Fall 22

EQUITY & EXCELLENCE IN EDUCATION, 48(1), 22–35, 2015 Copyright C© University of Massachusetts Amherst College of Education ISSN: 1066-5684 print / 1547-3457 online DOI: 10.1080/10665684.2015.991158

What Does Charter School Mean to You? A Look at Louisiana’s Charter Enrollment by Charter Type

Jandel Crutchfield University of Mississippi

This article examines the intersection of race, socioeconomic status (SES), and charter type/admission practices in Louisiana charter schools. This study used publicly available Department of Education data to compile the sample of charter school demographic information. A one-way Multiple Analysis of Variance (MANOVA) was conducted using race and socioeconomic status as dependent variables and charter type and admission practice as the independent variables. Findings suggest that there are significant differences in African American student enrollment and enrollment of low income students, particularly for Board of Elementary and Secondary Education (BESE) start-ups/conversions and forced conversions/Recovery School District (RSD) charters when compared to all other types. This is also true when comparing selective versus open enrollment charters. Results reveal a higher enrollment pattern of African American and low income students in BESE start-ups/conversions and forced conversion charters (Types 2 and 5) than in local school board start-ups and conversions (Types 1, 3, and 4). This indicates that race and SES are a function of charter type in Louisiana, which adds context to the charter school debate on enrollment patterns.

Since the enactment of No Child Left Behind (NCLB; 2001), the effectiveness of education reforms has become a major focus in education research. Market-based reforms, like charter schools, have seen tremendous flexibility under NCLB (2001). Effectiveness research, however, has failed to keep pace with the changing and diverse natures of charter schools and any resulting outcomes for learners. These outcomes are linked to admission practices, in which different types of charter schools disproportionately enroll high numbers of minority students, which leads to resegregation, isolation, and a lack of access to diverse schools (Frankenberg, Siegel-Hawley, & Wang, 2010). As schools of choice, proponents argue that charters should address resegregation, isolation, and lack of access, issues that some research suggest are exacerbated by charters. While resegregation, isolation, and lack of access have been identified by some as characteristic of charters nationally, the distinction between specific types of charters and enrollment patterns is understudied (Frankenberg, Siegel-Hawley, & Wang, 2010).

This study gives a brief history of the American charter school movement, leading to a discussion of the uniqueness of charter schools in Louisiana. Next, we delineate three major categories of charters across the nation, as well as the five subtypes as they exist in Louisiana. This is followed by a broad discussion of No Child Left Behind and its facilitation of Type 5

Address correspondence to Jandel Crutchfield, University of Mississippi, Department of Social Work, 303 Longstreet Hall, University, MS 38677. E-mail: [email protected]

WHAT DOES CHARTER SCHOOL MEAN TO YOU? 23

charters in the Louisiana Recovery School District. Lastly, using a one-way Multiple Analysis of Variance (MANOVA), an analysis of how enrollment plays out within each type of charter is detailed.

THE NATIONAL CHARTER SCHOOL MOVEMENT

Charter schools are one of the newest waves of market-based reforms that both politicians and scholars argue offer choice and stronger autonomy for schools and teachers (Frankenberg & Lee, 2003). There are currently more than 1.5 million students enrolled in over 5,000 charter schools across 39 states (Allen & Consoletti, 2010). Stemming from standards-based and achievement- based reforms, charters began in the late 1980s with a focus on parental choice in a free market. Minnesota was the first state to enact a law allowing the creation of charter schools (Shanker, 1988), followed by Georgia in 1993. By 1995, 18 more states joined Minnesota by creating charter laws. In the 2010–2011 school year, an additional 465 new charters opened (Allen & Consoletti, 2010). This process was heavily influenced by Allen St. Shaw, former president of the American Federation of Teachers, who proposed that teachers be given increased flexibility and autonomy, which are the hallmarks of charter schools (Shanker, 1988). Louisiana joined the charter school movement in 1995 with its charter school law (Act 192) that allowed eight school districts to open charters. This was followed in 1997 by a law that expanded charter school access to all Louisiana school districts. In 2003, the creation of the Recovery School District expanded access to charters by granting the state school board the power to take over failing public schools and reopen them as charters.

