examining latino representation on california’s school boards latino...survey of school board...

Examining Latino Representation on California’s School BoardsIt’s impact on perceptions of District problems, priorities and policies

Max NeimanBelinda ReyesLuis FragaDaniel Krimm

October, 2010

César E. Chávez InstituteCollege of Ethnic StudiesSan Francisco State University

Produced with funding from The Spencer Foundation

Contents

Acknowledgments ii

Summary 1

Introduction 4

CHAPTER 1: SCHOOL BOARD MEMBERS AS ELECTED OFFICIALS 8

Background Characteristics 8

Characteristics of the Districts in which Latinos Serve 10

Political and Institutional Characteristics of the Districts 11

Why They Decided to Run 14

Factors Making It Difficult to Run 15

CHAPTER 2: SCHOOL BOARD MEMBERS AS POLICY MAKERS 18

The Challenges in the Districts Where Latinos Serve 18

Perceived Severity of District Problems 19

School Board Priorities 22

Are there Differences in Policy between Latino and white Board Members? 24

Conclusion & Policy Recommendations 26

References 28

About the Authors 31

Appendix A: Regression Models 33

Appendix B: Survey of School Board Members 2009 39

EXAMINING REPRESENTATION ON CALIFORNIA’S SCHOOL BOARDS ii

Acknowledgments

We would like to acknowledge the support of the Spencer Foundation and the Public Policy Institute of California. Debbie Reed and Hans Johnson provided indispensable guidance in the development of this project. We also benefited greatly from the research expertise of Monique Magana, Amrah Salome, Daniel Cornejos, and Ana Frost, and the editing and distribution support of Miguel Casuso.

EXAMINING REPRESENTATION ON CALIFORNIA’S SCHOOL BOARDS 1

Summary

Latinos are soon to be the majority of students in California‟s public schools. Meanwhile, they remain under-represented in school boards. In two thirds of school districts there is no Latino representation. There are many reasons for why we should care about a representation gap. For some, the mere under-representation of Latinos on school boards is, on its face, a problem. Insofar as democracy implies some accountability and responsiveness to constituents, when a substantial segment of constituents are unable to elect someone that represent them, it is possible that their views and concerns will not be considered when decisions are made about schools. The absence of Latinos, then, implies a deficit of concern for those matters that might be of most concern to Latino school children and their parents. But this may also matter in terms of policy. Latinos may have a different set of priorities or see issues differently than white board members, and may therefore advocate for a different set of policies. Moreover, Latino representation could have an impact on Latino student‟s performance or parental involvement in schools. If Latino board members are better able to engage Latino children and parents in schools, diversity in the board could have an impact on the educational progress of one of the most disadvantage groups in California.

Using secondary data from government sources, previous studies and an on-line, web-based survey of school board trustees,1 we explored barriers to Latino representation on school boards, and school board trustee beliefs and perceptions regarding a variety of policy and political issues in their district. The survey included items dealing with the trustees‟ backgrounds, political aspirations, perceptions concerning electoral competition in the district, parent engagement, problems facing the school district, and assessments regarding barriers to running for the school district, and policy priorities.

The general conclusions of this report are:

Latinos represent a relatively small proportion of school board members, only about 15 percent of board members in the state of California are Latinos. Latino representation increases with the number of Latino students in the districts, but even in districts where Latinos are between half and two-thirds of the students, the majority had no Latino representation. Our study highlights the following findings about the barriers confronting school board members and their views aspirations and policy choices:

BARRIERS TO REPRESENTATION

1 The details of the on-line, web-based survey are provided in the Appendix.


Close to 20% of Latino elected officials indicate that at-large systems2 posed a barrier (significantly more so than did their white counterparts). At-large elections may limit the number of Latinos in school boards in California and the likelihood of having more than one Latino on the board, even after controlling for other factors. However, we have no way of estimating how many potential Latino or other candidates were deterred from running due to having to compete in at-large systems. More important than the election system, higher proportions of Latino trustees consistently rate a host of other factors as barriers to overcome in running for office: prominent among these other factors are low turnouts, high numbers of ineligible residents, and the cost of running a school board election. They are also more likely to have expressed having difficulty filing candidate papers and finding information about their district and policy options.

ASPIRATIONS AND PRIORITIES

Although most trustees profess not to have ambitions beyond the school district board, Latino respondents are somewhat more likely to aspire to other elected office than white trustees – 28% of Latinos say they would consider running for another office and 11% indicated pursuing another political office was a factor in their decision to run for school boards; All board members are guided to seek school board because they are committed to maximizing the achievement of students. Latino trustees, however, also indicated higher levels of interest across a broader agenda (e.g., district hiring and administration issues) than did white trustees.

PERCEPTIONS OF PROBLEMS AND POLICIES

Compared to their white counterparts, and reflecting the conditions of their districts, Latino trustees are more likely to see a variety of problems as serious, particularly such issues as low levels of college attendance or school and community condition. Whites, by contrast, are more concerned with the involvement of the state and federal government in school policy.

Latino trustees, independently of district characteristics, tend to be more concerned about such matters as advocating on behalf of immigrant students, raising the percentage of minority college students, and increasing the number of trustees and teachers of color in their districts. There is evidence that variation in the number of Latino school board members makes a difference, not only with regard to perceptions and priorities, but also with regard to such matters as the districts‟ number of Latino teachers and administrators. Moreover, having more than one Latino on the board is associated with a small decline in dropout rates for students between the 9th and 12th grades; these differences remain even after adjusting for district characteristics.

In general, the survey findings point to important differences between Latino and white school board members in barriers to running for office (even among successful candidates), and to

2 Using the definition of at-large election systems in the Elections Code: refers to “any of the following

methods of electing members to the governing body of a political subdivision: (1) One in which the voters of the entire jurisdiction elect the members to the governing body. (2) One in which the candidates are required to reside within given areas of the jurisdiction and the voters of the entire jurisdiction elect the members to the governing body. (3) One that combines at-large elections with district-based elections.” (California Elections Code, Section 14026).


differences in priorities, preferences and policies, even after controlling for socioeconomic and other conditions in the districts. This research underscores potential benefits for increasing ethnic diversity in school boards.

The findings of this study also signal potential avenues for policy intervention. Non-citizen voting, caps on campaign spending, facilitating the candidate recruitment process, and increasing the amount of information available to potential candidates may help increase Latino representation on school boards. Moreover, while Latinos are as likely to get into school boards in at-large and district election systems, district elections may help elect more Latinos, allowing them to have greater influence in school board decisions. Parents also need to be made aware of the importance of voting in school board elections, since voter turnout is one of the largest barriers to getting Latinos on the board.


Introduction

On March 2010, Kingsburg High School's board, located south of Fresno in California‟s Central Valley, voted grudgingly to get rid of at-large elections in the school district (Hoagland, 2010).3 Board members were pressured to take this action in order to avoid a likely lawsuit. The Lawyers' Committee for Civil Rights had been filing lawsuits in the San Joaquin Valley over at-large elections and in one of the cases was seeking to collect $1.2 million in legal fees from the Madera Unified School District (Hoagland, 2010). The central claim of the cases is that at-large system of electing trustees violates the California Voting Rights Act of 2001.4

These lawsuits reflect a general concern that many scholars and policy makers have regarding the impact of institutions on the civic engagement of previously marginalized groups. Systems of election have come under persistent political and legal scrutiny, because there are reasons to believe that some kinds of election systems dilute the voting strength of racial and ethnic groups, leading to the under-representation of socially disadvantaged groups. But other factors, such as campaign costs, may potentially be barriers to representation for disadvantaged groups.

There are at least two important reasons for why a representation gap is important. On the one hand, it is possible that the perceptions and priorities of Latino school board members are different from their non-Latino (mostly white) counterparts. Insofar as democracy implies some accountability and responsiveness to constituents, when a substantial segment of constituents are unable to elect one of “theirs,” it is possible that their views and concerns will be discounted in the relevant decision making body. For some, the mere gap between the proportional percentage of Latinos on school boards relative to the proportion of students and families who are Latino is, on its face, a problem. For others, the issue is one of policy responsiveness, the assumption being that Latino school board trustees are more likely to highlight the needs and problems of Latino families and children. The absence of Latinos, then, implies a deficit of concern for those matters that might be of most concern to Latino school children and their parents. As discussed below, others have suggested that the representation gap is important because, through a variety of mechanisms, it might explain the student performance gap.

This issue is possibly critical in California, where Latinos are soon to be the majority of school children. As of 2008-09, the California Department of Education reported that Latinos comprised 49 percent of children in California schools, while only 27.9 percent were white. However, the political leadership in California schools still remains principally white non-

3 The form of local elections varies from school district to school district. Some cities elect their local

representatives by district, some have at-large elections, and some have both. In at-large system of voting all candidates for school board are elected by all of the city's voters. In district elections, council member are selected from a geographical section of the city, or ward. Some cities combine these two methods and elect some council members at large and some from districts.

4 Recent California law has made it easier to sue to challenge at-large elections. The California Voting Rights Act of 2001 expands the voting rights granted under the federal Voting Rights Act by, among other things, prohibiting at-large voting if evidence shows that such voting "impairs the ability" of a minority group "to elect candidates of its choice or its ability to influence the outcome of an election."


Hispanic. As of 2005, Latinos were less than 15 percent of the trustees in California schools, merely 15 percent of school administrators, and a scant 6 percent of schoolteachers.

