Why do they need an admission committee ?

Applying to College
101

<p>

</p>

<p>But that’s precisely the point: adcom officers who simply doesn’t want to admit somebody - perhaps because of prejudice - can always simply claim that they found the essays to be repulsive, and nobody can really challenge them. Similarly, if they want to admit somebody who happens to be the child of their friend, they can claim that his essays were ingenious, and nobody can challenge that. I remember movie trailers claiming that Gigli or Battlefield Earth would be the ‘blockbuster movie event of the year’. </p>

<p>As I heard recently, scientific knowledge - if narrowly defined to simply being journal published scientific papers - is simply 3 people agreeing: the journal editor and 2 out of the 3 peer review referees. Similarly, college admissions at schools that run purely human-based admissions decisions consists simply of a handful of adcom officers agreeing, or sometimes even only a single (head) officer deciding, and for reasons that nobody can ever be entirely sure about. </p>

<p>Hence, while I can agree that computerized algorithms may not be able to reliably judge subjective works such as essays (at least not yet), let’s face it, people don’t do so either. Shakespeare was not considered to be a truly great playwright until long after his death. Citizen Kane, now widely considered arguably the greatest movie ever made, languished in obscurity for years after its release until its fortuitous rediscovery, first in Europe and then when shown on US television.</p>

102

<p>remembering that the “holistic” system was born as a tool to discriminate against jews, then how can we even think that the same system is not still used as a discrimination tool.Maybe it is not a trivial discriminination against a race or a religion but a more sophisticated or subtle objective such as “shaping” a class.
A computer program would be less prone to let the adcoms friends’ children in</p>

103

<p>Letters of recommendation were quite openly born as a tool to let the friends of the professors in; does that mean that they should be abolished too?</p>

104

<p>LOR’s perhaps shouldn’t be abolished, but modified to select a random sample of a student’s teachers, not just those particular teachers that the student cherry-picks because he knows he’ll receive a top rec. If a student happens to have a a horrific relationship with all of his teachers except three, he’ll surely obtain letters from those particular three. Heck, some students deliberately and strategically cultivate relationships with only three teachers, knowing that that’s all they need.</p>

105

<p>So if I don’t happen to excel in a subject, I still might have to go ask that teacher to write a letter about what a good student I am? Why is this a good idea?</p>

106

<p><so if=“” i=“” don’t=“” happen=“” to=“” excel=“” in=“” a=“” subject,=“” still=“” might=“” have=“” go=“” ask=“” that=“” teacher=“” write=“” letter=“” about=“” what=“” good=“” student=“” am?=“” why=“” is=“” this=“” idea?=“”> It may not be a good idea to you, but it is to the admissions committee. They want to weed out people who aren’t good. People who can’t provide strong recs.</so></p>

107

<p>So because I excel in math and science, but am mediocre in english, I’m no better a student than someone who’s just mediocre in everything? I’m sorry, but I just don’t accept this idea that everyone needs to be “well-rounded”.</p>

108

<p><so because=“” i=“” excel=“” in=“” math=“” and=“” science,=“” but=“” am=“” mediocre=“” english,=“” i’m=“” no=“” better=“” a=“” student=“” than=“” someone=“” who’s=“” just=“” everything?=“”></so></p>

<p>No. Just get your rec letters from those subjects you are good in. But make sure the teacher knows you well.</p>

<p><i’m sorry,=“” but=“” i=“” just=“” don’t=“” accept=“” this=“” idea=“” that=“” everyone=“” needs=“” to=“” be=“” “well-rounded”.=“”></i’m></p>

<p>Good, because colleges don’t want well rounded students. They want someone who excelles in one area instead of being Jack of All Trades. It’s a misconception that colleges want well-rounded students.</p>

109

<p>I’m… not sure what we’re arguing about here. I have no problem with the existence of letters of recommendation, just the idea that they should be changed such that you have to get them from randomly selected teachers.</p>

110

<p>Don’t forget that LOR’s aren’t just about picking teachers who give you the highest marks. It’s more about commenting the applicant’s the character and the personality. </p>

