Interethnic Competition for Classroom Teacher,Assistant Principal,Principal,and Administrator Jobs in Multiethnic U.S. School Districts,2002-2008

Abstract

Previous research on interethnic competition has not examined public school jobs across the United States. We use U.S. Equal Employment Opportunity Commission data to answer two research questions. What are the levels of under- and over-representation in various public school jobs among African Americans, Latinos, and White non-Latinos? And, is there interethnic competition for these positions? Answers to these questions have implications for resource distribution and the future of representational equity. Based on analysis of 267 multiethnic U.S. school districts for 2002 and 2008, we find that Latinos are underrepresented in all four job categories and that non-Latino Whites are overrepresented or achieve near-parity in all four categories. On average, Blacks are overrepresented among assistant principals and principals but underrepresented among classroom teachers and administrators. Competition between minorities and non-minorities is more prevalent than competition between Blacks and Latinos. Among districts in which we find competition between Blacks and Latinos, competition does not heavily favor one group over the other for assistant principal, principal, and administrator positions; however, among classroom teachers, districts with Latino gains and Black losses outnumber those with Black gains and Latino losses by greater than 2 to 1. In low Latino growth districts, Latinos outpace Blacks in competition for teacher positions, but in high Latino growth districts, Blacks outpace Latinos for administrator positions.

Keywords

interethnic competition public school employment education policy race and ethnicity

Introduction

The U.S. Latino population will be approximately twice the size of the U.S. African American population by 2050 (Telles, Rivera-Salgado, Sawyer, & Zamora, 2011). As regions within the United States become more racially and ethnically diverse, interest in interethnic competition for political and economic resources has become widespread. Political scientists, public administration scholars, sociologists, and economists have contributed to the competition literature through investigations of conflict and cooperation between Hispanics and African Americans across a number of policy areas (Bean, Bachmeier, Brown, & Tafoya-Estrada, 2011; Falcon, 1988; Hall & Saltzstein, 1975; Hero, & Preuhs, 2013; Kerr, Miller, & Reid, 2000; MacManus & Cassel, 1982; McClain 1993; McClain & Karnig, 1990; Meier, McClain, Wrinkle, & Polinard, 2004; Meier & Stewart, 1991a; Morin, Sanchez, & Barreto, 2011; M. Oliver & Johnson, 1984; Rocha, 2007; Rodrigues & Segura, 2007; Rosenbloom, 1973; Sawyer, 2005; Telles et al., 2011). Themes covered in the broader literature on interethnic relations include changes in the labor market, the character of electoral coalitions/divisions, attitudes about politics and policy, resource distribution, and conflict/cooperation in different policy areas, primarily in urban settings.

Interethnic competition is typically defined as a situation in which two or more groups attempt to obtain scarce (or finite) resources—and where the success of one group reduces significantly the probability that other group(s) will achieve their goals (Blalock, 1967; McClain, 1993; McClain & Karnig, 1990). Some studies focus on our primary area of interest, interethnic competition for employment and public sector jobs (Hall & Saltzstein, 1975; Kerr et al., 2000; McClain, 1993; McClain & Karnig, 1990; Meier et al., 2004; Rocha, 2007; Rosenbloom, 1973)—and most studies of interethnic employment competition focus on levels of U.S. government that are the most diverse, U.S. municipalities (Hall & Saltzstein, 1975; McClain, 1993; McClain & Karnig, 1990; Kerr et al., 2000) and school districts (Meier et al., 2004; Rocha, 2007). Two of the central questions addressed by previous research are as follows: (a) Does the occupational progress of one minority group come at the expense of another minority group? and (b) Are occupational patterns complementary in the sense that each minority group enjoys gains at the expense of non-Latino Whites?

Research Questions

We are primarily interested in two questions, the first of which is necessary to establish the context for the second.

Research Question 1: In multiethnic U.S. school districts, what are the levels of under-and over-representation among African Americans, Latinos, and non-Latino Whites in classroom teacher, assistant principal, principal, and administrator positions?

Research Question 2: Is there interethnic competition for classroom teacher, assistant principal, principal, and administrator positions in multiethnic U.S. school districts?¹

These questions require us to access two related bodies of literature, research on descriptive representation, situated largely in public administration, and research on interethnic employment competition, which rests at the intersection of public administration and political science.

Literature, Theoretical Orientation, and Hypotheses

In this section, we review the research on representative bureaucracy and interethnic competition for government jobs, provide a discussion of our theoretical orientation, provide an assessment of the research designs in some of the extant literature, summarize our contribution to the literature on interethnic competition for government jobs, and provide a brief statement of our hypotheses.

Representative Bureaucracy and Studies of Interethnic Competition for Government Jobs

Significant scholarly attention has been devoted to the investigation of the underrepresentation of ethnic minorities in government jobs, often referred to as descriptive representation (among others, see Cayer & Sigelman, 1980; Cornwell & Kellough, 1994; Dolan & Rosenbloom, 2003; Dometrius & Sigelman, 1984; Dye & Renick, 1981; Eisinger, 1982; Kellough, 1990; Kellough & Elliot, 1992; Mladenka, 1989; Reid, Kerr, & Miller, 2003; Selden, 1997; Welch, Karnig, & Eribes, 1983). To review extensively the large literature on descriptive representation in bureaucracies would not materially serve our purpose. The research on representative bureaucracy is widely known, excellent reviews of the literature already exist (Bearfield, 2014; Dolan & Rosenbloom, 2003; Pitts, 2005; Selden, 1997), and there is widespread agreement among researchers about the patterns of representation. Although there has been modest improvement over time, there is overwhelming evidence that African Americans, Latinos, and women remain underrepresented at the highest organizational levels in public bureaucracies and are overrepresented in the lower levels (Cayer & Sigelman, 1980; Cornwell & Kellough, 1994; Dometrius & Sigelman, 1984; Kellough, 1990; Lewis & Nice, 1994; Reid et al., 2003). Based on previous research, our first hypothesis is that Blacks and Latinos will be underrepresented in all job categories, but we expect the ethnic proportion ratios of Latinos to be lower than those for Blacks. It is important for our purposes to emphasize that when interethnic competition occurs, it does so in a representational context.

The general topic of interethnic competition has received a great deal of academic attention, much of it in the form of local-level case studies (among others, see Browning, Marshall, & Tabb, 1984; Cohen, 1982; Falcon, 1988; Hutchinson, 2007; Kaufmann, 2003; MacManus & Cassel, 1982; M. Oliver & Johnson, 1984; E. Oliver & Wong, 2003; Sawyer, 2005; Telles, Sawyer, & Rivera-Salgado, 2011; Vaca, 2004; Warren, Corbett, & Stack, 1986); however, as ethnic group competition for government jobs was first raised as a possibility by Rosenbloom (1973), and shortly afterwards examined by Hall and Saltzstein (1975), the issue has received surprisingly little academic attention. This is probably due to the lack of available data and not to a lack of scholarly and practical interest in the topic.