Meanwhile, Latinos have some of the poorest educational outcomes. They have some of the highest dropout rates, some of the lowest college-going and college completion rates. Scholars in education, and more generally in the social sciences, have consistently shown that aspects of family structure, income, schools and neighborhoods have a direct impact on patterns of educational achievement for students. There is also an extensive literature examining the impact of particular policy approaches on student outcomes from class-size reduction, teacher training, to bilingual education. What is less clear is the role of school boards in policy development and whether the perceptions, beliefs, and choices of school board trustees and local elites are associated with variation in school district school policy.

This project examines Latino representation on California school boards, the implications of a lack of representation on school policy, and the potential barriers to representation. The research addresses issues of representation and policymaking in a multicultural society, both of which are related to policy legitimacy, which is often the first ingredient in advancing needed policy changes.

Research on Latino school board representation has reached four consistent conclusions. First, levels of Latino representation are greater in single-member district systems of election as compared to elections at-large, controlling for relevant factors such as percent of population, political resources, Latino non-citizen population, Latino education, and median family income (Fraga, Meier, and England 1986; Meier and Stewart 1991; Leal, Meier, and Martinez-Ebers 2004; Meier and Juenke 2005). This is especially the case when Latinos are a minority of the overall population (Meier and Juenke 2005).5 Although it is rare to find districts where Latino representation is proportional to the population, and even less prevalent using Latino student enrollment as a baseline, research has consistently found a representational bias in the use of at-large elections in districts where Latinos are not a majority of the population. 6

The assumed reason that at-large systems of election produce fewer Latino representatives than do single-member district systems of election is the presence of voter polarization where, for example, an Anglo/Caucasian majority of the electorate votes as a block against the preferences of Latino voters, who also vote as a block but, as a result of being a minority of the electorate, are unsuccessful in getting their first-choice candidates elected. Under district systems, where a school district is subdivided into smaller geographical representational units, it is more likely that ethnic and racial communities will comprise a majority of the electorate in a specific number of these units. It is, of course, possible for a system of single-member districts to be gerrymandered to the advantage or disadvantage of specific communities. Interestingly, one study of Texas school districts finds that this structural bias in the system of representation does not exist in districts where Latinos are over 50 percent of the population. In these districts, at-large election systems are just as effective as ward systems in facilitating Latino representation on school boards (Meier and Juenke 2005).

5 This discussion is largely taken from Luis R. Fraga and Roy Elis, ““Interests and Representation: Ethnic

Advocacy on California School Boards.” Teachers College Record 111 (3): 659-682. 6 The same relationship between school board representation and educational outcomes is found for

African Americans (Meier and England 1984; Meier, Stewart, and England 1989).


A second conclusion of the research is that higher levels of Latino representation on school boards are associated with larger numbers of Latino school administrators (Meier and Stewart 1991; Leal, Meier, and Martinez-Ebers 2004; Meier and Juenke 2005). Enhanced political representation, it is claimed, leads to greater descriptive bureaucratic representation, especially in systems where Latinos are a majority of the population. One study found that in school districts where Latino school board members are elected from single-member districts, more Latino administrators are hired as compared to those districts where Latinos are elected at large (Meier and Juenke 2005).

Third, greater Latino representation on school boards and enhanced administrative representation are both associated with greater Latino presence in teaching staffs (Meier and Stewart 1991; Leal, Meier, and Martinez-Ebers 2004; Meier and Juenke 2005). However, the influence of Latino administrators has a noticeably greater effect. Interestingly, Meier and Juenke (2005) found that this is especially the case in Texas districts where Latinos comprise a majority of the population. Districts with Latino population majorities and significant Latino administrative presence are especially likely to have higher numbers of Latino teachers.

The fourth conclusion reached in existing research is that higher levels of Latino presence among district teaching staffs are associated with more favorable educational experiences by Latino students as measured by outcomes such as graduation rates, dropout rates, enrollment in advanced placement classes, suspensions, expulsions, and, more recently, standardized test scores (Fraga, Meier, and England 1986; Meier and Stewart 1991; Leal, Meier, and Martinez-Ebers 2004; Meier and Juenke 2005). This independent effect of the impact of Latino teachers holds even when appropriate controls are applied, such as white student performance, average teacher salary, class size, percent of noncertified teachers, percentage of African American students, percentage of Latino students, and percentage of students receiving free or reduced-price lunch.

Meier and Juenke provide a succinct summary of the primary findings in this scholarly work, stating:

A relatively simple choice – whether to hold school board elections at large or by wards – influences not only the ethnic composition of the school board but also the quality of the Hispanic representation on that board. That representational quality influences the number of Hispanic administrators employed by the school district and, indirectly, the number of Hispanic teachers in the classroom. Hispanic teachers, in turn, affect the quality of education provided to Hispanic students. Biases generated by political structures, therefore, reverberate throughout the system (2005: 222-223).

What underlies all of this research is an ethnic alignment assumption. It is assumed that Latino voters vote as a block and largely have the same electoral preferences. It is also assumed that it is Latino representatives on school boards who are responsible for more Latino administrators. Similarly it is assumed that Latino administrators are largely responsible for hiring more Latino teachers. Finally, it is assumed that Latino teachers are responsible for more favorable educational experiences of Latino students. None of the above studies, however, have actually examined the practices of Latino school board members, administrators, or teachers.


The claims made in this line of research are, while logical and reasonable, largely unexamined and untested. Stated differently, there is a black box of policy advocacy and decision-making that has been the foundation of over twenty years of research on the role of Latinos on school boards. The primary goal of our project is to open this black box and begin to explore the inner workings of policy agenda setting, policy interests, and policy decision-making by Latino and non-Latino school board members in California. Moreover, we are interested in the potential barriers confronting Latinos from getting elected onto school boards in a search for mechanisms to increase Latino representation in school boards.


Chapter 1

SCHOOL BOARD MEMBERS AS ELECTED OFFICIALS

Latinos constitute a relatively small percentage, approximately 14 to 15 percent, of the school board membership in California,7 despite the fact that nearly 50 percent of the state‟s public school children are Latino. Latino school board members are elected in certain types of districts and have different demographic characteristics from white board members, who are about 60 percent of school board members in California.

In the following sections we examine some of the reasons that our responding trustees provide for seeking the school board position and some of the barriers they face in running successfully. But first we start with a look at the characteristics of board members and the districts in which they serve.

BACKGROUND CHARACTERISTICS

Table 1.1 summarizes some of the differences in the background characteristics of the respondents to our survey of school board members.8 Over 30% of the respondents are Latino, and so we are clearly oversampling the Latino school board members, since only about 16 percent of the board members in schools with at least 10% Latino students are Latino.9 The data indicate that generally the school board members who are Latino are somewhat younger and have higher percentages of respondents who have had some college or less schooling. Of the Latino respondents, about 38% have less than a college education, although they appear to be somewhat overrepresented in the doctoral degree category.10 Part of the lower reported education levels might also be a result of the younger status of the Latino respondents, with 31% being 45 years old or younger.

7 According to Arrona and Donoso (2006), in California‟s “elementary school districts there are 1,736

school board seats available. Of these seats, Latinos represent only 15%, or 266 members. 8 Caution needs to be exercised, since our samples of Latino and non-Latino school board members are

smaller than we would have preferred. 9 We needed to oversample Latinos in order to capture a significant number of Latino board members.

Of our sample of 332 respondents, 124 are Latinos, which although small, still is large enough to generate some basic results.

10 Interestingly a higher percentage of Latino respondents have “some college” education, compared to about 25% of the whites. Also, nearly 12% of the Latino respondents have doctorates, while only 7% of the whites have them.


Table1.1 Background Characteristics of Latino and White Board Members White Latino

Gender

Female 50% 48%

Male 50% 52%

School Board Experience

Mean years served on school board 9 8

Level of Schooling*

Less than high school 0% 1%

High school 2% 7%

Some college 23% 30%

College graduate 44% 33%

Masters degree 24% 18%

Doctoral degree 7% 11%

Age*

25 or younger 0% 0%

26-35 2% 10%

36-45 14% 21%

46-55 39% 30%

56-65 25% 23%

Greater than 65 years old 20% 16%

Respondent Race/Ethnicity

White 60% -

Latino - 32%

*Indicates that the differences between Latino and Non-Latino School Board members is significant at a 5 percent level.

Of the respondents to our survey, 60% are Caucasian, 32% are Latino, and 8% are comprised of “others,” which include self-identified multi-racial, multi-ethnic categories, as well as, African-Americans, Asian and Pacific Islanders. In this study we concentrate on results for white and Latino board members, since the sample for “other” is too small.11

11 With only 28 total respondents in the “other” categories, the cell sizes are simply too small to make the

discussion of results for these respondents meaningful.


Characteristics of the Districts in which Latinos Serve

For this report we selected districts in which Latinos are at least 10 percent of the student population. Considering these districts as a whole, Latinos are 16 percent of trustees in school boards. In two-thirds of these districts, however, there are no Latino board members, even though Latinos are at least a third of students in 47 percent of districts. If one also assumes that a Latina will behave and see things differently if she has Latino colleagues on the board, the picture is a bit more daunting. Only 16 percent of the districts have more than one Latino on the school board. Figure 1.1 shows Latino representation on school boards by the enrollment of Latino students in the districts. It is not until Latinos are an overwhelming majority of the district‟s students that we see a large proportion of Latino board members.

