
PP Here: Stanford should read 18
Brown should read 8 |
And indeed these are strong numbers. I had previously found this information and but haven't analyzed it yet. It certainly will be included when I set up the non-NYC day school page (soon, I promise... and there will be as many DC area schools on it as I can manage - though not all at the first appearance of the list).
There actually is some additional information on the NCS website that helps fill in the picture. Elsewhere, it lists all the schools that at least ONE student matriculated at during 2006-2009. This doesn't completely nail down the NCS matriculation analysis, of course, but it does help me performa a better analysis. Of course, I always prefer a school to just give me the full data and account where every student matriculated over a specified period of time. On the other hand, incompete data like this can be sort of intriguing to determine the best way to use it (or even if it's usable) so that distortion of results is kept to a minimum. There are a number of schools that provide information of the sort: These are the schools that at least N of our graduates attended over the last so many years. Combined with a good estimate of the size of the graduating class, that's usually good enough as long as N isn't too big. The extra info that NCS provided helps even more. I do appreciate your posting the information. I'm still not so tuned in to the schools in this area and a surprising number of excellent schools don't seem to post adequate data (St. Alban's, Sidwell, Georgetown Prep). |
DC area schools that have now been included on the website http://www.matriculationstats.org/day-schools-outside-of-nyc are:
Holton-Arms School Landon School Maret School National Cathedral School St. Alban's St. Anselm's Washington International School Also, Episcopal High School is included on the boarding school page. Georgetown Day School, Potomac School and Sidwell Friends did not make either any or adequate data available on their websites to be included. Any other suggestions for schools to be included? Can anybody help with providing data for schools that I don't have it? I check in here occasionally or can be contacted via the website. |
It is simply do you want a small college or a large university. Turning down Yale for W&M makes no sense. One might say it was financial. But why not UVa. Both are great but UVa is better on the average. |
MatriculationStats, thanks for the link to your website -- it is really informative and clearly took a lot of work. I will make one comment to those perhaps too prone to assume admission to a school with good matriculation statistics means your specific child's path will be that much easier statistically. Computer programmers used to talk about "GIGO": "Garbage in, garbage out." I would change this to "DIDO" "Diamonds in, Diamonds out." Bright kids attend these schools, and they get the good grades and high standardized tests scores needed for college acceptance (and throw in a little bit of athletics in lacrosse, crew, squash, etc). Your kids will get great educations at these DC independent schools, and they will most likely do very well in college -- but what college they get into does, I think, depend on what they bring along at the start of the process. |
PP, no one's disputing that a student needs to be well-qualified (with good grades, high test scores, and perhaps athletic prowess) to be admitted to a top college. But I think the point is that a high school's college admissions success suggests it can provide an environment that allows such students to capitalize on those gifts.
Lance Armstrong will be fast no matter what bike he's riding, but you can be damn sure he's riding a top-of-the-line bike when he's in the Tour. |
PP, maybe the right analogy is not which bike he's riding, but with whom he's riding. Bright kids are stimulated by other bright kids, and teachers can teach to them differently. It's no surprise that those kids then matriculate disproportionately to selective colleges. Even if my kid can be one of the "top" students at a less selective private or a non-magnet public, I fear he will receive an inferior secondary education without the benefit of a quorum of intellectually curious peers. So the question is whether these matriculation figures really reflect where the brightest kids go to school, or if they reflect other factors, such as legacy, wealth, athletic recruitment, etc. |
That is so true. Has anyone taken a statistics class? There are lies, lies and statistics Are you really measuring what your childs success will be based on those figures |
First, thanks for the compliment. Yes, it has taken a fair amount of work by both me and my 8th grade son. As to your comment of DIDO, sure, the raw material (whether it be mental, physical or other characteristics) that the entering student brings with them is the most important factor that enables them several years later to gain admission to a selective college. And to a large extent, the lion's share of the results that you see on the website for each school are determined by the students admitted who choose to enroll at a particular school. Nonetheless, I do feel that if a school is doing its job properly it does add value to its students during their time on campus. That value can be added in many different ways, some of which have been mentioned above. Associating with other bright students, having the benefit of motivated teachers who have the time (because they aren't stretched too thin) to devote to individual students, access to numerous resources (academic or athletic), college placement counselors who can help manage the system to help place the students one notch up than they otherwise would be entitled. I'm sure there are other perfectly valid reasons, but I think I've given some sense of the flavor of the possibilities. A number of people have suggested that I should just account for such factors as legacy status, et al. Well, it just isn't that easy to do. I don't have access to adequate data to perform that kind of analysis. And even if I did, that kind of analysis isn't always so straightforward. In lieu of that, I've provided a summary of basic data and leave it to the user to interpret it correctly for their own purposes. |
