Tuesday, September 14, 2010

API scores are in

Here's a link to the latest API scores:



  1. There are 42 elementary schools w/API over 800:

    Elementary Schools 2010
    Alice Fong Yu 947
    Clarendon Alternative 944
    Sherman Elementary 941
    Robert Louis Stevenson 923
    Lawton Alternative 923
    Alamo Elementary 923
    Ulloa Elementary 914
    West Portal Elementary 904
    John Yehall Chin (Elem) 901
    Claire Lilienthal 901
    Yick Wo Elementary 897
    Jefferson Elementary 893
    Lafayette Elementary 891
    Sunset Elementary 890
    Argonne Elementary 888
    George Peabody 886
    Sutro Elementary 885
    Rooftop Elementary 884
    Francis Scott Key 882
    Commodore Sloat 882
    attan Elementary 873
    Miraloma Elementary 865
    Dianne Feinstein 861
    Jean Parker Elementary 859
    Edward R. Taylor 855
      Spring Valley 846
      George R. Moscone 843
      Frank McCoppin 841
      Alvarado Elementary 836
      Gordon J. Lau 835
      Garfield Elementary 835
      Visitacion Valley 826
      Sheridan Elementary 825
      McKinley Elementary 822
      SF Community 816
      Jose Ortega 811
      Harvey Milk Civil 810
      Monroe Elementary 808
      Longfellow Elementary 808
      Lakeshore Elementary 804
      Guadalupe Elementary 804
      New Traditions 802

  2. Top 10 increases from 2009 to 2010:

      Fairmount Elementary 78
      Sherman Elementary 74
      Grattan Elementary 51
      Charles Drew 43
      John Muir Elementary 43
      Buena Vista 38
      Leonard R. Flynn 38
      George Wash Carver 34
      Lafayette Elementary 32
      Paul Revere 31

  3. Top 10 decreases from 2009 to 2010:

      Sunnyside Elementary -44
      Glen Park Elementary -20
      Cleveland Elementary -19
      Sanchez Elementary -15
      Tenderloin Community -11
      Claire Lilienthal Elementary -10
      Edward R. Taylor Elementary -10
      Lakeshore Alternative -8
      Sheridan Elementary -7
      Clarendon Alternative -6

    What happened at Sunnyside?

  4. This is really impressive! Thanks for posting the scores.

  5. On another stand there is someone calling for the heads of the administration and the school board - I say that I am seeing some good work in the district (verified by these scores and by my personal observations).

  6. Truly amazing scores across the district. Right on to Roosevelt and Balboa, especially. There are lots of good schools out there people!

  7. Middle schools above 800.

    Bear in mind that scores reflect demographics as well as achievement. It is useful to look at the "similar schools" ranking to understand this. For example, A.P. Giannini is higher-ranked overall, but is only a 6 on similar schools rank compared to 8 for Aptos, because Aptos serves a more diverse and disadvantaged population than A.P. (and does relatively well).

    You should also look at the CST breakdown by sub-group--again, with Aptos, while the overall scores put it only 4th among the large middle schools, if you look at the sub-group for say white kids whose parents graduate from college, it has among the highest scores in the city.

    Of note, many of the middle schools had trouble making adequate yearly progress this year for all their subgroups as the percentage proficiency required by No Child Left Behind made another big leap. This is true across the state.

    Also, important thing to note: several schools' CST 8th grade algebra scores show a huge decline. That is because there has been a move to teach algebra to all 8th graders. So what was previously a test for honors kids is now a test of all kids, honors and gen ed. Not suprisingly, the scores show a decline that looks like the average of the previous two groups' math scores. Hope that makes sense....

    Anyway, here are the scores. Suddenly makes sense why the uproar over feeders, eh?:

    Alice Fong Yu 951
    Claire Lilienthal 911
    Lawton 911
    Rooftop 874
    Presidio 863
    A.P. Giananni 862
    Roosevelt 861
    Aptos 824
    Hoover 820
    SF Community 805

    Marina, Creative Arts, Francisco, and James Lick are in the 700s. It does drop from there.