THREE CHARTER CATEGORIES

Scholars who have written about charters across the nation categorize them into three broad categories: start-ups, voluntary conversions, and forced conversions (Arkin & Kowal, 2005). Start-up charters are schools opening for the first time with new facilities and staff. Like-minded individuals in the community may identify a specific need or niche that can be served in a charter school. There may be a start-up charter school focused on engineering for children who plan to be engineers, for example. In contrast, voluntary conversions are pre-existing public schools that become charter schools after the management organization is awarded the charter by the governing board. Voluntary conversions may retain their old building and some of their staff. Similar to start-ups, a central body or group of interested parents, educators, or business leaders agree to change the current school into a charter. The third charter category, forced conversion, has increased in number. Forced conversion charters are pre-existing public schools with chronic low performance. In the process of forced conversions, a state department of education or authorized agency is responsible for establishing a charter contract between themselves and the organization seeking the charter. The majority of charter schools across the nation are start-ups, followed in number by voluntary conversions (Allen & Conseletti, 2010). There are less than eight states that authorize forced conversion charters.

Louisiana leads the nation in terms of the number of forced conversion charters. Identifying the enrollment trends in these specific schools is paramount for a state with a growing number of this

24 CRUTCHFIELD

TABLE 1 Numbers of Charters by Type and Definition

Admission # of Schools Main Type Subtype Practice in Louisiana

Start-Up Type 1 (local school board) Selective 22 Start-Up or Voluntary Conversion Type 2 (BESE) Selective 24 Voluntary Conversion Type 3 (local school board) Selective 11 Start-Up or Voluntary Conversion Type 4 (local school board charter with BESE) Selective 4 Forced Conversion Type 5 (BESE) Open

Enrollment 59

Total 120

Source: LDOE, 2013. BESE = Board of Elementary and Secondary Education.

category of charters. However, Louisiana charters are very unique in that they are categorized by the Department of Education into five subtypes of charters within the three main categories. Table 1 describes the number of schools by charter subtype in Louisiana, including the corresponding main type and admission practice.

Type 1 charters are brand new start-up schools sought by parents or community groups and run by the local school board. They can receive students from anywhere in the district and may enforce selective admission requirements.

Type 2 charters can either be new start-ups or conversions, but are characterized by an agreement between the school and the state board of elementary and secondary education (BESE). The BESE board consists of 11 members, 8 of whom are elected and 3 who are appointed by the Louisiana governor. BESE oversees and implements laws governing K-12 education in Louisiana. Type 2 charters can enroll students from anywhere in the state and can employ admission requirements based on the school’s vision and mission.

Type 3 charters are voluntary conversions granted by a local school board, whereas Type 4 charters are granted by BESE. Type 4 charters may include new start-ups or voluntary conversions. The difference between Type 2 and Type 4 charters lies in the type of Education Management Organization (EMO) that runs the charter. An EMO is an entity that provides management functions for the school, such as professional development, hiring, and community relations. In Type 2 charters, the EMO is a community group or organization, and in Type 4 charters, the EMO is the local school board.

Type 5 charters are the forced conversions of previously low performing public schools. The charter is between the agency and the state’s takeover entity, the Recovery School District. These are the only charters that may not select students based on specific criteria, such as grade point average (GPA) or test history. Type 5 charter schools are considered open enrollment, meaning that any student can attend (Cowen Institute, 2011).

Considering the linkage between charter type and admission practice (selective versus open enrollment) in Louisiana, it is valuable to examine how admission practices influence enrollment patterns in charters. Thus, the charter type not only is linked to enrollment but also to admission requirements.

WHAT DOES CHARTER SCHOOL MEAN TO YOU? 25

Charter schools are considered schools of choice because families exercise personal choice in choosing whether to attend charter schools (Allen & Consoletti, 2010). Unlike traditional public schools, charter student attendance is not legally dictated by student residence (Allen & Consoletti, 2010), although enrollment is sometimes defined by school districting and admission practices. Charter schools possess increased flexibility and autonomy in making decisions about school structure, culture, and standards (Allen & Consoletti, 2010) when compared to traditional public schools. Charters are considered market-based reforms because their characteristics of choice and autonomy are thought to increase competition in the free market of schools, so that lower performing schools will be pressured to improve student performance to compete with charters.

