student assignment 2nd annual report

STUDENT ASSIGNMENT

4/17/2013 Annual Report: 2012-13 School Year

Student Assignment 2012-13

i | P a g e

Preface

The purpose of this report is to help the Board of Education of the San Francisco Unified School

District (the Board) monitor its student assignment policy. On March 9, 2010, the Board

unanimously approved a new student assignment policy (Policy P5101), and this policy calls on

the Superintendent to present the Board with an annual report on the student assignment

system.

This is the second annual report since P5101 was approved by the Board; the first annual report

was shared with the Board and public on March 5, 2012.

The following SFUSD staff worked together to develop this report:

• Tomas Beccar Varela, Education Integration Specialist, Educational Placement Center

• Arcadio Fokin, Director, Educational Placement Center

• Chris Frazier, Senior Programmer, ITD

• Hans Gong, Education Integration Specialist, Educational Placement Center

• Jack Huang, Senior Business Analyst, ITD

• Darlene Lim, Executive Director, Educational Placement Center

• Orla O’Keeffe, Executive Director, Policy and Operations

• Alana Ramirez, Operations Manager, Educational Placement Center

A group of advisors voluntarily help staff monitor the Board’s student assignment policy.

• Prudence L. Carter, Associate Professor of Education & (by courtesy) Sociology at

Stanford University

• Michal Kurlaender, Associate Professor of Education at UC Davis

• Sean F. Reardon, Professor of Education at Stanford University

The advisors will review this annual report and they will share their thoughts about key findings

in this report with the Superintendent and his staff.


ii | P a g e


iii | P a g e

EXECUTIVE SUMMARY

Introduction

The majority of school enrollments are managed centrally by SFUSD’s Educational Placement Center (EPC), and are administered by EPC according to the Board’s student assignment policy (P5101) which was unanimously approved by the Board in March 2010. The student assignment system provides significant opportunities for family choice in enrollment. For the 2012-13 school year, 80% of K-12 applicants (11,139) received one of their choices compared with 78% in 2011-12, and 60% received their first choice compared with 58% in 2011-12. The student assignment system is designed to be flexible so it can be monitored and adjusted if it is not working in alignment with other SFUSD initiatives designed to:

• reverse the trend of racial isolation and the concentration of underserved students in the same school; and

• provide equitable access to the range of opportunities offered to students. The student assignment policy requires the Superintendent to conduct an annual assessment of the student assignment system and to present an annual report to the Board. This is the second annual report since P5101 was approved by the Board; the first annual report was shared with the Board and public on March 5, 2012.

Amendments to Date

The Board approved amendments to P5101 on September 28, 2010 and June 14, 2011. They revised P5101 to approve the design for new elementary school attendance areas, to approve middle school feeder patterns, to improve the assignment process for language pathways, and to eliminate the density tie-breaker. The Superintendent implemented a series of administrative changes since the policy was approved, including establishing a TK tie-breaker and adjusting the attendance areas for Miraloma and Sunnyside.

Questions Explored

Staff explored five questions to help monitor the impact of the current student assignment system. Findings from the exploration of the five questions will help the Superintendent and the Board to determine if future adjustments need to be made to the current student assignment system.

1. To what degree is SFUSD reversing the trend of racial isolation and the concentration of underserved students in the same school?

• Our current student assignment system has been in place for two years, which means it has been used to assign students to six out of 13 grades (kindergarten, 1st grade, 6th


iv | P a g e

grade, 7th grade, 9th grade, and 10th grade). Over these two years, there has not been a shift in the number of schools with an enrollment of more than 60% of a single race/ethnicity. Despite SFUSD’s diverse student population, in October 2009, 24 schools (about a fifth of our schools) had an enrollment of more than 60% of a single race/ethnicity compared with 23 schools in October 2012.

• Of the 23 schools in 2012 with more than 60% of a single race/ethnicity, 11 had an API of 1, 2, or 3, which means, according to the definitions in Board policy P5101, 11 schools were racially isolated with high concentrations of historically underserved students. Two schools with an enrollment of more than 60% African American and an API of less than 4 experienced a decrease in the percent African American between 2009 and 2012. Eight schools with more than 60% Latino offer Spanish language pathways that reserve up to three quarters of the seats for Spanish speakers, and the same is true for the school that was more than 60% Chinese and had an API of less than 4.

• With any choice based student assignment system, diverse applicant pools are critical to achieving diverse enrollment. The applicant pools for the 11 racially isolated schools with high concentrations of underserved students were not diverse – they reflect the lack of diversity of enrollment.

• Applicant pools for most elementary schools are not diverse, and there are geographically distinct patterns for the racial/ethnic composition of students applying to and enrolled in elementary schools. Kindergarteners enrolled in and applying to schools in the north east are primarily Chinese; in the south east they are primarily African American; in the Mission they are primarily Latino; on the west side they are primarily Chinese and White; and in the center of the city they are primarily White.

• Middle school and high school applicant pools are more diverse than elementary applicant pools, and as a result enrollments at the middle and high school levels are more diverse.

• Student assignment has a role to play in reversing the trend of racial isolation and the concentration of underserved students in schools; however student assignment alone cannot overcome the complex elements that contribute to the current state. For example, the demographics of the city, parent request patterns, and language pathways all have an impact on the demographics of our schools.

2. Does the CTIP1 tie-breaker help reverse the trend of racial isolation and the concentration of underserved students in the same school, and does it help provide equitable access to the range of opportunities offered to students?

• In October 2012, the two largest cohorts of students living in CTIP1 were Latino (503 kindergartners) and African American (266 kindergartners), and they tend to be enrolled in schools near CTIP1 that are predominantly Latino and African American. The two smallest cohorts were White (88 kindergartners) and Chinese (66 kindergartners), and they tend to be enrolled in schools with a high percent of White and Chinese students.

• It takes a while for new policies to show results; therefore it is too soon to identify trends. It is not yet clear from the data that the CTIP1 tie-breaker is helping reverse the trend of racial isolation and the concentration of underserved students in the same


v | P a g e

school. We plan to continue to monitor demand and enrollments, and expect it may take several more years before trends are revealed.

3. Do middle school feeders help reverse the trend of racial isolation and the concentration of underserved students in the same school, and do they help provide equitable access to the range of opportunities offered to students?

• The middle school feeders had a bigger impact on the choice making process than expected during the first year of implementation. Most middle schools experienced a significant increase in the percent of 5th graders from elementary feeder schools requesting their school, and the number of elementary schools feeding into each middle school decreased.

• 6th grade enrollment experienced both increases and decreases in the percent of students from a single race/ethnicity.

• Staff will continue to monitor the impact of the feeder patterns in 2013-14 to see if the shifts noted in year one continue in year two.

4. What percent of applicants submitted their enrollment application on time for the first round of assignments, and how does this vary by race/ethnicity?

• With the exception of Other Asian, the percent of late applicants in 2012 increased for every racial/ethnic group compared with 2011.

• Participation rates appear to be impacted by the timeframe for families to submit applications and the resources available to support outreach and recruitment. In 2011, the enrollment deadline was February 18, and in 2012 the enrollment deadline was a few weeks earlier - January 27. At the same time, EPC experienced a reduction in staff due to budget cuts. With less time and fewer staff, the number of outreach events went from 150 in 2011 to 100 in 2012, and on-time participation rates declined.

5. What is important to families when choosing schools?

• To get a better understanding of what is important to families when choosing schools, a family survey was included on the 2012-13 application form. The survey asked families to rate on a scale of 1 to 5 (1=not important, and 5=very important) how important 15 factors are when requesting schools. Over 10,000 families of incoming kindergartners, 6th graders, and 9th graders responded to the family survey.

• Neighborhood safety and school quality surfaced as paramount in the minds of families when choosing schools for their children. Over 75% of families rated neighborhood safety, teachers and principals, and schools academic reputation as very important. These findings are consistent with findings from various community engagement activities that have taken place over the years.

Attendance Areas

There is a mismatch between where students live and where schools are located; 35 of 58 attendance areas had more kindergarten residents than seats. At the same time, more than


vi | P a g e

half of kindergarten applicants did not request their attendance area school anywhere among their choices, and only 26% requested it as a first choice. Requests for schools relieve the mismatch between where students live and where schools are located. Staff is not recommending any changes to the attendance areas for the 2014-15 enrollment cycle.

General Education Transportation Services

The Board approved a new policy for general education transportation services in December 2010 (see Appendix 5), and in February 2011, the Superintendent presented the Board with a proposal to change general education transportation routes over a three year period. The proposal was designed to bring general education transportation services in alignment with the Board’s general education transportation policy, and with the district’s budget reduction goals. The first series of changes were implemented in the 2011-12 school year, and the final changes will be implemented on time for the 2013-14 school year. The bus fleet has been reduced from 44 buses in 2010 to 25 buses in 2013.

Next Steps

Discuss Annual Report with the Board of Education

SFUSD staff plan to discuss this report with the Board of Education at a public meeting in the spring of 2013. This public forum will provide an opportunity for the Board to give feedback to staff, and for the Board and staff to gather feedback from all stakeholders.

Review and Possibly Revise the CTIP Classifications

SFUSD staff are working with demographers to review the CTIP classifications. Our tentative timeline is to share findings with the Superintendent and Board at a public meeting by August 2013, and to confirm the CTIP1 areas for the 2014-15 enrollment cycle by September 2013.

Recommend Elementary Feeders for Willie Brown, Jr. Middle School

Willie Brown Jr. middle school is scheduled to open on time-for the 2015-16 school year. This means we must identify elementary feeder schools for Brown middle school by the time the enrollment cycle kicks-off in the fall of 2014. Staff plans to share recommendations with the Board of Education by the spring of 2014.

Explore Additional Research Questions

Through our partnership with Stanford, and under the supervision of Sean Reardon, Professor of Education, Stanford is exploring research questions related to SFUSD’s student assignment system. This research involves the analysis of all K-12 assignments made through the Educational Placement Center (EPC) beginning with the 2004-05 school year through the 2012-13 school year.


1 | P a g e

Table of Contents

INTRODUCTION ........................................................................................................................ 3

AMENDMENTS TO DATE .......................................................................................................... 5

QUESTIONS EXPLORED .......................................................................................................... 7

1. To what degree is SFUSD reversing the trend of racial isolation and the concentration of underserved students in the same school? ............................................ 9

2. Does the CTIP1 tie-breaker help reverse the trend of racial isolation and the concentration of underserved students in the same school, and does it help provide equitable access to the range of opportunities offered to students? .......... 23

3. Do middle school feeders help reverse the trend of racial isolation and the concentration of underserved students in the same school, and do they help provide equitable access to the range of opportunities offered to students? .................. 37

4. What percent of applicants submitted their enrollment application on time for the first round of assignments, and how does this vary by race/ethnicity? ..................... 45

5. What is important to families when choosing schools? .................................................. 47

ATTENDANCE AREA BOUNDARIES...................................................................................... 53

GENERAL EDUCATION TRANSPORTATION SERVICES ...................................................... 61

NEXT STEPS ........................................................................................................................... 65

APPENDIX ............................................................................................................................... 67

1. Historical Context .......................................................................................................... 69 2. Student Assignment Tie-Breakers: 2012-13 School Year .............................................. 73 3. Census Tract Integration Preference (CTIP) .................................................................. 77 4. Middle School Feeders .................................................................................................. 79 5. General Education Transportation Policy ....................................................................... 83 6. Glossary of Terms ......................................................................................................... 85


2 | P a g e


3 | P a g e

INTRODUCTION

Founded in 1851, the San Francisco Unified School District (SFUSD) was the first public school

district established in California. Today, SFUSD is the eighth largest school district in California

educating over 56,000 PreK-12 students in 112 schools within the 49 square mile area of the

City and County of San Francisco.

SFUSD is governed by a Board of Education (Board) comprised of seven elected members, and

is subject to local, state, and federal laws. SFUSD’s mission is to provide each student with an

equal opportunity to succeed by promoting intellectual growth, creativity, self-discipline, cultural

and linguistic sensitivity, democratic responsibility, economic competence, and physical and

mental health so that each student can achieve his or her maximum potential.

In May 2008, the Board approved a strategic plan, Beyond the Talk: Taking Action to Educate

Every Child Now, which named the achievement gap as the greatest civil rights issue facing

SFUSD and determined that SFUSD’s three goals are:

1. Access and equity – making social justice a reality by ensuring every student has access

to high quality teaching and learning;

2. Student achievement – creating learning environments in all our schools that foster

highly engaged and joyful learners and that support every student reach her/his

potential; and

3. Accountability – keeping our promises to students and families and enlisting everyone in

the community to join us in doing so.

Student Assignment Policy

The majority of school enrollments are managed centrally by SFUSD’s Educational

Placement Center (EPC). SFUSD’s Student, Family, and Community Support Department

manages enrollment in continuation schools, and each charter school has its own enrollment

process. All other school enrollments are processed centrally by the EPC.

• Admissions to Lowell High School and Ruth Asawa School of the Arts are administered

according to an October 16, 2001 proposal developed by a Taskforce on Admissions to

Lowell High School and School of the Arts and approved by the Board on October 23,

2001.

• All other school enrollments processed centrally by the EPC are administered according to

Board Policy P5101, which was unanimously approved by the Board on March 9, 2010.


4 | P a g e

Student Assignment System

2012-13 is the second school year students were enrolled in schools according to the

policy guidelines set forth in P5101; it was first used for the 2011-12 school year. Between

2002 and 2010, SFUSD’s assignment system gave families choice and used a ‘diversity index’

to ensure equitable access to all schools and to promote diversity without using race/ethnicity.

P5101 maintained choice as a tactic for achieving the Board’s goals, but it simplified the

choice system and differentiated it for elementary, middle, and high school.

To enroll in schools, families submit enrollment applications and they are offered their highest

ranked requests as long as there is space. If there are more requests for a school than

openings, the student assignment system sorts all requests using a series of preferences,

called tie-breakers, to assign applicants to schools. A high-level overview of the technical

aspects of tie-breakers used for the 2012-13 school year is provided in Appendix 2, and

additional information is available on the web at www.sfusd.edu.

The current choice assignment system continues to provide significant opportunities for

family choice in enrollment. For the 2012-13 school year, 80% of K-12 applicants (11,139)

received one of their choices compared with 78% in 2011-12, and 60% received their first

choice compared with 58% in 2011-12.

While choice is a tactic for achieving the Board’s goals, it is not the focus of the Board’s student

assignment policy. The policy’s focus is working in alignment with other SFUSD initiatives

designed to:


• provide equitable access to the range of opportunities offered to students.

Monitoring Student Assignment

Board Policy P5101 calls on the Superintendent to conduct an annual assessment of the

student assignment system governed by P5101, and to develop an annual report and present it

to the Board of Education each year.

This is the second annual report since P5101 was approved by the Board; the first annual

report was shared with the Board and public on March 5, 2012.


5 | P a g e

AMENDMENTS TO DATE

This section of the report describes any amendments to policy P5101, as well as changes to the

procedures for administering the policy, since it was approved by the Board on March 9, 2010.

Policy Amendments

September 28, 2010 Board Meeting (108-24SpE Superintendent’s Proposal)

The Board approved the design for new attendance areas for elementary schools; an

extension for implementation of elementary-to-middle school feeder patterns; and

middle school attendance areas until the 2012-13 school year, and a temporary student

assignment process for middle school students for the 2011-12 school year.