Figure 1.1 Latino Trustees by the Proportion of Latino Students in the District

Source: Number generated from the 969 districts in California with Latino enrollment of at least 10 percent.

Latino representation is highest in the Central Valley and in Los Angeles (See Table 1.2). It is lowest in the “Rest of State” category, which includes counties in the Northern and Sierra counties of the state. Latinos serve in larger districts with more Latino students. Districts where Latinos do not serve are, on average, 24 percent Latino and 58 percent white. Where there is at least one Latino trustee, on average the Latino student population is 63 percent and the white student population is 22 percent. The largest of the districts with some Latino representation are Los Angeles and San Diego Unified. In contrast, Long Beach, Capistrano, Garden Grove, Riverside and Fontana Unified are large districts with no Latino representation. All of the districts mentioned above, except Capistrano Unified, have Latino student populations of over 50 percent.

9.3%

34.4%

46.2%

86.2%

2.1%

8.6%

18.9%

58.7%

less than1/3 1/3 to half half to 2/3 2/3 & more

Latinos in the Board

More than 1 Latino

Proportion of Latino Students in the District


Table 1.2 Characteristics of All Districts With at Least 10% Latino Students

No Latinos on Board

At least 1 Latino

More than 1 Latino

Percent of Districts 67.2 32.8 16.4

Average number of Latinos in the Board 0 1.97 2.95

Regions of the State

Los Angeles 5.07 14.8 20.8

Rest of Southern CA 13.4 11.95 10.1

Bay Area 19.5 11.01 10.1

Central Valley 32.3 43.1 40.3

Rest of State 29.8 19.2 18.9

Total number of students in the District 2007-08 4362 10,496 13,349

Average school enrollment in the district in 2008 378.650 548.665 591.525

% of District Students who are Latinos 24.3% 63.1% 74.1%

% of District Students who are white 58.4% 22.4% 13.4%

Number of Latinos in the average school in the District 109.950 351 440.9

Note: Number generated from the districts in California with more than 10% Latino student enrollment.

Political and Institutional Characteristics of the Districts

The survey included a number of questions regarding the institutional, political, and formal context of the respondents‟ local school boards. The questions covered such topics as the system of school board elections, preferences regarding seeking higher office, perceived competitiveness of school board elections, amounts of time spent on school board matters, monetary compensation and benefits for serving as a school board trustee, as well as party identification among respondents.

The results in the following table indicate a few significant differences between Latino and white respondents in terms of their political affiliation, aspiration, and the political environment in their district. Seventy-seven percent of the Latino respondents report being either a Democrat or Strong Democrat, compared to only about 36% of the white respondents. However, the same proportion of white and Latino board members were elected (90%) as opposed to appointed (10%) to the board. Thirty percent of feel there was a fair or a lot of competition in their election.

While large majorities of the respondents indicated that they have no intention of seeking other elected office, there is a significant difference in the percentage of Latino respondents who indicate that they intend to seek other elected office, with more than one in five indicating their intent to do so, compared to 14% among the white school board respondents.


There seems to be no statistically significant difference across racial or ethnic categories when it comes to whether or not the respondent is from a district or at large/mixed election district. Nevertheless, when we assessed a model of the effect of the election system on the number of Latino school board members, the results indicated that the election system might be correlated with the number of Latinos on a school board. We modeled the probability that a district would have a Latino or more than one Latino on the board, and the number of Latinos on the board, while controlling for enrollment in the district, level of segregation in the district, percent of children on school lunch programs, the number of English Learners in the average school, the system of elections, whether the district is majority Latino, and controlling for being in Los Angeles.12 Although significant at only a 10 percent level, on average, the number of Latino board members and the probability of having more than one Latino on the board are higher with district elections, controlling for other factors in the district.

In terms of the compensation and time required to serve in school board, the respondents across ethnic racial categories report spending about the same amount of time on school board work, 9 to 10 hours on average. Latino respondents are far more likely to receive compensation for school board work, and among all respondents receiving compensation, Latino respondents receive nearly twice the monthly compensation as that reported by the white respondents. Once we control for district characteristics, such as district size, proportion of immigrants, and poverty levels, the differences in benefits and pay between Latinos and others disappear.13

12 Results from these models are available in Appendix A Tables A.1. 13 Results from the model are available upon request.


Table 1.3 Political and School Board Context White Latino

Seek Other Elected Office*

Yes 14% 28%

No 86% 72%

System of School Board Elections

At-Large 91% 88%

District 9% 12%

Whether Respondent Appointed or Elected

Appointed 10% 10%

Elected 90% 90%

Perceived Competitiveness of Open Seats*

Very little competition 36% 33%

Some competition 34% 36%

Fair amount of competition 23% 17%

A lot of competition 7% 14%

Average time on school board matters

Mean time in hours per week 9 10

Receives compensation for school board work*

Yes 50% 69%

No 50% 31%

Average amount of monetary compensation*

Dollars/month $331 $628

Receive medical benefits as board member*

Yes 53% 74%

No 47% 26%

Partisan Identification*

Strong Democrat 14% 30%

Democrat 22% 47%

Independent 16% 14%

Republican 39% 8%

Strong Republican 9% 1%


WHY THEY DECIDED TO RUN

Although we only surveyed people who won their election to their school board, and we are missing people who are discouraged from running for office or lose their election, we can still learn about factors that might have affected the school board trustees‟ decisions to run for office. In our survey we asked board members to rate items in terms of their individual importance for why they ran for office, with a “1” indicating “not at all important” and “5” indicating “very important.” The summary in the following table indicates the factors trustees rated as either as “important” (rated a 4) or “very important” (rated a 5).

Table 1.4 Percent Indicating Whether Factors Were Important or Very Important in School Board Members’ School Board Seat

*Indicates that the differences between Latino and white school board members is significant at 5 percent level.

Unsurprisingly, the factor that was by far most mentioned as important or very important by nearly all school board members was “to improve the achievement level of students,” “to help make districts more responsive to community” and to “set budget priorities.” Inasmuch as there were statistically significant differences between Latino and white respondents in their ranking of factors that affected their decision to run for the school board, they tended to be the following:

To decide what curricula schools use, with Latino respondents citing this factor with 23% more frequency than their white counterparts.

To improve board/administration relations, with Latinos citing this factor with 16% more frequency than white respondents.

To influence the hiring of principals, teachers, and superintendents, with Latinos citing this factor with 22% more frequency.

To improve employment conditions for teachers, with Latinos citing this factor with 15% more frequency.

The results also indicate that relatively few trustees see running for school board trustee as a way to prepare themselves for another political office. Only 11% of the Latino respondents

White Latino

Factors Affecting Decision to Run for School Board

To help improve the achievement level of students 96% 98%

To help make district more responsive to community* 78% 90%

To help set budget priorities 89% 88%

To help improve school facilities 71% 81%

To help decide what curricula schools use* 52% 75%

To help improve board/administration relations* 56% 72%

To influence hiring of principals, teachers, super.* 48% 70%

Improving employment conditions for teachers* 43% 58%

To help prepare yourself for another political office* 1% 11%


indicate such a motivation as being important or very important in their decision to run for the trustee position. However, this ratio is much higher than is the case for the white trustees, with only 1% of them indicating that “preparing yourself for another political office” was an important or very important factor in propelling them to run for the school board. But even if they do not plan to run for other office, once they get into office one in five Latino board members expressed a desire to run for another political office. School boards could serve as a “gateway” to other public offices for Latinos.

FACTORS MAKING IT DIFFICULT TO RUN

What do our school board trustee respondents say about the barriers that they face when choosing to stand for election? We asked the survey respondents to rate a series of “barriers” to running for the school board. Each factor was ranked, with a “1” indicating “not an important factor” and a “5” labeled as a “very important factor.” The following table reports the percent of the school board trustees who indicated that a barrier was either an important (rated a “4”) or very important (rated a “5”) factor. For example, if the kinds of households from which one expects support are registered to vote in lower numbers, or if one‟s likely supporters vote less often than those who are likely to support an opponent, then candidates would finding it more difficult to win office. Additionally, if there are high campaign costs typically associated with one‟s school board elections, then it might be more daunting to run, and campaign spending caps may alleviate the cost for lower income candidates. On the other hand, if candidates need to get the support of different interest groups (e.g., teachers union or the business community) and Latinos had a more difficult time gaining their support, this would make them unelectable. Finally, if one has to compete district-wide rather than in a part of the school district (e.g., the issue of at-large versus district election systems), then campaign costs and coalition-building are all more complex and might pose hurdles for those running for political office.

Table 1.5 indicates stark differences in the rates at which white and Latino respondents assess potential barriers. But before we look as such barriers, we most keep in mind that our respondents are the winners of the election. We did not interview those who decided not to run because of the barriers they experienced or those who lose the election because of institutional barriers they confronted. This is the story from the perspective of winners, and perhaps an under-estimate of barriers confronting candidates for school board. The differences between whites and Latinos are significant for each of the factors listed in the table. Interestingly, the largest gaps between white and Latino responses are with regard to parents being ineligible to vote. Only 7% of the white respondents think of parents‟ voting eligibility as a barrier to their running for school board, while 44% of the Latino respondents believe this is a barrier to their election, a 37-point difference. There is a 31% spread between white and Latino respondents regarding assessment of the lack of parents voting, even when eligible, with 52% of the Latino respondents indicating that this issue is an important or very important barrier to running for


school board, as compared to only 21% of white respondents. Non-citizens are a pool of individuals who could provide a larger source of election support for Latino candidates.14

Table 1.5: Percent Indicating Whether Factors Were Important or Very Important Barriers to Running for School Board

*Indicates that the differences between Latino and Non-Latino School Board members are significant at a 5% level.