  8. 12:38, I think you posted the list of '09 scores, not '10, for middle schools.

    I was doing a temporary job for SFUSD Student Nutrition last spring, which among other things involved doing a survey at Francisco Middle School about the cafeteria. A surprising number of students wrote their comments in Chinese, indicating that they're obviously not yet comfortable writing in English. That provides a small hint as to why some schools would show lower test scores than others!

    But also, parents need to understand how closely test scores correlate with demographics, with the resources and advantages each student brings with him/her (or doesn't). That means a new group in a middle school -- a different demographic breakdown even just in 6th grade -- has a significant impact on achievement.

    As a parent who started my older child at Aptos when it was scorned as a "dirty," "dangerous" "ghetto" school -- the disdained reject compared to shining Hoover -- I can tell you about turnarounds firsthand. And I can't resist a little gloating over Aptos' spot on the list ahead of Hoover's, even all these years later. That would have been unimaginable at the time (2002).

  9. Be careful how you are interpreting overall API scores. While many schools made strides, Flynn for instance has a gain, but the API for "white" students is 878 and for all other groups it ranges between 608-679. The overall API went up, but there is a huge disparity.

    It is more accurate to look at schools percentages of proficient students by statistically significant subgroup to determine how a school is doing in relation to similar schools. Schools with large sub-groups of students with traditionally high needs may score lower overall in their API but actually be serving kids better than a school with an overall high API that has a large group of kids who generally perform well, meaning a sub-group may have a higher percentage proficient at a school with a low API than the same sub-group at a school with a high API if the high API school has another subgroup that performs really well.

  10. I think it's pretty good that 42 of 67 SFUSD elementary schools have an API over 800. That's 62% of schools with really good scores. There are even some good schools in the 700s. Which for reasons pointed out above have lower overall scores, but are still doing a good job disadvantaged kids.

  11. Caroline, oops, right you are and thanks for pointing out that those were 2009 API scores for the middle schools.

    Here are the 2010 API scores for several 6-8 schools (leaving out the K-8s for now as most 5th grade parents looking will care more about the 6-8s).

    All have risen except Hoover which stayed the same

    A.P. Giannini 874
    Presidio 871
    Roosevelt 864
    Aptos 832
    Hoover 820
    Marina 805 (new to the 800s!)

    Lick, Francisco, Denman and MLK all in the 700s--Denman and MLK are new to that centile.

    Francisco had a 42 point rise and Denman 36 points; MLK rose 14 points.

    So the news is actually better than I posted earlier.

  12. As an educator and a parent, I'm disheartened by the emphasis placed on test scores. API scores DO NOT reflect best practices, authentic inquiry based learning and teaching. To begin testing children in 2nd grade is ridiculous, as are the aggressive literacy expectations of Kindergarters who are being prepped to take the STAR test in 2nd grade.. Unfortunately, the STAR test isn't going away anytime soon but I hope school communities will begin to push back finding more authentic and meaningful ways to assess learning and give voice to multiple intelligences. I don't want my child to be groomed to be a good test taker but instead be in an environment that nutures a real love for learning where all children get a chance to succeed. API scores reflect a child's socioeconomic status more than anything.

  13. I saw this in the comments section of the Chronicle article and thought it was an interesting take on the situation.

    "Great news that the API for African American kids in SF went up 20 points, but the API for SF's Asian students rose 14 points, and for SF's white students, 12 points. That means AA students moved just 6 points closer to Asians and just 8 points closer to whites. But with a gap of 261 points separating AA students from whites, at the rate of 8 points a year, it will take 33 years to finally close the achievement gap; with 254 points separating AA and Asian students, progress of 6 points a year will mean closing that gap in 42 years. How is it that our school board, led for years by so called progressives, hasn't been able to do better than this? Are we willing to wait until the grandchildren of current African American students are in school to see that gap close? Throw the bums out and let's elect some new school board members!"