MatriculationStats: Do you have data on SAT scores for these schools? Seeing how well differences in SAT scores predict college placement should enable you to say something about the role of these other factors. |
Matriculation Stats: Your website is terrific, but I would caution you on your methodology: You include Top 25 National Universities and only Top 15 Liberal Arts Colleges, yet the #25 National University (UCLA, Overall Score of 73) is actually rated lower than the #33 Liberal Arts College (Trinity College Conn., Overall Score of 74). You might want to stick to Top 25 for both categories, or just use only all "Overall Scores" over 70 (you could call this cut-off the "passing grade" for top-school consideration), for example (all from the 2010 US News Rankings). Obviously, one can slice and dice the data however one pleases; it just might make more sense to compare apples to apples in the two categories, rather than use arbitrary and different cut-offs that don't correlate, just to include more "larger" national universities...You're nevertheless still performing a useful service, and I commend you... I'd be interested in how many unique visitors have come to your site. |
I'm not exactly sure where you would be going with such an analysis. Let's assume you have SAT information for all the schools and you can determine how well SAT scores predict college placement. Then each school presumably does somewhat better or worse than that prediction and you ascribe that difference to something. Fine. But what is that something? It's everything else which includes GPA/Class rank, legacy status, prestige of the school, athletic ability, other special skill or talent, etc. But what have you discovered? I'm not really sure. It seems that you still need some more data to make some meaningful conclusions. |
Agreed. But suppose you plot college placement on the y-axis and SATs on the x-axis, and then put a regression line. Then the regression line is a benchmark (assuming that you have enough data points, which might be an issue). For each school you can then find the deviation from the benchmark --- how it has placed students, relative to predicted given the students SAT. What do the deviations look like? Are the schools at the top or bottom of your lists there because they have students with SATs that look different from those of schools in the middle (selection effect) or are the SATs all similar but all the other factors kick in (better academics and athletics or more legacies etc). |
I am certainly aware that my methodology is not perfect. I don't believe the perfect methodology can exist for this task. And the part of the methodology that has been taken to task most has been the decision to use the Top 25 National Universities and Top 15 Liberal Arts Colleges for the "Top Schools" (and something analagous for the "Strong Schools"). I describe that on the home page of my website. I felt it very important to attempt to unify those two rankings lists in some reasonable way. I believe I accomplished that. Of course, there was an element of arbitrariness to those choices. The issue with the Overall Score that you point out may isn't quite the problem that you seem to think it is, though. The US News Rankings assigns a score of 100 to the top school in each of the National U. and Liberal Arts categories and then works down from there relative to the top schools. But I'm not assuming that the top school in each category is identical. For better or worse, I'm placing more value on Harvard and Princeton than Williams. You're free to disagree with that, of course. And yes, even after making that assumption, there are some other issues in the methodology. As I said, I admit it's not perfect. The reaction I've generally received, however, is similar to yours - appreciation for creating something more or less reasonable. As to the visitors to the website, through yesterday there have been about 1,100 unique visitors (thank you google analytics). Upon setting up the website, I wasn't sure what to expect. 79 unique visitors have come to the website through links on this discussion forum. |
9:18 the admission officers at the top 25 wish they could read files during the morning or work day but actually do that from the time they get home until their eyes won't stay open and all weekend long. It is too busy during the day to concentrate and get that done. When you are reading thousands of files a really unique essay just might wake you up and get your attention and you will go to bat for that child when the committee meets. I love to hear the stories from my daughter and her coworkers about the things that really get their attention after midnight. Often they remember these students from the months they were on the road recruiting then are thrilled to see what they send in. There are so many impressive students all over the country they wish they could offer admissions to more than they do. |