Friday, November 2, 2012

Data: Richest and Poorest Elementary Schools, API and CST

I've been looking at the data on academic performance and income level for SFUSD. Am I naive to be shocked that our public elementary schools are so segregated by income level?


Schools with the Lowest Percentage of Economically Disadvantaged Students


Not Econ
Disadvantaged
Economically
Disadvantaged
Miraloma
86%
15%
Clarendon
84%
16%
Chinese Immersion
84%
16%
Grattan
81%
19%
SF Pub Montessori
79%
21%
Lilienthal
79%
21%
Creative Arts
73%
27%
Sunset
72%
28%
Peabody
71%
29%
McKinley
71%
29%

Source: CBEDS Information Day data from SNS Student files, published at http://sfusdfood.org/freereduced.html

Schools with the Highest Percentage of Economically Disadvantaged Students


Not Econ
Disadvantaged
Economically
Disadvantaged
Chavez
10%
90%
Hillcrest
10%
90%
Redding
10%
91%
Lau
9%
91%
Moscone
9%
91%
Cleveland
9%
91%
Tenderloin
8%
93%
Malcolm X
8%
93%
Harte
5%
95%
Chinese Ed Ctr
3%
97%
Mission Ed Ctr
0%
100%
Source: CBEDS Information Day data from SNS Student files, published at http://sfusdfood.org/freereduced.html

    It's true that at a school where everyone is poor, teachers have to slow down the academic pace because so many students have barriers to learning, like the lack of a quiet place to study at home or parents working long hours who aren't around to help with homework.  But the "richer" schools don't have higher test scores when it's broken down by income level. Miraloma is a great example.  For students who are not economically disadvantaged, Miraloma does no better than SFUSD as a whole on STAR testing. In fact, Miraloma's CST math scores are about 20 points lower than Starr King's and Glen Park's for students who are not economically disadvantaged.  Creative Arts Charter performs significantly below the SFUSD average for students who aren't economically disadvantaged.

    My bottom line: look at CST scores broken down by parent education and income, not API scores, when comparing schools' academic performance.

Here's the complete list of elementary schools from richest to poorest.  

All SFUSD Elementary Schools by % Economically Disadvantaged

Not Econ
Disadvantaged
Economically
Disadvantaged
Miraloma
86%
15%
Clarendon
84%
16%
Chinese Immersion
84%
16%
Grattan
81%
19%
SF Pub Montessori
79%
21%
Lilienthal
79%
21%
Creative Arts
73%
27%
Sunset
72%
28%
Peabody
71%
29%
McKinley
71%
29%
Feinstein
71%
29%
Lafayette
70%
30%
Yu
68%
32%
New Traditions
64%
36%
Rooftop
62%
38%
Argonne
62%
38%
West Portal
60%
40%
Jefferson
60%
40%
Alvarado
59%
41%
Alamo
58%
42%
Sherman
58%
42%
Sloat
57%
43%
Milk
55%
45%
King
55%
45%
Key
52%
48%
Sunnyside
51%
49%
Lakeshore
48%
53%
Lawton
46%
54%
Stevenson
46%
54%
Ortega
43%
57%
Yick Wo
43%
57%
Webster
40%
60%
Ulloa
40%
60%
Parks
39%
61%
Fairmount
37%
63%
Sutro
33%
67%
McCoppin
33%
67%
Buena Vista/Mann
32%
68%
S.F. Community
31%
69%
Flynn
29%
71%
Garfield
29%
72%
Revere
25%
75%
Monroe
24%
76%
El Dorado
23%
77%
Guadalupe
22%
78%
Glen Park
22%
78%
Cobb
21%
79%
Carmichael
21%
80%
Longfellow
20%
80%
Taylor
19%
81%
Spring Valley
17%
83%
Sanchez
16%
84%
Drew
15%
85%
Chin
15%
85%
Parker
14%
86%
Sheridan
13%
87%
Marshall
13%
87%
Carver
12%
88%
Bryant
12%
88%
Vis Valley
12%
88%
Serra
11%
89%
Muir
11%
89%
Chavez
10%
90%
Hillcrest
10%
90%
Redding
10%
91%
Lau
9%
91%
Moscone
9%
91%
Cleveland
9%
91%
Tenderloin
8%
93%
Malcolm X
8%
93%
Harte
5%
95%
Chinese Ed Ctr
3%
97%
Mission Ed Ctr
0%
100%

Source: CBEDS Information Day data from SNS Student files, published at
 http://sfusdfood.org/freereduced.html


 How much does family income matter?