NCLB AND CHARTER SCHOOLS

The most recent educational reform, the 2001 No Child Left Behind (NCLB) Act, served as a facilitator of forced conversion charters, or Type 5 charters in Louisiana. The NCLB Act (2001) built on the Elementary and Secondary Education Act of 1965 and established strin- gent accountability requirements for schools receiving federal funds (Rosenbaum, 2005). Under NCLB (2001), schools receive a yearly performance score based on the national measure of achievement known as adequate yearly progress (AYP) (NCLB, 2001). Schools that receive Title I funding and consistently fail to meet minimum AYP standards must develop plans to address low performance or face a restructuring mandate (NCLB, 2001). If a school fails to meet AYP for five consecutive years, five options of restructuring are available. Available op- tions include reopening the school as a public charter school; replacing all of the staff; allow- ing the state to take over the school; contracting the agency out to an external management organization; or majorly restructuring the school (NCLB, 2001). Of the restructuring options employed by states, the most recent options include charter schools and state takeovers (Steiner, 2005) .

This new latitude has led to the adoption of forced conversions by several states. In 23 states, state law permits forced conversion, but only Maryland, Rhode Island, Louisiana, Alabama, Virginia, Tennessee, and Massachusetts have implemented it (Steiner, 2005). As yet, however, no comprehensive studies examine the nature of forced conversion charter schools.

LOUISIANA’S CHARTER SCHOOLS

Louisiana leads the nation in the number of forced conversion charters in operation. Charter schools in Louisiana are having a large and growing influence, not only in the state but also across the nation. Charters are distinguished by the 5 subtypes. Over 115 charter schools are operational in Louisiana and have an enrollment of nearly 60,000 students as of Fall 2013 (Louisiana Department of Education [LDOE], 2013). Since 1995 there has been an exponential growth in Louisiana charters; before Hurricane Katrina (2005), there were just 17 charter schools in Louisiana (LDOE, 2013).

The growing influence of Louisiana charters is due mainly to the influence of the Recovery School District (RSD) in New Orleans, which oversees more than 50 charter schools as part of

26 CRUTCHFIELD

its turnaround model. The RSD model is serving as a model for the nation. As recently as 2012, the Louisiana Recovery School District was hailed by the Brown Institute as having a choice and competition index letter grade of A. This was the only A awarded to any school system in the country. Researchers at the Thomas Fordham Institute produced a similar study in which they summarized the overall academic achievement gains for RSD schools in New Orleans (Smith, 2012). Press releases from the Louisiana Department of Education tout the “incredible growth” (para. 7). for New Orleans Recovery School District students (RSD, 2013). As recently as 2011, the Ohio State Board of Education chose to model its school reform efforts after the Louisiana RSD (Smith, 2012). Several other reports tout the school improvement efforts of the Louisiana RSD and how these efforts can and should be used as a model for other states in addressing educational equality in their public school systems (Hill & Murphy, 2011). RSD charters, however, are not the traditional start-up or voluntary conversions that one might think of when imagining charter schools. They are forced conversions (Type 5), of previously low performing schools that were taken over by the RSD and chartered out to an EMO at the state’s discretion. The RSD consists only of Type 5 charter schools in addition to several non-charter schools. The RSD charters, or Type 5 charters, compose the largest number of charters in Louisiana. While RSD reports have used snapshots of school performance scores (Buras, 2012), the current study goes further than previous work and is the first to examine the variation in student enrollment patterns by charter school type in Louisiana. In a recent presidential mandate, President Barack Obama announced the ability for states to apply for and obtain waivers from federal NCLB policies, due to stringent mandates that states claim have hindered real progress. Without overwhelming supporting evidence to suggest the efficacy of NCLB, several states have opted out of NCLB policies. Louisiana has not chosen to do so.

One area in which Louisiana chooses to continue NCLB policies is in charter operation. This analysis sheds a small light on some of these policies, including the intersections among charter type, admission practices, and enrollment of African American students and low income students. According to Tulane’s Cowen Institute (2011), Louisiana charters have a larger average enrollment of African American students and students eligible for free and reduced-fee lunch than the average charter school across the nation (Cowen Institute, 2011). Historically underserved groups in the traditional public schools have been disproportionately enrolled in Type 5 charters. Thus, when examining the opportunities for access and diversity that charters promise, data describing the enrollment of African American students and free/reduced lunch students in charters must be delineated.