  14. Re Sunnyside -- I remember they had a huge leap upward between 2008 and 2009 (bigger than the drop this year if I remember right), so maybe last year was a bit of a blip.

  15. re: 4:24, yes the achievement gap is stubborn. not clear what "throw the bums out" would get us in terms of closing the gap for real. does someone have a magic formula out there? that is scalable? and doesn't cost anything close to what we spend on wars abroad? do tell who can do this job....

  16. I don't think charter schools work miracles just by being charter schools, but as they are much maligned, I found it interesting that both of the KIPP charter schools are now above 800 and KIPP Bayview in particular had a big leap of 69 points in its scores to 825. Are they under new management or something? Also Edison and Creative Arts saw increases though still falling short of the 800 goal.

  17. Regarding the low scores of African-American students in SFUSD: Test scores correlate closely with income level. There has been extensive coverage and discussion of the fact that SF's high housing prices tend to drive African-American families out to lower-income suburbs where they can get more (or any) house for the money. That tends to drive out working-class and middle-class African-Americans, leaving mostly the lowest-income in housing projects. That's widely believed to explain the low scores for African-Americans in SFUSD -- the fact that higher-income African-Americans whose kids would tend to score higher have largely left the city. Obviously, throwing the school board out would have no effect on that -- the school board can hardly be blamed for the housing market.

    Regarding "solutions": The "magic" of KIPP is undoubtedly heavily affected by the fact that applicants have to jump through a number of hoops, which needless to say would discourage the families who can't or don't take an interest in their kids' education. Students must take a test, which supposedly is not used strictly as an *admission* test, but would clearly discourage non-compliant students who are just not willing to take the test, as well as students who don't expect to do well on the test. Students and families must also go through a counseling session and sign commitments to this and that (homework, attendance, behavior, dress code). That process would obviously discourage the unmotivated and non-compliant.

    Also, the KIPP schools' attrition rate is eye-popping -- they work at least some of their "magic" by simply tossing out the unsuccessful students. A 2008 study by the organization SRI International confirmed that 60% of students in Bay Area KIPP schools leave before finishing 8th grade, and that those 60% are overwhelmingly the less successful students. When a student leaves a public school, that student is replaced, but KIPP doesn't replace the students who leave. So only the most successful 40% remain. What would happen if the public school down the street could also kick out its bottom 60% and not replace them?

    Charter schools are free to pick and choose their students -- they're not supposed to, but nobody is overseeing their admissions processes, so they can do what they want. Maybe some/many/most run very pure lotteries and don't pick and choose at all, but they are free to pick and choose if they want. So clearly that confounds the comparison with non-charter schools.

    Lowell and SOTA pick and choose too -- openly and honestly, with transparent processes -- and obviously that boosts their test scores. But it's obvious that it's not fair to compare their scores to those of a school that admits all comers, or enrolls by lottery if there are more applicants than openings.

  18. Also, 5:05, I would vigorously disagree that charter schools are much maligned. Charter schools have all the powerful and wealthy forces lined up promoting them -- the Obama administration; previously the Bush administration; the Schwarzenegger administration; the mighty think tanks, an array of billionaires such as Bill Gates, Eli Broad, the Walton (Wal-Mart) family, and many more; the editorial voices of all the major media; and now filmmakers like Davis Guggenheim. They are "maligned" by small and powerless dissenters armed with blogs and g-mail, plus a very few wise and experienced voices with larger forums, such as Diane Ravitch, Richard Rothstein and Valerie Strauss. It's true that many of the nation's most prominent civil rights organizations got together to speak out against the Obama administration's destructive education policies, declaring that low-income minority children suffer the most when schools are being attacked and harmed; but the administration pretty much stifled their complaint. The only way anyone could call charter schools overly maligned is if you required 100% compliance and objected to even a tiny voice of dissent being allowed to squeak.


  19. See what you started, 5:05 pm?
    Once she gets going, it's hard to get her to stop.

  20. Hey, nothing wrong with a real argument (as opposed to ad hominem attack). If you disagree with Caroline's argument, say why.

