of california california countsamong californians are fairly sim-ilar to the rates of people from...

California has a long tradition of policies aimed at providing equi-table, quality education for all students. Yet, not all racial and ethnicgroups fare equally well in California’s education system. Less than20 percent of Hispanic, black, American Indian, and Pacific Islanderstudents earn bachelor’s degrees by ages 25 to 29, compared to 31 percent of white students, 40 percent of Filipino students, and

over 60 percent of other Asian students. Despite substantial investments in public higher edu-cation, college completion in California is not notably higher than in the rest of the countryfor any of these racial and ethnic groups.

This issue of California Counts explores educational resources and outcomes across racialand ethnic groups in the state. We examine family and school resources, student outcomes,and public policy initiatives affecting California’s students from early childhood through university. We describe several factors that potentially contribute to racial and ethnic gaps incollege completion. Hispanic, black, American Indian, and Pacific Islander children ages 0 to 5are less likely than white and Asian children to be in families with income above poverty, to be in two-parent families, and to have a mother who has finished high school. Hispanic chil-dren are also less likely to have a mother who speaks English and less likely to attend preschool.In the public K–12 system, Hispanic and black students, and to a lesser extent American Indian and Pacific Islander students, are more likely than white and Asian students to be inlow-performing schools. Hispanic and black students are more likely to be in overcrowdedschools and in schools with lower shares of fully credentialed teachers.

Public Policy Institute of California

California CountsPOPULATION TRENDS AND PROFILES

Educational Resources andOutcomes in California,by Race and EthnicityBy Deborah Reed

Vo lume 6 Number 3 • February 2005

�ummary

Hans P. Johnson, editor

California Counts Educational Resources and Outcomes in California


2

At every step—eligibility, admission, enrollment, and graduation—Hispanic and black students fare worse than white and Asian students inthe University of California system. Hispanic students are also underrepre-sented in the California State University system. First-time freshmen in thecommunity college system are representative of the racial and ethnic mix of the state’s high school seniors. However, Hispanic and black students areunderrepresented among students transferring to four-year programs fromCalifornia community colleges.

A number of new policy initiatives seek to improve educational equityin California. The First 5 commissions are working to promote schoolreadiness through parental support, health, and preschool programs. ThePublic School Accountability Act of 1999 has focused attention andresources on low-performing schools and student test scores have substan-tially improved. Following settlement of a lawsuit in 2004, new state fundswill be allocated to provide qualified teachers, textbooks, and facilities inlow-performing K–12 schools. Overcrowded schools have priority access tosome of the new state bond money for school facilities. Federal No ChildLeft Behind legislation puts an emphasis on qualified teachers and providesoptions for school choice. University of California programs such as Eligi-bility in the Local Context increase opportunities for students from highschools and communities that have had low admission rates.

Despite considerable public policy effort and a variety of approaches,disparities in educational outcomes by race and ethnicity have proven difficult to eliminate. Even with strong new policy initiatives, we cannotexpect these disparities to disappear quickly. However, Californians havedemonstrated their resolve to work toward quality and equity in education,a resolve that has not been thwarted by the difficulty in eliminating thedisparities.

Deborah Reed is a research fellow and director of the population program at PPIC. Weiyi Shiand Richard Van Swearingen provided valuable research assistance. The author acknowledgesthe helpful comments and thoughtful reviews of Gary Bjork, Raymond Colmenar, EricHanushek, Hans Johnson, Heather Rose, and David Roth. Views expressed are those of theauthor and do not necessarily reflect the views of PPIC.

Despite substantialinvestments in publichigher education,college completion in California is notnotably higher than inthe rest of the countryfor any of these racialand ethnic groups.



3

college completion among youngadults. We use attainment of abachelor’s degree as a key indicatorbecause it measures a high level of success that, for many people,marks the completion of formal

education. We then examine racialand ethnic differences in factorsthat influence college completion,beginning with early childhoodthrough high school and college.

Introduction

College education is a key factorin economic success. Workers

with a college education are morelikely to be employed and havehigher earnings capacity (Reed,2003; Betts, 2000). Childrenwhose parents have a college edu-cation are more likely to liveabove poverty and to go on to col-lege themselves. College-educatedworkers provide the state with askilled workforce. Recognizing theimportance of education, Califor-nia makes the public educationsystem from kindergarten throughuniversity its largest investment inchildren and youth.

Providing an equitable educa-tional playing field has been animportant policy goal in Califor-nia for many decades. Yet, not all racial and ethnic groups fareequally well in the educationalsystem. Understanding the factorsthat contribute to educational dif-ferences is particularly importantbecause education plays a role indetermining racial and ethnic dif-ferences in other areas of socialand economic well-being such aspoverty, health status, employ-ment, home ownership, and civicparticipation (Reyes et al., 2001).

This issue of California Countsexplores the connections betweenfamily resources, school quality,public policy, and racial and eth-nic differences in educational out-comes. We begin by examiningracial and ethnic differences in

Providing an equitable educational playing field hasbeen an important policy goal in California for manydecades. Yet, not all racial and ethnic groups fareequally well in the educational system.

The data sources used for this study are not consistent in the cate-gories used to describe racial and ethnic groups. Identification ofrace and ethnicity differs depending on the nature of questionsposed, including whether respondents were allowed to choose mul-tiple racial categories or a “some other race” category. In reportingthese data, we seek to provide as much consistency as possibleacross sources. Where possible, we use the major racial and ethnicgroups typically reported by the California Department of Educa-tion (CDE): Hispanic, white (non-Hispanic), Asian, black (non-Hispanic), Filipino, American Indian, and Pacific Islander. Wefollow CDE convention in using “Asian” to refer to combinedAsian groups excluding Filipinos.

The 2000 Census permitted respondents to choose multipleracial categories and to separately identify a Hispanic ethnicity. In our analysis, Hispanics of any race are included as Hispanics,and we exclude non-Hispanics identifying more than one race. Ingeneral, the results would not differ substantially if we included in each non-Hispanic race group all those identifying that groupalone or in combination with other races except in the case ofAmerican Indians, as is noted throughout the text.1

Information from the University of California, Office of thePresident, and the California Postsecondary Education Commis-sion is based on the categories reported by those entities.

Identification of Race and Ethnicity



4

Disparities in College Graduation

Analysis of bachelor’s degreecompletion demonstrates the

tremendous variation in educa-tional attainment across majorracial and ethnic groups in Cali-fornia (Figure 1). Among U.S.-born Californians ages 25 andolder, college completion is lowfor American Indians (14 per-cent), Hispanics (13 percent),blacks (17 percent), and PacificIslanders (22 percent).2 Collegecompletion is substantially higherfor whites (33 percent) and Fil-ipinos (36 percent) and even high-er for Asians (51 percent).

Foreign-born Hispanics havethe lowest levels of college gradua-tion, with only 5 percent of thoseages 25 and older holding a bach-elor’s degree. Immigrants fromSoutheast Asian refugee-sendingcountries also have low collegecompletion at 17 percent.3 Becausethe purpose of this study is toinvestigate racial and ethnic differ-ences in the California educationsystem, it is appropriate not tocombine U.S.-born and foreign-born populations in measures ofcollege completion. Among immi-grants ages 25 to 29, for example,almost three in four (73 percent)came to the United States at ages16 or older. For Hispanic immi-

grants who arrived at ages 16 orolder, even those under age 19have a low share attending schoolin California—only 37 percent.Thus, the educational attainmentof these immigrants is primarilydetermined by education andother experiences in their countryof origin as well as the determi-nants of migration to the UnitedStates.

Interstate migration also playsa role in determining the educa-tional attainment of adults in Cal-ifornia because the state tends toattract college-educated migrantsfrom other states (Johnson, 2000).For this reason, to measure collegecompletion for Californians whorecently moved through the edu-cation system, we use the 2000Census sample for people ages 25to 29 who were born in California(regardless of state of current resi-dence).4 Some young adults whowere born in California left forother states at young ages and werenot primarily educated here—about 20 percent of 18-year-oldsborn in California were living inother states in 2000.5 However, in the 2000 Census, college com-pletion rates for those born inCalifornia are not substantiallydifferent from the college comple-tion rates of those born in Califor-nia and living in California in1995 (Figure 2).