June 14, 2011 Board Meeting (1115-24-Sp1 Superintendent’s Proposal)

The Board amended student assignment policy P5101 to include elementary-to-middle

school feeder patterns that operate as a tie-breaking factor in the choice process starting

in the 2012-13 school year and for four years thereafter. Beginning with enrollment for the

2017-18 school year, fifth graders will receive an initial assignment to the feeder middle

school based on the elementary school they attend, and they will have subsequent

opportunities to participate in a choice process. Additional information about the middle

school feeders is available starting on page 37 and in Appendix 4.

The Board also amended P5101 to improve the assignment process for language

pathways, and to eliminate the density tie-breaker since it was not working as originally

intended.


6 | P a g e

Administrative Changes

• Tie-breakers for Transitional Kindergarten (TK)

At the August 27, 2012 Ad Hoc Committee on Student Assignment, staff shared that,

beginning with the enrollment process for the 2013-14 school year, students enrolled in an

SFUSD TK and applying to kindergarten will receive a TK tie-breaker that mirrors the Pre-K

tie-breaker policy guidelines set forth in P5101.

o For attendance area schools, students who live in the attendance area of the school,

and are enrolled in an SFUSD TK in the same attendance area, will get a TK tie-

breaker if they apply to kindergarten at that attendance area school.

o For city-wide schools (i.e., elementary schools that do not have an attendance area),

students enrolled in an SFUSD TK at a city-wide school will get a TK tie-breaker if

they apply to kindergarten at that city-wide school.

• Adjustments to the elementary attendance areas: Miraloma and Sunnyside

At the August 28, 2012 Board meeting (Agenda U - Other Informational Items), the

Superintendent shared that staff planned to adjust the elementary attendance areas prior to

launching the enrollment process for the 2013-14 school year. The adjustment reduces the

size of Miraloma’s attendance area and increases the size of Sunnyside’s attendance area

by moving the Sunnyside boundary north to encompass Sunnyside Playground and

Sunnyside Conservatory. These modifications were recommended by the community and

reviewed and approved by staff. Additional information is available starting on page 58.


7 | P a g e

QUESTIONS EXPLORED

The Board’s student assignment policy (P5101) was designed to be flexible so it can be

monitored and adjusted if it does not work in alignment with other SFUSD initiatives designed

to:



This section of the report explores five questions intended to help monitor the impact of the

current student assignment system.






Findings from the analysis will help the Superintendent and the Board determine if adjustments

need to be made to the current student assignment system.


8 | P a g e


9 | P a g e


For the purpose of this report, all racial/ethnic categories reported by families and subsequently

captured in our Student Information System (SIS) are organized into six categories: African

American; Latino; Chinese; Other Asian (Filipino, Japanese, Korean, South East Asian); Other

(anything not captured in other categories); and White.

Chart 1 illustrates the racial/ethnic breakdown of all students (K-12) enrolled in SFUSD in

October 2012. This graph indicates that SFUSD’s student population is very diverse, with no

majority group.

As the Board was developing student assignment policy P5101 during the 2009-10

school year, they spent time reviewing enrollment and achievement data, and they expressed

concern that the assignment system at the time (i.e., the diversity index lottery) was not meeting

the District’s longtime goals of reducing racial isolation and improving educational opportunities

and outcomes for all students.

• A quarter of schools had more than 60% of a single racial/ethnic group, even though the

District was racially/ethnically diverse and did not have a majority group.

• The number of schools with high concentrations of a single racial/ethnic group had

increased since the implementation of the diversity index lottery.

African American

9%

Latino25%

Chinese32%

Other Asian8%

Other13%

White13%

Chart 1: K-12 enrollment, October 2012


10 | P a g e

• Although SFUSD’s standardized test scores had steadily increased over the years, the

achievement gap persisted for African American, Latino, and Samoan students.

• Racial isolation and the concentration of underserved students (students whose

educational needs have not been met) in the same school was correlated with other

school factors that define the quality of a school, such as average years of teacher

service, teacher turnover, attendance, and suspension rates.

The Board included the following definitions for racial isolation and underserved students in

Board policy P5101.

• Racial Isolation: Although SFUSD’s enrollment is racially/ethnically diverse and does

not have a majority group, many of our schools have more than 60% of a single

racial/ethnic group, more than 70% of a single racial/ethnic group, and more than 80% of

a single racial/ethnic group. Some schools with more than 60% of a single racial/ethnic

group also have an Academic Performance Index (API) of 1, 2, or 3. The Board

considers these schools racially isolated.

• Underserved Students: Students performing Below Basic or Far Below Basic on the

California Standards Test or other equivalent assessments administered by SFUSD.

Taking the concerns noted above, and the definitions of racial isolation and underserved

students, the analysis in this section of the report focuses on schools where the

enrollment is more than 60% of a single race/ethnicity and the Academic Performance

Index (API) is 1, 2, or 3.


11 | P a g e

Table 1 includes the Academic Performance Index (API), number of students enrolled, and

racial/ethnic breakdown of enrollment for any school that had an enrollment of more than 60%

of a single race/ethnicity between 2009 and 2012. Schools with an API of 1, 2, or 3 in 2012

are highlighted since they represent racially isolated schools with a high concentration

of historically underserved students.

Table 1: Schools with an enrollment of > 60% of a single racial/ethnic group (2009-2012)

School Name October API

# Enrolled

% African American

% Chinese

% Latino

% Other Asian

% Other

% White

Elementary Schools

Bryant ES 2009 1 241 3% 0% 90% 2% 5% 0%

2010 1 240 2% 0% 89% 2% 6% 0%

2011 1 248 4% 0% 87% 2% 6% 1%

2012 1 249 2% 1% 90% 2% 4% 0%

Carver ES 2009 1 283 70% 0% 6% 1% 23% 1%

2010 1 282 72% 0% 6% 1% 20% 1%

2011 1 249 70% 0% 9% 1% 20% 1%

2012 1 242 66% 0% 12% 2% 19% 1%

Chavez ES 2009 1 495 2% 1% 87% 2% 8% 0%

2010 1 471 3% 1% 86% 3% 7% 0%

2011 1 480 4% 2% 83% 4% 7% 1%

2012

465 3% 2% 85% 2% 8% 1%

Chin ES 2009 9 263 4% 78% 2% 2% 10% 5%

2010 9 257 2% 80% 2% 3% 8% 5%

2011 9 251 0% 85% 1% 3% 6% 4%

2012 10 264 1% 83% 1% 3% 8% 5%

Cleveland ES 2009 1 307 7% 7% 71% 7% 8% 0%

2010 1 335 5% 6% 73% 8% 7% 2%

2011 1 338 4% 5% 77% 6% 7% 1%

2012 1 346 3% 3% 80% 6% 7% 1%

Drew ES 2009 1* 256 77% 0% 8% 1% 13% 1%

2010 1* 270 78% 0% 10% 2% 10% 1%

2011 1 276 76% 0% 11% 1% 12% 0%

2012 1 286 74% 0% 12% 2% 12% 0%

Fairmount ES 2009 3 364 8% 1% 64% 3% 10% 14%

2010 1 368 5% 1% 67% 2% 9% 17%

2011 4 386 5% 1% 67% 2% 8% 17%

2012 3 391 4% 2% 67% 2% 6% 19%

Flynn ES 2009 1 478 16% 0% 53% 3% 9% 18%

2010 1 482 16% 0% 53% 3% 9% 19%

2011 1 478 15% 0% 57% 3% 6% 19%

2012 1 471 13% 0% 63% 2% 6% 16%


12 | P a g e


# Enrolled

% African American

% Chinese

% Latino

% Other Asian

% Other

% White

Garfield ES 2009 8 228 7% 60% 4% 4% 14% 11%

2010 6 231 7% 58% 4% 3% 16% 12%

2011 7 249 7% 55% 7% 6% 12% 12%

2012 6 255 9% 50% 12% 5% 13% 12%

Lau ES 2009 8 670 0% 82% 11% 1% 6% 0%

2010 7 666 0% 84% 11% 1% 4% 1%

2011 7 668 0% 83% 12% 1% 4% 0%

2012 6 647 0% 87% 8% 1% 5% 0%

Malcolm X ES 2009 1* 104 64% 0% 6% 1% 28% 2%

2010 1* 87 56% 0% 13% 3% 28% 0%

2011 5* 106 67% 1% 7% 5% 20% 1%

2012 4* 112 72% 0% 10% 1% 17% 0%

Marshall ES 2009 2 240 3% 0% 79% 2% 8% 9%

2010 3 239 3% 0% 81% 2% 6% 8%

2011 3 244 3% 0% 83% 2% 5% 7%

2012 4 249 3% 0% 82% 2% 4% 8%

Moscone ES 2009 9 340 2% 30% 58% 2% 7% 2%

2010 8 331 2% 28% 60% 2% 7% 2%

2011 7 336 2% 25% 63% 2% 7% 2%

2012 7 351 1% 27% 63% 1% 6% 1%

Parker ES 2009 8 263 3% 83% 5% 1% 6% 2%

2010 7 272 4% 82% 4% 2% 6% 3%

2011 8 292 3% 81% 5% 1% 7% 2%

2012 7 273 4% 81% 7% 2% 5% 2%

Sanchez ES 2009 1 268 4% 1% 83% 2% 9% 2%

2010 1 260 5% 1% 82% 1% 9% 2%

2011 1 280 6% 1% 81% 1% 7% 4%

2012 1 258 5% 0% 81% 1% 9% 4%

Serra ES 2009 5 267 7% 6% 65% 6% 13% 4%

2010

278 5% 6% 63% 6% 13% 7%

2011 1 282 6% 5% 69% 4% 11% 5%

2012 2 282 7% 5% 71% 3% 10% 5%

Stevenson ES 2009 10 462 2% 68% 1% 5% 17% 6%

2010 10 473 3% 71% 2% 5% 15% 5%

2011 10 469 2% 73% 2% 5% 14% 5%

2012 10 493 1% 75% 2% 3% 15% 4%

Sutro ES 2009 8 238 2% 66% 6% 3% 12% 11%

2010 8 247 2% 66% 3% 5% 12% 11%

2011 9 240 1% 67% 3% 5% 14% 10%

2012 8 242 0% 66% 4% 6% 15% 9%

Ulloa ES 2009 10 508 3% 67% 3% 5% 16% 7%

2010 10 500 3% 69% 3% 5% 15% 6%

2011 10 508 2% 71% 2% 5% 14% 6%

2012 10 507 1% 73% 3% 4% 14% 5%


13 | P a g e


# Enrolled

% African American

% Chinese

% Latino

% Other Asian

% Other

% White

K8 Schools

Lawton K8 2009 10 609 4% 57% 4% 7% 22% 7%

2010 10 613 4% 57% 4% 6% 21% 7%

2011 10 611 3% 61% 4% 6% 20% 6%

2012 10 607 3% 60% 5% 6% 20% 7%

Buena Vista ES 2009 1 375 6% 0% 66% 1% 8% 18%

2010 3 393 5% 0% 70% 1% 7% 17%

Mann MS 2009 1 238 16% 1% 68% 3% 10% 3%

2010 1 198 14% 1% 72% 3% 10% 1%

BV/Mann K8 2011 1 615 7% 0% 72% 1% 6% 14%

(merged in 2011) 2012 2 590 4% 0% 74% 1% 6% 15%

Middle Schools

Brown MS 2009 1 181 73% 0% 4% 1% 22% 1%

(closed in 2011) 2010 1 166 69% 0% 5% 1% 24% 1%

Francisco MS 2009 3 640 11% 59% 13% 3% 11% 3%

2010 2 570 13% 58% 13% 3% 9% 4%

2011 3 587 13% 62% 11% 2% 10% 3%

2012 3 547 10% 64% 10% 2% 11% 3%

Lick MS 2009 4 593 8% 1% 66% 2% 11% 12%

2010 4 571 7% 1% 66% 2% 11% 13%

2011 4 554 7% 2% 64% 2% 11% 14%

2012 4 561 7% 2% 67% 1% 10% 14%

Marina MS 2009 8 904 8% 62% 12% 2% 12% 4%

2010 6 881 9% 54% 14% 4% 14% 6%

2011 7 832 10% 53% 14% 3% 13% 6%

2012 6 818 10% 48% 16% 6% 13% 7%

High Schools

Galileo HS 2009 7 2092 5% 61% 10% 5% 15% 3%

2010 7 2149 5% 61% 10% 6% 16% 3%

2011 7 2133 5% 61% 12% 5% 15% 3%

2012 7 2100 5% 61% 14% 4% 13% 3%

O'Connell HS 2009 1 623 11% 3% 70% 6% 8% 2%

2010 1 604 16% 5% 61% 8% 9% 2%

2011 1 475 16% 7% 58% 8% 9% 3%

2012 1 453 16% 6% 59% 8% 8% 4%

Note: The data above reflects whole school enrollment. However, the current student

assignment system has only been in place for two enrollment cycles, which means only the first

two grades at each school were impacted by the current enrollment system (6 of the 13 grades).


14 | P a g e

Observations

• There was not a shift in the number of schools with an enrollment of more than 60% of a

single race/ethnicity between 2009 and 2012.

o In 2009, 24 schools had an enrollment of more than 60% of a single race/ethnicity:

� 4 African American

� 9 Chinese, and

� 11 Latino

o In 2012, 23 schools had an enrollment of more than 60% of a single race/ethnicity:

� 3 African American

� 9 Chinese, and

� 11 Latino.

• The following is a summary of changes between 2009 and 2012:

o Brown, which was more than 60% African American in 2009, closed in 2011.

o Buena Vista and Mann, which were both more than 60% Latino, merged in 2011.

o Three schools saw a decrease in the percent of a single race/ethnicity.

� Garfield’s enrollment grew from 228 students to 255 students, and the

percent Chinese decreased from 60% to 50%.

� Marina’s enrollment decreased from 904 students to 818 students, and the

percent Chinese decreased from 62% to 48%.

� O’Connell’s enrollment decreased from 623 students to 453 students, and

the percent Latino decreased from 70% to 59%.

o Four schools saw an increase in the percent of a single race/ethnicity.

� Flynn increased from 53% Latino to 63% Latino.

� Moscone increased from 58% to 63% Latino.

� Lawton’s increased from 57% to 60% Chinese.

� Francisco’s enrollment declined from 640 students to 547 students the

percent Chinese increased from 59% to 64%.

Chart 2: # schools with more than 60% of a single race/ethnicity 2009-2012

African American

Chinese

LatinoTotal

20092010

20112012

43

33

97

9 9

11 11 10 11

2421 22 23

African American

Chinese

Latino

Total


15 | P a g e

• Of the 23 school with an enrollment of more than 60% of a single race/ethnicity in 2012,

11 schools had an API of 1, 2, or 3, which means, according to the definitions in Board

policy P5101, 11 schools were racially isolated with high concentrations of historically

underserved students.

• School with more than 60% African American experienced a decrease in the percent African

American between 2009 and 2012.

o More than 60% African American and an API of 1, 2, or 3

School Observations

1. Carver Carver has had an API of 1 for each of the past four years. Its enrollment decreased from 283 students in 2009 to 242 students in 2012, and the percent of African Americans decreased from 70% in 2009 to 66% in 2012.

2. Drew Drew has had an API of 1 for each of the past four years. Its enrollment increased from 256 students to 286 students in 2012, and the percent of African Americans decreased from 77% to 74%.