Figure 1.2 reports a simulation of results of a set of models we estimated to capture the importance of particular barriers for running for office, while controlling for the characteristics of the district.15 Often barriers are more severe in districts with larger numbers of children in school lunch programs, and districts with higher levels of segregation. As we showed earlier, Latinos tend to concentrate in these districts. But even after controlling for the conditions of the districts, Latinos are significantly more likely to see the limited voting of parents and the impact of people‟s ineligibility to vote. Moreover, filing papers is more of a barrier for Latino board members than for whites and so is the cost of running a school board election. Budget cuts is the highest rates for whites, with 25 percent of them seeing it is a barrier, but it is even a greater barrier for Latinos, as 54 percent of Latinos see this as an important or very important barrier. The impact of budget cuts could be more severe in poor communities, where Latino board members concentrate, making it more difficult for board members to implement needed policy changes and discouraging many to run for office. Non-resident voting, caps on campaign spending, ways to facilitate access to filing election and campaign materials, and changes in the

14 This raises the issue of non-citizen, legal resident voting in school elections. In school board elections, it

is not just a question of taxation without representation, but it is also about the fact that one is obligated to send children to schools and children pose a special case, where many parents are stripped of their capacity to have legitimate voice with officials who are affecting the lives of their children.

15 Results for this model are available at Appendix A on Table A.2.

White Latino

Barriers to Running for School Board

Many parents don‟t vote even when eligible* 21% 52%

Budget cuts* 26% 50%

Many parents are not eligible to register to vote* 7% 44%

Cost of running an election campaign* 25% 41%

Need to have support of other school board members* 10% 34%

Too difficult to take time from work* 17% 34%

Need to have local business support* 7% 28%

Need to have support from various racial communities* 10% 28%

Need to have teacher union support* 9% 27%

Lack of a per diem payment for service on the board* 7% 22%

Having to run campaign at-large, rather than district* 7% 19%

Inconvenient meeting schedules* 3% 15%

Complexity of filing candidate‟s papers* 2% 14%

Difficult to get info about district and policy options* 4% 14%

Lack of language translation resources at meetings* 3% 10%

Gaining support of Republican or Democrat party* 3% 10%


system by which board members are elected (see Figure 1.2) could address some of the barriers faced by Latinos.

Figure 1.2 Probability of a Barrier being Somewhat Important, Important or Very Important

Note: Based on a simulation from the model of survey participants responses regarding barriers to running for office. The simulation controls for school enrollment, whether there was a Latino majority in the district, English Learners, proportion of students in school lunch programs, and levels of concentration of the Latino population as compare to whites.

Overall, Latino trustees rate potential barriers to running for office consistently higher than other board members. If one takes all of the items on our list of barriers to running for school board and computes a simple additive index,16 we find that there is a statistically significant difference between the scores of the Latino and non-Latino trustees. The mean score for non-Hispanic respondents was 27, while it was 38 for Latinos.17

16 The composite measure is simply the rating value (1 through 5) of a particular barrier. Each barrier has

a maximum value of 5 and a minimum value of 1. The composite score, for all trustees, varied from a low of 6 to a high of 77, with a mean value of 31.

17 We ran a standard difference of means test, and the difference was significant at less than .000 level

43%

59%

29%

43%

12.5%

31.1%

14.2%

34.4%32.7%

47.0%

39.1%

54.0%

10.8%

25.5%

5.4%

14.0%

40.0%

58.8%

28.7%

46.3%

white Latino white Latino

no latino majority Latino Majoirty

Budget cuts Parent no eligible to register to vote

Parents don't vote even when they are eligible Complexity in filling candidate paper work

Cost of the election


Chapter 2:

School Board Members as Policy Makers

There is a long tradition in the United States of seeing school board members as objective overseers of the education process, substantially separated from many of the traditional aspects of politics and policy making. This view is grounded in the need of school systems in the U.S. to be insulated from the presumably crass politics of elections, partisan bias, and patronage that characterized many school systems and the politics of cities in the late 19th and early 20th centuries (Wirt and Kirst 2005: 34-38; Tyack 1974).

This view, however, has often led to a misunderstanding of how important school board members have always been, and continue to be, as active public policy decision-makers (Wirt and Kirst 2005: 140-148). For example, one critical policy decision school board members make is whom to hire as superintendent. The superintendent is not only the chief executive officer of the school system, she is the person most responsible for the selection of other school leaders such as deputy superintendents and school principals, with the latter being now viewed as the central figures in school-level improvements in teacher and student performance.

Moreover, school board members make critical decisions regarding the distribution of resources through adopting annual school budgets, curriculum design regarding interpretation of state and national standards, disciplinary policies, negotiation of teacher contracts, and relations with parent teacher associations. School board members are especially critical during times of direct challenge, such as when they were required to address school diversity and segregation or when they are required to change their systems of election as a result of lawsuits. It is school board members who ultimately decide how to initially respond to such issues as well as how to respond once the court has made its decision in cases where lawsuits have been litigated or settled (Fraga, Erlichson, and Lee 1998). The following sections, then, examine perceived and concrete contexts in which school board members work.

THE CHALLENGES IN THE DISTRICTS WHERE LATINOS SERVE

Before we examine policy priorities, we thought it would be important to examine the problems board members report confronting in their district. The decisions board members face are framed by the particular educational and economic environment in their district. Latino school board trustees are more often located in districts with lower graduation rates on average,18 lower family income, higher poverty rates, more English Learners, and greater levels of housing segregation. There are, however, a greater proportion of 12th grade students who graduate UC or CSU eligible in districts with Latino representation and less experienced teachers.

18 12th grade graduation rates exaggerate success, since the question is how many students that entered

high school subsequently graduate. If there are high dropout rates, the 12th grade graduation rate likely exaggerates the “success” of the school.


Table 2.1 School District Characteristics

White Latino

Average graduation rate, grade 12 students 87 73

% of grade 12 students that are UC/CSU qualified 14 17

Average % of graduates that are UC/CSU qualified graduates 16 20

Average number of teachers in the schools in the district 24 27

Average number of Latino teachers in the schools 3.4 8.3

Average total teacher experience 13.1 12.5

% of teachers who have one year or less of experience 14 12

%of teachers who have a BA or less 11 14

% of teachers who are authorized bilingual teachers 2008 11 14

Level of Segregation (Dissimilarity index 2008) 19 22

% of students that are on free and reduced meal programs 50% 70%

Median family income , 2000 $47,000 $39,000

PERCEIVED SEVERITY OF DISTRICT PROBLEMS

Aware about the conditions in the districts, we examined the views of board members about what they see as critical issues confronting their district. We asked trustees in our survey to rate the severity of particular problems in their district as a whole, rather than just a single school. In other words, many districts that are comparatively well off might contain a problem school. So our effort was directed at getting the trustee respondents to characterize the district as a whole.

The following table summarizes the results in terms the percent of respondents indicating that a problem was “sometimes a problem”, “a problem”, or “a very serious problem” in their districts as a whole. Of the 20 items rated, there were a number where white and Latino respondents significantly differed, although white and Latino respondents agreed on the top two problems: insufficient funds, and federal and state mandates. .

Latino school board members are more likely to rate dropout rates, lack of parental involvement, insufficient representation of various groups, substandard buildings and school overcrowding as problems when compared to their white colleagues. However, Latinos are more likely to be elected in district with types of problems, so we need to control for district characteristics to be able to assess whether white and Latino board members would see different issues as a problem when we control for district characteristics.


Table 2.2 Rating the Severity of Problems Throughout the School District (Percent Indicating a Sometimes a Problem, a Problem, or a Very Serious Problem)

*Indicates the differences between Latino and Non-Latino School Board

Figure 2.1 shows simulations of a set of models that explore board members‟ perceptions about the problems in their district, while controlling for the conditions of the district.19 In general, board members see more issues as important in poorer districts (for example, board members see community conflicts, community conditions, gang prevention, student absenteeism, dropout rates, parent‟s involvement and teacher salaries as more of a problem in poorer districts). Student language issues are more of a problem in schools with large enrollments and many English Learners, but even after controlling for district characteristics, white and Latino board members rate different issues as problems in their district. More whites see federal/state government mandates and insufficient funds as serious problems, while Latinos are significantly more likely than whites to see representational issues as a problem in their district. While only 9 percent of white board members see a lack of representation of groups in the board as a problem or a serious problem, 29 percent of Latinos feel that is a problem in their district. Moreover, in districts with more than one Latino on the board, both white and Latino trustees are more likely to see dropout rates of students of color as a serious problem in their district. This could suggest a shift in priorities for board members – both white and Latino –

19 The results are available in Appendix A Table A.3.

White Latino

District Problem

Insufficient funds 48% 45%

Mandates from federal or state governments* 49% 40%

Students not going to college* 24% 39%

Lack of parental involvement* 25% 35%

Conditions in your community (e.g., gangs)* 23% 28%

Problems meeting federal or state standards 26% 28%

Language limitation of students 22% 26%

Dropout rates* 13% 24%

Gang prevention 14% 17%

Lack of representation for various groups* 7% 16%

Substandard buildings* 8% 16%

Student absenteeism 16% 15%

Student discipline 9% 12%

Teacher salaries 13% 12%

School over-crowding* 5% 10%

Class size 6% 7%

Community conflicts 2% 6%

Lack of on-going training for teachers 3% 6%

Teacher quality 4% 5%

Poor classroom materials 2% 4%


when a large number of Latinos are on the board. However, the finding is only significant at a 10 percent level.