Among young adults born inCalifornia, American Indians havecollege completion rates of 11 per-cent. If we were to include all peo-

0

10

20

Perc

enta

ge

30

40

U.S.-born Foreign-born

50

60

70

Figure 1. Bachelor‘s Degree Completion in California byAge, Race, Ethnicity, and Nativity, 2000

Amer

ican In

dianAsia

n

Filip

ino

White

Pacifi

c Isla

nder

Other

Asia

n

Filip

ino

South

east

Asian

Hispan

icBlac

k

Hispan

ic

Source: Author’s calculations from the 2000 Census.

Notes: Among the U.S.-born, Asian excludes Filipinos. Among the foreign-born, “Other Asian” excludes Southeast Asians and Filipinos. See the text box for details on defining racial and ethniccategories.

Ages 25 and olderAges 25–29Ages 30–34Ages 35–39



5

0

10

20

Perc

enta

ge

30

50

40

60

70

80

FilipinoWhitePacificIslander

BlackHispanicAmericanIndian

Asian

Figure 2. Bachelor’s Degree Completion in California andthe Rest of the United States, by Race, Ethnicity, andNativity, Ages 25 to 29, 2000

Source: Author’s calculations from the 2000 Census.

Born in California and lived in California in 1995Born in CaliforniaBorn in any other stateLived in CaliforniaLived in any other state

ple identifying as non-HispanicAmerican Indian alone or in com-bination with other races, we findthat 13 percent have a bachelor’sdegree—a similar share as Hispan-ics and slightly below that ofblacks (15 percent) and PacificIslanders (18 percent). Whiteshave substantially higher comple-tion at 31 percent, as do Filipinosat 40 percent; Asians have evenhigher completion at 62 percent.

Among the states, California isranked 14th highest in state bud-get allocation per capita for highereducation (Palmer, 2004).6 Never-theless, college completion ratesamong Californians are fairly sim-ilar to the rates of people fromother states, with the exceptionthat Filipinos and Asians fromother states have higher comple-tion rates than their Californiancounterparts (Figure 2).7 BecauseCalifornia is able to attract college-educated people from other states,young white adults living in Cali-fornia have substantially highercollege completion than theircounterparts in the rest of theUnited States.

Educational attainment for the major racial and ethnic groupsmasks the variation within groups.8

Among Hispanics, for whom theshare with a bachelor’s degree was13 percent, the share was 12 per-cent for Mexican Americans, who make up 80 percent of theCalifornia-born Hispanic popula-tion ages 25 to 29 (subgroup sta-tistics are not shown in Figure 2).

Among Puerto Ricans, the sharewith a bachelor’s degree was 16percent and among Cubans, theshare was 34 percent.9 AmongCalifornia-born Asian subgroups,the share completing college was40 percent for Filipinos, 61 per-cent for Japanese, 64 percent forIndians, 68 percent for Koreans,and 70 percent for Chinese.10

During the 1970s, after theMaster Plan, civil rights legislation,and the initiation of affirmativeaction policies, college completionimproved more among blacks andHispanics than among whites inCalifornia (Table 1). It is difficultto gauge progress over the 1980sbecause of changes in the decennialCensus survey question between1980 and 1990. During the 1990s,bachelor’s degree completion im-

During the 1970s,after the Master Plan,civil rights legislation,and the initiation ofaffirmative action policies, college completion improvedmore among blacksand Hispanics thanamong whites in California.



6

changes in affirmative action pro-grams. In California, race-basedadmission policies at public uni-versities in California ended withthe passage of Proposition 209 in1996.11

Early ChildhoodEducation

The previous section documentedthe substantially lower college

completion rates of Hispanics,blacks, American Indians, andPacific Islanders relative to thoseof whites, Filipinos, and Asians.We now turn to the factors thatinfluence racial and ethnic gaps in education, beginning with earlychildhood education. Disparitiesin school preparation begin beforechildren enter elementary school.Research at the state and nationallevels has found that black andHispanic children are more likelythan white or Asian children to

Table 1. Four-Year College Completion Rates, by Race, Ethnicity, and Place of Birth, Ages 25 to 29, 1970, 1980, 1990, and 2000 (percent)

Born in California American Indian Hispanic Black Pacific Islander White Filipino Asian All

Born in any other state American Indian Hispanic Black Pacific Islander White Filipino Asian All Sources: Author’s calculations from the 1970, 1980, 1990, and 2000 Censuses.Notes: For some groups, the measured change between decades is affected by rounding. The 1970 sample was too small to compute California statistics for American Indians and Filipinos. Statistics for American Indians in 2000 are not comparable to those in 1990 because of respondents’ ability to identify more than one race in the 2000 Census (see the text box). Statistics for Pacific Islanders before 2000 are not provided because of changes in identification across Census surveys.

Bachelor’s Degree

1970 Change1980

45

21

4119

556

18142817

58

2

112

255

62

136

71012

23285321

61211

25304222

1113151831406225

1015151832496728

33

71210

4

34

71925

6

16 Years of Education

1990 Change2000

81013

24185222

81011

25174222

proved substantially for whites,Filipinos, and Asians but muchless so for Hispanics and blacks.For American Indians, the mea-sured change in college comple-tion over the 1990s depends onhow multiracial American Indiansare classified. The 1990 Censusrecorded only one race for eachperson, and 7 percent of those whoidentified as American Indian hadcompleted college. The 2000 Cen-sus allowed respondents to selectmore than one race and of thosewho identified as American Indianalone, 11 percent completed col-lege, whereas those who identifiedas American Indian alone or in com-bination with other races completedcollege at a rate of 13 percent.

Thus, despite improvements inthe share with a bachelor’s degreeduring the 1990s, the gap withwhites grew larger for Hispanics,blacks, and American Indians.The same was true nationally.Growth in the education gapsduring the 1990s may reflect

During the 1990s,bachelor’s degree completion improvedsubstantially forwhites, Filipinos, andAsians but much less so for Hispanics andblacks.



7

enter kindergarten with fewerschool-related skills (Bridges et al.,2004; U.S. Department of Educa-tion, 2004, p. 118).12

Several family factors arebelieved to be related to youngchildren’s educational develop-ment, preparation for school, andearly literacy. These include hav-ing a mother who has completedhigh school, having a mother whospeaks English, living with bothparents, and having family incomeabove the poverty level.13 Familyresources available to young chil-dren differ substantially acrossracial and ethnic groups (Table 2).Parental education, particularlymaternal education, is stronglylinked to a child’s cognitive devel-opment and school success (Have-man and Wolfe, 1995). Less thanhalf of young Hispanic childrenhave a mother with a high schooldiploma and less than 10 percent

have a parent with a bachelor’sdegree. Black, American Indian,and Pacific Islander children alsohave low parental education.Among children in most racialand ethnic groups in California,over 90 percent have a motherwho speaks English, but the shareis substantially lower among Asianand Hispanic children. The shareliving with two parents is particu-larly low among black childrenand also relatively low amongAmerican Indian, Hispanic, andPacific Islander children.14

Black, Hispanic, and Ameri-can Indian children are less likelythan Asian and white children tobe growing up in a family withincome above the poverty level.15

Less than half of Hispanic, black,and American Indian young chil-dren are growing up in familieswith incomes above twice the fed-eral poverty level.

Preschool attendance is par-ticularly low among Hispanic,Pacific Islander, and Filipino chil-dren in California. Less than halfof these four-year-olds attend pre-school, whereas for other groupsover 60 percent attend preschool(Table 3).16 Rates of preschoolattendance are fairly similar inCalifornia to those in the rest of

Table 2. Family Resources of California Children, by Race and Ethnicity, Ages 0 to 5, 2000 (percent)

American IndianHispanicBlackPacific IslanderWhiteFilipinoAsianAll

Source: Author’s calculations from the 2000 Census.Notes: Parents include coresident biological, adoptive, and stepparents. “Mother Speaks English” is the percentage who speak only English at home plus those whose mother speaks English “well” or “very well.” The low-income threshold is twice the federal poverty threshold for each family.

