre stats reporting those stats on the grad cafe results page.

 

But, of course, the main issue is this: 1) The quality of the writing sample can't really be quantitatively measured, and 2) The quality of the writing sample is, according to all the admissions advice and information I've read on every department page, an extremely important part -- if not the most important part -- of one's application. 

No, I don't think people give false information on the results page.  I think they neglect to provide information in many cases, and I am speculating that people with lower stats are less likely to be among those providing information, since it is publicly viewable.  People filling out the survey, however, have a clear interest in creating an accurate data bank, and may be questioning their own results.  Since the information from that will never be public aside from summary statistics, those people are likely to provide information, regardless of the excellence of the stats. 

 

While writing samples cannot really be quantified easily, pretty much every other part of the application can be.  As such, if the quantitative data does not explain my results, then I can more reliably assume that my writing sample was the difference.

 

Honestly, I don't understand why you feel the need to question me on this.  I see no harm in me being excited to get data and see where I fall statistically on various measures.  Big data sets can provide a lot of information, so even if I were not wondering why I got the results I did, I think that I would be interested in seeing the results of this survey.  If you are so sure the writing sample is the most important component and that it can't be quantified, then you are of course not required to view the results.  Your sentiments do not apply to me, however, and there is no need to badger people over opinions that literally have nothing to do with you.

[humean_skeptic]

No, I don't think people give false information on the results page.  I think they neglect to provide information in many cases, and I am speculating that people with lower stats are less likely to be among those providing information, since it is publicly viewable.  People filling out the survey, however, have a clear interest in creating an accurate data bank, and may be questioning their own results.  Since the information from that will never be public aside from summary statistics, those people are likely to provide information, regardless of the excellence of the stats. 

 

While writing samples cannot really be quantified easily, pretty much every other part of the application can be.  As such, if the quantitative data does not explain my results, then I can more reliably assume that my writing sample was the difference.

 

Honestly, I don't understand why you feel the need to question me on this.  I see no harm in me being excited to get data and see where I fall statistically on various measures.  Big data sets can provide a lot of information, so even if I were not wondering why I got the results I did, I think that I would be interested in seeing the results of this survey.  If you are so sure the writing sample is the most important component and that it can't be quantified, then you are of course not required to view the results.  Your sentiments do not apply to me, however, and there is no need to badger people over opinions that literally have nothing to do with you.

 

Not sure what the 'badgering' and 'this has nothing to do with you' attacks are about. But I do apologize if I've offended you in disagreeing with you and providing what I think are good reasons for doing so -- it was my intention to clarify, not to offend. Given the amount of money and time I've invested in my applications, I do think the application process has something to do with me. 

 

About your conjecture that people with low stats are less likely to post their stats on the grad cafe results page: Sure, you might be right. But given the anonymity of posting on the grad cafe results page, it seems more likely than not that people would just go ahead and post the stats if they care to. If they don't care, they won't post them. Further, I know that I personally *definitely* have an interest in posting accurate stats when I post results on the grad cafe results page: Everyone else will post accurate stats too. 

 

About your claim that every single part of the application other than the writing sample can be quantified easily: Letters are hard to quantify, as is fit. But suppose you're right, and every part of everyone's application is easy to quantify *except* the sample. Well, even in that case, it's still true that the sample is probably the important part of the application. (I'm going by what it says on pretty much every Leiter-ranked department's admissions-info page.) Seems to follow, then, that you still won't be able to know a ton about the overall quality of your application. Still, the stats could be helpful, especially when they are -- as you point out -- from a large data set. 

Edited April 17, 2014 by humean_skeptic

[primenumbers]

Not sure what the 'badgering' and 'this has nothing to do with you' attacks are about. But I do apologize if I've offended you in disagreeing with you and providing what I think are good reasons for doing so -- it was my intention to clarify, not to offend. Given the amount of money and time I've invested in my applications, I do think the application process has something to do with me. 

 

About your conjecture that people with low stats are less likely to post their stats on the grad cafe results page: Sure, you might be right. But given the anonymity of posting on the grad cafe results page, it seems more likely than not that people would just go ahead and post the stats if they care to. If they don't care, they won't post them. Further, I know that I personally *definitely* have an interest in posting accurate stats when I post results on the grad cafe results page: Everyone else will post accurate stats too. 

 

About your claim that every single part of the application other than the writing sample can be quantified easily: Letters are hard to quantify, as is fit. But suppose you're right, and every part of everyone's application is easy to quantify *except* the sample. Well, even in that case, it's still true that the sample is probably the important part of the application. (I'm going by what it says on pretty much every Leiter-ranked department's admissions-info page.) Seems to follow, then, that you still won't be able to know a ton about the overall quality of your application. Still, the stats could be helpful, especially when they are -- as you point out -- from a large data set. 

I was not attacking you.  I was simply pointing out that what you arguing against is literally my personal rationale for being interested in something.  While the application process surely has to do with you, my personal reasons for being interested in a set of data do not.

 

Again, I point out that my worry about the results page is that relatively few people do supply their stats there, and those that are provided are very high.  I think it is very possible that this is artificial, due to the fact that the data on the results page is not in a large data-set and is presented individually, though anonymously.  It is likely, then, that people with lower stats may not share those.

 

Parts of the application other than the writing sample can at least be somewhat quantified, but the writing sample cannot easily be quantified at all.  Additionally, I have heard from several Leiter-ranked school professors that the importance of the writing sample has declined following an increase in faculty assistance in writing said samples.  I think that the overall quality of my application was shown by my results.  What I don't know is what part of my application made it a better application than I had expected it to be, and this data will at the very least, provide some insight there.

[humean_skeptic]

I was not attacking you.  I was simply pointing out that what you arguing against is literally my personal rationale for being interested in something.  While the application process surely has to do with you, my personal reasons for being interested in a set of data do not.

 

Again, I point out that my worry about the results page is that relatively few people do supply their stats there, and those that are provided are very high.  I think it is very possible that this is artificial, due to the fact that the data on the results page is not in a large data-set and is presented individually, though anonymously.  It is likely, then, that people with lower stats may not share those.

 

Parts of the application other than the writing sample can at least be somewhat quantified, but the writing sample cannot easily be quantified at all.  Additionally, I have heard from several Leiter-ranked school professors that the importance of the writing sample has declined following an increase in faculty assistance in writing said samples.  I think that the overall quality of my application was shown by my results.  What I don't know is what part of my application made it a better application than I had expected it to be, and this data will at the very least, provide some insight there.

 

It's extremely disconcerting, but here's a piece of evidence suggesting that your guess might be right: Some departments think that the importance of the writing sample should be decreased: http://leiterreports.typepad.com/blog/2014/02/phd-admissions-writing-samples-and-ma-programs.html

 

Fortunately, the overwhelming majority of the 81 responses posted publicly on the Leiter Blog seem to suggest that most people in philosophy (including profs) disagree with that opinion [in italics] and the justification provided for that opinion.

[ianfaircloud]

Now that it's May, I'm checking in.  We received 101 responses, though some of these responses are incomplete enough to be excluded from the data.  In any case, I'll post the raw data soon.  I'll give a few thoughts and ask others to give their thoughts, as well.

[AcademicX]

hey Ianfaircloud, are you posting the results soon? I'm very curious to see them/I need something to do after getting done with college...