Relatively few studies focus specifically on questions about interethnic or interracial competition for employment/jobs (Hall & Saltzstein, 1975; Kerr et al., 2000; McClain, 1993; McClain & Karnig, 1990; Meier et al., 2004; Rocha, 2007; Rosenbloom, 1973). In his path-breaking examination of federal government jobs in 86 agencies from 1967 through 1969, Rosenbloom (1973) finds that competition between minorities and non-minorities is much more likely to occur than is competition between Blacks and Hispanics. Rosenbloom’s (1973) examination of the proportion of positions held by different groups suggests some evidence of employment competition between Blacks and Hispanics but in only a few federal government services/agencies including the Air Force, Army, Equal Employment Opportunity Commission, Federal Deposit Insurance Corporation, and General Accounting Office. McClain and Karnig’s (1990) study of political and socioeconomic competition in 49 U.S. cities, based largely on 1980 census data, focuses mostly on political competition. Using correlation analysis, they find that Hispanics are less likely to prosper politically—that is, are less likely to have larger shares of city council seats in municipalities with Black majorities or pluralities; however, McClain and Karnig (1990) find very little evidence of general competition between Blacks and Hispanics. With regard to employment, McClain and Karnig (1990) find that when one group prospers, all other groups also increase their employment share, a finding that provides support for the positive co-variation model. It is important to note that McClain and Karnig’s (1990) employment indicator is the percentage employed as reported by the U.S. Bureau of the Census. Using this measure, which combines private and public sector jobs, they do not find evidence of interminority employment competition.

Municipal Employment and School District Employment

Hall and Saltzstein (1975), McClain (1993), and Kerr et al. (2000) conduct systematic studies of interethnic municipal employment competition. In their examination of 26 Texas cities, based on correlation analysis, Hall and Saltzstein (1975) find higher proportions of one minority group coincide with lower proportions for the other group. Hall and Saltzstein (1975) suggest that the capture of the recruitment process by one group is a common occurrence in municipal employment. They argue that relative to Blacks, Mexican Americans experience an advantage in securing and retaining municipal government jobs regardless of their relative proportions of the labor force (Hall & Saltzstein, 1975). By contrast, McClain (1993), based on a national, cross-sectional sample of 41 cities, finds that Black and Hispanic municipal employment outcomes generally indicate that Blacks and Hispanics gain at the expense of non-Hispanic Whites. McClain (1993) also uncovers evidence to suggest that the size of the Black workforce has negative consequences for Hispanic employment outcomes and that Hispanics are typically more underrepresented than are Blacks. Studies in the literature tend to assume that high Latino population growth in many areas will result in fewer public resources for Blacks (de la Garza, 1997), but in the area of employment, this possibility may be tempered by the extent and nature of local-level African American political power as well as by the power of the dominant political coalition (McClain, 1993). Kerr et al. (2000), based on analysis of changes between 1985 and 1993 in 65 multiethnic U.S. cities, find that (a) Latinos in multiethnic cities experience very low levels of representation for managerial positions in municipal government and (b) widespread competition between Latinos and Blacks for municipal-level managerial positions is limited to agencies with redistributive policy commitments such as public welfare, health, community development, and parks. They find little evidence of competition between Blacks and Latinos for managerial positions in departments with non-redistributive functions (e.g., sanitation and sewage, utilities and transportation, fire, and police).

The only systematic examinations of interethnic employment competition in school districts focus exclusively on the State of Texas. The authors of these studies report that they find evidence of competition between African Americans and Latinos for administrator positions and classroom teacher positions (Meier et al., 2004; Rocha, 2007). Meier et al. (2004) examine 194 Texas school districts from 1997 to 1999 and find that school districts with greater percentages of Latino administrators have lower percentages of African American administrators and that districts with greater percentages of Latino teachers have lower percentages of teachers who are African American. Meier et al. (2004) maintain that the pattern of findings is consistent with the power thesis, specifically that non-Latino Whites are more likely to agree to benefits targeted at Latinos than those targeted at African Americans.

Rocha’s (2007) analysis of more than 300 multiracial school districts in Texas from 1994 to 2001 is similar in design to Meier et al. (2004). Rocha’s findings on school district employment do not differ from those of Meier et al. (2004), as he also finds an inverse relationship between percentage of Latino administrators and percentage of African American administrators as well as between percentage of Latino teachers and percentage of African American teachers. Rocha (2007) acknowledges that we know relatively little about how competition manifests itself in bureaucratic and policy making processes. Specifically, he says, “. . . Research should continue to use measures of different representational and policy outputs to fully understand the dynamics of rainbow coalitions” (Rocha, 2007, p. 176). Rocha’s recommendations are sound. Clearly, more research is needed; however, we also think that it is crucial to focus more attention on how interethnic competition is measured and evaluated in the first place. Following a brief discussion of our theoretical orientation, we discuss some measurement and research design limitations of previous research on interethnic employment competition.

Theoretical Orientation

Although our analysis is primarily concerned with competition, a few remarks on interethnic cooperation will help focus the discussion. The potential for interethnic cooperation varies across policy areas, a generalization about which there is agreement in the literature (Browning et al., 1984; de Leon, 2003; Hero & Preuhs, 2013; Meier et al., 2004; Meier & Stewart, 1991a; Segura & Fraga, 2008). As suggested by Meier et al. (2004) and Hero and Preuhs (2013), government employment differs from policies like bilingual education, government assistance to needy families, living wage ordinances, local government services, and voting rights/civil rights. The latter group of policies are not considered zero-sum, and as such, are policies where (a) one group is typically characterized as having low or no stakes or (b) cooperation between Latinos and African Americans can be expected. The inclination and ability to cooperate on policy is often linked to the rainbow coalition framework, based on the idea that on policies that benefit all traditionally disadvantaged groups, there is potential for coalition-building between African Americans and Latinos (Browning et al., 1984; Meier et al., 2004; Meier & Stewart, 1991a; Rocha, 2007). In other policy areas—those which are zero-sum, involve finite resources, or are unlikely to provide public goods (e.g., immigration reform, redistricting of legislative seats, and government jobs)—competition between African Americans and Latinos is more likely (Hero & Preuhs, 2013; Meier et al., 2004). Researchers sometimes find support for the power thesis, which is generally linked to the proposition that because of social distances, African American–Latino coalitions are less likely than are coalitions between Anglos and Latinos (Meier & Stewart, 1991b; Rocha, 2007). What do these generalizations about cooperation and competition mean for local level government employment?

Mucciaroni (1990) argues that we do not have centralized employment policy in the United States. State and local governments have largely retreated from passing and implementing affirmative action policies. The concomitant shift from group rights to the ethic of individualism/individual rights means that now, perhaps more than at any time in the last 30 years, competition between the traditionally disadvantaged is likely. Local government jobs, such as public school positions, are scarce resources for which competition is likely in a culture of individualism. Furthermore, pressure for efficiency in the administration and delivery of all government programs has become common and widespread, thereby increasing resource scarcity. Along these lines, more than 20 years ago, McClain (1993) urged scholars to take the possibility of interethnic competition more seriously, suggesting that because scarcity is a factor, cooperation between Blacks and Hispanics is unlikely. The power thesis is linked to the proposition that African American–Latino coalitions are less likely to form than are coalitions between Anglos and Latinos because Whites are more likely to agree to benefits targeted at Latinos (Meier & Stewart, 1991b; Rocha, 2007). In a sense, the power thesis and the rainbow coalition framework both focus on aspects of coalition building and cooperation. This focus cannot explain a scenario where all groups compete. In the area of government employment, we think that it makes the most sense to view Blacks and Latinos as in competition with White non-Latinos as well as in competition with each other. In the absence of a centralized employment policy that benefits underrepresented groups, we assume that there is little to no incentive in the area of government jobs to cooperate or to build ethnic coalitions. We agree with Meier et al. (2004) that when scarce resources are involved, competition/conflict is likely to be the norm. Our understanding of the literature leads us to believe that all three groups will compete against each other for public school positions. We examine our second and primary hypothesis—the interethnic competition hypothesis—across all four job categories. We expect an increase in the Black or Latino job share to be associated with a decrease in the other group’s ethnic proportion ratio. Alongside the interethnic competition hypothesis, we also examine a third hypothesis, the possibility of competition between Blacks and Whites as well as between Latinos and Whites for each of the four job categories.