  21. @ 11:01, For top 10 increases from 2009-2010, you forgot Malcolm X Academy. They increased 108 points from 692 to 800. Oh yeah, this school is also in BVHP.

  22. Go Malcolm X and now it makes sense it was going be being fed into Hoover!

  23. I don't WANT my kid going to a school with a bunch of kids and families who are not motivated to do well in school so if the charters are weeding them out, good for them.

  24. Caroline,

    We understand that as far as you are concerned there is no amount of charter school criticism that would be enough to suit you.

    We've heard all the arguments a million times. Yes, it is true that KIPP creams. So does SOTA. If they are more transparent about it does that make it moral? That is your point isn't it? That it is not morally acceptible to cream kids and then to compare it like apples to apples? I agree. Who wouldn't?

    But The net effect on the down side, tranparent or not, is the same - more under-performers elsewhere. So schools like SOTA can feel good about their transparency while the recipients institution of their chaff suffer, despite their transparency as well.

    But on the up side children who make the sacrifice and the "cut" are rewarded in manner that might not be possible if it werem't for schools like KIPP. True, though, that some charters claim successes that are not borne out by the facts.

    We know the traditional schools cannot compare their data in a meaningful manner with schools like SOTA (trad) or KIPP (charter). But when it comes to educating, they can focus on a highly motivated core group and, in the better cases, get impressive results. This indicates that perhaps schools should do more creaming.

    Doesn't producing results mean anything? The kids who are potentially being lifted out of poverty at a school like KIPP are from BVHP. Why deny them this opportunity just because all kids can't make the grade? Lift out as many as are willing to put in the effort. The more motivated group should not be held back by the less motivated as a result of institutional and behavioral neglect. It is a matter of giving everyone the best possible chance.

  25. I'm 11:01 that posted the stats at the top of this thread, I didn't leave Malcom X out on purpose. It wasn't in the list with the other elementary schools, it was listed way below under a category called "Smaller Schools." Daniel Webster was accidentally left off too. Below is the revised list of

    Top 10 increases from 2009 to 2010:
    Malcolm X Academy 108
    Fairmount Elementary 78
    Sherman Elementary 74
    Daniel Webster Elementary 57
    Grattan Elementary 51
    Charles Drew 43
    John Muir Elementary 43
    Buena Vista 38
    Leonard R. Flynn 38
    George Wash Carver 34

  26. About time Flynn showed the pop from the introduction of the immersion program.

    Interesting to see Revere and Fairmount making gains also.

    Revere should not have been in the "failing" category last year, but I guess as Revere gets Federal $$$ because of it (at the cost of losing its principal) it all works out.

    AFY stays on top. Eliana Szeto would have been *pissed* if Clarendon had overtaken AFY.

  27. "There are 42 elementary schools w/API over 800:"

    You missed Malcolm X Academy.

    Up 108; highest gain of any SFUSD ES. API score of 800. Pretty frickin' stunning.

    "Top 10 increases from 2009 to 2010"

    You also missed Webster: gain of 58 points, and that's before the immersion program would kick into the API scores.

    These scores are looking for schools in the SE.

  28. "These scores are looking for schools in the SE."

    I meant these scores are looking good for school in the South-East.

  29. There are 42 elementary schools w/API over 800:"

    And the average API for the district is a shade under 800. About the same as San Jose Unified.

  30. Hurrah to Daniel Webster - long way to go but if that huge gain is before immersion students kicked in then that's going to really help them continue their rise.

  31. If you account for Alamo's relatively high ELL population, they do better than Clarendon or AFY. But I'm with the teacher that wrote in to say that this emphasis on test scores is not good for education and our children. That's why I like a school like an Alamo over an AFY which scorns children that don't meet their academic expectations. I have friends who moved their kids out just to maintain their sanity.

  32. Don, I don't advocate shutting down KIPP.

    I just support honesty and transparency about why it's successful. (And conversely, I oppose the constant blizzard of lies and hype surrounding it.)