Share ofChildren

Mother Has High School

Diploma

<148

6<13227

100

149

161646535627

9665999298987482

5066386580798871

7372677991948580

Mother Speaks English

Lives withTwo Parents

7449807292938271

Parent Has Bachelor’s

Degree

Family Income Above Poverty

Family Income Above

Low Income

4638435077786955

. . . black and Hispanicchildren are more likely than white orAsian children to enterkindergarten withfewer school-relatedskills.



8

a wide variety of school readinessprograms including provision ofearly childhood education pro-grams, parenting support services,health services, and training ofearly childhood educators. Severalof the commissions are developingplans for universal preschool. Theannual cost of a universal pre-school for California has been esti-mated at between $3.2 billion and$5 billion.18

Head Start is a federal schoolreadiness program targeted towardyoung children in low-incomefamilies. In 2003, Head Start allo-cated over $811 million to Cali-fornia and enrolled almost 99,000children in the state.19 Numerousevaluations show that Head Startimproves children’s developmentand school readiness, but the evi-dence for a lasting effect of HeadStart on academic achievement ismixed.20

Elementary School

In this section, we describe schoolquality, school resources, and

student outcomes by student raceand ethnicity for California’s pub-lic elementary schools. Through-out the section, we discuss policiesthat affect resources for K–12schools. In the next section, wedescribe similar measures for highschools as well as recent policymeasures designed to improveequity in future K–12 resourcesand outcomes.

Table 3. Preschool and Kindergarten Attendance in California and the Rest of the United States, by Race and Ethnicity, 2000 (percent)

American IndianHispanicBlackPacific IslanderWhiteFilipinoAsianAll Source: Author’s calculations from the 2000 Census.

PreschoolAge 4

California CaliforniaRest of U.S.

3724462345214234

6143634565466354

5947674962506260

6663645957636762

4651535644565447

PreschoolAge 3

Rest of U.S. Rest of U.S.California

3226462237273937

KindergartenAge 5

the United States. In contrast,kindergarten attendance by five-year-olds is relatively similar acrossracial and ethnic groups at about60 percent and is higher in Califor-nia than in the rest of the nation.California offers noncompulsorypublic kindergarten for childrenwho are age 5 by December 2 ofthe school year—allowing youngerchildren than most states.17

Improving school readiness isa major goal of local, state, andfederal policies. In 1998, Califor-nia voters passed an initiative totax cigarettes to provide fundingfor health and school readinessprograms for children ages 0 to 5.In 2002–03, the tax brought inapproximately $580 million dollars—20 percent of which goesto the statewide commission and80 percent to county commissions.The First 5 commissions support

Improving school readiness is a majorgoal of local, state, andfederal policies.



9

In California, school perfor-mance is measured by a standard-ized test from which the AcademicPerformance Index (API) is calcu-lated and then converted into astate rank from 1 to 10. Schoolswith an API rank of 3 or belowfall in the lowest 30 percent andare considered “low-performing.”21

Many low-performing schoolshave high shares of students fromlow-resource families (Rose et al.,2003). Thus, low performancemay be related to less academicpreparation among these students.Whether this is viewed as a failureof the schools to meet studentneeds, a failure of parents and theearly childhood education systemto adequately prepare students, or both, it is clear that studentsbear the costs of attending a low-performing school. Researchshows that students in classrooms

with more academically skilledpeers tend to learn more (Betts,Zau, and Rice, 2003).22 Thus,attending a low-performing schoollikely has adverse effects on stu-dent learning.

In California, only 10 per-cent of white elementary studentsattend low-performing schools(Table 4). In contrast, 52 percentof Hispanic students and 43 per-cent of black students attend low-performing schools. The share inlow-performing schools is also rel-atively high for American Indians(29 percent) and Pacific Islanders(26 percent).23 There is a strongcorrelation between low-performingschools and those with large sharesof low-income students, and wefind that Hispanic and black stu-dents, and to a lesser extent Amer-ican Indian and Pacific Islanderstudents, are more likely than other

Table 4. School Resources of California Elementary Students, by Race and Ethnicity (percent)


Sources: Author’s calculations from data from the CDE. Information on school performance is based on the API from 2003. Information on low-income students is based on the CalWORKs data collection from 2002. Information on teacher credentials and experience is from the Professional Assignment Information Form (PAIF) for 2003–04. Information on overcrowded schools is from the CDE list of critically overcrowded schools in 2004.

Share ofStudents

Low-PerformingSchool

149

81

3028

100

5572645434484256

9691909497949693

9089888991888989

2629282931313029

Teacherswith a FullCredential

Teacherswith 3+ Yearsof Experience

2952432610161534

Low-IncomeStudents

Teacherswith a Master’s

DegreeOvercrowded

School

7272412

4121418

In California, only 10 percent of whiteelementary studentsattend low-performingschools. In contrast,52 percent of Hispanicstudents and 43 per-cent of black studentsattend low-performingschools.



10

Hispanic and black students are muchmore likely than whitestudents to be in overcrowded schools.

school resources.26 Although evi-dence suggests that high-qualityteachers make a difference,researchers have not found a con-sistent, strong, causal link betweenstandard measures such as creden-tial status and student achieve-ment.27 Indeed, a principal mightprefer to hire high-quality teacherswho meet specific needs, regard-less of credential status. Neverthe-less, in light of the recent strongincentives to improve teacher cre-dential rates under the federal NoChild Left Behind legislation, lowcredential levels among teachersare likely to indicate a school withdifficulty attracting and retainingquality teachers.28 Across all racialand ethnic groups in California,elementary teacher credential ratesare 90 percent or higher, but thelowest rates are found for blacks(90 percent, see Table 4) and His-panics (91 percent) and the high-est rates are found for whites (97percent).29

Teacher experience shows rela-tively little variation across racialand ethnic groups. The share ofteachers with a master’s degreealso shows little variation,although it is lower for AmericanIndian students. On average, thereare 23 elementary students perteacher, although Hispanic stu-dents are in schools where theaverage is 24 students per teacher(not shown in the table).

School resources may varymore widely in the area of schoolfacilities, because facilities funding

students to be in schools with ahigh share of low-income students.24

During the 1970s, Californiamade major reforms in schoolfinancing in an attempt to addressequity issues. In 1971, the Cali-fornia Supreme Court in Serranovs. Priest determined that the sys-tem of local school finance deniedconstitutional rights to equal pro-tection. A major element of theplaintiff ’s argument was thatinequalities in revenues per stu-dent were systematically related torace and wealth because the localproperty tax base was much lowerin black, Hispanic, and low-incomeneighborhoods. The Serrano casewas followed by a series of effortsto redesign the school finance sys-tem. Then, in 1978, Proposition13 essentially turned the propertytax into a statewide tax. The statenow controls 90 percent of schooldistrict revenue (Sonstelie, Brun-ner, and Ardon, 2000).

As intended by law, “revenuelimit” funds are relatively equallydistributed across school districts.State and federal “categorical pro-grams” (such as state Economic

Impact Aid and federal Title I)bring further revenue to high-poverty schools. Considering allrevenue sources in 2001–02, Roseet al. (2003) find that unified dis-tricts (those providing elementarythrough high school) received anaverage augmentation of $1,018per low-income student or 17 per-cent more than the base revenueof $6,019 per student. In elemen-tary districts, the average augmen-tation was lower at $451, and highschool districts actually receive an average of $301 less per low-income student. However, theseaverages mask substantial variationin revenues, with some high-poverty school districts receivingless than $6,000 per student.25

District finances provide anincomplete picture of schoolresources. Districts face differentteacher salary costs, related tolocal costs of living and other fac-tors (Rueben and Herr, 2001).School finance reform did notinclude building-related funds andschools with limited capital fundsmay face greater operating costs,for example, because of mainte-nance needs for older or morecrowded facilities. Finally, school-level resources may not align withdistrict revenues as a result ofinterdistrict spending allocationsand other choices made by dis-tricts and schools.