Addressing Previous Studies

Empirically demonstrating interethnic competition is a difficult task. The most direct way to measure (or to assess) such competition would be to use small-N or case study designs that rely on difficult, perhaps impossible to obtain, in-depth information on discrete hiring scenarios for individual jobs. To be sure, with large-N designs it is probably not possible to measure interminority employment competition directly. That said, it is important to emphasize that the development of the interethnic competition literature has not proceeded in typical fashion. Typically, after defining the concept of interest, researchers attempt to develop an operational definition of the concept. It appears that previous research has overlooked or simply bypassed the problem/puzzle of measuring the concept of interminority competition. Researchers have used commonly accepted methods like multiple regression analysis on dependent variables that are not indicators of competition. The dependent variables typically used in most models are the percentage of employment positions held by members of different racial/ethnic groups. Then, correlation or regression is used on cross-sectional or cross-sectionally dominant data. We contend that questions about interethnic employment competition cannot be answered well by relying on analysis of cross-sectional data, even when multivariate analysis is used. Analysis on cross-sectional data cannot show that an increasing proportion of one group is associated with an increasing or decreasing proportion of another group. It can show that higher values are related to lower values across units, but it cannot show increases or decreases over time within units. The observation of a statistically significant, negative relationship in a cross-section means that across governmental units, higher percentages (or ratios) of Blacks are associated with lower percentages (or ratios) of Latinos, not that within units—whether the units are cities or school districts—the employment gains of one group are associated with the employment losses of the other (Kerr et al.,2000); however, it appears that coefficients are sometimes interpreted this way (McClain, 1993; McClain & Karnig, 1990). We think it is important to keep in mind that job competition almost always occurs within units over time. Rarely does competition occur across units. African Americans in Cleveland generally do not compete with Latinos in Houston for local-level government positions. Results from analyses on cross-sectional data may be suggestive, but they should not be considered highly so. These results may be suggestive in a very loose sense.

By the same token, the use of pooled time-series analysis on data that are cross-sectionally dominant (see Meier et al., 2004; Rocha, 2007) will not produce parameter estimates that isolate information about variation over time within units—unless the researcher uses an assumption of fixed-effects. In the absence of a fixed-effects assumption, pooled time series models using cross-sectionally dominant data will explain variation across units and over time, but most of the explained variance will be across units (Zhao, He, & Lovrich, 2005). Thus, pooled time-series analysis with a random effects assumption and the observation of a significant and negatively signed coefficient means that units with higher percentages of Latinos in a job category tend to have lower percentages of Blacks or vice versa. This method provides little, if any, information about variation within units over time. To sum up, we question whether previous models based on cross-sectional or cross-sectionally dominant data identify and evaluate interethnic employment competition within governmental units. Not only is this a plea to take time more seriously when examining interethnic competition, but it is also a plea to take more seriously the measurement of interminority employment competition as a concept.

Significance of the Research and Contribution to the Literature

One important limitation of work on local-level government jobs is that it often relies on EEO-4 reports and data as published in Job Patterns for Minorities and Women in State and Local Governments (U.S. Equal Employment Opportunity Commission [U.S. EEOC], various years). EEO-4 reports do not include U.S. public school system employees. The exclusion of these positions from much of past research on descriptive representation and interethnic employment competition neglects a sizable proportion of the local public sector workforce in the United States. In fact, about 30% of state government expenditures are dedicated to public education, and about 7 million of the approximately 21 million federal, state, and local government employees work in the fields of elementary and secondary education (Stanley & Niemi, 2010). Furthermore, previous work on school district employment competition uses data from only one state, Texas (Meier et al., 2004; Rocha, 2007). A rich source of national data only recently made available to scholars, EEO-5 reports remain largely unexplored. State and local governments are required to use EEO-5 reporting forms (Elementary-Secondary Staff Information) to report public school district employees to the EEOC. These data sets allow us to address some limitations inherent in previous research, one of which is the omission of millions of school district jobs from analyses of local-level interethnic employment competition.

Among the perennial topics in public administration, research on under- and over-representation is important because school districts are directly involved on a daily basis in the lives of minority and non-minority residents. Whether we consider school district management, school management, or rank-and-file classroom teacher positions, increased representation is likely to lead to greater responsiveness to the policy concerns of traditionally underrepresented groups such as African Americans and Latinos (Meier & Stewart, 1991b; Meier & Stewart, 1992; Selden, 1997; Selden, Brudney, & Kellough, 1998). In addition, three of the four positions that we analyze in this article—assistant principals, principals, and administrators—are sources of considerable economic progress as they provide high salaries, benefits, status, job security, and power. These positions, as well as classroom teacher positions, are awarded and retained in an environment of resource scarcity. Because all four of these positions are desirable and scarce, they are especially likely to be the objects of competition.

Questions about interethnic employment competition are especially important. Blalock (1967) argues that interethnic competition may account for some aspects of discrimination experienced by members of traditionally underrepresented groups. Also, the presence of interethnic employment competition suggests that affirmative action progress and equal employment opportunity goals face unanticipated hurdles (Kerr et al., 2000). A serious consequence of competition may be that qualified ethnic minority group members will seek employment in the private or non-governmental non-profit sectors, thereby shrinking the pool of applicants for local government positions (Waldinger, 1996). Struggling in competition with other groups as well as losing in competition with another minority ethnic group may restrict the capacity for descriptive and substantive representation of minority interests. A number of studies establish that the nature of representative bureaucracy affects certain public policy outputs/outcomes and may lead to policy representation in public bureaucracies (Grissom, Nicholson-Crotty, & Nicholson-Crotty, 2009; Hindera, 1993; Keiser, 1999; Keiser, Wilkins, Meier, & Holland, 2002; Meier & Stewart, 1992; Meier, Wrinkle, & Polinard, 1999; Selden, 1997; Selden et al., 1998).

We contribute to the literature in several ways. We extend analysis of interethnic competition for public school jobs beyond examination of the State of Texas to include multiethnic districts across the United States. Based on changes between 2002 and 2008, our analysis should be generalizable to all multiethnic school districts across the country. This research strategy also relies on a more direct measure of interethnic employment competition for public school districts so that we can better isolate employment competition within districts over time. As discussed below, we also rely on more than one measure to evaluate interethnic job competition in multiethnic U.S. school districts. This strategy provides an opportunity to examine competition from two vantage points: (a) a count of how many and what percentages of multiethnic districts experience competition and (b) an aggregate count of how many jobs are gained and lost by each group over time across all multiethnic school districts in the United States. In the context of elimination of affirmative action programs and retraction from equal employment opportunity programs across local governments, we also provide a theoretical argument that there is no reason to expect cooperation between African Americans and Latinos, Latinos and Whites, or African Americans and Whites in the distribution of the better paying, more prestigious school district positions.