Figure 2.1 Simulation of the Probability that Board Members see an Issue as a

Problem or a Very Serious Problem in their District

Note: Probability model for the sample respondents available in Appendix A Table A. 3.

A factor analysis of board members‟ views about problems in their district reveals three distinct categories of problems.20 The first factor is comprised of items that cluster together and suggest district resource issues, including poor classroom materials, lack of on-going teacher training, substandard buildings, teacher quality, and class size. The second factor tends to reflect disadvantaged families and indicators of social crisis (e.g., dropout rates, gang prevention, student absenteeism, and students not going to college). The third category of issues involves federal and state mandates, difficulty in meeting state and federal standards, teacher salaries, and lack of sufficient district funds. We assessed a difference of means test to determine whether Latino and white respondents were significantly different in their pattern of responses regarding these categories of problems.

20 Factor analysis is a standard statistical technique that has as one of its uses the capacity to summarize a

group of variables into a smaller number of “factors.” The technique groups variables in terms of clusters, and when these make substantive sense, it is possible to describe a larger group of variables in terms of a smaller, more parsimonious set of categories.


As could be expected, on the overall scores,21 on the first two factors (district resources and disadvantaged families), Latino respondents were higher, while with the third factor, reflecting concern about intergovernmental relations and mandates, it is the white respondents who scored significantly higher. Although clearly white and Latino trustees share concerns, the latter are noticeably more focused on the problems they face as a result of the conditions in their districts, while the former find the involvement of the state and federal governments in education policy to be more problematic.

SCHOOL BOARD PRIORITIES

Do Latino school board members have different priorities than white school board members? We asked our respondents how high a priority they assign to a series of issues. Each issue was rated from 1 (“not a priority”) to 5 (“very high priority”). The following table summarizes the results of the inquiry, with the proportion of Latino and non-Latino respondents indicating that an issue was either a priority (rated a “4”) or a very high priority (rated a “5”).

Table 2.3 Percent Indicating a Whether a District Responsibility is a Priority or Very High Priority

*Indicates that the difference between Latino and white School Board members is significant at a 5percent level.

21 Each category in the factor analysis can be represented by a factor score, which is computed by

summing the product of the each loading on a factor by the score for the item. So if for a particular item, the score was a 4 and the factor loading for that item on that factor (category) was .781, then the product 3.124. A given case (respondent) then is assessed in terms of the sum of all the products for that category, which then is used as a score for that individual on that category. Factor scores are computed for all the factors (categories) in the analysis (Kim and Mueller, 1978: 60-64)

White Latino

Listing of School District Responsibilities

Reducing the achievement gap 89% 95%

Supporting school and academic outreach program* 62% 81%

Encouraging parental involvement* 79% 91%

Advocating on behalf of underserved students 84% 91%

Improving the quality of teachers in the district 80% 89%

Advocating on behalf of immigrant students* 41% 71%

Growing percentage of minority students bound for college*

60% 83%

Reducing dropout rates* 13% 24%

Supporting community outreach programs* 64% 71%

Cutting costs 75% 76%

Gang prevention* 51% 74%

Improving district‟s physical plant* 52% 60%

Reducing class size* 43% 56%

Increasing teacher salaries 35% 40%

Increasing the number of teachers of color* 23% 40%

Recruiting candidates of color for school board* 14% 37%


There is a certain level of agreement among trustees in our survey, in that the relative ranking within Latino and white trustees is similar. For both Latino and white respondents, reducing the achievement gap between students is their highest priority issue. But there are significant differences between Latino and white trustees with regard to 8 of the 16 items. . In the case of advocating for immigrant students, 71% of the Latino respondents express this to be a priority or very high priority, while only 41% of white trustees report such a rating. Wide differences also exist regarding the issues of “growing percentage of African American/Latino students going to college,” and “recruiting candidates of color for the school board,” with Latino respondents far likelier to rate these issues as a priority or very high priority. Other differences of greater than 10% exist with regard to supporting community outreach programs, increasing the numbers of teachers of color, and reducing dropout rates.

Figure 2.2 Probability that Board Members see a Issue Some of a Priority, a Priority or

Very High Priority

Note: Probability model for the sample respondents available in Appendix A Table A.4.

Once we control for the conditions in the district in Figure 2.2, many of the differences in priority disappear, except for the recruitment of Asian/Latino/black board members and teachers. Latinos are significantly more likely to see this as a priority. Moreover, the importance of increasing the college going rate of Black and Latino students increases with Latinos in the board, although only at a 10% level of significance. Similar to the perception of board members about dropout rates of Latino and African American students as a serious problem, it is likely that with more Latinos in the board, the college going rate and dropout

32.5%

52.3%

28.7%

47.8%

41.7%

67.4%

37.8%

63.6%

white Latino white Latino

no latino majority Latino Majoirty

recruit enough asia/latino/black for the board

recruit enough asia/latino/black teachers


rates of student of color become more of a priority with more Latino representation. But again, a larger sample is needed to confirm these results.

ARE THERE DIFFERENCES IN POLICY BETWEEN LATINO AND WHITE BOARD MEMBERS?

So far we have seen significant differences in problems and priorities between Latino and white board members. Are there differences in policy? We use the results of a set of regressions to examine the impact of Latino representation on the districts‟ number of Latino teachers, administrators and student service personnel22; the experience and educational attainment of the teachers in the district; bilingual training of teachers in the district; and student-teacher ratios. We control for school demographics (total enrollment in the district and whether Latinos are the majority of the students in the district), proportion of students in school lunch programs, and the board members‟ ethnicity.23

Figure 2.3 Simulation of the Correlation between Latino Representation and Policy

Outcomes

Note: The results from this analysis are available in table A5 in Appendix A.

Figure 2.3 reports the results of a simulation of the percent of teachers, administrators and student services employees (for example, counselors, librarians and nurses) that are Latino among the districts in our survey. There are significantly more Latino teachers, administrators and student service providers in districts with Latino trustees.24 After controlling for other factors, on average, one in four schoolteachers, student service personnel, and administrators are Latinos in districts with Latinos on the board and a majority Latino student body. In

22 These are for example, counselors, librarians, or nurses. 23 The results from the analysis are available in table A5 in Appendix A. 24 Meier and Stewart 1991; Leal, Meier, and Martinez-Ebers 2004; Meier and Juenke 2005 encounter

similar results.

15.4%

8.9%

29.2%

22.7%

17.2%

9.8%

31.9%

24.4%

16.0%

8.7%

33.6%

26.3%

Latino on Board No Latino in Board Latino on Board No Latino on Board

No Latino Majority Latino Majority

latino teachers

latino administrators

student services latino


districts where Latinos are not the majority of students, the proportion of Latino staff members and administrators is about two times higher in those districts with Latino representation. Other factors affecting the recruitment of Latino teachers and administrators are the proportions of Latino students in the districts and the proportions of students in school lunch programs. There were more Latino teachers and administrators in districts with more poor and Latino children.

Affinity among board members and the ability to engender more influence in the board may require that there is more than one Latino in the board before Latino trustees are truly able to influence policy. A lone Latino school board member might be under more pressure “to go along to get along.” We examine the impact of having more than one Latino in the board on the same policy decisions (see Figure 2.4). There are, on average, more Latino teachers, administrators and in-student service providers where the boards have more than one Latino. In addition, there are modestly but significantly higher proportions of teachers authorized to teach bilingual education and less inexperienced teachers or teachers with a BA or less in districts with more than one Latino on the board, even after controlling for district conditions. Moreover, there is a significant but small difference in dropout rates correlated with the number of Latinos on the board. Again, having more than one Latino on the board may influence the decisions of all board members around particular issues. There is a slightly lower dropout rate between 9th and 12th grade in districts with more than one Latino in the board.

Figure 2.4 Simulation of the Correlation between Having More than One Latinos in the Board

and Policy Outcomes in Districts Where Latinos are the Majority of Students

Note: Simulation from the results of regression analysis. Results available in Appendix A.5 and A.6.

2.0%3.1%

2.0%3.1%

30.5%

16.6%

31.9%

14.1%

35.2%

22.0%

6.3%

11.8%

15.1%

10.9%

More than 1 Latino Not More than 1 Latino

teachers one year or less of experience drop out rates between 9 and 12th gradelatino teachers latino administratorsstudent services latino teachers BA or less% teachers authorized bilingual teachers


Conclusion

We focus on California‟s lack of Latino representation on school boards and (a) the role of at-large elections, and other barriers to running for office (such as the cost of the election and the need to gain interest group support in depressing the number of Latinos winning school board elections) and (b) the effects of such underrepresentation on perceptions, priorities, and policy choices.