• Schools with more than 60% Latino offer Spanish language pathways that reserve up to

three quarters of the seats for Spanish speakers.

o More than 60% Latino and an API of 1, 2, or 3

School Observations

1. Bryant All of these schools have Spanish language pathways that reserve seats for Spanish speaking students. 2. Chavez

3. Cleveland 4. Fairmount 5. Flynn 6. Sanchez 7. Serra 8. BV/Mann

• Schools with more than 60% Chinese offer Chinese language pathways that reserve up to

three quarters of the seats for Chinese speakers.

o More than 60% Chinese and an API of 1, 2, or 3

School Observations

1. Francisco Francisco’s enrollment decreased from 640 students to 547 students the percent Chinese increased from 59% to 64%. Francisco has a Chinese language pathway that reserves seats for Chinese speaking students.


16 | P a g e


17 | P a g e

Diversity of Enrollment Compared with Applicants and Attendance Areas

With any choice based student assignment system, diverse applicant pools are critical to achieving diverse enrollment.

Maps 1, 2, and 3 on the next few pages are designed to compare the diversity of enrollment

at the entry grades with the diversity of demand, and, at the kindergarten level, with the

diversity of kindergartners living in the attendance area.

The maps use 100% stacked columns for each school to show the racial/ethnic diversity of the

kindergarten/6th grade/9th grade students who applied to each school compared with the

kindergarten/6th grade/9th grade students enrolled in each school. For kindergarten, there is a

third stacked column showing the racial/ethnic breakdown of all children who live in the

attendance area regardless of the choices they made.

The stacked columns are intentionally the same size; they are not relative in size to the number

of students. Each racial/ethnic group is represented by a different color - the color key is

provided in a box on the right side of each map. Having 100% stacked columns makes it

possible to compare the diversity of the applicants with the diversity of enrollment, and for

kindergarten with the diversity of attendance area residents.

Looking at the stacked columns for the 11 racially isolated schools with high concentrations

of underserved students, we can determine if applicant pools for these schools are more/less

diverse than enrollment (Bryant, Chavez, Cleveland, Fairmount, Flynn, Marshall, Moscone,

Sanchez, Serra, Buena Vista/Mann K8, and Lick middle school). If applicant pools are more

diverse than enrollment, this may suggest the student assignment system is not maximizing the

opportunity to reverse the trend of racial isolation and the concentration of underserved students

in schools.


18 | P a g e

Kindergarten

Map 1: Racial/ethnic diversity of K enrollment, K applicants, and K attendance area residents


19 | P a g e

Observations

• Racially isolated schools with high concentrations of underserved students in 2012 (i.e.,

schools with more than 60% of a single race/ethnicity and an API of 1, 2, or 3) had racially

isolated applicant pools.

o The racial/ethnic composition of these kindergarten applicant pools is very similar

to the racial/ethnic composition of kindergarten enrollment. Applicant pools and

enrollment at schools with more than 60% Latino are predominantly Latino (Bryant,

Chavez, Cleveland, Fairmount, Flynn, Sanchez, Serra, and BV/Mann K8), and the

same is true for schools that are predominantly African American (Carver and Drew).

o While the applicant pools and enrollment at Fairmount, Flynn, Serra, and BV/Mann

K8 are predominantly Latino, the racial/ethnic composition of the kindergarten

applicant pools are slightly more diverse than enrollment at these four schools.

This may be attributed to the presence of Spanish language pathways at the schools

that reserve up to three-quarters of the seats for Spanish speaking students.

o The racial/ethnic composition of kindergartners living in the Carver, Cleveland,

Drew, and Serra’s attendance areas are more diverse than the kindergartners

enrolled in the schools.

• There are geographically distinct patterns for the racial/ethnic composition of students

enrolled in and applying to schools. Kindergartners enrolled in / applying to schools in the:

o north east are primarily Chinese;

o south east they are primarily African American;

o Mission, near Moscone and Chavez, they are primarily Latino;

o west side they are primarily Chinese and White; and

o center of the city, near Grattan and McKinley, they are primarily White.

• In 2012, 10 schools had more than 60% of a single race/ethnicity and an API of 4 or higher:

Chin, Lau, Malcolm X, Marshall, Moscone, Parker, Stevenson, Sutro, Ulloa, and

Lawton.

o The racial/ethnic composition of these kindergarten applicant pools is very similar

to the racial/ethnic composition of kindergarten enrollment.

o While the applicant pools and enrollment at Stevenson, and Sutro are

predominantly Chinese, the racial/ethnic composition of the kindergarten applicant

pools are slightly more diverse than enrollment at these two schools. For Sutro,

this may be attributed to the presence of a Chinese language pathway at the school

that reserves up to three-quarters of the seats for Chinese speaking students. It is

unclear why Stevenson’s enrollment is less diverse than the applicant pool.

o The racial/ethnic composition of kindergartners living in the Chin, Stevenson, Sutro,

and Ulloa’s attendance areas are more diverse than the kindergartners enrolled in

the schools.


20 | P a g e

6th Grade

Observations

• Middle school applicant pools are more diverse than elementary applicant pools. With the

exception of K8 schools, the diversity of each school’s 6th grade applicant pool is similar

to the diversity of the 6th graders enrolled in the schools.

o The variance at K8 schools is because there are very few openings for new students.

For example, only one student was assigned to Rooftop, and that one student is

represented by the column on the left. Alice Fong Yu has one column (applicant

pool) because there were no students assigned to 6th grade in 2012; all the 5th

graders promoted to 6th grade leaving no room for new assignments.

• Aptos, Denman, Visitacion Valley, and King have diverse applicant pools and enrollment.

• Everett and Lick’s applicant pools and enrollment are predominantly Latino. Both

schools have Spanish language pathways that reserve seats for Spanish speaking students.

• Francisco, Marina, Roosevelt, Presidio, Giannini, and Hoover’s applicant pools and

enrollment are predominantly Chinese.

Map 2: Racial/ethnic diversity of 6th grade enrollment and 6th grade applicants


21 | P a g e

9th Grade

Observations

• High school applicant pools are more diverse than elementary applicant pools.

• The diversity of each school’s 9th grade applicant pool is similar to the diversity of the 9th graders enrolled in the schools

• Galileo, Washington, Lincoln, and Lowell applicant pools and enrollment are predominantly Chinese.

• Academy of Arts and Sciences, O’Connell, SF International, and June Jordan applicant pools and enrollment are predominantly Latino.

• International Studies Academy is the only high school with an applicant pool and enrollment predominantly African American.

Map 3: Racial/ethnic diversity of 9th grade enrollment and 9th grade applicants


22 | P a g e

Conclusion

Our current student assignment system has been in place for two years, which means it has been

used to assign students to six out of 13 grades (kindergarten, 1st grade, 6th grade, 7th grade, 9th

grade, and 10th grade). Over these two years, there has not been a shift in the number of schools

with an enrollment of more than 60% of a single race/ethnicity. Despite SFUSD’s diverse student

population, in October 2009, 24 schools (about a fifth of our schools) had an enrollment of more

than 60% of a single race/ethnicity compared with 23 schools in October 2012.

Of the 23 schools in 2012 with more than 60% of a single race/ethnicity, 11 had an API of 1, 2, or

3, which means, according to the definitions in Board policy P5101, 11 schools were racially

isolated with high concentrations of historically underserved students. Two schools with an

enrollment of more than 60% African American and an API of less than 4 experienced a decrease

in the percent African American between 2009 and 2012. Eight schools with more than 60%

Latino offer Spanish language pathways that reserve up to three quarters of the seats for Spanish

speakers, and the same is true for the school that was more than 60% Chinese and had an API of

less than 4.

With any choice based student assignment system, diverse applicant pools are critical to

achieving diverse enrollment. Applicant pools for the 11 racially isolated schools with high

concentrations of underserved students were not diverse – they reflect the lack of diversity of

enrollment.

Applicant pools for most elementary schools are not diverse, and there are geographically distinct

patterns for the racial/ethnic composition of students applying to and enrolled in schools.

Middle school and high school applicant pools are more diverse than elementary applicant pools,

and as a result enrollments at the middle and high school levels are more diverse.

Student assignment has a role to play in reversing the trend of racial isolation and the

concentration of underserved students in schools; however student assignment alone cannot

overcome the complex elements that contribute to the current state. For example, the

demographics of the city, parent request patterns, and language pathways all have an impact on

the demographics of our schools.


23 | P a g e


The Census Tract Integration Preference (CTIP) operates as a preference/tie-breaking factor in

the choice student assignment process for children who live in areas of the city with the lowest

average test score. Areas of the city with the lowest average test scores are called CTIP1. A

detailed description of the method used to identify CTIP1 is provided in Appendix 3.

This section of the report includes an exploration of whether the CTIP1 tie-breaker is achieving

its intended goals, which are to help:

• reverse the trend of racial isolation and the concentration of underserved students in the

same school, and


Review CTIP Classifications

When the Board approved the method of assigning a preference/tie-breaker based on the

characteristics of where a student lives, it was understood that CTIP classifications would be

reviewed as additional test score data became available.

Currently, CTIP regions are based on the average score of SFUSD’s K-12 students in each

Census Tract on the 2006-2009 California Standards Test (CST) English Language Arts (ELA)

exams. That was the most current test score data available when the policy was approved.

Now that we have three additional years of CST data, we are preparing to analyze whether

CTIP1 classifications need to be revised.

Lapkoff & Gobalet Demographic Research, Inc. will construct CTIP1 regions based on the latest

student data, and they will re-evaluate the method used in 2010 to assign Census tracts. If

changes seem to be desirable, they will share their rationale and will explain the changes in

detail with the Superintendent and the Board.

Here’s our tentative timeline for reviewing and possibly revising the CTIP classifications.

• March - July 2013: Using the latest CST ELA scores and student addresses, review

CTIP classifications and re-evaluate the method used in 2010 to assign Census tracts.

• August 2013: Share findings with the Superintendent and Board at a public meeting.

• September 2013: Confirm the CTIP1 areas to be used beginning with the 2014-15

school year.


24 | P a g e

20% Guidelines for High School Choice Process

The high choice process outlined in P5101 gives preference to applicants in transitional years

(i.e., students transitioning from eighth to ninth grade) in the following order, except that

students applying for programs that have eligibility requirements must meet the applicable

requirements for those programs:

1. younger siblings of students who are enrolled in and will be attending the school during the year for which the younger sibling requests attendance;

2. CTIP1, with a minimum of 20% of seats reserved at each high school for students who live in CTIP1 census tracts;

3. all other students

If there are fewer requests than reserved seats for CTIP1 students, the Superintendent has the

discretion to determine whether and when to release reserved seats to other students.

For each of the past two years, the 20% reserve has been released in the March assignment

runs because holding 20% of the seats for CTIP1 residents would have left requested seats

unfilled. This is because the demand for seats from CTIP1 residents was less than the number

of seats available. By releasing the 20% reserve, we were able to offer more students a school

they requested without impacting students living in CTIP1.

Note: CTIP Analysis over Multiple Years

It is too soon to determine whether the CTIP tie-breaker is: (a) helping reverse the

trend of racial isolation and the concentration of underserved students in the same

school; and/or (b) helping provide equitable access to the range of opportunities

offered to students.

We plan to monitor the data over several years. The analysis on the next few pages

is designed to shed some light on the potential impact based on an initial review of

enrollment and demand data for the 2012-13 school year.

In future years we hope to: (a) separate out students with older siblings when

completing the analysis; (b) look at shares of students (percentages) instead of just

looking at the number of students; and (c) explore how many families living in CTIP1

are choosing and enrolling in schools they would not have otherwise enrolled in but for

the CTIP1 tie-breaker.


25 | P a g e

Race/Ethnicity and Number of K/6/9 Students Living in CTIP1

Chart 3 illustrates the number of students enrolled in our schools who live in CTIP1, and the

racial/ethnic breakdown of those students for each of the past four years - 2009 through 2012.

The CTIP1 tie-breaker/preference did not exist prior to 2011, so we thought it would be

interesting to compare the two years prior to the CTIP1 tie-breaker/preference (2009 and 2010)

with the two years the CTIP1 tie-breaker/priority has existed (2011 and 2012).

Observations

• In general, the number of K/6/9 applicants living in CTIP1 and the racial/ethnic diversity of

those applicants has not changed much since the tie-breaker was first used in 2011.

o African American students living in CTIP1 decreased from 761 in 2009 to 631 in 2012.

o Chinese students increased from 215 in 2009 to 260 in 2011, but decreased to 229 in

2012.

o Latino students decreased and increased slightly over the years. They are the largest

cohort of students living in CTIP1 - in 2012 1,166 Latino students lived in CTIP1.

o White students increased from 102 in 2009 to 148 in 2011, and while they decreased to

134 in 2012 the number is still higher than prior to the CTIP1 tie-breaker.

• In future analysis we might want to focus on shares (percentages) of students, rather than

numbers because numbers change over time. Also, the racial/ethnic mix of newborn

populations is changing over time and can affect the diversity of students over time.

2009 2010 2011 2012

White 102 110 148 134

Other Asian 169 146 173 165

Other 300 272 244 304

Latino 1124 1131 1114 1166

Chinese 215 241 260 229

African American 761 667 654 631

0

500

1000

1500

2000

2500

3000Before the CTIP1 Priority With the CTIP1 Priority

Chart 3: Racial/ethnic diversity of K/6/9 students that live in CTIP1 (2009-2012)


26 | P a g e

Kindergarten Enrollment by Race/Ethnicity

To get a better understanding of the impact of the CTIP1 tie-breaker, this section of the report

analyzes enrollment patterns by race/ethnicity for kindergartners living in CTIP1 areas of the

city. The analysis focuses on the four largest racial/ethnic groups (African American, Chinese,

Latino, and White) and it compares enrollment in 2010, the year before the CTIP1 tie-breaker

was introduced, with enrollment in 2012, the second year the CTIP1 tie-breaker was used.

AFRICAN AMERICAN Kindergartners Living in CTIP1 and Enrolled in School

2010

(before CTIP1 tie-breaker) 2012

(second year of CTIP1 tie-breaker)

Number of kindergartners

265 266

Number of schools

51 52

Number of schools with fewer than three kindergartners

32 schools had three or fewer kindergartners

• 15 schools had one kindergartner

• Nine schools had two kindergartners

• Eight schools had three kindergartners



• Four schools had two kindergartners

• Six schools had three kindergartners

Number of schools with four to nine kindergartners

12 18

Number of schools with more than 10 kindergartners

Seven schools had more than 10 kindergartners. 52% of the kindergarteners were enrolled in these seven schools (138 of 265 students):

• Drew – 46 kindergartners

• Carver – 35 kindergartners

• Muir – 14 kindergartners

• King – 11 kindergartners

• Cobb – 11 kindergartners

• Carmichael – 11 kindergartners

• Vis Valley – 10 kindergartners

Seven schools had more than 10 kindergartners. 45% of the kindergarteners were enrolled in these seven schools (119 of 266 students):

• Drew – 30 kindergartners (16 fewer)

• Carver – 25 kindergartners (10 fewer)

• Parks – 22 kindergartners (14 more)

• Malcolm X – 11 kindergartners (five more)

• Muir – 11 kindergartners (three less)

• Harte – 10 kindergartners (one more)

• King – 10 kindergartners (one less)

Observations

It is not yet clear from the data that African American kindergartners living in areas with the

lowest average test scores (CTIP1) are more likely to be enrolled in demographically different

schools because of the CTIP 1 tie-breaker.

• The number of African American kindergartners living in CTIP1 did not change: there were

265 in 2010 and 266 in 2012.

• African American kindergartners living in CTIP1 are enrolled in over 50 schools: in 2010

they were enrolled in 51 schools, and in 2012 they were enrolled in 52 schools.