  33. If you are saying the success of KIPP is simply a blizzard of lies and hype, why would you not oppose shutting it down? I have read much of what you have written on the subject. Are you giving in to charters or is this a calculated decision only to oppose the worst charter operators?

  34. Don, you love to argue for the sake of arguing, which is just a distraction from the topic under discussion. For one last response on this digression:

    I don't dispute that KIPP schools are (often) successful. They are successful due to operative procedures that probably would allow any public school to be equally successful (winnowing and getting rid of unsuccessful students, keeping only the more successful ones). Their claims of miracle success deny that aspect and are based on massive lies and hype. I object to the lies and hype.

    If KIPP schools would openly say "we screen for low-income students who are motivated and extremely docile and compliant, and we continue to get rid of the students who don't succeed, and in that setting, the remaining students do really well," that would be truthful and useful.

    It seems clear that in a setting surrounded only by other motivated, docile and compliant students, the KIPP students do very well. I readily acknowledge that; it's useful for educators and the public to know. It's hard for them to know it due to the lies and hype.

  35. "If you account for Alamo's relatively high ELL population, they do better than Clarendon or AFY. "

    ELLs scored 940 at Clarendon, 911 at AFY, but 891 at Alamo.

    It seems Alamo's scoring lower more because the gap between the ELLs and the non-ELLs is larger, rather than it being solely demographics. ELLs are a big chunk of AFY as well (20%, compared to 30% for Alamo).

  36. Malcolm X's 2007 API was 567. Someone's working miracles there.

  37. Looking at the API scores, it seems that there's a lot of strong elementary schools (API >800), and a lot of solid-to-OK elementary schools in the SE given their demographics (750-800 API), and a smaller number of mediocre elementaries.

    For middle schools, it seems you have an even split between strong Middle schools (API >800) and weak ones (API <750).

    But the suckitude seems to really kick in at the High School level: only two (SOTA and Lowell) out of 18 HSs with APIs >800, and 9/18 with APIs <700. Even Lincoln is <800.

    I thought that SFUSD had a problem with middle schools, but that it's elementary and high schools were good-to-OK. It seems that's wrong: the elementary schools are solid, the middle schools OK, but the high schools have real issues.

  38. "There are 42 elementary schools w/API over 800:

    And the average API for the district is a shade under 800. About the same as San Jose Unified."

    The average API for the district is for ES, MS and HS.

    The district did an excellent job turning around ES's (2/3 are above 800), and it is starting to show at MS level (1/2 above 800). However, The only HS's scored above 800 are Lowell and SOTA.

    As the current crop of ES kids enter MS, and then HS, I expect the MS and HS will improve too over the next 5 to 10 years.

  39. Re: API troubles

    Beginning this school year new law requires that drop outs by included in the API for 8th and 9th grade. Those are all zeros and will affect the averages in a big way at low SES schools in particular. Anyone know why they are including this data?

    Because the pro-charter advocates want traditional public schools to look as bad as possible. I'm sure Caroline can enlighten you on this subject.

    RE: 8:32's stats

    Where are you getting your numbers from?

    The SARC lists Alamo as 25% ELL and AFY as 13%. In addition, the SE disadvantaged is Alamo 30%, AFY 25%.

    Also, did you ever wonder how could ELL students score just as well in Language arts as the rest in elementary school? The disparity in the statistics don't usually begin to narrow until the latter part of elementary. I did not look at the breakdown by grade yet.

  40. Once you get to high school, you have a much greater population of students who are there because they law says they have to be there, not because they want to be there, and at that age they're not tractable any more. If they don't want to do well, nobody can make them. That affects test scores at any high school that's not a magnet for the motivated.

  41. what does this mean 11:15?

    Once you get to high school, you have a much greater population of students who are there because they law says they have to be there, not because they want to be there.

    It is mandatory to go to school from 1st grade.