<p>Though I do agree that getting a random sample of LOR’s isn’t such a great idea. It’s not just about having bad relationships with certain teachers; certain teachers simply don’t know how to write a good LOR, and that’s out of the applicant’s control. And some teachers even let the applicant write the LOR for them (or at least let the applicant tell the teacher what to write). </p>

<p>What I do think is that LOR shouldn’t have as much weight as the other components of the application. They’re just too… less representative of the true applicant than the other components.</p>

111

<p><i have=“” no=“” problem=“” with=“” the=“” existence=“” of=“” letters=“” recommendation,=“” just=“” idea=“” that=“” they=“” should=“” be=“” changed=“” such=“” you=“” to=“” get=“” them=“” from=“” randomly=“” selected=“” teachers.=“”></i></p><i have=“” no=“” problem=“” with=“” the=“” existence=“” of=“” letters=“” recommendation,=“” just=“” idea=“” that=“” they=“” should=“” be=“” changed=“” such=“” you=“” to=“” get=“” them=“” from=“” randomly=“” selected=“” teachers.=“”>

<p>Hell no. The idea of recs is so that they can get a sense of THE BEST YOU. Randomly selecting teachers would be unfair: someone who sucks compared to you might get a good rec while you end up getting a sucky one. </p>

<p>And why do you want them to be random? Don’t you want your best self to be portrayed to the admissions committee instead of being random?</p>
</i>

112

<p>I don’t want them to be random…</p>

113

<p>Then could you please elaborate on <…just the idea that they should be changed such that you have to get them from randomly selected teachers>?? Seems like you want them to be random.</p>

114

<p>No, that’s the idea that I have a problem with. I’m fairly sure that if I wanted them to be random I would know about it.</p>

115

<p>There, problem solved. Misunderstood. Sorry.</p>

116

<p>No problems.</p>

117

<p>

</p>

<p>

</p>

<p>I think your problem is with me: I was the one who proposed the idea, and I still endorse it. </p>

<p>

</p>

<p>I fundamentally disagree - adcoms don’t want to get a sense of ‘the best of you’ (or, at least, they shouldn’t). Instead, they should want to get a sense of the true you. </p>

<p>Consider an example. I can probably find a handful of people who think that Gigli is the greatest movie ever made. I can probably also find a few people who think that the Godfather was the worst movie ever made. But to pick out only those opinions would clearly present a tremendous distortion of what the true quality of those movies is. </p>

<p>Consider another example. Would you trust a medication if you knew that the pharmaceutical firm simply ran multiple clinical trials over and over again and was only showing you the results of the trial that demonstrated a result, and was not presenting you the data for the other trials that displayed no result (or won’t even tell you about those other trials at all)? I think we would all be deeply skeptical of any medication whose efficacy was ‘proven’ in this manner. That is precisely why the FDA demands to see the results of all clinical trials, not just the firms’ favorite trials. </p>

<p>

</p>

<p>That’s an issue that all studies involving sampling inference are faced with, and can be ameliorated with, say, increasing the sample size (i.e. asking 10 randomly selected teachers of yours as opposed to only 3). </p>

<p>But what’s far far worse is allowing the subjects to choose their recommenders, for that’s nothing more than sampling on the dependent variable. Then of course the vast majority of recommendations are going to be positive, simply because students will be able to cherry-pick them. </p>

<p>Again, consider the movie analogy. If the producers of Gigli were able to pick out their own reviewers, then they would surely pick only that particular (tiny) group of people who happen to believe it’s the greatest movie ever made, hence providing a deeply misleading view of the true quality of the movie. </p>

<p>The real question seems to be: what is the true purpose of LOR’s? Is it to provide a true view of the candidate (subject to sampling size error), or is it merely to serve as an advertisement for the candidate. If it is the latter, then one has to wonder whether it serves any real purpose at all. If I want to know which movies are worth my time watching, I am not going to take the word of the movie’s directors or producers, because obviously they are always going to tell me that their movies are stellar. Nor would I rely on the opinions of those particular ‘critics’ that the producers have specifically nominated. The conflict of interest is obvious.</p>

118

<p>I have to disagree. Your GPA, test scores, and (to some extent) extracurriculars are the things that give adcoms an unbiased view of how good of a student you are. The letters of recommendation are, like the essays, SUPPOSED to be basically an advertisement for you.</p>