Hypotheses

To summarize, we expect to observe underrepresentation of Blacks and Latinos in all job categories, but we expect the ethnic proportion ratios for Latinos to be lower than those for Blacks. We expect to observe overrepresentation of Whites (non-Latino) for all four job categories. We also examine across all four job categories our primary and second hypothesis, the interethnic competition hypothesis—that an increase in the Black or Latino job share will be associated with a decrease in the other group’s ethnic proportion ratio. Alongside the interethnic competition hypothesis for African Americans and Latinos, we also examine a third hypothesis, the possibility of competition between Blacks and Whites as well as between Latinos and Whites for each of the four job categories.

Data, Variables, and Method

Data

The data for our analysis were provided by the U.S. EEOC and are not publicly available.² We obtained all EEO-5 reports (i.e., EEOC Form 168A) for 2002 and 2008, the earliest and most recent years for which complete data are available. School districts with 100 or more employees are required to file reports every even-numbered year. The EEOC’s EEO-5 reports are the most complete and resolute source of U.S. school district employment data available.³ These data permit analysis of employment distribution by job category (e.g., Officials/Administrators/Managers, Principals, Assistant Principals, Elementary Classroom Teachers, Secondary Classroom Teachers, etc.), sex, and race/ethnicity for each U.S. school district with at least 100 employees.⁴ We omit from the analysis all educational units of government that do not employ classroom teachers, such as educational cooperatives and state/regional education agencies. This results in EEO-5 data on 5,843 school districts in 2002 and 6,153 districts in 2008.⁵ Following previous research (Kerr et al., 2000; Meier et al., 2004; Rocha, 2007), we select for analysis only those school districts that include populations that are at least 5% African American, 5% Latino, and 5% White non-Latino. This results in 419 multiethnic districts in 2002 and 638 multiethnic districts in 2008. Because we must calculate changes in ethnic proportion ratios and evaluate interethnic competition within districts over time, we only include for analysis districts that were multiethnic in both years. This results in a sample of 267 multiethnic U.S. school districts.

Variables and Method

Our first objective is to calculate by job category the ethnic proportion ratios for each group for each school district for 2002 and 2008. Groups’ ethnic proportion ratios are calculated by dividing for each job category (i.e., classroom teachers, assistant principals, principals, and administrators) the percentage of jobs held by each ethnic group by the percentage of their coethnic population in the school district.⁶ The denominator for 2002 is based on census data from 2000 (U.S. Bureau of the Census, 2000) and the denominator for 2008 is based on data from 2006 to 2008 (U.S. Bureau of the Census, 2008). Ratios greater than one indicate that the group is overrepresented in that job category in the school district; ratios less than one indicate that the group is underrepresented in that job category in the school district. We also difference the means of the ethnic proportion ratios for 2002 and 2008 to determine whether ethnic groups have, on average, increased or decreased their ethnic proportion ratios in multiethnic school districts. These over-time changes permit us to make inferences about the nature of interethnic competition; however, reliance on means can be misleading because of differences in variances across subgroups.

To achieve our primary objective, we report by job category the number and percentages of school districts that show employment competition between Blacks and Latinos, Blacks and Whites, and Latinos and Whites. Following previous research, our operational measure of interethnic competition, using the school district as the unit of analysis, is an over-time increase in the ethnic proportion ratio of at least 0.05 for one group and an accompanying over-time decrease in another group’s ethnic proportion ratio of at least 0.05 (Kerr et al., 2000). For districts that experience competition between Blacks and Latinos, we supplement the analysis with data on the raw number of jobs gained and lost by Blacks and Latinos in each job category. The aggregate number of jobs gained and lost provides a different perspective on competition. Although this measure does not control for each group’s population percentage in the school district, it does provide an account of jobs gained and lost across all multiethnic districts in the United States.

Findings

Table 1 reports means and standard deviations of ethnic proportion ratios by job category, ethnic group, and year. The means for both years show widespread overrepresentation for White non-Latinos among classroom teachers, assistant principals, and principals. On average, Blacks and Latinos are underrepresented among classroom teachers in 2002 and 2008; however, the extent of underrepresentation among Latinos is nearly double that of Blacks. The difference of means between these two groups is statistically significant at p < .01. Furthermore, based on the means of the proportion ratios, neither group made sizable gains between 2002 and 2008 in the classroom teacher category. Latinos are underrepresented in all other job categories (i.e., assistant principals, principals, and administrators) and made little over-time progress. This was not the case for Blacks.

Table 1.

Employment Proportionality in Multiethnic Districts by Ethnic Group and Job Category: Means of Ethnic Proportion Ratios and Changes in the Means of Proportion Ratios, 2002-2008.

Ethnic group and job category	2002	2008	Change in mean ratio, 2002-2008
Ethnic group and job category	M (SD)	M (SD)	Change in mean ratio, 2002-2008
Classroom teachers
Black	0.54^a (0.28)	0.56^a (0.33)	0.02
Latino	0.29^b (0.17)	0.32^b (0.18)	0.03
White (non-Latino)	1.55^c (0.57)	1.74^c (0.74)	0.19
Assistant principals
Black	1.33^a (1.11)	1.42^a (1.18)	0.09
Latino	0.32^b (0.54)	0.33^b (0.50)	0.01
White (non-Latino)	1.24 (0.54)	1.33 (0.60)	0.09
Principals
Black	1.09^a (0.75)	1.13^a (0.80)	0.04
Latino	0.22^b (0.33)	0.26^b (0.37)	0.04
White (non-Latino)	1.33^c (0.45)	1.42^c (0.56)	0.09
Administrators
Black	0.69^a (0.65)	0.81^a (0.84)	0.12
Latino	0.21^b (0.36)	0.24^b (0.31)	0.03
White (non-Latino)	0.71 (0.44)	0.87 (0.47)	0.16

Note. N = 267 multiethnic school districts. Ethnic proportion ratios are measured as the percentage of employment positions held by the group, divided by the group’s percentage of school district population. Standard deviations are in parentheses. T-tests are for differences of means.

Significant difference between Black and Latino.

Significant difference between Latino and White.

Significant difference between White and Black.

All significant results are at p < .01.

In multiethnic school districts, the means of ethnic proportion ratios for Blacks in the assistant principal job category are 1.33 in 2002 and 1.42 in 2008, well above parity. Among principals, on average, Blacks are represented at rates slightly above parity with little indication of change over time—1.09 in 2002 and 1.13 in 2008. The means of the ethnic proportion ratios for administrators reported in Table 1 are below 1.00 for Blacks, Latinos, and Whites. Because there are no administrators in districts for some ethnic groups in some years, we observe a large number of zeroes, which pulls down the means. Thus, we must focus on the relative sizes of the mean ratios across ethnic groups. The mean of the ethnic proportion ratios for Whites is the largest of all three groups, as is the change in the means, from 0.71 to 0.87. The means for Black administrators are similar to those for Whites, suggesting that Blacks are, on average, performing about as well as Whites in obtaining and retaining prestigious administrator positions. The difference in means for Blacks and Whites is not statistically significant. By contrast, we observe low means for Latinos—0.21 in 2002 and 0.24 in 2008, both of which are lower than the means for Blacks and Whites at p < .01. We also observe relatively small standard deviations, suggesting that Latinos achieve administrative parity in very few multiethnic districts. In the median districts, for administrators we observe Latino representation ratios of 0.00 in 2002 and 0.14 in 2008.