27 | P a g e

• The majority of schools have fewer than three kindergartners: 32 of the 51 schools in

2010 (63%), and 27 of the 52 schools in 2012 (52%) had 3 or fewer kindergartners.

• While the number of schools with more than 10 kindergartners did not change (7 in 2010

and 2012), the list of schools did change.

o Cobb, Carmichael, and Visitaction Valley had more than 10 kindergartners in 2010,

but fewer than 10 kindergartners in 2012.

o Parks, Malcolm X, and Harte had fewer than 10 kindergartners in 2010, but more

than 10 kindergarteners in 2012.

• While Drew and Carver continue to have the largest number of African American

kindergartners living in CITP1, they both saw a significant decrease in the number

between 2010 and 2012 – 16 and 10 fewer kindergartners respectively.

• Table 2 shows the Academic Performance Index (API) and enrollment demographics in

October 2012 for schools that in 2010 and/or 2012 had more than 10 African American

kindergartners living in an area of the city with the lowest average test scores (CTIP1).

Table 2: October 2012 enrollment demographics and API for schools with more than 10 African American kindergartners living in CTIP1 (2010 and/or 2012)

School African American Chinese Latino Other White API

Carmichael K8 11% 3% 26% 57% 3% 2

Carver ES 66% 0% 12% 20% 1% 1

Cobb ES 57% 2% 18% 17% 7% 3

Drew ES 74% 0% 12% 14% 0% 1

Harte ES 39% 0% 37% 23% 1% 1

King ES 20% 21% 19% 16% 24% 4

Malcolm X ES 72% 0% 10% 18% 0% 4

Muir ES 37% 0% 50% 12% 2% 1

Parks ES 34% 2% 15% 36% 13% 2

o All of the schools are located in/near CTIP1 areas.

o Seven of the schools (70%) have concentrations of historically underserved

students (i.e., an API of 3 or lower)

o At four of the schools (40%), more than 50% of the students enrolled are African

American.


28 | P a g e

CHINESE Kindergartners Living in CTIP1 and Enrolled in School

2010




69 86

Number of schools

29 28




• Six schools had two kindergartners

• One schools had three kindergartners



• Five schools had two kindergartners

• Four schools had three kindergartners

Number of schools with four to nine kindergartners

4 5


One school, Visitaction Valley, had more than 10 kindergartners. 11 of the 26 kindergartners (16%) were enrolled in Visitaction Valley.

One school, ER Taylor, had more than 10 kindergartners. 19 of the 86 kindergartners (22%) were enrolled in ER Taylor.

Observations

It is not yet clear from the data that Chinese kindergartners living in areas with the lowest

average test scores (CTIP1) are more likely to be enrolled in demographically different schools

because of the CTIP 1 tie-breaker.

• The number of Chinese kindergartners living in CTIP1 grew: there were 69 in 2010 and 86

in 2012. In both years, there were less than 100 kindergartners.

• Chinese kindergartners living in CTIP1 are enrolled in about a quarter of our schools: in

in 2010 they were enrolled in 29 schools, and in 2012 they were enrolled in 28 schools.

• Only one school had more than 10 kindergartners; in 2010 it was Visitaction Valley and in

2012 it was ER Taylor.

o ER Taylor saw the greatest increase in the number of Chinese kindergartners living

in CTIP1 – it grew from five in 2010 to 19 in 2012.

o The number decreased by two, from 11 to nine, at Visitaction Valley


October 2012 for schools that in 2010 and/or 2012 had more than four Chinese

kindergartners living in an area of the city with the lowest average test scores (CTIP1).


29 | P a g e

Table 3: October 2012 enrollment demographics and API for schools with more than four Chinese kindergartners living in CTIP1 (2010 and/or 2012)


Chin ES 1% 83% 1% 10% 5% 10

Hillcrest ES 9% 25% 43% 20% 3% 1

Lau ES 0% 87% 8% 5% 0% 6

Moscone ES 1% 27% 63% 7% 1% 7

Spring Valley ES 4% 44% 36% 11% 5% 7

Visitacion Valley ES 15% 47% 15% 24% 0% 5

Yu Alt K8 4% 59% 4% 25% 8% 10

o With the exception of Alice Fong Yu K8 and Hillcrest, the schools are in/near CTIP1

(Moscone and Visitaction Valley) or in Chinatown (Chin, Lau, Spring Valley)

o With the exception of Hillcrest, all the schools have an API of 5 or higher.

o At 3 of the schools (42%), more than 50% of the students enrolled are Chinese.


30 | P a g e

LATINO Kindergartners Living in CTIP1 and Enrolled in School

2010




481 503

Number of schools

49 62



• Eight schools had one kindergartner

• Three schools had two kindergartners




• 10 schools had two kindergartners


Number of schools with four to 14 kindergartners

23 20


Nine schools had more than 15 kindergartners. 51% of the kindergarteners were enrolled in these schools (244 of the 481 students):

• Chavez – 61 kindergartners

• Bryant – 29 kindergartners

• BV/Mann – 28 kindergartners

• Moscone – 28 kindergartners

• Flynn – 25 kindergartners

• Marshall – 24 kindergartners

• Harte – 17 kindergartners

• Tenderloin – 17 kindergartners

• Fairmount – 15 kindergartners

12 schools had more than 15 kindergartners. 58% of the kindergarteners were enrolled in these schools (293 of the 503 students):

• Chavez – 46 kindergartners (15 fewer)

• Moscone – 34 kindergartners (six more)

• BV/Mann – 31 kindergartners (three more)

• Flynn – 26 kindergartners (one more)

• Marshall – 25 kindergartners (one more)

• Sanchez – 23 kindergartners (10 more)

• Bryant – 20 kindergartners (nine fewer)

• Fairmount – 19 kindergartners (four more)

• Alvarado – 19 kindergartners (six more)

• Muir – 17 kindergartners (14 more)

• Hillcrest – 17 kindergartners (four more)

• Tenderloin – 17 kindergartners (same)

Observations

It is not yet clear from the data that Latino kindergartners living in areas with the lowest average

test scores (CTIP1) are more likely to be enrolled in demographically different schools because

of the CTIP 1 tie-breaker.

• The number of Latino kindergartners living in CTIP1 grew: there were 481 in 2010 and 503

in 2012.

• The number of schools with Latino kindergartners living in CTIP1 grew: in 2010 they were

enrolled in 49 schools, and in 2012 they were enrolled in 62 schools.

• The number of Latino kindergartners living in CTIP1 increased at some schools and

decreased at other schools.


31 | P a g e

o The number of kindergartners at Alvarado, Hillcrest, Muir and Sanchez increased

between 2010 and 2012

o Muir and Sanchez experienced the greatest increase: Muir grew from three to 17,

and Sanchez grew from 13 to 23.

o Chavez, Carmichael, and Bryant experienced the greatest decrease: Chavez went

from 61 to 46, Carmichael went from 14 to five, and Bryant went from 29 to 20.

o The number of kindergartners at Harte decreased from 17 in 2010 to 11 in 2012.


October 2012 for schools that in 2010 and/or 2012 had more than 15 Latino kindergartners

living in an area of the city with the lowest average test scores (CTIP1).

Table 4: October 2012 enrollment demographics and API for schools with more than 15 Latino kindergartners living in CTIP1 (2010 and/or 2012)


Alvarado ES 5% 1% 42% 13% 40% 7

Bryant ES 2% 1% 90% 6% 0% 1

Buena Vista/ Mann K8 4% 0% 74% 7% 15% 2

Chavez ES 3% 2% 85% 9% 1%

Fairmount ES 4% 2% 67% 8% 19% 3

Flynn ES 13% 0% 63% 7% 16% 1

Harte ES 39% 0% 37% 23% 1% 1

Hillcrest ES 9% 25% 43% 20% 3% 1

Marshall ES 3% 0% 82% 7% 8% 4

Moscone ES 1% 27% 63% 7% 1% 7

Muir ES 37% 0% 50% 12% 2% 1

Sanchez ES 5% 0% 81% 10% 4% 1

Tenderloin ES 16% 5% 33% 44% 3% 2

o 10 of the 13 schools (77%) are located in/near CTIP1 areas. The remaining three

schools (Alvarado, Hillcrest, and Fairmount) are located relatively close to CTIP1.

o 10 of the schools (77%) have concentrations of historically underserved students

(i.e., an API of 3 or lower)

o At nine of the schools (69%), more than 50% of the students enrolled are Latino.


32 | P a g e

WHITE Kindergartners Living in CTIP1 and Enrolled in School

2010




66 88

Number of schools

28 30




• Four schools had two kindergartners

• Two schools had three kindergartners

22 schools had 3 or fewer kindergartners


• Five schools had two kindergartners

• One school had three kindergartners

Number of schools with four to 14 kindergartners

6 5


No school had more than 10 kindergartners Three schools had more than 10 kindergartners. 36% of the kindergarteners were enrolled in these schools (32 of the 88 students):

• Clarendon – 12 kindergartners (10 more)

• McKinley – 10 kindergartners (two more)

• Lilienthal – 10 kindergartners (seven more)

Observations

It is not yet clear from the data that White kindergartners living in areas with the lowest average

test scores (CTIP1) are more likely to be enrolled in demographically different schools because

of the CTIP 1 tie-breaker.

• The number of White kindergartners living in CTIP1 grew: there were 66 in 2010 and 88 in

2012. In both years, there were less than 100 kindergartners.

• The number of schools with White kindergartners living in CTIP1 grew: in 2010 they were

enrolled in 28 schools, and in 2012 they were enrolled in 30 schools.

• The number of White kindergartners living in CTIP1incrased at some schools and

decreased at other schools.

o The number of kindergartners at Clarendon, Lilienthal, Rooftop, BV/Mann and

Miraloma increased between 2010 and 2012. Clarendon experienced the greatest

increase – it grew from two to 12 kindergartners. With the exception of BV/Mann, all

of these schools have an enrollment that ranges from 28% to 58% White, compared

with a district-wide enrollment that is 13% White.

o Parks and Flynn experienced the greatest decrease: Parks went from six to one,

and Flynn went from eight to two. Both schools have an API of 1 or 2, and an


33 | P a g e

enrollment that is less than 16% White and more than 50% African American and

Latino.


October 2012 for schools that in 2010 and/or 2012 had more than 10 White kindergartners

living in an area of the city with the lowest average test scores (CTIP1).

Table 5: October 2012 enrollment demographics and API for schools with more than 10 White kindergartners living in CTIP1 (2010 and/or 2012)


Clarendon ES 5% 13% 12% 37% 34% 10

Lilienthal K8 9% 8% 11% 37% 35% 9

McKinley ES 14% 3% 14% 14% 55% 6

o All of the schools are located far from CTIP1 and in areas of the city where the

residential demographics are predominantly White.

o All of the schools have an API of 6 or higher.

o The percent of White students enrolled in the schools ranges from 34% to 55%,

compared with a district-wide enrollment that is 13% White.


34 | P a g e

Schools with Greatest Change in Number of Kindergartners Living in CTIP1

To get a better understanding of the impact of the CTIP1 tie-breaker, this section of the report

analyzes enrollment patterns by race/ethnicity at schools that experienced the greatest increase

and the greatest decrease in the number of kindergartners living in CTIP1. Schools were

identified by comparing enrollment in 2010, the year before the CTIP1 tie-breaker was

introduced, with enrollment in 2012, the second year the CTIP1 tie-breaker was used.

Table 6: API and enrollment demographics for schools that experienced an

increase of 10 or more kindergartners living in CTIP1 between 2010 and 2012

School Name Year API African American Chinese Latino Other Asian Other White

Clarendon ES

2010 10 6% 15% 9% 19% 19% 33%

2012 10 5% 13% 12% 19% 18% 34%

Rooftop K8

2010 9 16% 14% 23% 7% 19% 22%

2012 8 15% 12% 23% 6% 16% 28%

Sherman ES

2010 8 5% 35% 11% 4% 15% 31%

2012 10 8% 26% 14% 5% 14% 32%

Taylor ES

2010 8 4% 53% 26% 4% 13% 1%

2012 9 3% 53% 29% 3% 12% 1%

Observations

• None of these schools were racially isolated with high concentrations of underserved

students in 2010 before the CTIP1 tie-breaker existed.

o All schools have an API of 8 or higher.

o Taylor is the only school where more than 50% of the enrollment is a single

race/ethnicity.

• The demographics of the schools did not change much between 2010 and 2012 (the second

year the tie-breaker was used).

o The percent African American decreased at all schools except Sherman.

o The percent Chinese decreased at all schools, except Taylor where it remained the

same.

o The percent Latino increased at all schools, except Rooftop where it remained the

same.

o The percent White increased at all schools, except Taylor where it remained the

same.


35 | P a g e

Table 7: API and enrollment demographics for schools that experienced a

decrease of 10 or more kindergartners living in CTIP1 between 2010 and 2012

School Year API African American Chinese Latino Other Asian Other White

Carver ES

2010 1 72% 0% 6% 1% 20% 1%

2012 1 66% 0% 12% 2% 19% 1%

Chavez ES

2010 1 3% 1% 86% 3% 7% 0%

2012

3% 2% 85% 2% 8% 1%

Drew ES

2010 1 78% 0% 10% 2% 10% 1%

2012 1 74% 0% 12% 2% 12% 0%

Vis Valley ES

2010 7 16% 42% 15% 9% 18% 1%

2012 5 15% 47% 15% 6% 18% 0%

Observations

• Three of the four schools were racially isolated with high concentrations of underserved

students in 2010 before the CTIP1 tie-breaker existed.

o Carver and Drew had APIs of 1 and enrollments that were more than 70% African

American.

o Chavez had an API of 1 and an enrollment that was more than 80% Latino.

• The demographics of the schools did not change much between 2010 and 2012 – Carver,

Drew, and Chavez continue to be racially isolated with high concentrations of underserved

students.

o The percent African American decreased from 72% to 66% at Carver, and from

78% to 74% at Drew.

o The percent Latino decreased from 86% to 85% Latino at Chavez.


36 | P a g e

Conclusion

In October 2012, the two largest cohorts of students living in CTIP1 were Latino (503

kindergartners) and African American (266 kindergartners), and they tend to be enrolled in

schools near CTIP1 that are predominantly Latino and African American. The two smallest

cohorts were White (88 kindergartners) and Chinese (66 kindergartners), and they tend to be

enrolled in schools with a high percent of White and Chinese students.

It takes a while for new policies to show results; therefore it is too soon identify trends. It is

not yet clear from the data that the CTIP1 tie-breaker is helping reverse the trend of racial

isolation and the concentration of underserved students in the same school. We plan to

continue to monitor demand and enrollments, and expect it may take several more years

before trends are revealed.


37 | P a g e


On June 14, 2011, the Board of Education approved staff’s recommendation to develop

elementary to middle school feeder patterns. Starting with the 2012-13 school year, and for four

years thereafter, the elementary-to-middle school feeders will operate as a preference/tie-

breaking factor in the choice process. A detailed description of the middle-school feeders is

provided in Appendix 4. This section of the report explores the impact of the middle-school tie-

breaker on middle school demand and enrollment patterns.

Requests for Middle School Feeders

Table 8 indicates the percent of students who requested their middle school feeder in the

first round of the assignment process for the 2012-13 school year (the first year the tie-breaker

was used as a factor in the choice process) with choice patterns from the same elementary

schools the year before (when the tie-breaker did not exist). For each year, it lists the percent

who requested the middle school anywhere among their choices and the percent who listed it as

a first choice. It also notes the differences between the two years.