<p>And the fact is, whether or not that’s an ideal system, it’s how the world works. In a job interview, you tell the interviewer why they want to hire you. As a salesperson, you tell the customer why they need your product. You just don’t get anything done by presenting the facts and letting other people decide.</p>

119

<p>I have two comments to make. </p>

<p>First, on the letter of recommendation issue. I do believe that where stats and hard and fast achievements speak to what is there and what is lacking at times, the focus is on what is there. And certainly the LOR is a bonus that is meant to distinguish a candidate. Thus, allowing recommenders to be handpicked is good. </p>

<p><em>HOWEVER</em> the issue is how much to factor in the recommender’s statements. I think the problem is that some random teacher going on about a student’s personality should probably be taken with a grain of salt, and only a teacher with a reputation for really churning out good courses and students and with some maturity going behind his/her statements should be taken seriously. Goodness, with graduate schools, it goes as far as only tenured, famous faculty’s words being taken extremely seriously, when many of the postdocs are brilliant people with exceptional credentials to assess students aspiring to be researchers.</p>

<hr>

<p>Second remark: I do think a computer can do quite a bit as Sakky seems to suggest. It would be a complex role-filling mechanism, rather than an absolute scoring machine. The biggest problem I have with admissions as it’s done is that almost any decision can somehow be justified because some adcom “saw the whole application” when someone else didn’t. This is ludicrous to me, as it always has been, because realistically there is only so much one can put on a piece of paper to submit to a person reading tons of applications. What I’m in favor of is an application that throws out the essays used to psychoanalyze character through asking unconventional questions, and in favor of a simple “What do you want out of our school” type of essay as the only thing an adcom has to read, backed up with credentials saying that this is something the student could plausibly want, which can be assessed in a pretty impersonal fashion. </p>

<p>In the end, the impersonal fashion can be achieved by humans to a good enough degree, and I think the randomness perceived has to do with silly means of assessing a student’s character in place with the current system. A well-done mechanical process could probably still fill roles, but would probably psychoanalyze a good bit less, and frankly, I have almost no faith in psychoanalysis done through a piece of paper – takes more than that, and even then it’s difficult stuff.</p>

120

<p>

</p>

<p>Then what purpose does that really serve? If I watch a movie trailer that promises “the white-knuckle thrill rider of the year”, should I believe it? According to certain beer commercials, merely by drinking their designated brand of beer, a bunch of gorgeous bikini models will instantly appear and think I’m the most desirable man in the world - should I believe that? Let’s face it: advertising is little more than ‘managed lying’. Surely none of us were born yesterday; we’ve all learned to discount the veracity of advertising. We all know full well that companies are not really telling us the truth through advertising, but are merely presenting information that places them in the best possible light. </p>

<p>So if I want to know whether a particular restaurant serves good food, of course I am not going to ask the restaurant itself, nor will I trust the opinions of those people who the restaurant specifically nominated. We all now that they are unsurprisingly going to tell me that the restaurant is unsurpassingly fantastic. I would want to obtain a random sample of those who actually ate at the restaurant, and then place bounds on the obtained results that depend on the sample size I chose. </p>

<p>If the goal of LOR’s is merely to present a cherry-picked advertisement of the candidate, then that means that they’re not meant to be taken as an objective characterization of the candidate, and everybody knows that they shouldn’t be taken as such. Hence, they serve as a game to be played between the candidate and the adcom: the candidate aims to present a deeply misleading portrait of himself through the LOR’s he presents, and the adcom knows that the presented portrait is deeply misleading. </p>

<p>

</p>

<p>No, that’s not how the world works. That’s why movie recommendation websites such as Metacritic and Rottentomatoes exist to provide a more objective view of the true quality of movies: certainly far more objective than what the movie studio (or its ‘favorite critics’) would propound. That’s why Consumer Reports and other product assessment organizations exist to determine the true quality of any product you may wish to buy, as opposed to simply relying on the dubious opinion of the salesman in question. If I want to know how popular Barack Obama’s policies are, I am not going to trust the words of anybody in his administration. I am going to rely on Pollster or other organizations that obtain random samples of the opinions of average Americans.</p>