The results provide modest support for the concern that at-large elections have a negative influence on the number of Latino school board members. We found strong support of the correlation between Latino representation on the school board and district hiring practices. Moreover, Latino trustees in this project do seem to have somewhat different agendas than do their white counterparts. It is also plausible that there are dynamic relationships among the system of election and the propensity of eligible Latinos to register and vote. Moving to district-level elections might increase the probability of Latinos electing a Latino trustee, and with the increased prospect of having desirable election outcomes, there might be increased motivation to register and vote.

However, as reported by our school board trustee respondents, at-large election systems are perhaps only one of a variety of more important barriers – such as low election turnout, low voter registration levels, and the cost of election campaigns – are even more critical barriers to representation.

POLICY RECOMMENDATIONS

Election system analysis There should be systematic analysis of school districts with high proportions of Latino residents and no Latino board members to determine whether at-large elections are in those particular circumstances an impediment to electing Latino school board members. If that election system is found to depress Latino representation, then a variety of ways to seek alternative systems should be explored.

District voter registration and turnout Particular efforts need to be directed towards increasing registration of eligible district voters and elevating voter turnout. Any increases in registration and turnout are disproportionately likely to be among Latino residents.

Campaign aids and spending caps Campaign spending caps could be an alternative to the higher burden election costs puts on Latino candidates. More information and training might help with running for school boards, as well as more information about the district and policy options.

Non-citizen voting Many of the Latino residents who are not registered to vote are ineligible because they are not as yet citizens, even while they might be legal permanent residents. Even though they cannot


vote, they pay taxes and are obliged to have their children schooled. If permitted to vote in at least local school elections, non-citizens may also become more engaged with their schools and teachers and may increase Latino representation in school boards.

Although voting by non-citizens was common in the United States before 1926, it is rare now, but still permitted in a number of places – e.g., Chicago, Cambridge and Amherst (Massachusetts), and Takoma Park (Maryland). There should again be consideration of extending voting rights for legal permanent residents. Although this has been considered a number of times in California localities, including Los Angeles and San Francisco, these efforts failed. While controversial, there is a strong argument to be made for having non-citizens vote in at least school board elections (Hayduk, 2003). By permitting non-citizen voting, Latino school board representation might increase.


References

Abney, F. Glenn and John D. Hutcheson, Jr. 1981. “Race, Representation, and Trust: Changes in Attitudes after the Election of a Black Mayor, “The Public Opinion Quarterly, 45, (Spring), pp. 91-101. Arrona, John and Raquel Donoso. 2006. Beyond the Classroom: An Analysis of California’s Public School Governance. San Francisco, California: Latino Issues Forum. Barreto, Matt. 2007. “Si Se Puede! Latino Candidates and the Mobilization of Latino Voters,” American Political Science Review, 101 (August): 425-441. Betts, Julian, Kim. S. Rueben, and Anne Danenberg. 2000. Equal Resources, equal outcomes? The Distribution of School Resources and Achievement in California. San Francisco: Public Policy Institute of California. Bezdek, RR, DM Billeaux, and JC Huerta, 2000. “Latinos, At-large Elections, and Political change: Evidence from the „Transition Zone‟," Social Science Quarterly 81 (1): 207-225. Bowler, Shaun, Todd Donovan, and David Brockington. 2003. Electoral Reform Minority Representation: Local Experiments and Alternative Elections. Columbus, Ohio: Ohio State University Press. Brewer, Dominic J. and Joanna Smith, 2006. Evaluating the “Crazy Quilt”: Educational Governance in California, Center on Educational Governance, Rossier School of Education, University of Southern California. Browning, Rufus P., Rogers Dale Marshall, and David H. Tabb. 1986. Protest Is Not Enough. Berkeley: University of California Press. California Department of Finance, "Economic and Demographic Statistics," 2002, available at www.dof.ca.gov. [Do not include the http://.] Canon, David T. “Electoral Systems and the Representation of Minority Interests in Legislatures,” 1999. Legislative Studies Quarterly, 24, (August): pp. 331-385. Danenberg, Anne, Christopher Jepsen, and Pedro Cerdán . 2002. Student and School Indicators for Youth in California's Central Valley. San Francisco, California: Public Policy Institute of California. Davidson, Chandler and George Korbel. 1981. “At-Large Elections and Minority-Group Representation: A Re-examination of Historical and Contemporary Evidence.” Journal of Politics 43: 982-1005. Davidson, Chandler and Barnard Grofman, eds. 1994. Quiet Revolution in the South: The Impact of the Voting Rights Act, 1965-1990. Princeton, NJ: Princeton University Press.

http://www.ppic.org/main/bio.asp?i=131

http://www.ppic.org/main/publication.asp?i=166

http://www.ppic.org/main/publication.asp?i=166


Engstrom, Richard L, and Michael McDonald. 1986. “The Effect of At-Large versus District Elections on Racial Representation in U.S. Municipalities,” in Electoral Laws and Their Political Consequences, Bernard Grofman and Arend Lijphart, eds. New York: Agathon Press. Engstrom, Richard L. and Michael D. McDonald. 1981. “The Election of Blacks to City Councils: Clarifying the Impact of Electoral Arrangements on the Seats/Population Relationship,” The American Political Science Review, 75 (June): pp. 344-354 Engstrom, Richard L, and Michael McDonald. 1982. “The Underrepresentation of Blacks on City Councils,” Journal of Politics, 44 (November): pp. 1088-1105. Engstrom, Richard and Michael McDonald. 1997. “The Election of Blacks to Southern City Councils: The Dominant Impact of Electoral Arrangements.” In Blacks in Southern Politics, Laurence Moreland, Robert Steed, and Tod Baker, eds. New York: Preager Publishers. Fraga, Luis Ricardo and Roy Elis, “Interests and Representation: Ethnic Advocacy on California School Boards,” Teachers College Record, vol. 111, No. 3 (March, 2009), pp. 659-682. Fraga, Luis Ricardo, Kenneth J. Meier, and Robert E. England. 1986. “Hispanic Americans and Educational Policy: Limits to Equal Access.” Journal of Politics 48 (November): 850-876. Hayduk, Ronald. 2003. “Noncitizen Voting: Expanding the Franchise in the U.S.” Prepared for delivery at the Annual Meeting of the American Political Science Association, Philadelphia, Pennsylvania, August 28-31, 2003. Hero, Rodney. Latinos and the Political System: A Two-Tiered Pluralism. 1992. Philadelphia: Temple University Press. Karnig, Albert K. 1980. Black Representation and Urban Policy. Chicago: University of Chicago Press. Karnig, Albert K. and Susan Welch. 1982. “Electoral Structure and Black Representation on City Councils.” Social Science Quarterly 63: 99-114. Kim, Jae-On and Charles W. Mueller, Factor Analysis: Statistical Methods and Practical Issues. Beverly Hills and London: Sage Publications, 1978. Leal, David L., Kenneth J. Meier, and Valerie Martinez-Ebers. 2004. “The Politics of Latino Education: The Biases of At-Large Elections.” Journal of Politics 66 (November): 1224-1244. McDonald, Michael D., and Richard L. Engstrom. 1992. "Minority Representation and City Council Electoral Systems: A Black and Hispanic Comparison." In Ethnic and Racial Minorities in Advanced Industrial Democracies, Anthony M. Messina, Luis R. Fraga, Laurie A. Rhodebeck, and Frederick D. Wright, eds., New York: Greenwood Press. Meier, Kenneth J., and Robert E. England. 1984. “Black Representation and Education Policy: Are They Related?” American Political Science Review 78 (June): 392-403.

http://www.jstor.org/search/BasicResults?Search=Search&Query=aa:%22Richard%20L.%20Engstrom;%20Michael%20D.%20McDonald%22&hp=25&si=1

http://www.jstor.org/view/00030554/di961001/96p0300x/0?currentResult=00030554%2bdi961001%2b96p0300x%2b0%2c00&searchUrl=http%3A%2F%2Fwww.jstor.org%2Fsearch%2FAdvancedResults%3Fhp%3D25%26si%3D1%26All%3Drichard%2Bengstrom%26Exact%3D%26One%3D%26None%3D%26au%3Don%26sd%3D%26ed%3D%26jt%3D%26dc%3DPolitical+Science



http://www.jstor.org/browse/00030554


Meier, Kenneth J., Joseph Stewart Jr., and Robert E. England. 1989. Race, Class, and Education: The Politics of Second Generation Discrimination. Madison, WI: University of Wisconsin Press. Meier, Kenneth J., and Joseph Stewart. 1991. The Politics of Hispanic Education. Albany: State University of New York Press. Meier, Kenneth J., and Eric Gonzalez Juenke. 2005. “Electoral Structure and the Quality of Representation on School Boards.” In William Howell (ed.), Besieged: School Boards and the Future of Education Politics. Washington, DC: Brookings Institution Press, pp. 199-227. Pitkin, Hanna. 1967. The Concept of Representation. Los Angeles: University of California Press. Polinard, J. L., Robert D. Wrinkle and Tomas Longoria. 1990. “Education and Governance: Representational Links to Second Generation Discrimination.” Western Political Quarterly: 631-646. Polinard, J. L. 1991. “The Impact of District Elections on the Mexcian-American Community – The Electoral Perspective,” Social Science Quarterly, 72 (no. 3): 608-614. Reyes, Belinda and Max Neiman, 2008. “System of Elections, Latino Representation, and School Policy in Central California Schools,” Working Paper, University of California, California Policy Research Center. Romo, Harriett, and Toni Falbo, Latino High School Graduation: Defying the Odds, University of Texas Press, Austin, Texas, 1996. Taebel, Delbert. 1978. “Minority Representation on City Councils.” Social Science Quarterly, 59: 142-152 Tyack, David. 1974. The One Best system: A History of American Urban Education. Cambridge: Harvard University Press. Warren, Robert, "Estimates of Undocumented Aliens in the 1980 Census," Demography, Vol. 24, No. 3, 1987, pp. 375-393.