In sum, the employment proportionality statistics reported in Table 1 indicate very low levels of representation for Latinos and relatively high levels of representation for Whites (non-Latino), findings that are generally consistent with our first hypothesis. In every single case, the difference in means between Latinos and Blacks and between Latinos and White non-Latinos is statistically significant. On average, for Blacks, we observe near-parity for assistant principal and principal positions, a finding that is not consistent with our first hypotheses. For classroom teachers and administrators, however, we observe means of ethnic proportion ratios that are less than one for Blacks, which is generally consistent with our first hypothesis. Our conclusion for administrators is contingent because several ethnic proportion ratios are zero. This is a consequence of many districts having a small number of administrator positions. Because many districts have only a few administrator positions, division of administrators into three ethnic groups will often lead to zero for some groups.

Because Latinos do not possess a significant number of positions in most multiethnic school districts (see Table 1), we think it is unlikely that we will observe for any of the job categories pervasive patterns of Black gains and Latino losses. Latinos simply do not have a large share of positions for which Blacks (and White non-Latinos) can compete.

We turn now to interethnic employment competition in multiethnic U.S. school districts, our primary area of interest. Table 2 reports frequencies and percentages of school districts that show evidence of interethnic employment competition. Recall that based on our operational measure, interethnic competition exists if between 2002 and 2008 we observe an increase in the ethnic proportion ratio of at least 0.05 for one group and an accompanying decrease in another group’s ratio of at least 0.05. Because de la Garza (1997) and McClain (1993) argue that high Latino population growth may result in fewer resources and opportunities for Blacks, we also split the sample into high and low Latino growth districts. Our cutoff for low and high Latino population growth is population growth between 2002 and 2008 in the median district. Population growth in the median district is 4.3 percentage points. We also conduct difference of binomial proportion tests for all ethnic/racial pairings reported in Table 2.

Table 2.

Frequencies and Percentages of Multiethnic School Districts With Competition Between Blacks, Latinos, and Whites (Non-Latino) by Job Category.

Job category	Black-Latino	Latino-Black	No competition	Black-White	White-Black	No competition	Latino-White	White-Latino	No competition
Classroom teachers
All multiethnic	13 (4.9)	29 (10.9) *	225 (84.3)	12 (4.5)	75 (28.1) *	180 (67.4)	13 (4.9)	48 (18.0) *	206 (77.2)
High Latino growth	9 (6.7)	9 (6.7)	116 (86.6)	2 (1.5)	32 (23.9)	100 (74.6)	2 (1.5)	30 (22.4)	102 (76.1)
Low Latino growth	4 (3.0)	20 (15.0)	109 (81.9)	10 (7.5)	43 (32.3)	80 (60.1)	11 (8.3)	18 (13.5)	104 (78.2)
Assistant principals
All multiethnic	40 (15.0)	35 (13.1)	192 (71.9)	38 (14.2)	94 (35.2) *	135 (50.1)	36 (13.5)	47 (17.6)	184 (69.0)
High Latino growth	20 (14.9)	15 (11.2)	99 (73.9)	17 (12.7)	48 (35.8)	69 (51.5)	18 (13.4)	30 (22.4)	86 (64.2)
Low Latino growth	20 (15.0)	20 (15.0)	93 (69.9)	21 (15.8)	46 (34.6)	66 (49.6)	18 (13.5)	17 (12.8)	98 (73.7)
Principals
All multiethnic	32 (12.0)	33 (12.4)	202 (75.7)	57 (21.4)	85 (31.8) *	125 (46.8)	25 (9.4)	44 (16.5) *	198 (74.2)
High Latino growth	21 (15.7)	18 (13.4)	95 (70.9)	22 (16.4)	46 (34.3)	66 (49.3)	12 (9.0)	28 (20.9)	94 (70.1)
Low Latino growth	11 (8.3)	15 (11.3)	107 (80.5)	35 (26.3)	39 (29.3)	59 (44.4)	13 (9.8)	16 (12.0)	104 (78.2)
Administrators
All multiethnic	28 (10.5)	24 (9.0)	215 (80.5)	36 (13.5)	59 (22.1) *	172 (64.4)	14 (5.2)	35 (13.1) *	218 (81.7)
High Latino growth	19 (14.2)	10 (7.5)	105 (78.4)	13 (9.7)	31 (23.1)	90 (67.2)	6 (4.5)	29 (21.6)	99 (73.9)
Low Latino growth	9 (6.8)	14 (10.5)	110 (82.7)	23 (17.3)	28 (21.1)	82 (61.7)	8 (6.0)	6 (4.5)	119 (89.5)

Note. N = 267 multiethnic school districts. The whole numbers are frequencies for the number of school districts that experience competition between 2002 and 2008. The numbers in parentheses represent the percentage of multiethnic districts that experience competition between 2002 and 2008. Latino-Black competition is an increase in the Latino proportion ratio of at least 0.05 and an accompanying decrease in the Black proportion ratio of at least 0.05. Black-Latino competition is an increase in the Black proportion ratio of at least 0.05 and an accompanying decrease in the Latino proportion ratio of at least 0.05, and so on. T-tests are for differences in binomial proportions. For example, in the first group of cells (upper left portion of table) there is a statistically significant difference between 4.9% and 10.9%. If *, significant at p < .01.

Although less prevalent than competition between Blacks and White non-Latinos or between Latinos and White non-Latinos, we do observe competition between Blacks and Latinos in many multiethnic school districts. A total of 42 multiethnic school districts (16% of the sample) show evidence of this type of competition, and thus support for our second hypothesis. Among classroom teachers; however, the number of districts with Latino gains and Black losses (n = 29) is greater than double the number of districts with Black gains and Latino losses (n = 13). The test for differences in binomial proportions (10.9% and 4.9%) is statistically significant at p < .01.

We observe in many districts—approximately 25% of those in our sample—employment competition between Blacks and Latinos for assistant principal, principal, and administrative positions, but the number of districts with Black gains and Latino losses is roughly equal to the number with Latino gains and Black losses (see Table 2). Based on a simple count of the number of school districts, competition for these three job categories does not seem in the aggregate to advantage one group over the other. By contrast, Latino gains and simultaneous Black losses is the prevailing pattern among classroom teachers by a ratio of greater than two to one. Still, it is important to emphasize that for most multiethnic districts, roughly 75%, we observe no Black-Latino or Latino-Black competition; these districts do not indicate support for our second hypothesis. The results of the analysis on the split sample of low and high growth Latino school districts indicates that in low Latino growth districts, Latinos dominate African Americans in competition for classroom teacher positions; Latino gains and Black losses occur in 20 of 24 districts with competition between these groups. In high Latino growth districts; however, Blacks outpace Latinos for administrator positions in about two thirds of the districts (19 of 29) in which we observe minority group competition.