Table 8: Percent of 5th grade families requesting the middle school feeder

2011-2012

(without tie-breaker) 2012-2013

(with tie-breaker) Difference

School % Any Choice

% First Choice

% Any Choice

% First Choice

% Any Choice

% First Choice

A P GIANNINI 89% 65% 96% 86% 7% 21%

APTOS 59% 28% 86% 51% 27% 24%

DENMAN 23% 10% 57% 19% 33% 9%

EVERETT 25% 14% 43% 24% 18% 11%

FRANCISCO 56% 25% 66% 35% 11% 10%

HOOVER 70% 29% 80% 44% 10% 15%

ISA 22% 14% 13% 9% -8% -5%

LICK 44% 22% 48% 26% 4% 4%

MARINA 53% 22% 68% 34% 14% 12%

ML KING 48% 33% 62% 43% 13% 10%

PRESIDIO 86% 67% 96% 84% 9% 17%

ROOSEVELT 67% 27% 88% 49% 21% 21%

VIS VALLEY MS 36% 22% 36% 27% 0% 5% Ref: q4_3_1mid


38 | P a g e

Chart 4 illustrates the percent 5th graders promoting to 6th grade who requested their middle

school feeder as a first choice in round one of the assignment process for the 2012-13 school

year compared with first choice patterns for 5th graders to promoting to 6th grade in 2011-12,

the year before the choice tie-breaker was implemented.

Chart 4: Percent of first choice request for middle school feeders

Observations

• The middle school feeders appear to have created a significant shift in the demand

patterns for students promoting from 5th to 6th grade.

• Denman, Aptos, and Roosevelt experienced the greatest change in the percent of 5th

graders who requested their middle school feeder among their choices.

o Denman experienced a 33% increase, Aptos experienced a 27% increase, and

Roosevelt experienced at 21% increase in the percent of families requesting their

middle school feeder.

• With the exception of ISA, every middle school experienced an increase in the number of

families who listed their middle school feeder as a first choice.

o Giannini and Presidio received the greatest percent of first choice requests – 86%

and 84% respectively.

o More than 44% of families that feed into Aptos, Roosevelt, or Hoover listed their

middle school feeder as a first choice.

• ISA experienced an 8% decrease in the percent of families from elementary feeders who

listed it as a choice, and a 5% decrease in the percent who listed it as a first choice.

65%

28%

10% 14%25% 29%

14%22% 22%

33%

67%

27% 22%

86%

51%

19% 24%35%

44%

9%

26%34%

43%

84%

49%

27%

0%

20%

40%

60%

80%

100%

2011-12

2012-13


39 | P a g e

Enrollment in Middle School Feeders

Table 9 and Chart 5 show the percent of students enrolled in 6th grade in 2012-13 who

attended 5th grade at one of the elementary feeder schools compared with 6th grade

enrollment in 2011-12 (i.e., before the tie-breaker for middle school feeders).

Table 9: Percent of students enrolled in 6th grade who attended an elementary feeder

Middle School

% enrolled in 6th grade from elementary feeder – 2011

(without tie-breaker)

% enrolled in 6th grade from elementary feeder - 2012-

(with tie-breaker)

Difference

A P GIANNINI 42% 85% 43%

APTOS 15% 46% 31%

DENMAN 14% 29% 15%

EVERETT 13% 22% 9%

FRANCISCO 25% 39% 14%

HOOVER 28% 55% 27%

ISA 12% 6% -6%

LICK 18% 23% 5%

MARINA 31% 43% 12%

ML KING 34% 44% 10%

PRESIDIO 45% 81% 36%

ROOSEVELT 26% 66% 40%

VIS VALLEY MS 14% 27% 13%

Chart 5: Percent of students enrolled in 6th grade who attended an elementary feeder

42%

15% 14% 13%

25% 28%

12%18%

31% 34%

45%

26%

14%

85%

46%

29%22%

39%

55%

6%

23%

43% 44%

81%

66%

27%

2011-2012 2012-2013


40 | P a g e

Observations

• The middle school feeders appear to have created a significant shift in the enrollment

patterns for students promoting from 5th to 6th grade.

• ISA is the only school with lower enrollment from the elementary schools that feed into

ISA since the middle-school tie-breaker was implemented.

o 12% of students enrolled in ISA in 2011 came from the elementary schools feeding

into ISA.

o 6% of students enrolled in ISA in 2012 (the first year the tie-breaker was used) came

from the elementary schools feeding into ISA.

• Giannini, Roosevelt, Presidio, and Aptos experienced the greatest increase in the percent

of enrollment from the elementary schools feeding into them.

o 42% of students enrolled Giannini in 2011 came from the elementary schools feeding

into Giannini, compared with 85% in 2012 (43% change).

o 26% of students enrolled Roosevelt in 2011 came from the elementary schools

feeding into Roosevelt, compared with 66% in 2012 (40% change).

o 45% of students enrolled Presidio in 2011 came from the elementary schools feeding

into Presidio, compared with 81% in 2012 (36% change).

o 15% of students enrolled Aptos in 2011 came from the elementary schools feeding

into Aptos, compared with 46% in 2012 (31% change).


41 | P a g e

Number of Elementary Schools Feeding into each Middle School

In creating middle school feeders, it was assumed the feeders would reduce the number of

elementary schools each middle school received students from, and this would provide an

opportunity to build connections between families and students and to create seamless

academic pathways for students transitioning from elementary to middle school.

Table 10 and Chart 6 show the number of sending elementary schools for each middle school in

2011-12, compared with 2012-13.

Table 10: Number of elementary schools feeding into each middle school

# ES Feeders

2011-2012 SY Enrollment

(without tie-breaker)

2012-2013 SY Enrollment

(with tie-breaker)

Difference

A P GIANNINI 6 50 43 -7

APTOS 5 49 45 -4

EVERETT 6 44 30 -14

FRANCISCO 5 36 23 -13

HOOVER 5 49 42 -7

ISA 2 24 22 -2

DENMAN 5 42 31 -11

LICK 6 25 27 +2

MARINA 5 52 50 -2

ML KING 3 38 29 -9

PRESIDIO 5 48 38 -10

ROOSEVELT 6 42 37 -5

VIS VALLEY 4 44 25 -19

AVERAGE 5 42 34

Chart 6: Number of elementary schools feeding into each middle school

0

10

20

30

40

50

60

2011-2012 SY 2012-2013 SY # ES Feeders


42 | P a g e

Observations

• The middle school feeder appears to have decreased the number of elementary schools

represented in the incoming class at each middle school. The average number of

elementary schools feeding into middle schools decreased from 42 in 2011 to 34 in 2012.

o In 2011, the number of elementary schools feeding into middle schools ranged from

24 to 52.

o In 2012, the numbers ranged from 20 to 50.

• With the exception of Lick, all schools saw a decrease in the number of elementary

schools feeding into them.

o Visitaction Valley experienced the largest decrease – going from 44 elementary

schools in 2011 to 25 in 2012.

o Everett, Francisco, Presidio, and Denman all experienced a decrease of at least

10 schools between 2011 and 2012.

• Lick, which had one of the lowest number of elementary schools feeding into it in 2011, (25

schools compared with a district average of 42 schools), increased by 2, growing to 27

schools in 2012.

Racial/Ethnic Diversity and Middle School Feeders

This section of the report explores whether the middle school tie-breaker impacts the

racial/ethnic composition of schools by comparing the racial/ethnic composition of incoming 6th

graders at each middle school in 2011-12 (before the tie-breaker) with the racial/ethnic

composition of incoming 6th graders in 2012-13.

Table 11: Race/ethnicity of 6th grade students in 2011 and 2012

Middle School Year African American

Chinese

Latino Other Other Asian

White

A P GIANNINI 2011-2012 7% 51% 10% 13% 7% 12%

2012-2013 7% 49% 9% 10% 11% 14%

A P GIANNINI difference 0% -2% -1% -4% 4% 2%

APTOS 2011-2012 11% 21% 32% 12% 9% 16%

2012-2013 11% 26% 32% 9% 10% 13%

APTOS difference 0% 4% 0% -3% 1% -3%

DENMAN 2011-2012 7% 21% 37% 11% 20% 5%

2012-2013 7% 15% 40% 7% 25% 5%

DENMAN difference 1% -6% 3% -4% 5% 0%


43 | P a g e

Middle School Year African American

Chinese

Latino Other Other Asian

White

EVERETT 2011-2012 16% 8% 42% 9% 10% 15%

2012-2013 10% 3% 52% 11% 5% 18%

EVERETT difference -6% -5% 10% 2% -5% 3%

FRANCISCO 2011-2012 10% 53% 11% 14% 7% 5%

2012-2013 11% 59% 10% 9% 8% 5%

FRANCISCO difference 0% 6% -1% -5% 0% 0%

HOOVER 2011-2012 6% 48% 23% 8% 9% 6%

2012-2013 4% 48% 23% 10% 10% 6%

HOOVER difference -2% -1% -1% 2% 1% 0%

ISA 2011-2012 41% 2% 31% 16% 0% 10%

2012-2013 18% 2% 43% 20% 11% 7%

ISA difference -23% 0% 12% 4% 11% -3%

LICK 2011-2012 10% 1% 69% 9% 1% 11%

2012-2013 4% 1% 70% 9% 4% 12%

LICK difference -6% 0% 2% 0% 3% 1%

MARINA 2011-2012 11% 51% 15% 8% 6% 9%

2012-2013 9% 41% 18% 12% 11% 10%

MARINA difference -1% -11% 3% 3% 6% 1%

ML KING 2011-2012 18% 33% 20% 8% 18% 3%

2012-2013 17% 26% 31% 10% 15% 2%

ML KING difference -1% -7% 11% 2% -3% -2%

PRESIDIO 2011-2012 7% 41% 10% 12% 12% 18%

2012-2013 7% 34% 9% 14% 14% 22%

PRESIDIO difference 0% -6% -2% 2% 3% 4%

ROOSEVELT 2011-2012 7% 33% 14% 12% 15% 20%

2012-2013 5% 42% 12% 14% 12% 15%

ROOSEVELT difference -2% 9% -2% 2% -3% -5%

VIS VALLEY MS 2011-2012 21% 17% 30% 10% 12% 10%

2012-2013 31% 17% 28% 11% 9% 4%

VIS VALLEY difference 10% 0% -2% 0% -3% -6%

OBSERVATIONS

• It is too soon to reach tentative conclusions about whether the middle school tie-breaker is: (a) helping reverse the trend of racial isolation and the concentration of underserved students in the same school; and/or (b) helping provide equitable access to the range of opportunities offered to students.


44 | P a g e

• 6th grade enrollment experienced both increases and decreases in the percent of

students from a single race/ethnicity.

o There was a decrease in the percent of African Americans enrolled in Everett, ISA, and Lick. ISA experienced the most significant decrease, going from 41% African American to 23% African American.

o There was an increase in the percent of African Americans enrolled in Visitaction Valley, going from 21% to 31%.

o There was a decrease in the percent of Chinese enrolled in Denman, Everett, Marina, King, and Presidio. Marina experienced the most significant decrease, going from 51% Chinese to 41% Chinese.

o There was an increase in the percent of Chinese enrolled in Francisco and Roosevelt, going from 53% to 59% at Francisco, and 33% to 42% at Roosevelt. Francisco is a racially isolated school (64% of all students enrolled in the school are Chinese) with concentrations of underserved students (it has an API of 3).

o There was an increase in the percent of Latinos enrolled in Everett, ISA, and ML King, going from 42% to 52% at Everett, 31% to 43% at ISA, and 20% to 31% at ML King.

Conclusion

The middle school feeders had a bigger impact on the choice making process than expected

during the first year of implementation. Most middle schools experienced a significant

increase in the percent of 5th graders from elementary feeder schools requesting their school,

and the number of elementary schools feeding into each middle school decreased.

6th grade enrollment experienced both increases and decreases in the percent of students

from a single race/ethnicity.

Staff will continue to monitor the impact of the feeder patterns in 2013-14 to see if the shifts

noted in year one continue in year two.


45 | P a g e


Outreach and Recruitment

The Educational Placement Center (EPC) employ various outreach strategies to encourage all

families to participate in the choice process, with a specific focus on reaching underserved

communities and groups who do not historically participate in the first round of assignments.

From November 2011 through the January 2012 deadline, EPC staff conducted close to 100

family education and enrollment workshops and outreach events at Head Starts, subsidized pre-

schools, Early Education Schools, elementary schools and numerous community-based

organizations.

Below is a summary of additional outreach strategies employed by EPC for the 2012-13 school

year.

• Established a Satellite Collection site at Carver elementary school in the Bayview for

families to turn in enrollment applications and documents.

• Established a Mobile Enrollment Center. A yellow school bus went to various areas of

the city to conduct outreach, provide counseling and collect applications. The Mobile

Enrollment Center targeted the Mission, Bayview/Hunter’s Point, Tenderloin, Western

Addition, Treasure Island, and Sunnydale/Visitacion Valley areas.

• Conducted “Walk-the-Beat” outreach events. EPC staff walked in high-traffic

neighborhoods (Mission, Chinatown, Visitacion Valley, Bayview) to pass out postcards

with key dates and reminders.

• Sent targeted mailings to the Samoan community through various churches and

community organizations.

• Partnered with the Housing Authority to send mailings to families in public housing.

• Trained community partners, including the Family Support Network and Parents for

Public Schools.

• Provided language assessment and counseling in neighborhoods and schools with high

populations of English language learners.

• Participated in community events, such as the backpack give away in the Bayview, and

Project Homeless Connect.

In addition to EPC’s outreach and recruitment efforts, school communities invest significant

time and energy promoting their schools. For example, all schools offer tours to families,

and every school participates in an annual school fair (that attracts over 10,000 people) to kick

off the enrollment cycle.


46 | P a g e

On Time Participation by Race/Ethnicity

Applications received by the first enrollment deadline in January are considered on-time, and

applications received after the first enrollment deadline through the first day of school are

considered late.

Chart 7: Percent of late applicants by race/ethnicity in 2010, 2011, and 2012

Observations

• With the exception of Other Asian, the percent of late applicants in 2012 compared with

2011 increased for every racial/ethnic group.

A couple of factors played a role in the decreased participation rates between 2011 and 2012.

• The application deadline for 2011 was February 18, and for 2012 it was January 27. This

change resulted in three fewer weeks in 2012 for outreach activities, and it also meant

families moving to the city in early February could not participate on-time.

• Due to ongoing budget cuts, EPC experienced a reduction in staff, including the elimination

of the outreach manager position, and this, in addition to having three fewer weeks,

impacted EPC’s capacity to host outreach events. In 2011 EPC hosted approximately 150

events, and in 2012 it hosted approximately 100 events.

24%

4%

13%

9% 9%

5%

20%

3%

10% 9% 9%

4%

23%

4%

11% 11%9%

6%

0%

5%

10%

15%

20%

25%

30%

AfricanAmerican

Chinese Latino Other Other Asian White

2010

2011

2012

Conclusion

Participation rates appear to be impacted by the timeframe for families to submit

applications and the resources available to support outreach and recruitment. In 2011,

the enrollment deadline was February 18, and in 2012 the enrollment deadline was a

few weeks earlier - January 27. At the same time, EPC experienced a reduction in

staff due to budget cuts. With less time and fewer staff, the number of outreach

events went from 150 in 2011 to 100 in 2012, and on-time participation rates declined.