About the Authors

Max Neiman is a Senior Resident Scholar at the Institute of Government Studies, University of California, Berkeley and Professor Emeritus, Political Science, at the University of California, Riverside. He was a senior fellow and an associate director of research at the Public Policy Institute of California, responsible for research in governance-related topics. His research interests include government performance, urban growth and development, disputes over managing growth, the politics of race, ethnicity and immigration, and political models of policymaking. He has received awards from the National Science Foundation, The Haynes Foundation, the California Endowment, and the Department of Housing and Urban Development. He is the author and editor of five books and numerous articles and conference papers, many of which are published in a variety of scholarly journals. He is a co-author (with Paul Lewis) of the forthcoming book, Custodians of Place. He holds a Ph. D. in political science from the University of Wisconsin, Milwaukee.

Belinda I. Reyes is the Director of the César E. Chávez Institute and Associate Professor in Latina/o Studies at the College of Ethnic Studies, San Francisco State University. She is an expert in demography, immigration policy, immigrant adaptation, race and ethnicity, urban economics, and social and economic progress of race/ethnic minorities. She was a research fellow at PPIC (1995-2004) and an assistant professor at UC Merced (2004-2007). She holds a PhD in Economics from the UC Berkeley. Her work has focused on immigration and migration, the economic and social progress of racial and ethnic groups, and issues of diversity in education. She has experience in many statistical and data management methods.

Luis Fraga is one of the nation‟s leading scholars in the study of school district representation and local education policy, and of local and ethnic/racial politics. He has extensive experience in designing and implementing surveys of local government and school boards. He is currently Associate Vice Provost for Faculty Advancement at the University of Washington and Director of the Diversity Research Institute, as well as the Russell F. Stark University Professor in the Political Science Department. His primary interests are in American politics where he specializes in urban politics, education politics, voting rights policy, immigration policy, and the politics of race and ethnicity. He has two recent books: United States Government: Principles in Practice (Holt McDougal 2010) and the co-authored Latino Lives in America: Making It Home (Temple University Press 2010). He has also published the co-authored book Multiethnic Moments: The Politics of Urban Education Reform (Temple University Press 2006). He was a member of the APSA standing committee on Civic Engagement and Education that co-authored Democracy at Risk: How Political Choices Undermine Citizen Participation, and What We Can Do About It (Brookings Institution Press 2005). He is also co-editor of Ethnic and Racial Minorities in Advanced Industrial Democracies (Greenwood 1992). He has received numerous awards and grants, including from such sources as the National Science Foundation and the Ford Foundation. He received his A.B., cum laude, from Harvard University (1978) and his MA (1981) and PhD (1984) from Rice University.


Daniel Krimm is a research associate at the Public Policy Institute of California. Before joining PPIC, he was a Global Policy Fellow at IP Justice, addressing issues in information and communication technology and Internet governance. He holds an M.P.P. degree from the University of Southern California and a B.A. from Princeton University in the History and Philosophy of Science.


Appendix A: Regression Models

Table A.1 Regression Results of Latino Representation in School Boardi

One Latino on

the Board

More than one

Latino in the Board

Number of Latinos on

the Board

Intercept -298***

(.451)

-4.59***

(.451)

-.494***

(.134)

Total Enrollment 3.8E-6

(9.6e-6)

2.1E-7

(4.4e-6)

2E-6

(.1.5e-6)

50% Latino 1.39***

(,277)

1.8***

(,35)

.611***

(.1329)

Number of English Learners in

Average School

.006***

(.00143)

.0048***

(.00126)

.0039***

(.00053)

% in Meal Programs 2.18***

(.538)

2.12***

(.697)

1.103***

(.2558)

Level of Segregation .0099

(.812)

.172

(.78)

-.077

(.3141)

District Election System .44

(.37)

.75*

(.40)

.2914*

(.167)

LA .938***

(.34)

1.149***

(.351)

.677***

(.141)

R-square 42.88***

Likelihood Ratio Test 2967*** 247***

Wald Test 183*** 140***

Observation 644 644 542

Note: Table presents coefficient estimates and in parenthesis are standard errors. *** significant at a 1 percent level ** significant at a 5 percent level * significant at a 10 percent level.

Note 1: These models estimate the probability of having one or more Latinos in the board using Logistic equations and we model the number of Latinos in the board using simple OLS equations.


Table A.2 Regression Results of What Board Members Consider “Somewhat Important Barrier,” “Important Barrier,” and “Very Important Barrier” to Running for School Board

Budget

Cuts

Parents not

Eligible to

Vote

Parents

not

Voting

Complexity

Filling candidate

papers

Hard Time

Getting Dist.

& Policy Info.

Cost to

Run An

Election

Intercept -.395

(.423)

-2.79*** (.59) -1.27

(.461)

-1.7***

(.63)

-1.86***

(.655)

-.51

(.47)

Total

Enrollment

-5.96E-6

(1.1E-5)

-2E-5

(1.4E-5)

-4.1E-6

(1.1E-5)

2E-5

(1.8e-5)

2.6E-5** (1.3E-

5)

5.5e-5***

(1.9e-5)

50% Latino -.6406*

(.333)

.148

(.39)

.2789

(.334)

-.744

(.489)

-.247

(.463)

-.5054

(.3681)

District Total

English

Learners

.001

(.0012)

.0017 (.0014) 9.2e-4

(.00128)

.0011

(.002)

-.0014

(.0016)

.0021

(,785)

% of Sudents

in Meal

Programs

1.04

(.685)

2.22**

(.901)

1.298*

(.72)

.404

(1.03)

1.82*

(1.02)

-.6698

(.785)

Level of

Segregation

.7099

(.8221)

4.03*** (.986) 2.037**

(.847)

2.14*

(1.15)

-.099

(1.86)

2.65**

(.92)

White Board -.633**

(.248)

-1.15***

(.289)

-.601**

(.25)

-1.04**

(.367)

-1.725***

(.3767)

-.076**

(.275)

District

Elections

.044

(.398)

-.0099 (.012) -.007

(.0097)

-.0085

(.0118)

.0245

(.072)

-.004

(.009)

Likelihood

Ratio Test

19*** 82*** 43*** 14** 44*** 70***

Wald Test 14** 61*** 38*** 13* 34*** 48***

Obs 355 355 355 355 355 355


Note 1: These are logit equations of the probability that a board member would consider the particular issue an important or very important barrier for running for office.


Table A.3 Regression Results of What Board Member Consider to be “A Problem” or a “Very Serious Problem” for them in the Districts

Language

Limitation

of Students

Community

Groups

Representation

on Board

Insufficient

Funds

Community

Conditions

Federal /

State

Mandates

Dropout

Rates

Intercept -3.05***

(.601)

-1.86***

(.683)

-.092**

(.43)

-4.39***

(.672)

-1.46***

(.453)

-3.9***

(.736)

Total

Enrollment

-7E-5**

(2.9E-5)

2.6E-5

(2.2E-5)

1.3E-5

(2E-5)

2.1E-5

(2.2(e-5)

2.1E-5

(1.9E-5)

6.4E-5***

(2.3E-5)

50% Latino .3199

(.406)

-.984*

(.522)

-.148

(.314)

.465

(.381)

.0108

(.321)

.0319

(.442)

District Total

English

Learners

3.7E-4***

(.00011)

-.2E-5

(7.7E-5)

4.5E-5

(6.6E-5)

4E-5

(7.1E-5)

2.6E-5

(6.2E-5)

-7E-5

(6.7E-5)

% of Students

in Meal

Programs

1.15

(.858)

1.56

(1.1)

.2987

(.6588)

3.5***

(.8995)

.729

(.681)

2.225**

(1.013)

White Board .8175**

(.342)

-1.35***

(.43)

.569**

(.267)

.692**

(.3299)

.8788***

(.278)

.123

(.371)

More than one

Latino in Board

.5202

(.3391)

-.838*

(.458)

.236

(.278)

.0533

(.3241)

-.378

(.278)

.625*

(.371)

Likelihood

Ratio Test

47*** 20.5*** 11.5* 48.2*** 23.9*** 33.9***

Wald Test 33.7*** 19.7*** 9.97 45.6*** 21.1*** 25.8***

Obs 355 355 355 355 355 355


Note 1: These are logit equations of the probability that a board member would consider the particular issue a problem or a very serious problem.