   The California Department of Education calls students getting free/reduced-price lunch "economically disadvantaged," and uses that as a proxy for income level. Students with free lunch have a household income under 130% of the Federal Poverty Limit, about $30,000 for a family of four. Students with reduced-price lunch have a household income under 185% FPL, about $42,000 for a family of 4.
    In SFUSD, 60% of the K-5 enrollment is economically disadvantaged, meaning they get free or reduced-price lunch. 51% get free lunch, and  9% get reduced-price lunch.
    Economically disadvantaged elementary school students in SFUSD score, on average, 50 points lower on CST test (STAR testing) in both English and Math.  Just over half the economically disadvantaged elementary school students score at grade level, compared to 80% of the non-economically disadvantaged students.  APIs don't take into account the poverty level or parent income level, so schools with more affluent, higher-educated parents have high APIs.

SFUSD STAR test scores, Grades 2-5, 2011-12


CST Test 
Not Econ
   Disadvantaged 
Economically
   Disadvantaged
English
398
351
Math
433
385
Source: California Department of Education, http://star.cde.ca.gov/
CST scores by grade were averaged for grades 2-5.

Percentage of Students Testing at or Above Grade Level, SFUSD Grades 2-5, 2011-12


CST Test 
Not Econ
   Disadvantaged 
Economically
   Disadvantaged
English
80%
52%
Math
82%
63%
Source: California Department of Education, http://star.cde.ca.gov/

Your take: useful or not useful?

7 comments:

  1. What are the top 12 schools for students who are not economically disadvantaged?
    Thanks!

    ReplyDelete
  2. You might also consider the number of English learners in each school, which significantly raises the bar for teachers.

    I was more surprised that, of the recently publicized 10 SFUSD schools with the highest API scores (assuming you buy into test scores as a measure of success), 4 of them have populations that are over 50% economically disadvantaged and the same 4 have English Learner populations over 25%, 3 of them over the district average of 29%. In fact, the highest API-ranked school in SFUSD is John Yehall Chin, with 80% English Learners and 84% economically disadvantaged! The question I ask is what are these successful schools doing - and could it be replicated among other schools?

    ReplyDelete
  3. Since you have been touting yourself as a geek, please avoid making blanket statements like "It's true that at a school where everyone is poor, teachers have to slow down the academic pace because..." unless you have some studies to support this. At my kid's school (high poverty), I can see this is not the case. From friends' experiences at other high poverty schools, I can see that this is not the case. I have no experience with colleagues or friends sending kids to high poverty schools outside of SFUSD, but within this city, I would have say that (in honor of my former profession) Objection, assumes facts not in evidence!

    I am going to assume that you are naive since you are surprised that schools here are segregated by income. So educate yourself. ;)

    ReplyDelete
  4. Liz,
    English learners is an interesting area. CST scores much more useful than APIs here.
    The percentage of English learners drops rapidly from 2nd to 5th grade as students learn English. FWIW, these former English learners, called "Reclassified Fluent English Proficient," consistently outscore "English-Only" students on the CST. Bilingual students who are already proficient in English when they enter school also outscore "English only" students on the CST.

    8:11,
    I will post the data that shows a trend toward lower test scores for non-economically disadvantaged students who attend poorer schools. If you'd like to check my numbers, I've been using the 2011-12 CST test scores by school for grades 2-5, "Non-Economically Disadvantaged" and "Economically Disadvantaged," available at the California Dept of Education STAR test website.

    ReplyDelete
  5. There is no doubt that economically disadvantaged kids have lower test scores on average, that's why everyone talks about the achievement gap.

    But you are making a big assumption that the teachers slow down the pace in lower income schools. It's not something I've seen in SFUSD. The only place that I know of with an accelerated curriculum is Clarendon - everyone else tries to cover the state mandated grade level content standards.

    ReplyDelete
  6. 8:11.

    Virtually every study strongly correlates socio-economic status with performance and more specifically, the lower the status the lower the performance. That's why SFUSD's social justice-oriented strategic plan speaks to "the predictive power of demographics."

    Where much of the classroom learning is considered remedial, it goes without saying that the teachers must slow down the pace.

    ReplyDelete
  7. I think the fairest way to do it is to look at CST scores for students whose parents share your educational background (e.g. college only, grad, etc.). You can do this on the CST website. I think this is a better predictor of a kid's performance than race or income. If kids with similarly educated parents are doing poorly, it almost doesn't matter if you want to blame it on bad teachers, bad facilities, or disruptive/ill-prepared kids from other groups.

    ReplyDelete