@ucbalumnus , you critique of the USSR is misguided. During the early years after its creation they only admitted children of workers and peasants to colleges and later that kind of background was still a big hook. And they had an affirmative action too - they has reserved spots for ethnic minorities. Very similar to what we have in the US colleges today.
How is what you describe of the USSR similar to US colleges today when US college enrollment is heavily skewed toward kids of the upper and upper middle classes, rather than “workers and peasants” that you say the USSR only admitted students from?
“Lab rats and singular mindedness gets a bad rap like the kids are anti social or some other connotations.”
Agree, but note that athletes, who get the largest admission preference are pretty single minded, the recruitable ones in high div 1 say, are singularly focused on improving themselves during the whole year, whether it’s the AAU and club for basketball or summer weight training for football.
“Picking a class should include some of these high stats stats too.”
It depends on your definition of high stats but at all selective colleges, most of the kids have high stats. If the ACT 25/75 band is 31-34, then 75% of the class is above a 31. If your definition of high stats is 34, then it’s 25% of the class.
@mdphd92, this is helpful. @Tanbiko, actually trying to patch together a list of reach/target/safe which is really driven by my son’s interest but also maximizes potential acceptances. Obviously there’s thousands of potential schools. Filtering for size, setting and location there’s still hundreds of schools. We can’t visit all of those, so we’ll read about as many as possible to further refine that list.
My thought (and it might be misguided or flatout wrong) is if we can run probability scenarios to figure out the most positive outcome of a reach/target/safe list, so we could narrow our list to that and concentrate on those schools.
On the other hand, the concept of knowing a college’s student body well enough to also predict a potential outcome is also interesting to me, but we don’t have that kind of time or money.
But the discussion here has pointed out that there are other “hidden” factors which can’t be accounted for: LOR, other applicants EC/essays/rigor/geography/etc making a prediction tough (however, that’s compensated with knowing the student body…). So, it seems to me that the BEST way to predict a positive outcome is to find true safe schools first which my kid loves and wants to go to and then build from there.
What a waste to apply scattershot, hoping some top school needs a bassoonist. And assuming you’re the only one, and/or the college is so desperate that they’d take ‘anyone’ with “nothing special” except one aspect.
That’s not the level of wisdom/smarts high reaches look for. It’s dismissive of a very real qualitative filtering. If your kid doesn’t have the academic level, pattern of actions, thinking, no small tip will overcome that for non recruits.
OP, you don’t need exhorbitant time or money to do research. One does need savvy. It’s not the “student body.” While tgere are going to be attributes in common, the students will show variety. You need a read on those qualities, what the colleges say. Or you’re wasting your app fees. Nor is the selection process all about institutional needs. Those form final choices, are not exclusive of “match.”
This idea there’s one trump card- bassoonist, some award, how your own hs loves you, etc- is what leads to so many kids being disappointed in late March.if you don’t care about the process, aim lower.
@NewEngParent Since you found my post #78 helpful, I’ll continue with some more thoughts.
- The idea of conditional probability explains why CC has so many "Chance me" posts. Students are, in fact, trying to estimate their probability of acceptance to a particular college, given their individual statistics, EC, and background. That information, if accurate, would help them decide whether to apply and how many colleges to apply to. The reason people often consider those threads a waste of time is that (1) the information given is incomplete, and (2) nobody really knows how admissions committees make their decisions, so the confidence interval for the probability estimate can be wide.
- Naviance statistics, if your school provides them, are useful, because they control somewhat for socioeconomic background and the rigor of your school in the eyes of the admissions committee for a given college. Of course, we don't know the individual background, ECs, letters of recommendation, or essays for the applicants, but school-specific statistics are more helpful than the national statistics, because they are better estimates of the conditional probabilities.
- The decreasing admission rates at highly selective schools over time explains the phenomenon of why students are applying to more of these schools over time. This is entirely rational, since they are trying to maximize their probability of getting into one of these schools.
- Probability is not the only consideration. Expected utility is really the quantity that people are trying to maximize. Otherwise, everyone would just apply to safety schools. The reason students apply to "reach" schools is because they have higher perceived utility for that student. This is why the standard advice is to make sure you have a high-probability event, which is a set of "safety" schools with sufficient utility that you would be happy attending one of those schools.
- Weighed against the expected utility is the cost of applying, in terms of both time and money. The application fee represents the monetary cost of applying, and supplemental essays represent much of the time cost. The Common App has made it easier to apply to additional schools, so that has eliminated the cost associated with duplicating the main essay.
- The utility question comes into play on CC in posts about "school X vs school Y", even after the student has been accepted to both, so probability is no longer an issue. In those posts, the student is trying to determine which school has the higher utility, and the cost of attendance, expected job prospects, and student life are components of utility. Of course, each person has different perceived utilities, which is why we see different opinions.
- Early Decision strategies are a trade-off between probability and utility. Students will apply to a school with lower perceived utility because their probability of acceptance is higher in the ED round.