Table A.4 Regression Results of What Board Member Consider to be “Somewhat a Priority”, “A Priority” and a “Very Important Priority” for them in the Districts

Recruitment of

Asian/Latino/Black

Board Members

Recruitment of

Asian/Latino/Black

Teachers

College-Going Rate of

Latino and Black Children

Intercept -.124

(.438)

.521

(.4415)

-.354

(.498)

Total Enrollment 1.4E-5

(1.1E-5)

3.9E-5***

(1.7E-5)

2.6E-5

(2E-5)

50% Latino -.18 (.336) -.164

(.342)

.111

(.4228)

District Total English

Learners

-.9.8E-4

(.00125)

-.00031

(.00128)

.0011

(.00202)

% of Students in Meal

Programs

.435

(.694)

-.4042

(.7084)

1.463*

(.788)

Level of Segregation .152

(.839)

.8502

(.8714)

-.1376

(1.1)

White Board -.082***

(.253)

-1.057***

(.2603)

.0909

(.331)

Number of Latinos in the

Board

.2991*

(.157)

Likelihood Ratio Test 15** 37*** 37***

Wald Test 14** 29*** 20***

Obs 355 355 355


Note 1: These are logit equations of the probability that a board member would consider the particular issue a priority.


Table A.5 Regression Results of Policy Choices across Districts in our Survey

Note: Table presents coefficient estimates and in parenthesis are standard errors. *** significant at a 1 percent level; ** significant at a 5 percent level; * significant at a 10 percent level.

Note 1: These are OLS equations of the percent of Latino teachers, administrators and pupil service providers in our districts.

% teacher

Latino % Latino Admin.

% Latino Pupil Services

Intercept -.0097 (.028)

-.017 (.036)

.02 (.048)

Total Enrollment 3.1E-7 (5.1E-7)

1.1E-6 (6.8E-7) -9.9E-7 (8.4E-7)

50% Latino .138*** (.018)

.1467*** (.023)

.176*** (.029)

% of Students in Meal Programs

.171*** (.038)

.2015*** (.05)

.1135* (.065)

White in Board -.004 (.02)

-.027 (,027)

.017 (.036)

Latino in Board .062*** (.021)

.047* (.028)

.09** (.037)

F Test 64.1*** 46.9*** 25.47***

R-square 47.6 39.96 29.46

Obs 358 358 311


Table A.6 Regression Results of Policy Choices across Districts in our Survey

% teachers

BA or less

% authorized bilingual teacher

% teacher Latino

% Latino Admin.

% Latino Pupil

Services

% teachers 1 yr or less experience

Intercept .066*** (.02)

.033 (.029)

-.015 (.022)

-.045 (.029)

.036 (.041)

.118*** (.019)

Total Enrollment 1.4E-6*** (4.2E-7)

1.32E-6 (1.2E-6)

-.85E-8 (4.6E-7)

6.5E-7 (6.1E-7)

1.4E-6 (8.1 E-7)

1.1E-5 (9.2E-6)

50% Latino .027* (.015)

.026 (.021)

.103*** (.016)

.099*** (.021)

.145*** (.029)

-.009 (.012)

District Total Spanish English Learners

-1.8E-6 (4.1E-6)

% of Students in Meal Programs

.044 (.032)

.069 (.045)

.158*** (.035)

.183*** (.045)

.097 (.064)

.078*** (.027)

White Board .004 (.013)

.0095 (.018)

.0008 (.014)

19.2E-5 (.018)

-.0003 (.026)

-.016 (.01)

More than one Latino in Board

-.012 (.013)

.042** (.019)

.138*** (.015)

.178*** (.019)

.132*** (.026)

-.041*** (.011)

F Test 4.84*** 4.43*** 94.01*** 74.2*** 30.7*** 5.26***

R-square 6.4 7.1 57.2 51.3 33.5 6.95

Obs 358 355 358 358 311 358


Note1: These are OLS equations of the percent of Latino teachers, administrators and pupil service providers in our districts, the percent of teachers with a BA or less, authorizations to teach bilingual education and their experience using the sample of districts in our survey.


Appendix B: Survey of School Board Members 2009

OVERVIEW

The key objective of the School Board Survey was to assess the barriers, experiences, and priorities of school board members in California. The survey targeted a sample of current (and recent former) trustees in school districts across the state of California at a sample of districts with more than 10 percent Latino students in elementary, high school, and unified districts in California. PPIC conducted the survey online over 17 weeks from June 24 to October 23, 2009. Online survey services were provided by SurveyMonkey.com. School board members were asked about issues facing their district, their priorities as board members, and their views on policies particularly as they relate to Latino students. The complete survey questionnaire is included at the end of this appendix.

SURVEY METHODOLOGY

Survey Design

The questionnaire was designed to elicit details of school board member characteristics and experiences and to probe the factors that might have an impact on district policy decisions. The survey was moderately lengthy compared to other similar surveys, but respondents could typically complete it in about 20 minutes. The survey began with a few questions about the respondent‟s demographic characteristics and the electoral process used in the respondent‟s school district. It continued with several questions about policy attitudes and judgments with respect to school district actions and issues, ending with a few questions about the structure of school board participation.

Pre-targeting and Sampling

A list of school districts as of the 2007-08 school year was acquired from the California Department of Education, and of the 969 elementary, high school and unified districts included in that list, 767 districts were identified that had at least 10% Latino student enrollment (using data from the California Basic Educational Data System (CBEDS) as our domain of inquiry. These districts contained 5,814,080 out of the total 6,163,434 students enrolled in all districts, or about 94.3% of total K-12 student enrollment in California.25

In order to identify trustees in each of the selected districts, we started with a list of Latino school board members and contact information provided by the National Association of Latino Elected and Appointed Officials (NALEO) as of January 1, 2008. We merged this list with the school district list according to school district name, as a starting point, and we flagged these names as Latino individuals. As a separate process, we researched each district manually to identify current board members and contact information, starting with the districts‟ public web

25 We were missing enrollment data for two districts, so these figures represent 765 districts with at least

10% Latino enrollment, out of 967 districts in total.


sites. These web sites provided trustee names for all but 102 of our selected districts, including matches for the Latino trustees identified from the NALEO list. The remaining districts required individual phone calls to acquire trustee names, and we ultimately failed to reach 9 of the smallest districts.

In order to confine the survey process to a manageable scale, we set a general target of roughly 1500 trustees to sample, out of roughly 5000 or so total school board members in the state. From the NALEO list, we had identified 544 Latino trustees, so in order to assure a robust response from Latino trustees we included all of them in our survey sample. Among the names we acquired manually from the districts, there were 75 additional individuals with Latino surnames, and for sampling purposes, we treated them as Latino as well, for a total initial estimate of 619 Latino trustees. (One Latino trustee passed away during the survey period without response, and was removed from the sample, yielding a total of 618 Latino trustees.) Our survey included a question about the respondent‟s ethnic group, so respondents also had an opportunity to provide their ethnic identification to us directly.

For the rest of the sample, we selected one non-Latino trustee from each district in our sample that had any non-Latino trustees (this includes most districts, but a small number had all-Latino school boards). We randomized the selection of these names with an algorithm that ensured order-independence of the names as they appeared in the data set and ultimately yielded a total of 734 non-Latino trustees in the sample, for a total of 1352 trustees in the full sample. Due to time lags in acquiring names, at least 98 of these trustees had left office by the time we initiated the survey. Our expectation was that recent former trustees could answer the particular questions we were asking as effectively as current trustees, so due to the randomization process we opted to retain them in the sample.

Because of the various methods of acquisition of contact information, and varying amounts of information provided on school districts‟ public web sites, we ended up with various types of contact information for different trustees. We use board member emails as the first contact with board members, but we also emailed, mailed, phoned the board members in our sample, their districts and district‟s superintendents to request participation in the survey. The active contact method for some trustees changed over the course of the survey period, as we send additional request for participation.

Survey period/method/sequence

The survey effort began June 24, 2009 with a bulk email letter to the email contacts inviting them to participate in the online survey. Several of these message generated “bounce” replies indicating that the messages had not been received. In such cases, efforts were made to identify a working email address or an alternative contact to the trustee. In many of these cases working email addresses or other new contact information were obtained through phone calls or emails to informational contacts at the districts. In the cases of district and superintendent contacts, we asked them to forward the email message to the trustee for us, in lieu of having direct contact information ourselves. In 26 cases, district representatives informed us that trustees did not have access to a computer or the Internet. In those cases, we emailed a printable copy of the actual questionnaire itself to the district as a PDF file, to print and forward to the trustee by


mail. Five of those paper surveys were completed and returned, for a response rate of 19.2% for PDF-delivered surveys.

Reminders were sent on July 24 and August 11, and finally on September 18 with the latter communication including a signed endorsement on letterhead from State Senator Gloria Romero, Chair of the California Senate Education Committee.

Table B.1 Response rates for the various contact categories

Contact Method Response Rate Number

Trustee Email 24.7% 129 / 523 District Email 25.0% 30 / 120 Superintendent Email 25.1% 169 / 673 Trustee Mail 0.0% 0 / 11 District Mail 16.0% 4 / 25

Total 24.6% 332 / 1352

The response rate for Latino board members was 20.1% (124 of 618) and for non-Latinos it was 28.3% (208 of 734), according to our pre-identification of Latino ethnicity.

c o l l e g e o f e t h n i c s t u d i e s • s . f . s t a t e u n i v e r s i t y1 6 0 0 h o l l o w a y a v e n u e • e p 1 0 3

s a n f r a n c i s c o c a 9 4 1 3 2c c i @ s f s u . e d u • c c i . s f s u . e d u

c é s a r e . c h á v e z i n s t i t u t e

César E. Chávez InstituteCollege of Ethnic StudiesSan Francisco State University1600 Holloway Ave. EP103San Francisco CA 94132

[email protected] / cci.sfsu.edu/schoolboards

examining latino representation on california’s school boards latino...survey of school board...

Documents