47 | P a g e


To get a better understanding of what is important to families when choosing schools, we

included a family survey on the 2012-13 application form.

The survey asked families to rate, on a scale of 1 to 5, how important each of the following 15

factors are when requesting schools (1=not important and 5= very important):

• Attend school with sibling

• Attend after school program

• Language pathway

• Special education services

• School bus transportation

• Near home

• Near childcare

• Near work

• School hours

• Diversity of students

• Neighborhood safety

• School’s academic reputation

• Teachers and principal

• Parent community

• Recommended by family/friends

Families were asked to rate, not rank, which means they could have rated all factors with an

equal degree of importance. For example, a family could have rated all factors as (1) not

important or (5) very important.


48 | P a g e

Kindergarten

4,307 families who applied to kindergarten between January 27 and August 20, 2012 responded

to the survey. Table 12 and Chart 8 show the rating for each factor on the scale of 1 to 5 (1=not

important, and 5= very important).

Table 12: % kindergarten families who rated factors 1(not important) to 5 (very important)

How important are the following factors when requesting schools?

1

2

3

4

5

Attend school with sibling 20% 1% 5% 5% 69%

Attend after school program at school 7% 4% 15% 16% 58%

Language pathway 13% 6% 18% 16% 47%

Special education services 39% 9% 18% 9% 24%

School bus transportation 35% 11% 19% 9% 26%

Near home 7% 4% 16% 16% 57%

Near childcare 27% 9% 20% 14% 30%

Near work 25% 10% 26% 14% 25%

School hours 5% 4% 19% 22% 49%

Diversity of students 6% 4% 23% 25% 42%

Neighborhood safety 1% 0% 4% 11% 84%

School’s academic reputation 1% 0% 5% 15% 79%

Teachers and principal 1% 0% 3% 13% 83%

Parent community 1% 1% 9% 23% 66%

Recommended by family/friends 4% 3% 17% 25% 51%

Chart 8: % kindergarten families rated factors 1(not important) to 5 (very important)

69%58%

47%

24% 26%

57%

30%25%

49%42%

84%79% 83%

66%

51%

0%10%20%30%40%50%60%70%80%90%100%

1 Not important 2 3 4 5 Very Important


49 | P a g e

6th Grade

2,579 families who applied to 6th grade between January 27 and August 20, 2012 responded to

the survey. Table 13 and Chart 9 show the rating for each factor on the scale of 1 to 5 (1=not


Table 13: % 6th grade families who rated factors 1(not important) to 5 (very important)


1

2

3

4

5






Near home 10% 5% 15% 16% 54%

Near childcare 32% 9% 21% 12% 27%

Near work 26% 9% 23% 15% 27%

School hours 7% 3% 17% 20% 53%







Chart 9: % 6th grade families who rated factors 1(not important) to 5 (very important)

62% 60%

42%32%

37%

54%

27% 27%

53%45%

87%80%

85%

61%54%

0%10%20%30%40%50%60%70%80%90%100%



50 | P a g e

9th Grade

3,519 families who applied to 9th grade between January 27 and August 20, 2012 responded to

the survey. Table 14 and Chart 10 show the rating for each factor on the scale of 1 to 5 (1=not


Table 14: % 9th grade families who rated factors 1(not important) to 5 (very important)


1

2

3

4

5






Near home 15% 7% 21% 14% 42%

Near childcare 53% 11% 17% 7% 12%

Near work 41% 10% 23% 10% 15%

School hours 13% 6% 23% 21% 37%







Chart 10: % 9th grade families who rated factors 1(not important) to 5 (very important)

40%36%

28%21% 24%

42%

12% 15%

37% 34%

79% 75% 76%

49%44%

0%10%20%30%40%50%60%70%80%90%

100%



51 | P a g e

Observations

• The three factors rated as very important by the largest percent of families were:

neighborhood safety, teachers and principals, and school’s academic reputation.

o Neighborhood safety was rated very important by more families than any other

factor. 84% of kindergarten families, 87% of 6th grade families, and 79% of 9th grade

families rated neighborhood safety as very important.

o Teachers and principals was rated very important by 83% of kindergarten families,

85% of 6th grade families, and 76% of 9th grade families.

o Schools academic reputation was rated very important by 79% of kindergarten

families, 80% of 6th grade families, and 75% of 9th grade families.

• A higher percent of families rated parent community as more important than near home.

o Parent community was rated very important by 66% of kindergarten families, 61% of

6th grade families, and 49% of 9th grade families.

o Near home was rated very important by 57% of kindergarten families, 54% of 6th

grade families, and 42% of 9th grade families.

• A higher percent of kindergarten families rated attend school with sibling (69%) than

near home (57%) as very important.

Conclusion

Over 10,000 families of incoming kindergartners, 6th graders, and 9th graders

responded to the family survey. Neighborhood safety and school quality surfaced as

paramount in the minds of families when choosing schools for their children. Over

75% of families rated neighborhood safety, teachers and principals, and schools

academic reputation as very important. These findings resonate with findings from

various community engagement activities that have taken place over the years.


52 | P a g e


53 | P a g e

ATTENDANCE AREA BOUNDARIES

In September 2010, the Board of Education approved the district’s elementary attendance areas

after more than a year of demographic analysis and a public engagement process that included

an evaluation of suggestions from the community. Throughout this development process, the

following factors were taken into account: neighborhood demographics, where students live now

and where enrollment changes are expected in the future; availability of school facilities; traffic

patterns; location of programs; and coherence of preK-to-K and elementary-to-middle school

pathways.

This section of the report explores four questions:

1. Do attendance areas have the capacity to accommodate kindergarten residents?

2. Do kindergarten applicants request their attendance area school?

3. What modifications were made to the attendance areas for the 2013-14 school year?

4. Does staff recommend any changes to the attendance areas for 2014-15?

Do Attendance Areas have the Capacity to accommodate Kindergarten

Residents?

Map 4 (on the next page) shows how many kindergarten applicants for the 2012-13 school year

live in each attendance area (on-time and late applicants, including those who did not request

their attendance area school), and how that compares to all kindergarten seats available in the

attendance area – including kindergarten seats for citywide schools and programs in the

attendance area. 13 attendance areas have citywide schools, and the kindergarten seats for

each citywide school are added to the total number of seats for that attendance area.


54 | P a g e

Each attendance area has a circle relatively sized according to the ratio of kindergarten

applicants living in each attendance area to kindergarten seats in the attendance area. Red

circles indicate there are more kindergarten residents than seats. Yellow circles indicate there

are fewer kindergarten residents than seat. The larger the circle the greater the ratio of

residents to seats.

Map 4: Kindergarten applicants living in each attendance area relative to

kindergarten capacity at all schools in the attendance area


55 | P a g e

Observations

• Overall, the number of students who applied for kindergarten in 2012-13 exceeded the

number of kindergarten seats available in the district. 5,558 students applied and there

were 5,525 seats, which mean the demand for kindergarten was 106% of seats. Because

100% of applicants don’t end up enrolling in our schools, we were able to accommodate all

kindergartners who wanted to enroll in our schools.

• During the development of the boundaries approved by the Board in September 2010, we

identified a mismatch between where students live and where schools are located, and

discovered that given the size and distribution of schools throughout the city it’s not possible

to create attendance areas that can accommodate all students living in them. The

residential patterns of kindergarten applicants for the 2012-13 school year is consistent with

those findings.

o Technically, 23 out of 58 of attendance areas (40%) had the capacity to

accommodate all kindergarten applicants living in the attendance area (yellow

circles), and 35 attendance areas (60%) had more residents than seats (red circles).

o Many attendance areas in the west, northeast, and central areas of the city with

more kindergarten residents than kindergarten seats (red circles) are located near

attendance areas that have more seats than residents (yellow circles). For example,

on the west side, Key and Jefferson have more residents than seats but they are

located near Stevenson and Sunset which have fewer residents than seats.

o The southeast has significantly more kindergarten residents than kindergarten seats.

Kindergarten applicants living in Carver’s attendance area (regardless of their

requests) are about four times the number of seats in Carver’s attendance area. The

schools near Carver (Harte, Malcolm X, and Drew) also have more residents than

kindergarten seats.


56 | P a g e

Do Kindergarten Applicants Request their Attendance Area School?

Map 5 has a color coded pie in each attendance area to illustrate request patterns for all

kindergarten applicants for the 2012-13 school year (both on-time and late) living in the

attendance area – regardless of the school they enrolled in. The larger the pie the greater the

number of kindergarten applicants living in the attendance area.

Red indicates the percent of kindergarten applicants living in the attendance area who

requested their attendance area school as a first choice; green indicates second or third choice;

blue indicates fourth or lower choice; and orange indicates the percent who did not request their

attendance area school.

Map 5: % Kindergartners who requested their attendance area school


57 | P a g e

Observations

51% of residents did not request their own attendance area school anywhere among their

choices.

o Over 50% of residents in 32 of the 58 attendance areas did not request their

attendance area school anywhere among their choices.

o Less than a quarter of residents in Drew, Muir, Cobb, Webster, and Carver’s

attendance areas requested their attendance area school as one of their choices.

o Drew (14%) and Muir (21%) had the lowest percent of requests from students living in

the attendance area.

Only 26% of applicants requested their attendance area school as their first choice.

o Lau had the greatest percent of residents requesting their attendance area school as a

first choice; 60% of all kindergarten residents of Lau’s attendance area listed Lau as a

first choice.

o More than half of the residents in Lau, Sherman, Grattan, Clarendon, and Sunset

requested their attendance area school as a first choice.

More than 75% of residents in 12 of the 58 attendance areas listed their attendance area

school somewhere among their choices: McKinley, Sunset, Grattan, Garfield, Sherman,

Alvarado, Clarendon, Lau, Alamo, Argonne, West Portal, and Parker.

o McKinley had the greatest percent of residents requesting their attendance area school;

86% of all applicants living in McKinley’s attendance area listed McKinley among

their requests.


58 | P a g e

What Modifications were made to the Attendance Areas for the 2013-14 School

Year?

Following the March 2011 assignment offers, community members recommended the following

changes to the elementary attendance area boundaries:

1. Alvarado (Glen Park) - move boundary south from 29th to 30th street;

2. Alvarado (Milk) - move boundary north nearer crest of hill;

3. Grattan (McKinley) - move boundary east to include Upper Haight;

4. Sloat (Miraloma) - move boundary north to include St. Francis Wood;

5. Sunnyside (Miraloma) – move boundary north to include Sunnyside Playground and

Sunnyside Conservatory; and

6. Parks (Sherman) – move boundary north of Geary Boulevard.

When reviewing the six recommendations received from the community in 2011, SFUSD staff

considered the same factors taken into account when developing the attendance area

boundaries approved by the Board in September 2010 (i.e., neighborhood demographics, where

students live now, where enrollment changes are expected in the future, etc.).

Staff reached the following conclusions.

• Alvarado’s attendance area should not be enlarged because the number of kindergarten

applicants in 2011-12 exceeded the number of kindergarten seats in the attendance area.

In addition, 50% of the seats at Alvarado are for the Spanish Immersion pathway and

therefore the attendance area tie-breaker only applies to 50% of Alvarado’s seats.

• Grattan’s attendance area should not be enlarged to encompass some of McKinley’s

attendance area because Grattan’s attendance area has more residents than seats; and

McKinley’s attendance area should not be made smaller because it has fewer residents than

seats.

• Both Sloat and Miraloma have more kindergarten residents than seats; therefore neither

attendance area should be enlarged.

• Sherman’s attendance area should not be reduced to increase Parks’ attendance area

because both attendance areas have fewer residents than seats. In addition, there are

concerns about moving the Parks attendance area north of Geary since the high traffic

patterns are considered a topographical barrier.

• Sunnyside has fewer residents than seats, and Miraloma has more residents than seats.

Therefore we should adjust the boundary to reduce the size of Miraloma’s attendance area

and increase the size of Sunnyside’s attendance area by moving Sunnyside’s boundary

north to encompass Sunnyside Playground and Sunnyside Conservatory.


59 | P a g e

Adjusting the Miraloma/Sunnyside Attendance Area Boundary for 2013-14

The specific changes to the boundary between Miraloma and Sunnyside were shared and

discussed in different ways during the 2011-12 school year.

• The March 2012 Annual Report on Student Assignment included an analysis of our

elementary attendance area boundaries, responses to recommendations from the

community to adjust certain attendance area boundaries, and the proposed adjustments to

Miraloma and Sunnyside’s attendance areas.

• This adjustment was discussed by the Ad Hoc Committee Meeting on Student Assignment

on March 12, 2012.

• Information was posted on the SFUSD web page.

Map 6: Modifications to Sunnyside and Miraloma

At the August 28, 2012 Board meeting, the Superintendent shared that the elementary

attendance area between Miraloma and Sunnyside would be adjusted prior to launching the

enrollment process for the 2013-14 school year.

This change is reflected in the maps developed by EPC for the 2013-14 enrollment cycle.


60 | P a g e

Does Staff Recommend any Changes to the Attendance Areas for 2014-15?

In the fall of 2012, the Japanese Bilingual Bicultural Program (JBBP) community at Rosa Parks

restated its request that staff consider moving the Rosa Parks boundary north of Geary

Boulevard. The JBBP community at Rosa Parks feels a strong connection to the Japantown

community north of Geary, and they are concerned that the attendance area for Rosa Parks will

divorce the Japantown community from the JBBP program.

Staff continues to be concerned about moving the Rosa Parks attendance area north of Geary

since the high traffic patterns are considered a topographical barrier.

1. The Board’s guidelines for drawing attendance area boundaries require staff to take

traffic patterns into consideration, and this adjustment would go against those guidelines.

2. The Board’s general education transportation policy guidelines indicate that we will

provide limited transportation to support reasonable access for attendance area

residents to attend their attendance area school. Geary Boulevard is a major traffic

thoroughfare that would create a topographical obstacle in the attendance area, and as

a result the District may have to consider providing school bus transportation to Rosa

Parks students living on the other side of Geary Boulevard.

Staff recognizes and appreciates the connection between the JBBP community at Rosa Parks

and the Japantown community north of Geary Boulevard. The JBBP program at Rosa Parks is

a city-wide program, which means for the purposes of enrollment it does not have an

attendance area. Students living in the Rosa Parks attendance area do not get an attendance

area tie-breaker for student assignment; all students in the city, regardless of where they live,

have the same opportunity to get assigned to the JBBP program at Rosa Parks.

Taking the Board’s policy guidelines for attendance area boundaries and general education

transportation into account, along with the recognition that the JBBP program at Rosa Parks

is a city-wide program that does not give any preference to students living in Rosa Park’s

attendance area, staff is not recommending any changes to the Rosa Parks attendance area at

this time.

Conclusion

There is a mismatch between where students live and where schools are located; 35 of 58

attendance areas had more kindergarten residents than seats. At the same time, more than

half of kindergarten applicants did not request their attendance area school anywhere among

their choices, and only 26% requested it as a first choice. Requests for schools relieve the

mismatch between where students live and where schools are located. Staff is not

recommending any changes to the attendance areas for the 2014-15 enrollment cycle.


61 | P a g e

GENERAL EDUCATION TRANSPORTATION SERVICES

The Board approved a new policy for general education transportation services in December

2010 (see Appendix 5), and in February 2011, the Superintendent presented the Board with a

proposal to change general education transportation routes over a three year period to

bring services in alignment with the Board’s general education transportation policy, and the

district’s budget reduction goals.