Table 2 also reports on the results of our third hypothesis, the frequencies and percentages of multiethnic school districts with employment competition between Blacks and Whites and between Latinos and Whites. Seventy-five multiethnic school districts indicate patterns of White non-Latino gains and Black losses among classroom teachers. However, we observe for only a small number of districts (n = 12) patterns of Black gains and White (non-Latino) losses for classroom teacher positions. The number of districts with Black gains and White losses among assistant principals (n = 38) is far less than the number with White gains and Black losses (n = 94), a pattern that is generally repeated for principal and administrator positions. We observe for pairings of Blacks and White non-Latinos for assistant principal and principal positions competition in a large number (more than 50%) of school districts, 132 for assistant principals and 142 for principals. For principals and administrators—among the most coveted and powerful positions in school districts—White gains and Black losses are much more likely to occur than are patterns of Black gains and White losses. The differences are statistically significant in every case. The findings on competition between White non-Latinos and Latinos resemble the findings for White–Black and Black–White competition. Of the 61 school districts in which we observe competition among classroom teachers, 48 follow a pattern of White non-Latino gains and Latino losses. Likewise, among assistant principals, principals, and administrators, the number of multiethnic school districts with White non-Latino gains and Latino losses is greater than the number with Latino gains and White losses; however, the difference for assistant principals is not statistically significant. Among administrators, the ratio of districts with White gains and Latino losses to Latino gains and White losses is well over 2 to 1 and the test of binomial proportions indicates that the differences are statistically significant. In the majority of multiethnic school districts, however, we observe no evidence of competition between Whites and Latinos, irrespective of job category.

In sum, the results reported in Table 2 show that group competition does not occur in most multiethnic school districts; however, it is not a rare occurrence. Furthermore, even though we know that Whites are generally overrepresented or well-represented in all job categories (see Table 1) and that the population percentages of Whites in multiethnic school districts (and cities) is generally declining, the results show that when we do observe competition between Whites on the one hand and Blacks or Latinos on the other, Whites consistently win. The pattern of success for White non-Latinos across all job categories is especially pronounced in school districts with high levels of Latino population growth. When competition emerges between Blacks and Latinos, Latinos fare well in competition for classroom teacher positions, especially in school districts with low Latino population growth. On the other hand, Blacks and Latinos perform at about the same level for assistant principal, principal, and administrator positions. Latino populations are generally growing faster than those of Blacks, so it is possible we are seeing an upward trend toward greater representation of Latino teachers as a result of a tipping phenomenon. Latinos may in many areas be reaching a critical proportion of the population that leads to greater representation, at least in one of the key entry-level positions of school district employment, classroom teachers. This is probably a necessary condition for improving the representation of Latinos in the upper levels of public school systems. Latinos tend to do better in competition with Blacks in school districts with low Latino population growth.

Interethnic employment competition is a difficult concept to measure and evaluate. As with many measures of complex concepts, there may be some distance between the definition of the concept itself and measures/methods in the literature used to identify it. We supplement our analysis with an additional measure, the absolute number of jobs gained and lost by Blacks and Latinos in districts with interethnic employment competition. This measure provides an opportunity to examine competition across school districts in the United States. Instead of assessing group gains and losses in relation to the same group’s proportion of the school district’s population, this measure is simply the raw number of jobs gained and lost.

Table 3 reports the number of positions gained and lost by job category for multiethnic districts that experienced competition between Blacks and Latinos between 2002 and 2008. In the 13 districts with Black gains and Latino losses based on ethnic proportion ratios reported in Table 2, Blacks gained 1,516 classroom teacher positions and Latinos gained 251 classroom teacher positions. Latinos can increase their number of jobs, but still experience declining proportional representation. Said differently, even though there was an absolute gain in Latino teacher positions between 2002 and 2008, there was not enough of an increase to keep up with the growth in Latino population. In the 29 districts with Latino gains and Black losses, Latinos gained 1,051 classroom teacher positions while Blacks lost a total of 2,144 teaching jobs. Across all 42 districts with interethnic competition, Latinos picked up a total of 1,302 classroom teacher positions and Blacks lost a total of 628.

Table 3.

Positions Gained/Lost Between 2002 and 2008 in Multiethnic Districts With Competition Between Blacks and Latinos.

	Black/Latino			Latino/Black			Latino change		Black change
Job category	N			N			N		N
Classroom teachers	13	1,516	251	29	1,051	−2,144	42	1,302	42	−628
Assistant principals	40	254	−29	35	93	−86	75	64	75	168
Principals	32	108	−32	33	80	−78	65	48	65	30
Administrators	28	172	−3	24	62	12	52	59	52	184

Note. N is the number of school districts. Cell entries represent the number of jobs. Latino and Black change columns represent the total change (gain or loss) in the districts which had competition.

Far fewer jobs were at stake in higher-level positions. The raw numbers reflect to some extent the fact that the number of school districts with Black gains and Latino losses is roughly equal to the number with Latino gains and Blacks losses. Both groups experience net gains in the total number of jobs across all districts with competition, but Blacks gained about 3 times more jobs than did Latinos for assistant principal and administrator positions. By contrast, Latinos gained 48 principal positions compared with an increase of 30 for Blacks, but these gains do not keep pace with the rate of Latino population increases.

Summary and Discussion

We provide a different look at interethnic job competition, with a largely unexplored national data set on positions in U.S. multiethnic school districts in 2002 and 2008. We also rely on measures of interethnic competition for public school positions that are more direct than those previously appearing in the research on local-level government employment. Latinos are underrepresented, and White non-Latinos are overrepresented or achieve near-parity in multiethnic school districts for all four job categories. On average, Blacks are overrepresented in assistant principal and principal positions, but underrepresented among classroom teachers and administrators. These results are generally consistent with our first hypothesis. In addition, the findings show that group competition does not occur in most districts, but neither is it uncommon. We find less support for competition between Blacks and Latinos (i.e., the second hypothesis) than we do for competition between White non-Latinos and Blacks and between White non-Latinos and Latinos (i.e., the third hypothesis). When competition between Blacks and Latinos occurs for classroom teacher positions, Latinos tend to be more successful than Blacks whether we control for population or simply examine the raw number of teaching positions. The picture is less clear-cut for Black-Latino competition for assistant principal, principal, and administrator positions. For these positions, at least in districts in which we observe competition, it appears that each group does roughly equally well. In low Latino population growth districts, Latinos outpace Blacks in competition for teacher positions; however, Blacks in high Latino population growth districts outpace Latinos for administrator positions.