Researchers disagree as towhether there is a consistent, systematic link between studentachievement and measures of



11

was not included in California’sschool finance reform.30 The Criti-cally Overcrowded School Facili-ties program set aside $4.1 billionof the $21.4 billion in new statebond funds for overcrowdedschools. The Department of Edu-cation measures overcrowdingbased on the number of studentsper acre.31 Hispanic and black students are much more likelythan white students to be in over-crowded schools (Table 4).32 How-ever, overcrowding is a limitedand problematic measure ofschool facilities. A rural schoolwith a leaking roof may not beovercrowded, whereas an urbanschool with an agreement to usecity parks as fields may measure asovercrowded, despite adequateclassroom space. Currently, nocomprehensive data are availableto assess school facilities.33

In light of the differences infamily and school resources, it isperhaps not surprising that His-panic and black elementary schoolstudents, and to a lesser extentAmerican Indian and PacificIslander students, are less likelythan white, Filipino, and Asianstudents to be proficient in Eng-lish and math. Only about one infour Hispanic and black studentsare proficient in English (Figure 3). A higher share are proficient inmath in the second grade, but bythe fifth grade proficiency falls toabout one in four.34

In recent years, there havebeen additional policy changes

aimed at reducing the gaps inK–12 public school resources andperformance in California. Beforeturning to a discussion of thesereforms, we first describe highschool conditions. For brevity, wedo not describe school resourcesfor middle schools, but the pat-terns are very similar to those forhigh schools (Table 5).

High School

For high schools, we use the same measures of school quality and

resources as we used for elemen-tary schools and find very similarpatterns (comparing Tables 4 and5). More than half of Hispanic stu-dents and over 40 percent of blackstudents are in low-performingschools, compared to only 11 per-

0

20

40

Perc

enta

ge

60

80

100



Asian

Figure 3. Proficiency in English and Math in the Secondand Fifth Grades in California, by Race and Ethnicity, 2004

Source: California Standards Test results from the CDE.

Note: The figure shows the percentage scoring at or above proficient.

Grade 2, EnglishGrade 5, EnglishGrade 2, MathGrade 5, Math

On average, Hispanicand black studentsattend schools whereabout 85 percent ofteachers have creden-tials, compared to anaverage of 92 percentcredentialed teachersin schools attended bywhite students.

cent of white students. On aver-age, Hispanic and black studentsattend schools where about 85percent of teachers have creden-tials, compared to an average of92 percent credentialed teachers inschools attended by white stu-dents. Teacher experience and theshare of teachers with a master’s



12

(including one after completion of grade 12). Results for tenthgraders in 2004 show relativelylow passage rates for Hispanicsand blacks at roughly 60 percentand somewhat higher passage ratesfor American Indians and PacificIslanders at about 70 percent (Fig-ure 4). Whites, Filipinos, andAsians passed at rates of over 85percent.

High school completion ratesfor young adults born in Califor-nia show the same pattern ofracial and ethnic disparities (Table6).35 The share of Hispanics, blacks,and American Indians with a highschool diploma was less than 80percent. The share of whites witha high school diploma was 89 per-cent and the share of Filipinos and Asians was even higher. Highschool completion rates for Cali-fornians are similar to those foryoung adults in the rest of the

United States for each of the racialand ethnic groups, with the excep-tion of American Indians, whohave higher completion rates inCalifornia. In contrast, during the1970s, Californians had substan-tially higher completion rates for every group. Perhaps in partbecause California had higherrates of completion in the 1970s,the state did not experience thesubstantial growth in high schoolcompletion rates that occurred inthe rest of the nation, especiallyamong Hispanics, blacks, andAmerican Indians.

The disparities in resourcesand outcomes in K–12 educationhave been the focus of severalrecent policies that seek to improvequality and equity. The PublicSchool Accountability Act of 1999included standardized perfor-mance assessments, performancetargets, rewards and sanctions

degree are also lower for Hispanicand black students, but the differ-ences are small. The number ofstudents per teacher also is fairlysimilar across racial and ethnicgroups (an average of 29 to 30students per teacher, not shown inthe table). More than 20 percentof Hispanic and black students arein overcrowded schools, comparedto only 5 percent of whites.

Beginning in the 2005–06school year, students must pass the California High School ExitExamination, in addition to ful-filling district requirements, tograduate from high school. Theexam is based on English-languagearts content standards throughgrade 10, and mathematicsthrough Algebra 1. The firstopportunity to take the test is inthe second half of tenth grade.After that, students have anotherfive opportunities to pass the test

Table 5. School Resources of California High School Students, by Race and Ethnicity (percent)


Source: Author’s calculations from the CDE.

Note: See Table 4 for information on measurement.

Share ofStudents

Low-PerformingSchool

141

81

3639

100

3146413322323035

9285848992899188

8987868789888888

3637363839383938

Teacherswith a FullCredential

Teacherswith 3+ Yearsof Experience

2352433111271631

Low-IncomeStudents

Teacherswith a Master’s

DegreeOvercrowded

School

6222412

5151515



13

based on performance, and assis-tance for low-performing schools(funds and expert assistance).Since implementation of the act,API scores have increased substan-tially, particularly in low-perform-ing elementary schools (Rose etal., 2003). However, there remainsconcern about the capacity of low-performing schools to continue toimprove student learning (O’Dayet al., 2004).

In May 2000, the AmericanCivil Liberties Union and othersfiled a lawsuit, known as theWilliams case, against the state ofCalifornia. The plaintiffs arguedthat the state was failing to pro-vide public school students, par-ticularly low-income students andstudents from racial and ethnicbackgrounds other than white,

with educational necessitiesincluding textbooks, trainedteachers, and adequate facilities.In settling the case in August2004, the state agreed to providequalified teachers, to end short-ened school years in overcrowdedschools, and to fund instructionalmaterials and school facilities.

The focus on qualified teach-ers in the Williams case is alsofound in the federal No ChildLeft Behind legislation, whichrequires qualified teachers in everyclassroom. This task may provedifficult because it requires attract-ing and retaining qualified teach-ers in low-resource schools.Improving the facilities and otherconditions in these schools may behelpful. Supportive teacher devel-opment programs may also be

0

20

40

Perc

enta

ge

60

80

100



Asian

Figure 4. Passage Rates for the California High SchoolExit Examination in the Tenth Grade, by Race andEthnicity, 2004

Source: California Department of Education (2004).

English-language artsMath

helpful. In addition, programssuch as the Assumption Programof Loans for Education (APLE)provide incentives for teachers towork in low-performing schools.Participants in APLE who obtainan initial teaching credential areeligible for up to a total of$11,000 in educational loanassumption payments with fourconsecutive years of teaching ser-vice in a qualifying Californiapublic school. The Governor’sTeaching Fellowship, which pro-vided similar incentives, has notbeen funded since 2002.

In contrast to the equity-based reforms of the 1970s, new approaches are considering“adequacy-based” reforms (Rose,2001). The California QualityEducation Commission, created by

The disparities inresources and outcomesin K–12 education havebeen the focus of several recent policiesthat seek to improvequality and equity.



14

In light of the challenges faced byschools that serve high concentrations of disadvantaged students, one approach is to redistribute thestudents.

tion, districts are required to allow students in failing schools to transfer to other schools in thedistrict. That legislation also sup-ports magnet schools, charterschools, and other school choiceprograms. Research on the effec-tiveness of school choice programsprovides mixed evidence. Aca-demic improvements have beenshown for some programs, but thequestion remains whether theseprograms could be effectivelyimplemented on a broad level(Fuller et al., 1999; Gill et al.,2001).

College

The goals of California’s 1960Master Plan for Higher Educa-

tion were access, quality, andaffordability. Under the plan, thetop 12.5 percent of high schoolgraduates would be accepted intothe University of California (UC),the state’s premier research system.The top third of high school grad-uates would be accepted into Cali-fornia State University (CSU),which had as its primary missioneducation through the master’sdegree level, including teaching

Table 6. High School Completion Rates, by Race, Ethnicity, and Place of Birth, Ages 20 to 24, 1970, 1980, 1990, and 2000 (percent)


Born in any other state American Indian Hispanic Black Pacific Islander White Filipino Asian All Sources: Author’s calculations from the 1970, 1980, 1990, and 2000 Censuses.Notes: In 1990 and 2000, the high school diploma category includes those earning a General Equivalency Diploma. See the notes to Table 1 for measurement issues.