IMO it’s more than that. To prevent students from comparing recs with each other, for one.
Maybe so. But my memory is that my kids’ school teachers pretty much insisted on the waiver.
You dont increase your chances by applying to more. Rather, applying with more savvy, more info, a better sense of what makes YOU a match to what they want, then offering a bang up app and sipps. Not resting on kids on Naviance who applied before you, their full presentation unknown.
It doesn’t make sense to apply to more reaches, because they’re now more competitive. Imagine a kid without the stats or ECs- or other issues- thinking that’s all it takes. It’s so incomplete. It’s not the level of thinking and processing high reaches want.
Sure, many families insist there’s nothing they can do, that it’s only a crapshoot. Don’t think so narrowly. You’re safer trying the right steps than assuming.
For a kid who is a reasonable applicant, though, there’s probably an inverted U curve to applications. Going from 0 to 1 clearly increases your chances. IMHO going from 1 to 2 does, as well. Going from 10 to 20 probably decreases them, because there’s less time available for each additional one.
The question is, where’s the sweet spot at the top of the inverted U?
I’d rather kids and families invest more into learning and processing what makes one “a reasonable applicant” than playing with numbers and speculating. Most kids don’t know more than their stats/rigot/ECs and their own high school greatness. It takes more.
Sure, there are perfect kids, perfect apps/supps and you know those kids will have many tippy top admits, in the end. But that’s so few. It makes playing with probability just an acadeic exercise.
@lookingforward I think we can agree that candidates should not apply to just one school. I also think we can agree that candidates should of course try to maximize their probability of getting into a certain college (as you say by finding a “match” and being savvy).
We also agree that if the conditional probability of acceptance to each reach school is near zero (what you mean by a kid without the stats or ECs), then applying to more of them won’t help the overall acceptance rate much, since it will still be near zero.
Also, if the conditional probability of acceptance to each reach school is near certainty (like the Houston senior who got accepted to 20 reach schools), then applying to more of them won’t help, since the overall probability of acceptance is still essentially 1.0.
But there is a middle ground, where the conditional probability of acceptance is reasonable but not certain. The middle ground (which is what people loosely mean by “matches” as opposed to “reaches” or “safeties”) is going to be different for each candidate. In that case, I do think it is rational for that candidate to apply to multiple colleges where he or she has a reasonable chance of acceptance, depending on the effort and cost involved in applying, since there is always an element of randomness, even if you think you’re a pretty good match for a given college.
Sorry, but my head spins when folks try to make some mathematical formula out of a very qualitative process. That Houston kid, eg, had no certainty until his results came in. That’s the rub. There are so many ways to blow an app to a high reach with thousands of top applicants to choose among. Your wildcard is your actual presentation in the app and supp, not the surface details. Those are just the bones, the rest fleshes you out. Or not.
I think “randomness” is the wrong word or thought. When a top college has a type they seek (more than stats, some ECs,) you either show that or not. The issue after passing that muster is how instituional needs play. Maye they need X majors, aim for geo diversity, need a tuba player or you have a strong side interest in continuing your music sideline and the music dept is interested-- all things you cannot control for, much less predict. Instead of random, I see it as “beyond your control.”
I told mine, after good matching and apps/supps, a real personality match, that their chances were now 50-50, get in or not. The final decisions were beyond their control. Rather than stats (and we never looked at Naviance,) we used the calmom approach. It worked. It could not have been predicted via numbers.
“The question is, where’s the sweet spot at the top of the inverted U?”
Totally agree with the theory of the inverted U. The sweet spot will vary by kid, though. I think the important variables that would determine where the sweet spot is for any particular kid include:
- How quickly and well the student researches. Doing the research to figure out what a certain school culture and fit plus what that school values in admissions takes time. The better and faster researchers would have more ability to do this well for a number of schools, where less experienced or slower researchers might do best if they stick to a few schools so they can do their best work.
- How quickly and well the student writes essays. Great essays incorporate some of the key points that the college is looking for and since that varies by college, so will the essays (at least slightly, there are obviously some things that would be valued universally.) Especially important is doing a good, targeted job on the "Why this college" essays.
- Commonality of the target colleges. Using the top 10 colleges in the USNWR as an example, those are all outstanding colleges but some value different things. So if a kid applied to all the top 10, to maximize chances, some customizing of essays and ordering of the app would be helpful since those colleges value different things. To the extent that a student is applying to colleges that are looking for similar things, less work will be needed in customizing.
My kid is a good researcher and very practical, but takes a long time to write good essays. It was helpful for him to group the colleges into ones that were looking for similar things and knock off a those apps at the same time then do the outliers later. He was not a kid who would have done well with 20 apps since the quality of his writing went down as he got “over” the process. Think he ended up with 7 or 8 really good, targeted applications which was great for him, so using the inverted U theory - his sweet spot was 8. YMMV
I think that a mathematical perspective and estimation of probabilities help, not because they are precise, but because they can guide our decision-making process. Also, they do explain human behavior to some extent. All of these notions about “reaches”, “matches”, and “safeties” are very informal ways of thinking about probabilities and utilities. If it helps to think about these concepts in a qualitative sense, then that’s fine.