  42. "The SARC lists Alamo as 25% ELL and AFY as 13%."

    Greatschools.net gives a 19% for AFY.

    Data from the State's Dataquest site indicates 122 ELLs at AFY, which with an enrollment of 544 is 22%.

    For Alamo, the state's data shows 181 ELLs, with an enrollment of 549: 32%

    For Free/Reduced price meals, it's 33% for Alamo and 27% for AFY by the State's statistics.

    I think the data from the SARC is out of date.

  43. FYI, found this on a website that explains API:

    "A caveat: Although the API is meant as a measure of academic growth, it is not intended to track the school’s growth over several years. The meaningful comparison is within each annual API cycle, between the Base API and the Growth API, because the computation of the API is kept as similar as possible within each cycle.

    The incorporation of new elements into the index at the beginning of an API base/growth cycle can lead to unintentional or confusing fluctuations in API scores compared with scores from the previous cycle. The state adjusts API scores to compensate for the effect of those new elements. The mechanism for that technical adjustment is called the scale calibration factor. Even so, the CDE warns against tracking the scores year to year. A better comparison is to look at whether a school or district consistently meets its growth targets or to consider the amount of growth year to year."

    I also read recently that some of the huge increases in API this year are a result of adding the California Modified Assessment (CMA) for students for whom the CAPA and CSTs are not appropriate.

    It seems to me, this made a huge difference for schools with large population of students with IEPs.

  44. Oops, here is the link to my source re: API


  45. Also, here is the link regarding the addition of California Modified Assessment (CMA) to the API over the past few years.

  46. Is there any place that shows how many kids used the modified test in each school?

  47. Re: 9:24am and the 'suckitude' at the high school level.

    It just isn't fair to compare Lowell and SOTA's test scores to the comprehensive high schools (Lincoln, Gal, Bal etc...) because they have self-selected populations of motivated students. If Lowell did not exist (wishful thinking I might add), and those students were spread throughout the District, the scores of all of the comprehensive high schools would rise accordingly. The 2000 or so of Lowell's highest achieving students spread throughtout the district would mean 150-200 students at each school, assuming that they would be distributed evenly. That would make a significant diference in those schools test scores.

    If you look at the breakout test scores of the comprehensive high schools, and take the GATE identified students for example, you will see that the GATE students in the coprehensives test scores are not that different from Lowell's students.

    The comprehensive high schools educate (and test) students af widely varying abilities, and viewed in this light, the test scores don't look so bad.

  48. Or the students at Lowell and SOTA could spread out among the comprehensive high schools and get socialized into thugs. It happens, even in families with very involved parents. I know SOME kids at the comprehensive schools do extremely well and all credit to them for having such strong character. However, I don't feel obligated to put my kids, who are more followers than leaders, in a comprehensive high school. I want them in a place where the whole culture is about working hard and attaining goals. Socioeconomic diversity by all means. Academic diversity, no thank you.

  49. To 9:36am

    Have you been to visit any of the comprehensive high schools you are so down on or are you just going by what you have heard? Because I have a child at Balboa (a 'follower" BTW not a 'leader')and I can tell you from personal experience that the halls are not full of thugs looking to socialize my child into a thug. The culture there IS about self respect, working hard and attaining goals, both academic and social. They don't tolerate bad behavior there and when academic or behavioral issues appear, they are dealt with immediately.

    Go to the schools and visit them yourself and then come back here and tell us what you see. Judging a school by relying upon stories of what your neighbor or cousin experienced is narrow-minded and lazy.

  50. Uh, then why are half or more of the kids at Balboa not proficient for grade level in all subjects tested on the 1009 SARCs?

  51. Sorry I meant to say 2009 SARCs.

  52. umm, what was your point?

    I thought you were worried about the thugs in the schools?

    Back to the issue you raised, have you been to visit any of schools and seen these alleged thugs with your very own eyes? Really, it is tiresome to listen to opinions based on air.

  53. MLK's high scores probably have to do with its ridiculously small class size. The school is very underenrolled.