High School Diploma

1970 Change1980

7181

87

9886

555761

82828979

21

–1

–1–1

111112

3644

777277

87899483

737275

88949485

7974778589929683

7473758689949685

20

3311

10

1020

12 Years of Education

1990 Change2000

737481

87909785

666873

86889283

Assembly Bill 2217 in 2001–02, is charged with examining what itwill take “so that the vast majorityof pupils can meet academic per-formance standards established bythe state” for K–12 education. Thegovernor has not yet appointedcommissioners and the CaliforniaPerformance Review recommendedagainst having a commission, butcommission funding is availablefrom private foundations. If thecommissioners are appointed, they will have the opportunity toconsider what resources are neces-sary to provide adequacy in educa-tion for low-performing schools in terms of teacher-pupil ratios,counseling staff, special programs,and other features.

In light of the challenges facedby schools that serve high concen-trations of disadvantaged students,one approach is to redistribute thestudents. Within-district transfersare allowed in California, subjectto space constraints.36 Under fed-eral No Child Left Behind legisla-



15

degrees. Any student who would“benefit from instruction” wasaccepted into the California Com-munity Colleges (CCC) system.From the community colleges, allwho qualified were to be offeredtransfers into public university.

Eligibility into the UC andCSU schools is based on merit,particularly high school grades,required coursework, and testscores. Eligibility rates are low forHispanic, black, and AmericanIndian high school students (Table7). Race-based admissions policiesended after the passage of Proposi-tion 209 in November 1996. How-ever, the UC system has severaladmissions policies designed topromote educational opportunities.With the entering class of Fall2001, UC began implementationof “Eligibility in the Local Con-text” (ELC). ELC guaranteesadmission to junior-level highschool students in the top 4 per-cent of their high school class toincrease opportunities for studentsfrom high schools and communi-ties that generally have had lowadmission rates.37 In addition,starting with the entering class of 2002, the most selective UCcampuses used a “comprehensivereview” admissions strategy wherebyacademic and nonacademic “lifechallenges” were considered forstudents who met the eligibilityrequirements. UC systemwide,Berkeley, and UCLA campusadmissions data for the periodsuggest that these policies have

not substantially increased admis-sions for traditionally underrepre-sented students (Figure 5).38

In addition to affecting eligi-bility, differences in academicpreparation can affect the decisionto prepare for, apply for, andenroll in a university. Students inlow-performing and low-resourceschools are less likely to follow acollege preparatory track in highschool. For example, Hispanic andblack high school students aremuch less likely to participate inadvanced placement courses (Col-lege Board, 2001; EducationTrust, 2004). The fact that manyof these students’ own parentshave not graduated from college(as shown in Table 2) may lead tolower expectations of their ownpotential to graduate from college(U.S. Department of Education,1997, Table 18).

In light of the lower incomelevels of Hispanic, black, andAmerican Indian students (asshown in Table 2), the cost of uni-

Table 7. Eligibility Rates for California’s High School Seniors, by Race and Ethnicity, 2003 (percent)

American IndianHispanicBlackWhiteAsianAll Source: California Postsecondary Education Commission (2004b).Note: Asian includes Filipinos and Pacific Islanders.

CSU

19.716.018.634.347.528.8

6.66.56.2

16.231.414.4

UC

versity is likely another factor inracial and ethnic differences incollege completion. Although stu-dent fees in California publichigher education are low relativeto national standards (CaliforniaPostsecondary Education Com-mission, 2004a), the intent of theMaster Plan was to provide freepublic higher education. Duringthe state’s current fiscal crisis, feeshave been rising. Under a recentagreement with the governor,undergraduate fees increased by

Eligibility rates are lowfor Hispanic, black, andAmerican Indian highschool students.



16

If the transfer functionis not effective, theresult is a serious chal-lenge to the MasterPlan’s promise of accessand affordability.

Differences in academic prepa-ration, eligibility, expectations,and financial resources lead to UCclasses that are remarkably differ-ent from the population of highschool graduates, with the share ofHispanics and blacks lower bymore than half (Table 8).39 Thedemographic differences betweenCSU enrollees and public highschool graduates are not as great,but Hispanic students are under-represented. The demographics ofthe graduating classes in UC andCSU and enrolling freshmen lookfairly similar. However, the proba-bility of graduating varies greatlyby race: Graduation rates in theUC system are lower for Hispan-ics (70 percent), American Indians

(66 percent), and especially blacks(62 percent) than for whites (80percent), Filipinos (76 percent)and Asians (80 percent).

First-time freshmen studentsenrolled in the CCC system haveroughly the same racial and ethnicdistribution as California’s highschool graduates. The Master Planenvisioned university access forstudents who were not eligible for admission into UC or CSUimmediately after high schoolthrough taking lower divisioncourses at CCC and then transfer-ring to a university. The impor-tance of the transfer function andconcern over the low number oftransfers have led to a number ofpolicies to increase transfers (see

14 percent in 2004–05 and willincrease by 8 percent in each ofthe next two years. Growing feesreduce access for students whoface financial hardships—althoughfees are relatively low compared toother costs of attending universitysuch as room and board and theloss of earnings for students whowould otherwise be working full-time. Financial aid can offset someof these costs. For example, forfinancially needy students, federalPell grants provided up to $4,000per student in 2002–03 to covercosts of attendance. Substantialtuition fees could provide addi-tional funding for grants to low-income students to cover fees,books, and cost of living.

0

5

10Perc

enta

ge

15

20

25

30

200019991998199719961995 2001 2002 2003

Figure 5. Underrepresented Minorities as a Share of UCAdmissions Among California Resident Freshmen,1995–2003

Source: University of California (2004).

Notes: The University of California, Office of the President (UCOP) uses the term “underrepresented minorities” to include American Indian, black, Chicano, and Latino students. Over this period, the number of admitted students of “unknown” racial and ethnic background increased substantially.

SystemwideBerkeleyLos Angeles



17

Table 8. Racial and Ethnic Distribution of Higher Education in California (percent)

American IndianHispanicBlackWhiteFilipinoAsian

Sources: Author’s calculations from CPEC data on public high school graduates for 2002, on college enrollments in Fall 2002, on bachelor’s degrees for 2002, and on CCC transfers in Fall 2002. UC graduation rates are from the University of California (2003).Notes: CCC transfers include transfers to private universities and colleges. Pacific Islanders are included with Asians. Columns do not add to 100 percent because some students do not identify their race or ethnicity.

Public High School Graduates UC Enrollees

133

743

312

112

340

430

667062807680

123

739

614

120

543

414

UC Graduation

RateCSU

Enrollees

114

337

533

UC Graduates

CSU Graduates

CCC Enrollees

133

739

410

CCC Transfers

120

542

314

California Postsecondary Educa-tion Commission, 2002, for asummary). Nevertheless, as of2000–01, transfers remained fairlyconsistent at about 48,000 peryear to CSU, about 11,000 peryear to UC, and about 8,000 ayear into private institutions. Inthe context of growing enrollmentat CCC, this means that the shareof CCC entering students whosuccessfully transfer to universityhas declined. The share transfer-ring to four-year institutions isparticularly low for Hispanics(Table 8, final column). If thetransfer function is not effective,the result is a serious challenge tothe Master Plan’s promise of accessand affordability. On the otherhand, one explanation for lowtransfer rates is that studentsattend CCCs for other programssuch as workforce training, voca-tional or occupational education,

and remedial education. Thus, theCCC system provides importantopportunities for educationalattainment beyond simply thetransfer function.

Perhaps because of the exten-sive public higher education sys-tem and especially the CCCsystem, the share of Californianswho have attended college is higherthan the share in the rest of thenation among blacks and Ameri-can Indians, and to a lesser extent,Hispanics (Table 9). This is incontrast to the finding aboutbachelor’s degrees (Table 1), whichshowed similar completion in Cal-ifornia and the rest of the nation.Nevertheless, the share who haveattended college is relatively low,49 to 55 percent, among Ameri-can Indian, Hispanic, black, andPacific Islander Californians ages25 to 29. Among whites, 70 per-cent have attended college and

among Filipinos and Asians, theshares are over 80 percent. Duringthe 1970s, the growth in collegeattendance in California wasgreater for Hispanics and blacksthan for whites. During the1980s, there was very little growthfor any group. During the 1990s,growth was greater for whites thanfor blacks and Hispanics.