CURRENT LITERATURE ABOUT CHARTER SCHOOL ENROLLMENT

Researchers generally agree that, nationally, charter schools enroll a disproportionately high number of non-Caucasian students, particularly African American and low income students (Frankeburg & Lee, 2003; Renzulli, 2006). There are dissenting voices, however, such as those of Zimmer, Gill, Booker, Lavertu, Sass, and Witte (2009), who examined long-term effects of charter schools in eight states. Zimmer et al. (2009) reported that charter school students who transfer into charters from traditional public schools share similar demographics to those students in surrounding traditional public schools, suggesting that the charters may not be more segregated than traditional public schools.

WHAT DOES CHARTER SCHOOL MEAN TO YOU? 27

In contrast to Zimmer and colleagues (2009), authors such as Renzulli (2006) argue that enrollment of African Americans in charters is higher than enrollment of all other ethnic groups. Renzulli goes further to determine that the enrollment is higher particularly in already segregated school districts, which is often characteristic of large urban districts. Louisiana, for example, has 40 of its 69 school districts still operating under a 40-year-old desegregation order for its public school systems, including its two largest urban districts. From Renzulli’s work, one would then expect higher charter enrollments of African Americans in the two largest urban districts in Louisiana.

Likewise, in their edited volume of essays, Lubienski and Weitzel (2010) reviewed the history and outcomes of the charter movement. Evidence from their work also suggests that charter school students of minority ethnic groups are highly segregated in charter schools, compared to the traditional public schools in the same school districts. They argue that this segregation is particularly disturbing because of the student outcomes associated with attending schools with high percentages of racial minorities, including the fact that schools with high percentages of racial minorities are 22% less likely to be high performing than more affluent schools. Similar trends are present for high poverty schools. These findings are indicative of small improvements in competition between charters and traditional public schools, as well as a lack of the promised increase in empowerment for charter school teachers outlined by proponents like Shanker (1988).

Lacireno-Paquet (2006) identified the relationship between charter school enrollment of low income students belonging to minority ethnic groups and the organizational characteristics of schools and state policies governing charter schools. She examined whether EMOs, school history, and school characteristics of charter schools influence the enrollment of low income students belonging to minority ethnic groups in charter schools. The study population was drawn from the 1999–2000 school and staffing survey of public charter schools. Her ordinary least squares regression demonstrated that EMOs can have positive or negative effects on enrollment, but that enrollment also depends on state policy characteristics. The coefficient indicating whether the school was run by a large EMO in an urban area was (20.03). Thus, schools operated by a large EMO in an urban location had a large positive and significant effect, meaning that these schools enrolled higher percentages of free and reduced-fee lunch students. Similarly, Lacireno- Paquet also found that small-EMO operated schools enroll about 10% more low income students, compared to similar independent schools. Lastly, she found that schools in states with local school districts as the only charter school authorizer also served much lower percentages of free and reduced-fee lunch students. Miron, Urschel, Mathis, & Tornquist (2010) also considered the type of EMO in their analysis of charter school enrollment practices. The results of their analysis of variance revealed that charters managed by EMOs were more highly racially (1.47 percentage points) and economically (1.95 percentage points) segregated than traditional public schools in the surrounding districts. This suggests that the type of charter (EMO managed versus non-EMO managed) was related to enrollment patterns in this study.

Current Literature: Influence of Charter School Type

A comprehensive study done by the RAND Corporation (Zimmer & Buddin, 2005) examined all California charters in comparison to a matched group of traditional public schools. The study provided a description of the charter schools’ demographic composition and achievement as

28 CRUTCHFIELD

compared to traditional public school students. Zimmer and Buddin (2005) argued that a simple comparison of charters versus traditional public schools is misleading because charter school students may not be a representative sample of public school students. Instead, they argued that comparisons of a charter and a traditional public school should be made after identifying the specific category of charter. Their analysis demonstrated that elementary conversion charters had a small positive effect on reading and a small negative effect on math as measured by standardized test scores on the California Standard 9 (CS9) achievement test when compared to traditional public schools. They found that start-up charter school students scored five to seven percentage points lower in reading and math than traditional public school students at the elementary level. In high schools, start-up charter students scored two percentage points higher in reading and math than conversion charter schools, but they scored lower, overall on the CS9, than traditional public school students. These results demonstrate the complexity of student performance across different types of charters.

