Sunday, April 30, 2006

Newsweek's Top 1000 High Schools

The Complete List: 1,000 Top U.S. Schools

Public schools are ranked according to a ratio, devised by Jay Mathews, that is the number of Advanced Placement and/or International Baccalaureate tests taken by all students at a school in 2005 divided by the number of graduating seniors. All of the schools on the list have an index of at least 1.000; they are in the top 5 percent of public schools measured this way.


That seems a ridiculously simplistic way to rank high schools across the country. I thought it might be a fun little exercise to identify all the schools on this list that are in the Sacramento area, and list both their Newsweek rank and their state-calculated Academic Performance Index (API). Here goes:

Newsweek Rank........School...............API (higher is better)
(lower is better)
241........................... Mira Loma........... 750
589 ...........................Davis.................... 845
672........................... Granite Bay ...........814
702........................... Rio Americano.......777
727............................Kennedy................704
893............................Elk Grove..............761

Admittedly that's a very small sample size. California had dozens of schools listed in the top 1000, though; it would be interesting to see if there's any correlation between Newsweek's ranking, which looks only at AP and IB tests, and a school's API, which factors in performance data from all students.

Did I miss any nearby schools? If so, let me know and I'll add them.

5 comments:

KauaiMark said...

(from our area in CA)
Newsweek...School...........API
--------------------------------
217.....San Jose HS.........557
261.....Leland HS...........819
420.....Silver Creek........742
472.....Lynbrook............472
684.....AP Hill.............654
784.....Pioneer.............784
795.....A Lincoln...........673
(nr)....Evergreen Valley....813

Anonymous said...

Actually one big problem with the methodology is that schools with lots of hispanic students will score artificially well because of the Spanish AP test. I suspect that this might explain a lot of the non-correlation (e.g. Lybrook HS with an API score of 472 being in the top 500 public high schools in the US).

Two more obvious problems:

1) AP tests *taken* count, so a high failure rate can still generate a high school rank.

2) Actual college courses taken don't seem to count, so taking a US history class at a local college (and passing and being able to transfer the units) doesn't count while taking the AP test (even if not passing it!) does.

-Mark Roulo

Anonymous said...

I think that the Lybrook above might have an API score of 892:

http://www.lhs.fuhsd.org/school_info/

-Mark Roulo

Onyx said...

I know two of the Sac area schools. Mira Loma and Rio, way to go! I'm an Encina baby myself

Ellen K said...

I laughed and laughed at the Newsweek article. Being a Dallas native who has pretty much been in education in one way or another since childhood, I found the listings hilariously skewed. I wouldn't argue with most of the top two hundred, but after that there are a raft of local schools who manage to run kids out of the schools via dropping out and such which makes the key number-graduating class number-a diminished chord at best. I saw this up close when my daughter, who had been top ten percent (a key factor in our state...)until senior year, when due to the many MANY dropouts she ended up in the top TWELVE percent even though her actual grades were higher her senior year. So go figure. I saw some glaring omissions on the list of schools that work hard to retain students and still have solid AP programs. I also saw some schools that I would never teach in, send my kids to or even recommend to my worst enemy that were on the list. So as with most polls and statistics, this was slanted towards making as many of the entries urban as suburban, and whenever you do a poll, which this is, with an outcome in mind, then it's only a list that is as good as your numbers. In this case, I really think some of the numbers bear some second checking.