Conclusion

Among young adults educatedin California, the share who

have achieved a bachelor’s degreeis remarkably low for AmericanIndians, Hispanics, blacks, andPacific Islanders. State and federalpolicies of the 1960s and 1970s—including the California MasterPlan for Higher Education, civilrights, and affirmative action—were followed by a closing of col-



. . . strategies that go beyond early childhood development and school reform will likely prove important for reducing racial and ethnic education gaps.

Table 9. Share with Some College Coursework, by Race, Ethnicity, and Place of Birth, Ages 25 to 29, 1970, 1980, 1990, and 2000 (percent)


Born in any other state American Indian Hispanic Black Pacific Islander White Filipino Asian All Sources: Author’s calculations from the 1970, 1980, 1990, and 2000 Censuses.Notes: For measurement issues, see the notes to Table 1. The table shows the share of people ages 25 to 29 who have ever attended college. Unlike high school and college completion, college attendance can be compared across all four decades of the Census.

1970 1980

2434

53

7850

181617

37365334

454453

61708458

364241

56707253

5049545570838962

4648485665858662

1990 2000

454353

61628859

363437

52467249

lege education gaps. However,during the 1990s, the gaps grew ascollege attendance and completionincreased more for whites than forblacks and Hispanics.

Several factors contribute toracial and ethnic education gapsamong Californians. Black, His-panic, American Indian, and PacificIslander children are less likely thanwhite and Asian children to be infamilies with resources that are asso-ciated with educational develop-ment, preparation for school, andearly literacy—having a motherwho has completed high school,living with both parents, and havingfamily income above the povertylevel. These children, particularlyblacks and Hispanics, are morelikely than whites and Asians to bein low-performing K–12 schools.As young adults, they are less likelyto graduate from high school, andeven among high school graduates,they are less likely to be eligiblefor California’s public universitysystem.

We have described a numberof new education policy initiativesthat seek to improve equity inCalifornia. In addition, strategiesthat go beyond early childhooddevelopment and school reformwill likely prove important forreducing racial and ethnic educa-tion gaps. Policies that supportfamilies such as Food Stamps,Medi-Cal, and housing subsidiesprovide basic needs, enabling chil-dren and families to focus ondevelopment and education

18



Notes1 See Hill, Johnson, and Tafoya (2004) foranalysis of the multiracial California popula-tion.

2 U.S.-born includes those born in the 50states and those born abroad of American par-ents. Among U.S.-born non-Hispanic Cali-fornians ages 25 and older who identified asAmerican Indian alone or in combinationwith other races, the college completion ratewas 18 percent.

3 The Southeast Asian category includes onlythose countries from which many refugee-status immigrants have come to the UnitedStates: Laos, Vietnam, and Cambodia.

4 Among people ages 25 to 29 who had notcompleted a bachelor’s degree, 14 percentwere enrolled in college in 2000. Among theU.S.-born, college completion rates by ages30 to 34 are similar to the rates of people ages25 to 29 (Figure 1).

5 The calculation is based on the 2000 Censusand does not include people born in Califor-nia who were not living in the United Statesin 2000.

6 State budget comparisons are based on stateallocations and not actual expenditures. In2002, the most recent year for which stateand local appropriations are available, Califor-nia ranked fourth for total per capita appro-priations (after Wyoming, New Mexico, andNebraska).

7 College completion rates are statistically sig-nificantly higher (at the 95 percent level) forpeople from the rest of the nation relative topeople from California for whites, Hispanics,Filipinos, and Asians.

8 For whites, Hispanics, and blacks, the shareof women with a bachelor’s degree is 4 or 5percentage points higher than that of men(for ages 25 to 29). Among Filipinos andAsians, the share for women is about 10 per-centage points higher than the share for men.Among American Indians, the share is aboutthe same for women and men. Among PacificIslanders, men appear to have substantiallyhigher rates by about 10 percentage pointsbut the difference is not statistically signifi-cant because of the small sample size.

9 Among California-born Hispanics ages 25to 29, 13 percent did not identify a specificHispanic subgroup. Puerto Ricans were thesecond-largest identified group (less than 2percent) and Cubans were the third-largest(about 1 percent).

10 Among California-born Asians ages 25 to29 who identified a single Asian subgroup,the largest groups were Filipino (32 percent),Chinese (26 percent), Japanese (18 percent),Korean (7 percent), and Indian (5 percent).

11 See Horn and Flores (2003) for analysis ofthe effects of ending affirmative action in Cal-ifornia, Texas, and Florida.

12 Jencks and Phillips (1998) show lowerschool skills in kindergarten for black childrenthan white children when comparing familieswith similar parental education and socioeco-nomic status.

13 Racial and ethnic minority is also consid-ered an independent risk factor. For a discus-sion of family factors associated with childrenbeing at risk of low reading scores and schoolfailure, see Denton, West, and Walston (2003).Mayer (1997) finds that family income hasonly a small effect on educational attainmentnationally. She calculates that if familyincome were doubled for poor families, theoverall high school dropout rate would fallfrom 17.3 to 16.1 percent. She argues thatthe effect is small because many basic necessi-ties are met through government programsincluding Food Stamps, Medicaid, and hous-ing subsidies.

14 If we include all children who identify asnon-Hispanic American Indians regardless ofother racial identification, the overall share ofthe young child population is slightly over 1percent and each of the family resource mea-sures would be somewhat higher, particularlythe share whose mother has a high schooldiploma (81 percent).

15 The 2000 Census measures poverty in1999. Since 1999, a recession and unstableeconomic conditions have led to mild growthin poverty (Reed, 2004). See Reed andSwearingen (2001) for a study of povertytrends adjusted for higher rental costs in Cali-fornia.

(Mayer, 1997). Adult educationprograms in English language,vocational skills, and academicscan help improve earnings capacityand other resources for parentsand families. Economic develop-ment in disadvantaged communi-ties would also improve familyresources. Economic developmentand affordable housing programsthat successfully reduce economicstratification across neighborhoodswould likely lessen the concentra-tion of disadvantaged students inlow-performing schools.

Looking over the last fewdecades, California’s experience hasshown that leveling the educationalplaying field is a complex problemthat will not be easily solved. Yet,the demonstrated resolve of Califor-nians to continue working towardquality and equity in educationfuels the hope that the future willbe brighter for all our students. ◆

19



24 In Table 4, the number of “low-income students” is measured by the percentage ofstudents in the school who are enrolled in thefree or reduced-price meals program.

25 See also Education Trust (2004).

26 In his reviews of the literature, Hanushek(1997, 2003) concludes that researchers havenot found a consistent link between schoolresources and student achievement. Kruger(2003) concludes that improved resources,particularly smaller class sizes, are systemati-cally related to higher student achievement.

27 For discussions of the relationships betweenteacher quality and student outcomes, seeEducation Trust (1998); Walsh (2001); Eschand Shields (2002); Rivkin, Hanushek, andCain (2002); and Darling-Hammond (2002).

28 See Harris (2002) for a survey of Californiateachers that suggests that teacher staffing andretention is linked to other measures of schoolquality.

29 Teacher credential levels are measured inthe PAIF. Alternative measures from the APIdata show lower credential levels but similarpatterns across racial and ethnic groups, withrates of 95 percent for white students, 87 per-cent for Hispanic students, and 86 percent forblack students.

30 See Legislative Analyst’s Office (2001) for adiscussion of equity in school facilities finance.

31 Schools are “critically overcrowded” accord-ing to the Department of Education if theyhave 115 or more students per acre for ele-mentary schools or 90 or more students peracre for middle and high schools.

32 If we add multitrack schedule schools tothe measure of overcrowding, the share ofwhites in overcrowded elementary schools is16 percent and is 42 percent for Hispanicsand 40 percent for blacks.

33 Several recent reports have called for astatewide inventory and prioritization of needincluding the Legislative Analyst’s Office(2001), the Little Hoover Commission(2000), and the Joint Committee to Developa California Master Plan for Education (2002).