Whites are overrepresented in areas that are likely to have declining percentages of White (non-Latino) residents, and they are quite successful at maintaining or protecting this advantage over time. Rosenbloom (1973) found that competition between minorities and non-minorities for federal government positions was much more prevalent than competition between Blacks and Hispanics. Roughly two generations later, although there is evidence of competition in a fairly large number of school districts, we do not observe competition in a majority of multiethnic school districts. The exception to this generalization is for assistant principal and principal positions for pairings between Blacks and White non-Latinos where we observe competition in a majority of districts. White non-Latinos were generally overrepresented in multiethnic school districts in 2002, school districts in which percentages of White populations tend to be on the decline. Being the dominant economic and political group in most school districts, it is not surprising that Whites managed, on average, to increase their ethnic proportion ratios over time. We do find some evidence to suggest that Latinos are making important gains, but these gains frequently come at the expense of Blacks, a finding that provides support for the interethnic competition hypothesis. The evidence is most convincing in districts with Latino gains and Black losses, especially for classroom teacher positions. The findings suggest that gains in the Latino share of classroom teacher positions reflect the increase in Latino populations in urban centers. Growth in these positions may establish a solid base that will have implications for increased administrative and political power for Latinos in public school districts. The findings are important because in the roughly 25% of districts in the sample in which we find support for the interethnic competition hypothesis, gains made by one traditionally underrepresented group are accompanied by losses of another traditionally underrepresented group. Patterns such as these present hurdles to equal employment opportunity, may reduce the pool of qualified minority applicants for public sector positions, and may further restrict the capacity for substantive representation of minority interests (Blalock, 1967; Meier & Stewart, 1992; Waldinger, 1996). In the context of resource scarcity, rapid increases in Hispanic populations in urban areas, and pressures for greater efficiency in the delivery of public education, McClain’s (1993) call for more attention to interethnic job competition appears to have been prophetic. Our findings further suggest that researchers need to pay closer attention to the overrepresentation of White non-Latinos in urban centers as well as to the ability of White non-Latinos to prevail in employment competition with Blacks and with Latinos in cities in which the shares of White non-Latino populations are decreasing.

Future research should continue to address with different measures and methods whether patterns of interethnic competition in school districts/urban areas that have experienced high levels of Latino population growth differ from patterns of interethnic competition in areas that have not. The implications for human resource practice in public school settings very probably include the need for managers to become more aware of metrics for representation. Such awareness is critical because, as we noted in our literature review, representation can have very significant impacts on education outcomes. Our finding that there are districts with very similar minority constituencies but very different levels of minority representation leads us to think that the structure, culture, and human resource practices of the districts may have a real impact on representation. In those situations where a group’s representation is weak, human resource practice should include the accepted methods for addressing this in hiring and other practices. These methods should include a critical examination of any possible organizational structure or cultural impediments to better representation, a strong focus on developing hiring pools that are more representative, the careful construction of job descriptions so as not to exclude specific minority groups unintentionally, and the use of mentoring programs to increase the pool of qualified underrepresented groups.

Moving beyond simply identifying which districts experience competition, it would be useful in expanding theory and informing practice if future research explored the conditions that result in competition. Given the importance of these and related questions to resource distribution, the interests of members of traditionally disadvantaged groups, and power, it would be desirable to devise more and better measures of interethnic employment competition—measures that closely correspond to the concept of interethnic employment competition and are suitable for use as a dependent variable in multivariate analyses. Several questions need to be addressed. Why do some districts—even those having similar minority population characteristics—experience interethnic competition while others do not? Are districts in high-poverty areas more likely to experience competition than districts in affluent areas? Does the nature of interethnic competition vary by region? Are minorities in Texas more competitive with each other than minorities in Illinois? Does the presence of a successful dominant group result in more interethnic competition? How much impact does human resource policy and practice have on representation and interethnic competition? In this exploration, we believe it is essential to use representational measures (e.g., group percentage of job held/group population percentage) rather than a simple measure of percentage jobs held by minority groups. The latter approach may misidentify as competition patterns that result from simple demographic differences in the minority populations of one district when compared with another. Representation measures “control for” population differences as well as getting at central questions on representative bureaucracy and interethnic competition.

Footnotes

Acknowledgements

We are grateful to the U.S. Equal Employment Opportunity Commission (EEOC), Office of Research, Information, and Planning, for providing data for this study and to Dr. Ronald Edwards of the EEOC for helpful advice and comments. We also thank Dr. Rafael Jimeno for helpful comments and Paul Valentine for research assistance.

Authors’ Note

A previous version of this article was presented at the annual meeting of the Midwest Political Science Association, April 12-15, 2012, Chicago, Illinois.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Brinck Kerr is a professor of political science and director of the Public Policy PhD Program at the University of Arkansas. His research focuses on representation, employment policy, and education policy. His work has appeared in many political science and public administration journals.

Will Miller is a professor of public administration and chair of the public administration program at the University of Illinois at Springfield. His research focuses on political and economic minorities and public policy analysis. His work has been published in Public Administration Review, American Review of Public Administration, State and Local Government Review, and elsewhere.

Grace Rusk Kerr is a clinical instructor of childhood education at the University of Arkansas. Her research interests include ethnic and gender employment patterns in public schools, education policy, literacy, and language variation and change.

Renee Deshommes earned her PhD in public policy from the University of Arkansas–Fayetteville in 2013. Her area of specialization is education policy, and her scholarly interests include policy issues that target at-risk students. She is the elementary/middle-school principal of the Arkansas Arts Academy in Rogers, Arkansas.

References

Bean

F. D.

Bachmeier

J. D.

Brown

S. K.

Tafoya-Estrada

(2011). Immigration and labor market dynamics. In Telles

E. E.

Sawyer

Rivera-Salgado

(Eds.), Just neighbors? Research on African American and Latino relations in the United States (pp. 37-60). New York, NY: Russell Sage Foundation.

Bearfield

(2014). It’s been a long time comin’: An examination of public personnel research in PAR and ROPPA in celebration of the fiftieth anniversary of the Civil Rights Act of 1964. Review of Public Personnel Administration, 34, 59-74.

Blalock

H. M.

(1967). Toward a theory of minority-group relations. New York, NY: John Wiley.

Browning

R. P.

Marshall

D. R.

Tabb

D. H.

(1984). Protest is not enough: The struggle of Blacks and Hispanics for equality in urban politics. Berkeley: University of California Press.

Cayer

N. J.

Sigelman

(1980). Minorities and women in state and local government: 1973-1975. Public Administration Review, 40, 443-450.

Cohen

(1982). Alliance and conflict among Mexican Americans. Ethnic and Racial Studies, 5, 175-195.

Cornwell

Kellough

(1994). Women and minorities in federal government agencies: Examining new evidence from panel data. Public Administration Review, 54, 265-276.

de la Garza

(1997). Latino politics: A futuristic view. In Garcia

F. C.

(Ed.), Pursuing power (pp. 448-456). South Bend, IN: University of Notre Dame Press.

de Leon

(2003). San Francisco: The politics of race, land use and ideology. In Browning

R. P.

Marshall

D. R.

Tabb

D. H.

(Eds.), Racial politics in American cities (pp. 167-198). New York, NY: Longman.

10.

Dolan

Rosenbloom

D. H.

(2003). Representative bureaucracy: Classic readings and continuing controversies. Armonk, NY: M.E. Sharpe.

11.

Dometrius

Sigelman

(1984). Assessing progress toward affirmative action goals in state and local government. Public Administration Review, 44, 241-246.

12.

Dye

Renick

(1981). Political power and city jobs: Determinants of minority employment. Social Science Quarterly, 62, 475-486.

13.

Eisinger

P. K.

(1982). Black employment in municipal jobs: The impact of Black political power. American Political Science Review, 76, 380-392.

14.

Falcon

(1988). Black and Latino politics in New York City. In Garcia

(Ed.), Latinos in the political system (pp. 171-194). Notre Dame, IN: Notre Dame University Press.

15.

Grissom

Nicholson-Crotty

(2009). Race, religion, and representative bureaucracy. Public Administration Review, 69, 911-919.

16.

Hall

Saltzstein

(1975). Equal employment in urban government: The potential problem of interminority competition. Public Personnel Management, 4, 386-393.