Due to ongoing state budget cuts, the district reduced the fleet of buses offering general

education transportation from 44 buses in 2010 to 25 buses by August 2013.

The table below summarizes the changes over the past three years; it describes the size of the

fleet, the number of schools served, and the number of students receiving general education

transportation services.

School Year

# Buses

# Schools # Elementary riders

# Middle riders

2010-11 44 59 elementary / five middle 3,300 358

2011-12 38 48 elementary/ five middle 2,400 379

2012-13 30 42 elementary/ six middle 1,618 309

2013-14 25 39 elementary/ six middle unknown unknown

The reductions in service between 2010 and 2013 reduce the general fund contribution to

transportation by about $1.9 million a year; it costs about $100,000 per bus and we will have

reduced the fleet by 19 buses by August 2013. A fleet of 25 buses by 2013 means the cost of

providing general education transportation services in 2013 should be aligned with the funds the

district will receive for home to school transportation.

General Education Transportation Services and Middle School Feeders

Over the past couple of years staff worked to align middle school services with the middle

school feeders. In 2012-13, services were added:

• from Carver and Starr King elementary schools to Aptos middle school, and

• from Moscone and Serra elementary schools to Hoover middle school.

Beginning in August 2013, SFUSD will no longer provide services from:

• the Bayview to Hoover middle school (services will be maintained from the Bayview to

Aptos, Giannini, and Lick middle schools); or

• the Mission area to Aptos middle school (services will be maintained from the Mission

area to Hoover middle school).


62 | P a g e

Changes Implemented for the 2011-12 School Year (6 fewer buses)

11 schools lost general education transportation services in August 2011. • Chavez • Garfield • Grattan • Key • Marshall • Miraloma

• Parker • Peabody • Sunset • Sutro • Yick Wo

Four schools experienced a reduction in general education transportation services, but kept transportation from areas of the city with the lowest average test score and stops to after school providers.

• Alice Fong Yu • Clarendon • Lilienthal • Rooftop

Changes Implemented for the 2012-13 School Year (8 fewer buses)

• 6 elementary schools lost general education transportation services in 2012-13:

o Bryant, Cleveland, SF Community, Cobb, McKinley, and Sheridan.

• 42 elementary schools received modified services in 2012-13.

o Stops with low ridership and stops within ½ mile of each other were deleted.

o One bus instead of two buses per school was available to provide services to Hillcrest, Ortega, Parks, Taylor, and Ulloa.

o Two buses instead of three buses per school was available to provide services to Sherman and Rooftop.

o One bus instead of three buses was available to maintain services for Carver and Malcolm X.

o One bus instead of two was available to provide services from Mission to Aptos.

• Services from Carver and Starr King to Aptos and services from Treasure Island to Marina were added, providing new services to students who are going to attend their middle school feeder.


63 | P a g e

Changes for the 2013-14 School Year (5 fewer buses)

All decisions regarding changes for the 2013-14 school year were guided by the Board’s

transportation policy guidelines and budget reduction goals. Within this context, every effort

was made to:

• avoid changes to bell schedules;

• minimize the number of schools impacted;

• minimize the number of riders impacted;

• align middle school services with the middle school feeders; and

• align with the enrollment cycle.

Summary

• There will be five fewer buses (going from 30 buses in 2012-13 to 25 buses in 2013-14).

• We will no longer provide general education transportation services to ER Taylor, Gordon J

Lau, or New Traditions.

• We will no longer provide general education transportation from the Bayview to Hoover

Middle School (we will maintain services from the Bayview to Aptos, Giannini, and Lick).

• We will no longer provide general education transportation from the Mission to Aptos Middle

School (we will maintain services from the Mission area to Hoover Middle School).

• The reduction of five buses in August 2013 will reduce the use of general fund budget

contributions to transportation by approximately $500,000 ($100,000 per bus).

General Education Transportation Services for the 2013-14 School Year

Note: Per State and Federal law, transportation will be provided to Special Education students

who have this service written into their Individualized Educational Program (IEP).

Families who want to receive general education transportation services in 2013-14 will have to

submit a Request for General Education Transportation Services. The request process for

the 2013-14 school year will begin in April 2013.

• There is no guarantee that general education services will be available to all students who request the services next year.

• To ensure equitable access to our limited transportation services, the following factors will be considered if there are more requests than our limited services can accommodate: Program Improvement status, public housing, areas of the city with the lowest average test scores, grade, enrollment in an Early Education School, and 6th graders attending their middle school feeder.


64 | P a g e

The following general education transportation services will be available for the 2013-14 school

year.

• 39 elementary schools will receive limited transportation:

1. ALAMO

2. ALVARADO

3. ARGONNE

4. BV/MANN K-8

5. CARMICHAEL K8

6. CARVER

7. CHIN

8. CLARENDON

9. DREW

10. EL DORADO

11. FAIRMOUNT

12. FLYNN

13. GLEN PARK

14. HILLCREST

15. LAFAYETTE

16. LAKESHORE

17. LAWTON K-8

18. LILIENTHAL K-8

19. MALCOLM X

20. MEC

21. MILK

22. MUIR

23. ORTEGA

24. PARKS

25. REVERE K-8

26. ROOFTOP K-8

27. S KING

28. SANCHEZ

29. SHERMAN

30. SPRING VALLEY

31. SPRING VALLEY

32. STEVENSON

33. TENDERLOIN

34. TENDERLOIN

35. ULLOA

36. VIS VALLEY

37. WEBSTER

38. WEST PORTAL

39. YU K-8

• 6 middle schools will receive limited transportation:

1. APTOS

2. FRANCISCO

3. GIANNINI

4. HOOVER

5. LICK

6. MARINA

Bus schedules for 2013-14 have been established and posted at www.sfusd.edu/transportation.

They were also distributed to each school with transportation services, and they have been

shared with the Placement Counselors in EPC.

Information about alternatives to yellow bus transportation has been consistently shared with

families over the past three years. For example, multilingual materials sent to families and

information posted on the web has directed families to the following organizations.

o Safe Routes to School: www.sfsaferoutes.org

o Walking School Bus: www.walksf.org

o Family Biking: www.sfbike.org

o SchoolPool: www.WePool2School.org

o MUNI: www.sfmta.com


65 | P a g e

NEXT STEPS

Discuss Annual Report with the Board of Education

SFUSD staff plan to discuss this report with the Board of Education at a public meeting in the spring of 2013. This public forum will provide an opportunity for the Board to give feedback to staff, and for the Board and staff to gather feedback from all stakeholders.

Review and Possibly Revise the CTIP Classifications

SFUSD staff are working with demographers to review the CTIP classifications. Here’s our tentative timeline for reviewing and possibly revising the CTIP classifications.

• March - July 2013: Using the latest CST ELA scores and student addresses, review CTIP classifications and re-evaluate the method used in 2010 to assign Census tracts.

• August 2013: Share findings with the Superintendent and Board at a public meeting.

• September 2013: Confirm the CTIP1 areas to be used beginning with the 2014-15 school year.

Recommend Elementary Feeders for Willie Brown, Jr. Middle School

Willie Brown Jr. middle school is scheduled to open on time-for the 2015-16 school year. This means we must identify elementary feeder schools for Brown middle school by the time the enrollment cycle kicks-off in the fall of 2014.

SFUSD staff plan to develop recommendations to share with the Board during the 2013-14 school year. The goal is to finalize the elementary feeders for Brown by the spring of 2014 so the information can be incorporated into the 2015-16 enrollment materials (i.e., on time for the fall 2014 enrollment fair).

Explore Additional Research Questions

Through our partnership with Stanford, and under the supervision of Sean Reardon, Professor of Education, Stanford is exploring research questions related to SFUSD’s student assignment system. This research involves the analysis of all K-12 assignments made through the Educational Placement Center (EPC) beginning with the 2004-05 school year through the 2012-13 school year.

Stanford Research Organization is investigating how district policies shape the distribution of students among K12 schools in SFUSD, which a specific focus on the following.

1. How school preferences/rankings are affected by the location of programs, characteristics of programs, information available, and the student assignment system. This involves looking at what schools/programs families list on their forms, how it has changed over time as the location of programs have changed, program characteristics have changed, information has changed, and the assignment system has changed.


66 | P a g e

2. How choice patterns affect the size and diversity of school enrollment. This involves looking at how the size and diversity of the student population in schools changed over time, and how the diversity of students assigned to schools and ultimately attending those schools compares to the diversity of students who requested schools.

3. How the student assignment system affects the distribution of students among schools and programs, and how this distribution of students among schools and programs affects academic outcomes. This involves looking at how school and program assignments affect families’ decisions about enrolling their students in SFUSD, how well the student assignment system produces equitable distributions of students among schools and programs, and how changes in policy might affect both the distributions of students among schools and programs and achievement patterns/gaps.


67 | P a g e

APPENDIX


68 | P a g e


69 | P a g e

1. Historical Context

Consent Decree

In 1978, the San Francisco NAACP brought a case against SFUSD and the State of California.

The NAACP argued that the SFUSD and the State engaged in discriminatory practices and

maintained a segregated school system in violation of the U.S. Constitution, federal statues, and

the State of California Constitution.

In 1983, the U.S. District Court approved a type of agreement between the parties called a

“Consent Decree” which had two primary goals for the SFUSD:

• continued and accelerated efforts to achieve academic excellence for all students with a

particular focus on African American and Latino students; and

• elimination of racial/ethnic segregation or identifiability in any school, program, or

classroom to the extent practicable.

In implementing the 1983 Consent Decree, SFUSD created a student assignment plan and a

transportation system designed to support SFUSD’s efforts to desegregate its schools. The

student assignment plan used a combination of schools with both contiguous and

noncontiguous attendance areas, alternative schools (without attendance areas), and optional

enrollment requests which allowed students to transfer to schools outside of their attendance

area school. In addition, no school could have fewer than four racial/ethnic groups, and no

racial/ethnic group could constitute more than 45% of the students at attendance area schools

or 40% at alternative schools.

In 1994, a group of San Francisco parents sued the SFUSD for using race as a factor in school

assignment, and as part of a 1999 settlement, SFUSD was prohibited from using race or

ethnicity as a consideration in student assignment. In attempting to comply with that

agreement, SFUSD initially proposed an assignment plan that used a lottery process in which

race/ethnicity was one factor, but the Court rejected that plan.

In 2001, the Court approved a settlement agreement that included a new student assignment

method called the Diversity Index, which was implemented for the 2002-03 school year and

was used through the 2010-11 school year. The Diversity Index was designed to:

• give families choice;

• ensure equitable access; and

• promote diversity without using race/ethnicity.

On December 31, 2005, the Consent Decree expired, and for the first time in 22 years the

SFUSD student assignment process was not regulated by the courts.


70 | P a g e

Diversity Index

The Board had many concerns about the diversity index, not least of which was that it was not

meeting the Board’s longtime goal of reducing racial isolation and improving educational

opportunities and outcomes for all students. The number of schools with high concentrations of

a single racial/ethnic group increased over the years under the diversity index. In 2008, a

quarter of SFUSD’s schools had more than 60% of a single racial/ethnic group, even though

SFUSD’s overall enrollment was racially/ethnically diverse and did not have a majority group.

In addition, although SFUSD had opened, closed, merged, and redesigned schools, the

attendance area boundaries had not been revised since the 1980s. The Board was also

concerned that some schools were over enrolled while others were under enrolled, and that the

participation rates in the choice process varied greatly by race/ethnicity. Finally, many families

reported finding the system time consuming, unpredictable, and difficult to understand.

Hearing from the Community

SFUSD partnered with different community members and organizations over the years to gather

feedback on the student assignment system. SFUSD has heard from thousands of families and

other community members regarding their experience, concerns, and suggestions for student

assignment. Families consistently report wanting quality schools and a fair and equitable

system that is easy to understand.

While families consistently report wanting quality schools, there are many divergent

perspectives on what student assignment should prioritize and support. For example, some

families want a school close to home, while others feel that choosing a school with particular

programmatic features is more important than having a school close to home.

Here is a high-level summary of key findings from the community reports. For the purposes of

the summary, we have pulled out findings that relate only to the student assignment system.

• Most families want their school communities to reflect San Francisco’s socioeconomic

and cultural diversity. But for families across the city, diversity is often trumped by a

school’s location, academic quality, and their own feeling of belonging.

• Even families who are happy with their children’s schools want more predictability in the

enrollment process and are uncomfortable with a process that feels excessively

complicated or random.

• Families want SFUSD to provide clear and accessible information that will help them

choose a school that is a good fit for their child.

• Families want to participate fully in the enrollment process, but many encounter

significant language, time, and information barriers.


71 | P a g e

Current Policy

In December 2008, the Board convened an Ad Hoc Committee on Student Assignment to

provide a regular and public way for the Board to conduct public policy discussions with staff

about the redesign of student assignment. Between December 2008 and January 2010, the

Board held monthly Ad Hoc Committee meetings, and staff, with assistance from local and

national partners and guidance from the Board, analyzed current conditions, explored different

student assignment options, and gathered additional feedback from the community.

Key findings from the research and analysis captured the complexity of designing a student

assignment system that could meet the Board’s goal of reversing the trend of racial isolation

and the concentration of underserved students in the same school.

• Neighborhood schools are limited in their ability to reverse the trend of racial isolation and the concentration of underserved students in the same school, although under a neighborhood system some schools might be less racially concentrated than they are today, and many schools might have a more robust enrollment.

• Different choice systems are limited in their ability to reverse the trend of racial isolation and the concentration of underserved students in the same school because the applicant pools for individual schools are racially isolated, and all families do not have the same opportunity to understand which schools they like and to submit their choices on-time for the assignment process.

• To reverse the trend of racial isolation and the concentration of underserved students in the same school through student assignment alone, the Board would need to assign students to schools they have not historically requested and to schools far from where they live. For example, some students living on the west side of the city and in the north of the city would need to be assigned to schools on the east side of the city and the southeast side of the city, and vice-versa.

SFUSD staff concluded that a new student assignment system is one part of creating

educational environments in which all students can flourish. School quality is the paramount

concern, and a student assignment system alone cannot ensure school quality, although it does

have a role to play in creating diverse learning environments and robust enrollments in all

SFUSD schools.

In March 2010, the Board unanimously approved a student assignment policy (P5101) that

maintained choice as a tactic for achieving its goals, but that simplified the system and

differentiated it for elementary, middle, and high school. The student assignment system

places students in their highest ranked requests as long as there is space. If there are

more requests for a school than openings, the student assignment system sorts all

requests using a series of preferences, called tie-breakers, to assign applicants to

schools.

An overview of the student assignment tie-breakers used for 2012-13 school year enrollments is provided in Appendix 2, and Board policy P5101 is available on the web at www.sfusd.edu.


72 | P a g e


73 | P a g e

2. Student Assignment Tie-Breakers: 2012-13 School Year

SFUSD’s student assignment system is a school choice system designed to place students in

schools within SFUSD in adherence to Board of Education Policy 5101. This summary provides

a high-level overview of the tie-breakers used to make school assignments in March 2012 for

the 2012-13 school year.

Tie-Breakers

Students are placed in their highest ranked choice as long as there are openings. If there are

more requests for a school than openings, the student assignment system sorts requests using

a series of preferences, known as tie-breakers, to place applicants in schools.

The following tie-breakers are request level tie-breakers, which means they are applied to

specific requests from students.

• AA & PreK. Requests from students who live in the attendance area of the school and

are also enrolled in an SFUSD preK in the same attendance area.

• AA. Requests from students who live in the attendance area of the school requested.