16 See also Lopez and de Cos (2004). Bridgeset al. (2004) measure attendance in center-based programs and find higher overall atten-dance but similarly low attendance rates forHispanic children (37 percent attend in theyear before kindergarten).

17 See Stipek (2002) for a review of researchon entry age to kindergarten. White childrenare less likely than those in other groups to beattending kindergarten at age 5 but are morelikely at age 6 (41 percent for whites versus anoverall rate of 36 percent).

18 Legislative analysis of Assembly Bill 56 from2004 estimated operating costs at $3.2 billionplus facilities costs of $4.7 billion by 2014.O’Brien-Strain, Moyé, and Sonenstein (2003)estimate the annual cost of universal pre-school at up to $5 billion.

19 At the time of the 2000 Census, there wereclose to 600,000 poor children ages 0 to 5 in California and of those ages 3 to 5, about145,000 were not enrolled in school. HeadStart statistics are from the Head Start Bureauof the Administration for Children and Fami-lies at the U.S. Department of Health andHuman Services (www.acf.hhs.gov).

20 For a recent evaluation, see U.S. Depart-ment of Health and Human Services (2003).For a brief review of the research, see Barnett(2002).

21 A state rank of 3 or below is low-perform-ing as defined in the settlement agreement ofthe Williams case (discussed in the next sec-tion). Incentive programs for teachers definelow-performing as a rank of 5 or below. The Public School Accountability Act defines low-performing as failure to meet performancegoals and no significant improvement in per-formance.

22 See also Hanushek et al. (2003) and Hoxby(2000) for evidence on the effects of peers’academic skills on student learning.

23 We rely on the API rank, which, by con-struction, includes 30 percent of schools each year. Since 1999, the API scores of low-performing schools has increased (Rose et al.,2003).

34 Statistics for Asian and Pacific Islander sub-groups are available at http://star.cde.ca.gov.Results for California from the NationalAssessment of Educational Progress (2003)also show large gaps between blacks and His-panics compared to whites and Asians(National Center for Education Statistics,2004).

35 Dropout information from California highschools is somewhat inaccurate becauseschools track when students leave, but theyare not consistently able to track whetherleaving students enter other high schools orreturn in later years.

36 Under the Public School AccountabilityAct, the Superintendent of Public Instructionmay permit cross-district transfers to studentsin a low-performing school. See Fuller et al.(1999) for a discussion of school choice policyin California.

37 In 2003–04, UC had a “dual admissionsprogram” whereby students who were in therange of the top 4 to 12.5 percent of theirhigh school class were guaranteed admissionto a specific UC campus if they successfullycompleted a transfer program at CCC. Thisprogram has been discontinued because offunding cuts.

38 Compared to other university systems inthe nation, UC has a strong record for enroll-ment of low-income students (Mortenson,2004). The evidence on racial and ethnicadmissions for specific campuses is mixed (seeUniversity of California, 2004). UCOP calcu-lates underrepresented minority rates as ashare of domestic admissions (as opposed tousing the California resident admissions, as inFigure 5). Statistics by race are available forblack, American Indian, Asian, Chicano, EastIndian/Pakistani, Filipino, Latino, and whitestudents (University of California, 2004).

39 Enrollment data include first-time fresh-men enrollees from private high schools andfrom out of state. If limited to graduates fromCalifornia public high schools, the share ofwhites would be slightly lower (34 percent)and the share would be higher for Latinos (15percent) and Asians (35 percent).

20



Denton, Kristin, Jerry West, and Jill Walston,Reading—Young Children’s Achievement andClassroom Experiences, U.S. Department ofEducation, National Center for EducationStatistics, NCES 2003-070, Washington,D.C., 2003.

Education Trust, “Good Teaching Matters:How Well-Qualified Teachers Can Close theGap,” Thinking K–16, Vol. 3, No. 2, Wash-ington, D.C., Summer 1998.

Education Trust, Education Watch California:Achievement, Attainment, and Opportunityfrom Elementary School Through College,Washington, D.C., Spring 2004.

Esch, Camille E., and Patrick M. Shields,Who Is Teaching California’s Children? Centerfor the Future of Teaching and Learning,Santa Cruz, California, 2002.

Fuller, Bruce, Elizabeth Burr, Luis Huerta,Susan Puryear, and Edward Wexler, SchoolChoice: Abundant Hopes, Scarce Evidence ofResults, Policy Analysis for California Educa-tion, University of California, Berkeley, andStanford University, 1999.

Gill, Brian P., P. Michael Timpane, Karen E.Ross, and Dominic J. Brewer, Rhetoric VersusReality: What We Know and What We Need toKnow About Vouchers and Charter Schools,RAND Corporation, Santa Monica Califor-nia, 2001.

Hanushek, Eric A., “The Economics ofSchool Resources and Student Performance:An Update,” Educational Evaluation and Poli-cy Analysis, Vol. 19, No. 2, Summer 1997, pp.141–164.

Hanushek, Eric A., “The Failure of Input-Based School Policies,” Economic Journal, Vol.113, February 2003, pp. F64–F98.

Hanushek, Eric A., John F. Kain, Jacob M.Markman, and Steven G. Rivkin, “Does PeerAbility Affect Student Achievement?” Journalof Applied Econometrics, Vol. 18, No. 5, Sep-tember/October 2003, pp. 527–544.

Harris, Louis, A Survey of the Status of Equalityof Public Education in California: A Survey of aCross-Section of Public School Teachers, March2002, available at www.publicadvocates.org.

Haveman, Robert, and Barbara Wolfe, “TheDeterminants of Children’s Attainments: AReview of Methods and Findings,” Journal ofEconomic Literature, Vol. 33, No. 4, 1995.

Hill, Laura E., Hans P. Johnson, and SonyaM. Tafoya, “California’s Multiracial Popula-tion,” California Counts, Vol. 6, No. 1, PublicPolicy Institute of California, San Francisco,California, August 2004.

Horn, Catherine L., and Stella M. Flores,“Percent Plans in College Admissions: AComparative Analysis of Three States’ Experi-ences,” The Civil Rights Project at HarvardUniversity, Cambridge, Massachusetts, 2003.

Hoxby, Caroline, “Peer Effects in the Class-room: Learning from Gender and Race Varia-tion,” National Bureau of Economic ResearchWorking Paper 7867, Cambridge, Massachu-setts, 2000.

Jencks, Christopher, and Meredith Phillips,The Black-White Test Score Gap, BrookingsInstitution Press, Washington, D.C., 1998.

Johnson, Hans P. “Movin’ Out: DomesticMigration to and from California in the1990s,” California Counts, Vol. 2, No. 1, Public Policy Institute of California, SanFrancisco, California, August 2000.

Joint Committee to Develop a CaliforniaMaster Plan for Education, California MasterPlan for Education, Sacramento, California,2002.

Krueger, Alan B., “Economic Considerationsand Class Size,” Economic Journal, Vol. 113,February 2003, pp. F34–F63.

Legislative Analyst’s Office, A New Blueprintfor California School Facility Finance, Sacra-mento, California, 2001.

Little Hoover Commission, To Build a BetterSchool, Sacramento, California, 2000.

Lopez, Elias S., and Patricia L. de Cos,Preschool and Childcare Enrollment in Califor-nia, California Research Bureau, Sacramento,California, January 2004.

Mayer, Susan E., What Money Can’t Buy:Family Income and Children’s Life Chances,Harvard University Press, Cambridge, Massa-chusetts, 1997.

ReferencesBarnett, W. Steven, The Battle Over HeadStart: What Research Shows, National Institutefor Early Education Research, The State Uni-versity of New Jersey, Rutgers, New Jersey,2002.

Betts, Julian R., The Changing Role of Educa-tion in the California Labor Market, PublicPolicy Institute of California, San Francisco,California, September 2000.

Betts, Julian R., Andrew C. Zau, and LorienA. Rice, Determinants of Student Achievement:New Evidence from San Diego, Public PolicyInstitute of California, San Francisco, Califor-nia, 2003.