17.

Hero

Preuhs

R. R.

(2013). Black-Latino relations in U.S. national politics: Beyond conflict or cooperation. New York, NY: Cambridge University Press.

18.

Hindera

(1993). Representative bureaucracy: Further evidence of active representation in the EEOC district offices. Journal of Public Administration Research and Theory, 3, 415-429.

19.

Hutchinson

E. O.

(2007). The Latino challenge to Black America: Towards a conversation between African-Americans and Hispanics. Chicago, IL: Middle Passage Press.

20.

Kaufmann

K. M.

(2003). Cracks in the rainbow: Group commonality as a basis for Latino and African American political coalitions. Political Research Quarterly, 56, 199-210.

21.

Keiser

(1999). State bureaucratic discretion and the administration of social welfare programs: The case of social security disability. Journal of Public Administration Research and Theory, 9, 87-106.

22.

Keiser

Wilkins

Meier

Holland

(2002). Lipstick and logarithms: Gender, institutional context, and representative bureaucracy. American Political Science Review, 96, 553-564.

23.

Kellough

J. E.

(1990). Integration in the public workplace: Determinants of minority and female employment in federal agencies. Public Administration Review, 50, 557-566.

24.

Kellough

J. E.

Elliot

(1992). Demographic and organizational influences on racial/ethnic and gender integration in federal agencies. Social Science Quarterly, 73, 1-11.

25.

Kerr

Miller

Reid

(2000). The changing face of urban bureaucracy: Is there competition for municipal government jobs? Urban Affairs Review, 35, 770-793.

26.

Lewis

Nice

(1994). Race, sex, and occupational segregation in state and local governments. American Review of Public Administration, 24, 393-410.

27.

MacManus

Cassel

(1982). Mexican Americans in city politics: Participation, representation, and policy preferences. Urban Interest, 4, 57-69.

28.

McClain

(1993). The changing dynamics of urban politics: Black and Hispanic municipal employment—Is there competition? Journal of Politics, 55, 399-414.

29.

McClain

Karnig

(1990). Black and Hispanic socioeconomic and political competition. American Political Science Review, 84, 535-545.

30.

Meier

McClain

Wrinkle

Polinard

(2004). Divided or together? Conflict and cooperation between African Americans and Latinos. Political Research Quarterly, 57, 399-409.

31.

Meier

Stewart

(1991a). Cooperation and conflict in multiracial school districts. Journal of Politics, 53, 1123-1133.

32.

Meier

Stewart

(1991b). The politics of Hispanic education. Albany: State University of New York Press.

33.

Meier

Stewart

(1992). The impact of representative bureaucracies: Educational systems and public policies. American Review of Public Administration, 22, 151-171.

34.

Meier

Wrinkle

Polinard

(1999). Representative bureaucracy and distributional equity: Addressing the hard question. Journal of Politics, 61, 1025-1039.

35.

Mladenka

(1989). Blacks and Hispanics in urban politics. American Political Science Review, 83, 165-191.

36.

Morin

Sanchez

Barreto

(2011). Perceptions of competition. In Telles

E. E.

Sawyer

Rivera-Salgado

(Eds.), Just Neighbors? Research on African American and Latino relations in the United States (pp. 96-124). New York, NY: Russell Sage Foundation.

37.

Mucciaroni

(1990). The political failure of employment policy, 1945-1982. Pittsburgh, PA: University of Pittsburgh Press.

38.

Oliver

Wong

(2003). Intergroup prejudice in multiethnic settings. American Journal of Political Science, 47, 567-582.

39.

Oliver

Johnson

J. H.

Jr. (1984). Inter-ethnic conflict in an urban ghetto: The case of blacks and Latinos in Los Angeles. Social Movements, Conflict, and Change, 6, 57-94.

40.

Pitts

(2005). Diversity, representation, and performance: Evidence about race and ethnicity in public organizations. Journal of Public Administration and Theory, 15, 615-631.

41.

Reid

Kerr

Miller

(2003). Glass walls and glass ceilings. New York, NY: Praeger Press.

42.

Rocha

(2007). Cooperation and conflict in multiracial school districts. In Espino

Leal

Meier

(Eds.), Latino politics: Identity, mobilization, and representation (pp. 161-176). Charlottesville: University of Virginia Press.

43.

Rodrigues

H. A.

Segura

G. M.

(2007). A place at the lunch counter: Latinos, African Americans, and the dynamics of American race politics. In Espino

Leal

Meier

(Eds.), Latino politics: Identity, mobilization, and representation (pp. 142-160). Charlottesville: University of Virginia Press.

44.

Rosenbloom

D. H.

(1973). A note on inter-minority competition for federal positions. Public Personnel Management, 2, 43-48.

45.

Sawyer

(2005). Racial politics in multi-ethnic America: Black and Latino identities and coalitions. In Dzidzienyo

Oboler

(Eds.), Neither enemies nor friends: Latinos, Blacks, and Afro-Latinos (pp. 265-280). New York, NY: Palgrave Macmillan.

46.

Segura

Fraga

(2008). Race and the recall: Racial and ethnic polarization in the California recall election. American Journal of Political Science, 52, 421-435.

47.

Selden

S. C.

(1997). The promise of representative bureaucracy. Armonk, NY: M.E. Sharpe.

48.

Selden

S. C.

Brudney

Kellough

J. E.

(1998). Bureaucracy as a representative institution: Toward a reconciliation of bureaucratic government and democratic theory. American Journal of Political Science, 42, 717-744.

49.

Stanley

Niemi

(2010). Vital statistics on American politics, 2009-2010. Washington, DC: CQ Press.

50.

Telles

Rivera-Salgado

Sawyer

Zamora

(2011). Introduction. In Telles

E. E.

Sawyer

Rivera-Salgado

(Eds.), Just neighbors? Research on African American and Latino relations in the United States (pp. 1-33). New York, NY: Russell Sage Foundation.

51.

Telles

Sawyer

Rivera-Salgado

(Eds.). (2011). Just neighbors? Research on African American and Latino relations in the United States. New York, NY: Russell Sage Foundation.

52.

U.S. Bureau of the Census. (2000). Census 2000 school district tabulation. Washington, DC: Government Printing Office.

53.

U.S. Bureau of the Census. (2008). American Community Survey, 2006-2008. Washington, DC: Government Printing Office.

54.

U.S. Equal Employment Opportunity Commission. (Various years). U.S. EEOC Form 168a, Elementary-Secondary Staff Information. Washington, DC: Government Printing Office.

55.

Vaca

(2004). The presumed alliance: The unspoken conflict between Latinos and Blacks and what it means for America. New York, NY: Harper Collins.

56.

Waldinger

(1996). Still the promised city? Cambridge, MA: Harvard University Press.

57.

Warren

Corbett

Stack

J. F.

Jr. (1986). Minority mobilization in an international city: Rivalry and conflict. PS: Political Science & Politics, 19, 626-634.

58.

Welch

Karnig

Eribes

R. A.

(1983). Changes in Hispanic local employment in the Southwest. Western Political Quarterly, 36, 660-673.

59.

Zhao

Lovrich

(2005). Predicting the employment of minority officers in U.S. cities: OLS fixed-effect panel model results for African American and Latino officers for 1993, 1996, and 2000. Journal of Criminal Justice, 33, 377-386.