• CL. Requests from students who are enrolled in and wish to continue in a language

program.

• CLS. Request from students who are enrolled in and wish to continue in a language

program AND who are the younger sibling of a students who is enrolled in and will be

enrolled in the language program at the school at issue.

• MSF. Requests from students who attend an elementary K-5 which is identified as a

school that feeds into a specific middle school.

• PreK. Requests from students who attend an SFUSD preK program at the city-wide

school they are applying to.

• Sibling. Requests from a younger sibling of a student who is enrolled in and will be

attending the school.

The following tie-breakers are student level tie-breakers, which means they are applied to all

requests submitted by a student who meets the parameters for the tie-breaker in question.

• CTIP1. Students who lived in areas of the city with the lowest quintile of average test

scores.

• NCLB. Students who attended a Program Improvement school or an Open Enrollment

School.


74 | P a g e

Cohorts Sets

To determine the order in which requests are sorted, each request is assigned to one or more

“cohort”. A cohort is a group of students or requests that shared a tie-breaker. For example,

requests submitted to a school by students who have older siblings attending the school are

part of the “sibling” cohort.

For each type (citywide or attendance area) and level (elementary, middle, high) of school and

program, a different list of cohort orders is used – these were are “Cohort Sets.”

The following is a list of the 10 different Cohort Sets used in the March 2012 student

assignment run.

1. Kindergarten, non-citywide

a. Sibling

b. AA & PreK

c. CTIP 1

d. AA

2. Kindergarten, citywide

a. Sibling

b. PreK

c. CTIP 1

3. Kindergarten, citywide language pathway

a. CLS

b. CL

c. Sibling

d. PreK

e. CTIP 1

4. Non-transitional elementary grades, non-citywide

a. Sibling

b. NCLB

c. CTIP 1

d. AA

5. Non-transitional elementary grades, citywide

a. Sibling

b. NCLB

c. CTIP 1


75 | P a g e

6. Non-transitional elementary grades, citywide language pathway

a. CLS

b. CL

c. Sibling

d. NCLB

e. CTIP 1

7. Middle transitional grade

a. Sibling

b. MSF

c. CTIP 1

8. Non-transitional middle

a. Sibling

b. NCLB

c. CTIP 1

9. High transitional grade

a. Sibling

b. CTIP 1

10. Non-transitional High

a. Sibling

b. NCLB

c. CTIP 1

For more information about other aspects of SFUSD’s student assignment system, please visit

our website at www.sfusd.edu.


76 | P a g e


77 | P a g e

3. Census Tract Integration Preference (CTIP)

The Census Tract Integration Preference (CTIP) operates as a preference/tie-breaking factor in

the choice student assignment process for children who live in areas of the city with the lowest

average test score. Areas of the city with the lowest average test scores are called CTIP1.

Students who live in CTIP1areas of the city receive a preference, known as the CTIP1 tie-

breaker, in the choice assignment process. The CTIP1 preference is intended to help:

• reverse the trend of racial isolation and the concentration of underserved students in the

same school, and

• provide equitable access to the range of opportunities offer d to students.

Method Used to Identify Areas of the City with the Lowest Average Test Scores

In December 2010, Lapkoff & Gobalet Demographic Research Inc. (LGDR) identified areas of

the city with the lowest average test scores.

• They computed the average 2006-2009 CST English Language Arts score for each

combined Census Tract, using records for 144,830 K-12 students. They adjusted the

quintiles slightly to spread numbers of K-12 students across the quintiles fairly evenly

(approximately 20 percent of students per quintile). They then gave students in the lowest

quintile a Census Tract Integration Preference (CTIP) score of 1. Residents of the highest

quintile received a CTIP score of 5.

• The average test scores for each of the CTIP regions ranged from 297 to 407. Table A

shows the average score of residents of each type of CTIP region as well as the ranges of

scores in each. CTIP 1 regions had average scores between 297 and 331; CTIP 2 regions

had average scores between 332 and 345, etc.

Table A: Description of CTIP Regions (shown in Map A)

Lapkoff & Gobalet Demographic Research, Inc., computations. % may seem not to total 100 percent because of rounding.

CTIP region Description

Shading on map

Lowest

average CST-ELA score of

tracts in CTIP region

Highest

average CST-ELA score of

tracts in CTIP region

Number of

2006-2009 K-12 student

residents with CST scores

Percent of K-12 test-

taking students who

were in CTIP region

Percent of all K-12 students

who were in CTIP region

CTIP 1 lowest quintile darkest

green

296.7 331.4 29,783 20.6% 21.3%

CTIP 2 next-to-lowest

quintile

medium

green

331.5 345.0 29,105 20.1% 19.8%

CTIP 3 middle quintile lighest green 345.2 357.4 28,700 19.8% 19.7%

CTIP 4 next-to-highest

quintile

lighter purple 358.1 376.4 28,770 19.9% 19.3%

CTIP 5 highest quintile dark purple 376.4 407.0 28,472 19.7% 19.9%

Total 296.7 407.0 144,830 100% 100%


78 | P a g e

Map A: All CTIP Regions (1-5)

Map B: CTIP1 Regions

Map A (left) shows the geographic patterns of these CTIP regions. Observation: Most CTIP 1 regions are located in the city’s southeastern and east central areas as well as Yerba Buena/Treasure Island. Most CTIP 5 regions are found in the western and central portions of the city. Map B (left) shows the geographic pattern for CTIP1 (i.e., areas of the city with the lowest average test scores)


79 | P a g e

4. Middle School Feeders

On June 14, 2011, the Board of Education approved staff’s recommendation to develop

elementary to middle school feeder patterns.

o For enrollment of students starting in the 2012-13 school year and for four years

thereafter, the elementary-to-middle school feeders operate as a tie-breaking factor in the

choice process.

o Beginning with the enrollment of students for the 2017-18 school year, fifth graders will

receive an initial assignment to their feeder middle school based on the elementary school

they attend, and they will have subsequent opportunities to participate in a choice process.

The 2017-18 school year was chosen for full implementation to allow time to bring the vision

behind staff’s proposal to fruition.

Below is a summary of the rationale and assumptions behind the staff proposal.

o Reducing the number of sending elementary schools and knowing who incoming students

will be provides each middle school with an opportunity to:

• build a professional community and accountability between elementary teachers and administrators and middle school teachers and administrators to benefit students; and

• create a community among smaller groups of students and families for better articulation among the grade levels; and

• place an emphasis on early identification and proactive intervention to actively address each student’s academic needs; and

• develop a middle school mission that will prepare incoming students academically for the future; and

• coherently implement cohesive policies and strategies to support incoming students; and

• use assessment data for incoming students with student outcomes as the focus when developing the Balanced Score Card and allocating limited resources during the budget development process.

o There is a link between robust enrollment and a school’s ability to provide equitable

access to an enriched learning environment.

• Under-enrolled middle schools have fewer teachers, fewer parents, smaller budgets, and therefore less opportunity to make sure all the students enrolled in the middle school have equitable access to electives, athletics, and enrichment programs available to children enrolled in large middle schools.

• The elementary-to-middle school feeders will facilitate improved communication with families, including targeted outreach and recruitment to help build robust


80 | P a g e

enrollment across all middle schools. The feeders provide an opportunity for parents and staff at elementary and middle schools to work closely together and share specific information that will help families understand how the school can support their child’s academic and enrichment needs.

o The feeder patterns provide an opportunity to build connections between families and

students to facilitate the transition between elementary and middle school. They also

offer families a degree of predictability regarding where their children will attend school

and minimize the degree of effort families must invest to enroll their children in school.

o Feeders support the strategic use of limited resources to provide a continuity of

curriculum and academic programs from elementary to middle school, and they permit

the efficient and cost-effective use of school facilities and transportation.

o Working in alignment with other District initiatives the feeders can help to decrease racial

isolation and the concentration of underserved students in the same school.

o Middle schools are going to change – we do not expect status quo for our middle

schools. Middle school principals, staff, and parent groups are waiting to build

relationships with elementary communities so they can plan and work together to align

services and resources to better meet the needs of all students.


81 | P a g e

Elementary to Middle School Feeders

Sorted in alphabetical order by middle school

Middle School Elementary

Aptos MS

Carver

Feinstein

King ES

Ortega

Sloat

Denman MS

Lakeshore

Longfellow

Miraloma

Sheridan

Sunnyside

Everett MS

Chavez

Fairmount

Marshall ES

McKinley

Milk

Sanchez

Francisco MS

CEC

Chin

Garfield

Parker

Tenderloin

Yick Wo

Giannini MS

Drew

Grattan

Jefferson

Key

Stevenson

Sunset

Hoover MS

Monroe

Moscone

Serra

Ulloa West Portal

Middle School Elementary

ISA MS

Bryant

Webster

King MS

Hillcrest

Malcolm X

Taylor

Lick MS

Alvarado

Flynn

Glen Park

Harte

MEC

Muir

Marina MS

Lau

Montessori

Redding

Sherman

Spring Valley

Presidio MS

Alamo

Argonne

Clarendon

Lafayette

Parks

Roosevelt MS

CIS @ DeAvila

Cobb

McCoppin

New Traditions

Peabody

Sutro

Vis Valley MS

CEC

Cleveland

El Dorado

Guadalupe

Longfellow

Vis Valley ES


82 | P a g e


83 | P a g e

5. General Education Transportation Policy

(108-24Sp1, Adopted by the Board on 12/14/2010)

The Superintendent, or his/her designee, shall have the discretion to construe all terms in this

General Education Transportation Policy and to approve specific transportation plans that he or

she deems necessary to meet the requirements of the Board’s strategic plan, student

assignment policy (P5101), and vision for after school services, and to comply with transfer and

transportation obligations under No Child Left Behind or similar state or federal laws.

As school bus transportation service is not a mandate under California law, the goals and

objectives set forth below will guide the strategic use of resources for limited general education

school bus transportation services.

• Support choice in school assignment as a tactic for creating diverse learning environments. (a) Maintain school bus transportation services that help create diverse enrollments. (b) Provide school bus transportation to racially isolated schools that have historically

been under enrolled.

• Support equitable access to the range of opportunities offered to students. (a) Provide English Learners with reasonable access to language programs. (b) Provide newcomers with reasonable access to newcomer programs. (c) Provide low-income students living in areas of the city with the lowest average test

scores (e.g., CTIP1 for the 2011-12 school year) with reasonable access to city-wide schools and programs.

(d) Provide students living in densely populated attendance areas with reasonable access to schools in less densely populated areas of the city.

• Provide limited school bus transportation to support reasonable access for attendance area residents to their attendance areas school.

• To support the middle school assignment process outlined in P5101, provide limited transportation for middle school students in cases where a middle school attendance area is not reasonably accessible to the middle school, taking into account factors including the availability of reasonable MUNI routes.

• Transportation is not contemplated for general education high school students, and/or will only be considered when the Superintendent determines it is necessary to provide equitable access and reverse the trend of racial isolation and the concentration of underserved students in the same school.

• Support SFUSD’s vision for after school services. (a) Provide limited school bus transportation to after school programs if feasible and

necessary to support SFUSD’s vision for after school services.

• Minimize the use of unrestricted general fund budget contributions for general education school bus transportation.

This transportation policy overrides all terms in Board policies and administrative regulations

that are inconsistent with any of its provisions, including but not limited to 5117.1


84 | P a g e


85 | P a g e

6. Glossary of Terms

Academic Performance index (API): The API rank of a school is established by deciles,

ranking schools from the lowest of 1 to the highest of 10. Two types of API ranks are reported,

a statewide rank and a similar schools rank. A school’s base API is used to determine its rank

and is used to compare to all other schools in the state of the same type (elementary, middle or

high school) to determine statewide ranking. The similar schools rank is determined by

comparing the school’s API to 100 other schools of the same type with a mix of similar

demographic characteristics. For the purposes of this report, the API ranking refers to the

statewide rank.

Attendance Areas: Boundaries drawn around individual schools.

Census Tract Integration Preference (“CTIP”): A preference in program or school

assignment based on a demographic value that is assigned to each combined census tract.

This preference is designed to facilitate attendance at the same schools by students who live in

demographically different areas of the city. Currently, the CTIP value is based on average K-12

California Standards Test (“CST”) scores of students who reside in the combined census tract.

An average K-12 CST score was computed for each combined census tract, and those

combined census tracts were divided into quintiles based on average CST scores so that

approximately 20% of all SFUSD students live within each quintile.

Combined Census Tracts: Geographic areas containing one or more adjacent census tracts

from the federal decennial Census.

CTIP1: CTIP1 tracts are the combined census tracts with the lowest average CST scores.

Designated assignments: Students who do not receive an assignment offer a school listed as

one of their choices receive an assignment offer to the school closest to their home that has

openings after choice assignments are made. These assignment offers are called designated

assignments.

Elementary City-Wide Schools: Elementary schools (K-5 and K-8 schools) that do not have an

attendance area and therefore do not offer any local preference to students. The purpose of the

city-wide school designation is to facilitate equitable access to the range of opportunities offered

by SFUSD.

Elementary City-Wide Programs: Programs that are (a) clearly defined and listed on the

SFUSD application form as a discrete choice, (b) are available at a limited number of

elementary attendance area schools, and (c) have a separate enrollment capacity with seats

reserved specifically for students enrolled in the program (for example, the Cantonese

Immersion program at West Portal), are designated city-wide programs, and they do not offer

any local preference to students. The purpose of the city-wide program designation is to

facilitate equitable access to the range of opportunities offered by SFUSD.


86 | P a g e

English Learner (“EL”): Students who are in the process of acquiring English as a second

language and have not yet reached Fully English Proficient (“FEP”) status.

Enrolled: Students are enrolled in a school or program if they have accepted an assignment to

and actually begun attending that school or program.

Feeder Patterns: Beginning with the assignment of students for the 2017-2018 school year,

SFUSD fifth graders will receive an initial assignment to middle school based on the feeder

pattern for the elementary school they attend, regardless of their residence.

Middle School Feeder Tie-breaker: A preference category used in student assignment until

the 2017-2018 school year, based on designated elementary-to-middle school feeder patterns.

On Time Applicants: All applicants who submitted an enrollment application during the first

placement period (November 15th, 2010 through February 18th, 2011) and received an

assignment offer in March 2011.

Program Pathway: A program that is listed as a discrete choice on the SFUSD enrollment form

and continues from pre-K to kindergarten, elementary to middle school, and/or middle school to

high school. Spanish Immersion is an example of a program pathway. General Education is

not considered a program pathway.

Racial/Ethnic Group: For the purposes of this report, the 20 racial/ethnic groups identified by

the California Department of Education have been condensed to these six categories – African

American, Chinese, Latino, White, Other Asians, and Other.

Racial Isolation: Although SFUSD’s enrollment is racially/ethnically diverse and does not have

a majority group, many of our schools have more than 60% of a single racial/ethnic group, more

than 70% of a single racial/ethnic group, and more than 80% of a single racial/ethnic group.

Some schools with more than 60% of a single racial/ethnic group also have an Academic

Performance Index (API) of 1, 2, or 3. The Board considers these schools racially isolated.

Tie-breakers: A set of preferences used to assign students when there are more requests than

available seats. Tie-breakers work in hierarchical order depending on the school, grade, or

program requested.

Transitional Grades: The first grade of enrollment at any particular school. For example, in

middle school, sixth grade is a transitional grade.

Underserved Students: Students performing Below Basic or Far Below Basic on the California

Standards Test or other equivalent assessments administered by SFUSD.

student assignment 2nd annual report

Documents