Bridges, Margaret, Bruce Fuller, Russell Rumberger, and Loan Tran, “Preschool forCalifornia’s Children: Promising Benefits,Unequal Access,” Policy Brief, Policy Analysisfor California Education, University of Cali-fornia, Berkeley, and Stanford University, September 2004.

California Department of Education, Califor-nia High School Exit Exam (CAHSEE) InternetReports, Sacramento, California, 2004, avail-able at http://cahsee.cde.ca.gov/reports.asp.

California Postsecondary Education Commis-sion, Student Transfer in California Postsec-ondary Education, Sacramento, California,February 2002.

California Postsecondary Education Commis-sion, Resident Undergraduate Charges at Cali-fornia’s Public Colleges and Universities,Sacramento, California, January 2004a.

California Postsecondary Education Commis-sion, University Eligibility Study for the Classof 2003, Sacramento, California, May 2004b.

College Board, 2001 Advanced PlacementState and National Summary Reports, The Col-lege Board, Princeton, New Jersey, 2001.

Darling-Hammond, Linda, “Research and Rhetoric on Teacher Certification: AResponse to ‘Teacher Certification Reconsid-ered,’” Education Policy Analysis Archives, Vol.10, No. 36, September 2002.

21



22

Rivkin, Steven G., Eric A. Hanushek, andJohn F. Kain, “Teachers, Schools, and Acade-mic Achievement,” unpublished manuscript,July 2002.

Rose, Heather, “The Concept of Adequacyand School Finance,” in Jon Sonstelie andPeter Richardson (eds.), School Finance andCalifornia’s Master Plan for Education, PublicPolicy Institute of California, San Francisco,California, 2001.

Rose, Heather, Jon Sonstelie, Ray Reinhard,and Shermaine Heng, High Expectations,Modest Means: The Challenge Facing Califor-nia’s Public Schools, Public Policy Institute ofCalifornia, San Francisco, California, 2003.

Rueben, Kim, and Jane Leber Herr, “TeacherSalaries in California,” in Jon Sonstelie andPeter Richardson (eds.), School Finance andCalifornia’s Master Plan for Education, PublicPolicy Institute of California, San Francisco,California, 2001.

Sonstelie, Jon, Eric Brunner, and KennethArdon, For Better or For Worse? School FinanceReform in California, Public Policy Institute ofCalifornia, San Francisco, California, Febru-ary 2000.

Stipek, Deborah, “At What Age Should Children Enter Kindergarten?” Social PolicyReport, Vol. XVI, No. 2, Society for Researchin Child Development, University of Michi-gan, Ann Arbor, Michigan, 2002.

University of California, Office of the Presi-dent, University of California: GraduationRates for First-Time Freshmen, UCOP StudentAcademic Services, Information Digest 2003, available at www.ucop.edu/sas/infodigest03/Persistence_Freshmen.pdf.

University of California, Office of the President, “Final Summary of FreshmanApplications, Admissions and Enrollment,Fall 1995–2003,” 2004, available at www.ucop.edu/news/studstaff.html.

U.S. Department of Education, NationalCenter for Education Statistics, Access to Post-secondary Education for the 1992 High SchoolGraduates, NCES 98-105, Washington, D.C.,October 1997.

Mortenson, Thomas, “Pell Grant Shares ofUndergraduates Enrollments at the 51 BestNational Liberal Arts Colleges 1992–93 and2001–02,” Postsecondary Education Oppor-tunity, Oskaloosa, Iowa, 2004.

National Center for Education Statistics, The Nations Report Card: State Profiles, Washington, D.C., 2004, available atwww.nces.ed.gov/nationsreportcard/states/.

O’Brien-Strain, Margaret, Laura Moyé, FreyaLund Sonenstein, Arranging and Paying forChild Care, Public Policy Institute of Califor-nia, San Francisco, California, December2003.

O’Day, Jennifer, Catherine Bitter, MikeKurst, Martin Carnoy, Elisabeth Woody,Melissa Buttles, Bruce Fuller, and DavidRuenzel, “Assessing California’s AccountabilitySystem: Successes, Challenges, and Opportu-nities for Improvement,” Policy Brief 04-2,Policy Analysis for California Education, Uni-versity of California, Berkeley, and StanfordUniversity, February 2004.

Palmer, James C. (ed.), Grapevine: An AnnualCompilation of Data on State Tax Appropria-tions for the General Operation of Higher Edu-cation, Center for the Study of EducationPolicy, Illinois State University, Normal, Illinois,2004, available at www.coe.ilstu.edu/grapevine/.

Reed, Deborah, “The Growing Importance of Education in California,” Occasional PaperSeries, Public Policy Institute of California,San Francisco, California, July 2003.

Reed, Deborah, “Recent Trends in Incomeand Poverty,” California Counts, Vol. 5, No.3, Public Policy Institute of California, SanFrancisco, California, February 2004.

Reed, Deborah, and Richard Van Swearingen,“Poverty in California: Levels, Trends, andDemographic Dimensions,” CaliforniaCounts, Vol. 3, No. 3, Public Policy Instituteof California, San Francisco, California,November 2001.

Reyes, Belinda (ed.), A Portrait of Race andEthnicity in California: An Assessment of Socialand Economic Well-Being, Public Policy Insti-tute of California, San Francisco, California,2001.

U.S. Department of Education, NationalCenter for Education Statistics, The Condi-tion of Education 2004, NCES 2004-077,Washington, D.C., 2004.

U.S. Department of Health and Human Services, Head Start FACES 2000: A Whole-Child Perspective on Program Performance,Fourth Progress Report, Washington, D.C.,May 2003.

Walsh, Kate, Teacher Certification Reconsid-ered: Stumbling for Quality, Abell Foundation,Baltimore, Maryland, 2001.



23

The Public Policy Institute of California is a private, nonprofit research organization established in 1994 with an endowment from WilliamR. Hewlett. The Institute conducts independent, objective, nonpartisan research on the economic, social, and political issues affecting Californians. The Institute’s goal is to raise public awareness of these issues and give elected representatives and other public officials inCalifornia a more informed basis for developing policies and programs. PPIC does not take or support positions on any ballot measure oron any local, state, or federal legislation, nor does it endorse, support, or oppose any political parties or candidates for public office.

PUBLIC POLICY INSTITUTE OF CALIFORNIA500 Washington Street, Suite 800 • San Francisco, California 94111Telephone: (415) 291-4400 • Fax: (415) 291-4401 • www.ppic.org

Board of DirectorsCheryl White Mason, ChairChief, Civil Liability ManagementOffice of the City AttorneyLos Angeles, California

Edward K. HamiltonChairmanHamilton, Rabinovitz & Alschuler, Inc.

Gary K. HartFounderInstitute for Educational ReformCalifornia State University, Sacramento

Walter B. HewlettDirectorCenter for Computer Assisted Research in the Humanities

David W. LyonPresident and CEOPublic Policy Institute of California

Arjay MillerDean EmeritusGraduate School of BusinessStanford University

Ki Suh ParkDesign and Managing PartnerGruen Associates

Constance L. RiceCo-DirectorThe Advancement Project

Thomas C. SuttonChairman & CEOPacific Life Insurance Company

Raymond L. WatsonVice Chairman of the Board EmeritusThe Irvine Company

Carol WhitesidePresidentGreat Valley Center

ISSN #1554-401X



NON-PROFIT ORG.U.S. POSTAGE

PAIDBRISBANE, CAPERMIT #83

PUBLIC POLICY INSTITUTE OF CALIFORNIA500 Washington Street, Suite 800San Francisco, California 94111

CaliforniaÕsRacial and Ethnic Educational Gaps

In This Issue

New Series from PPIC

California Economic Policy

This quarterly series consists of timely reports that analyzeand discuss policy issues affecting the California economy.The analyses are presented in a non-technical style to improve understand-ing of complex economic issues amongpolicymakers, the media, and the gen-eral public. There is no charge for thisseries.

All CEP issuesare available on our websitewww.ppic.org


If you would like to receive future issues of California Economic Policy, please call our toll-freenumber, 800-232-5343, or go to our website at www.ppic.org/main/mailinglist.asp

Upcoming Topics: Technology Exports

Living Wage

of california california countsamong californians are fairly sim-ilar to the rates of people from...

Documents