I agree that the notion of “randomness” is from the perspective of the candidate, not the admissions committee, which is what I think you mean by “beyond your control”. Perhaps the word “uncertainty” is better. Even if a given admissions committee had a deterministic formula, the fact that we do not know the formula makes the probability for the candidate between 0.0 and 1.0. If we did know the formula, then we could act accordingly, and apply to a single college.
The meanings of probability and uncertainty have a long-running debate in the mathematics and statistics, ranging from a frequentist to a Bayesian interpretation. For decision making, it probably makes sense to think of probability in the Bayesian sense, which is our belief about how likely something is to occur. Given that belief, even if it is not entirely accurate, we can then act in a way that is rational or not.
“Sorry, but my head spins when folks try to make some mathematical formula out of a very qualitative process.”
Ok but admissions is by definition a numbers game, it involves all sorts of acceptance rates, ED, RD, EA, SCEA and at the end you have to maximize your chance (again a number) of getting in to your top colleges. I don’t think anybody on cc was surprised by the Houston kid, if we had know about him before the news story, most of us would have predicted he would have gotten into everywhere he applied.
The inverted u-curve is a really good way to look at it, agree with the other posters.
If an applicant applies to two colleges whose admissions are completely correlated (i.e. their criteria and processes are identical), then the probability of acceptance by either one is not increased at all. In reality, however, the two admissions are not fully correlated (i.e. the admission by one doesn’t guarantee admission by the other) and the joint probability of acceptance by either one is higher. The less correlated the two admissions are, the higher the joint probability. To maximize this joint probability, an applicants should choose a small set of less correlated colleges. As others have said, if the set is too large, the probability of acceptance to each individual college would likely suffer due to limitation on resources, which in turn negatively affects the joint probability of acceptance by any one college.
However, this may not result in the desired improvement in admission chances, if the additional schools are less correlated because they emphasize applicant aspects that the applicant is weaker in, with less emphasis on aspects that the applicant is stronger in (presuming that the applicant already has on the list those schools which favor the applicant’s strengths).
For example, if a California resident who is GPA-heavy (relative to test scores) has several UCs on his/her application list, adding USC may not improve chances much, since USC tends to favor test scores. But a test-score-heavy applicant may benefit from adding USC to a list that otherwise is made up of UCs, which tend to favor GPA.
@ucbalumnus Yes, if the probability of acceptance to an individual college is zero, adding that college, whether correlated or not, to the set obviously won’t improve the joint probability. Applicants still need to find colleges that “fit”. Assuming s/he can find a number of colleges that “fit”, s/he should apply to a subset of these colleges that maximize the joint probability while minimizing the amount of resources required, instead of applying to every one of these colleges or to a less than optimal subset.
@lookingforward “…contrary to CC’s belief, it’s not so much something that “impresses” adcoms as holding to a standard or the standard expectations.”
I agree that it is reasonably structured, which points to a specific expectation as opposed to an unstructured impression. Still, if two applications grade out similarly, the Adcom may believe that one of the two candidates is just more compelling. Something about the aggregate of the application made more of an impression.
None of this is easy. And if college A is overwhelmed by X sorts and B needs them, chances shift. On some thread, this or another, calmom explained her experience researching dept needs. In ways, you can try to get inside the heads of adcoms. As opposed to having developed intuitive insight into what your hs teachers and admins like.
Becuase no one outside admissions has the broad insight into what comes across in apps/supps, just going by numbers is so incomplete. If two apps grade out equally, yes, one may be more compelling. But that’s based on what’s in the app- not a summary of stats or rigor, Naviance, the bare bones in a Chance Me thread. And those institutional wants. It’s possible, eg, that the one chosen, Much2Learn, wants and fits an underenrolled major or geo area or is at a high school they want to encourage or a host of other uncontrollables.
So saying a kid perfect for, say, Stanford, may as well apply to USC misses not only possible different priorities, but also factors in the pool itself. You can’t know that.
That’s part of what I told my kids. You did your best, but we can’t know who else is applying, from our town or area, from some needed state, with some edge you don’t have. Or if lots of kids apply for your major this year or many with your pattern of ECs.
One reason I nag to research what the college does look for/value, is because so many kids don’t. Instead, they rely on simplistic advice or assumptions like you need national awards or “passion” or spike or they don’t want rounded or they do or they only fill RD with URMs or all sorts of bosh.
And a holistic truth, for tippy tops: any flaw in yhour app/supp could doom you (or push you much down the list.) A “perfect” candidate is in the execution, not just the obvious ingredients.