Another study, conducted by Allen and Consoletti (2010), delineated between charter types but defined the categories as those charters associated with a public school district versus those unas- sociated with a public school district. They also found several significant factors that influenced achievement, including which agency granted the charters. Those granted charters by a state operating agency—such as forced conversions in Louisiana—and other school characteristics, such as freedom to choose instructional curriculum and and increase or decrease in instructional time explained 11% of the variance in this study. Concerning the policy of allowing multiple authorizers to manage charter schools, it was found that 78% of charter schools are located within states that allow multiple authorizers, which other research has demonstrated increases student achievement. This research suggests that charters granted by state entities, like the Recovery School District (Type 5) or the BESE board (Type 2), only account for a small percentage of the increase in achievement.

A final study also conducted by the RAND Corporation (2004) examined all California charters to determine basic information about who attends charter schools and how student achievement is affected by attendance in the charter schools. The researchers used case studies of single schools and secondary data on students, teachers, and schools to draw conclusions. Results demonstrated that in the Los Angeles Unified District, elementary school reading scores for voluntary conversion and start-up schools were not statistically different from scores for conventional public schools. Math scores for conversion schools were 0.6 percentile point above the scores for conventional public schools, but start-up schools scored 1.5 percentile points below conventional public schools. In the San Diego City Unified District, the gap between voluntary conversion and start-up schools was wider. In reading, voluntary conversion students scored 2.2 percentile points higher than comparable students in traditional public schools students, whereas start-up students scored 1.7 percentile points lower than comparable students in traditional public school.

Summarizing across studies, authors generally found comparable scores for charter schools relative to conventional public schools. Only when charter schools were broken down by charter type did significant differences in enrollment and achievement appear (Allen & Consoletti, 2010; Lacireno-Paquet, 2006; Zimmer & Buddin, 2005). Start-ups scored higher than traditional public schools in elementary and higher than conversion charters in high school.

A study done by Arkin and Kowal (2005) delves deeper into the nature of conversion char- ters and deals briefly with forced conversion charters. The study reviews methods implemented

WHAT DOES CHARTER SCHOOL MEAN TO YOU? 29

across states when closing a public school and re-opening it as a charter school. They concluded that forced conversion charters are new phenomena but face difficult challenges in terms of performance. In their study, practitioners provided data through telephone interviews. Results demonstrated that governance structure, environmental factors, school-level governance, lead- ership, and organizational factors all present significant challenges to those attempting to close public schools and re-open them as charters.

Limitations of Charter School Literature

Forced conversion charters are characteristically different from start-up charters and voluntary conversion charters (Ziebarth, 2004). However, there are a severely limited number of studies that examine the nature of forced conversion charters (Type 5) singularly. This lack masks the true influences of charter types on achievement. Though charter school research is plentiful, it is narrow in scope and often excludes a classification of charters by type.

The state takeover literature indirectly offers a picture of what characteristics may define forced conversion charter schools. States that exercise takeover power commonly target schools that are failing, have high racial minority enrollments and low income students, and are already segregated. Offering charters and choice is a delicate balance for these schools. Enrollment characteristics are compounded by the stigmatization inherent in the state takeover process due to being labeled as underperforming. The nature of the state takeover process curbs the ability of schools to operate with autonomy and, instead, encourages heavy bureaucratic influence. Thus, forced conversion charters often lack the autonomy and choice inherent in theoretical assumptions of charter school success. Descriptive data in the literature must be extended beyond start-up (Types 1, 3, and 4) and voluntary conversions (Types 2, 3, andr 4) for states with forced conversions, like Louisiana.

The single study examining outcomes for Louisiana charter schools was conducted by the Center for Research on Education Outcomes (2009) and fails to make distinctions among charter types. It presented results for 16 states, including Louisiana. The authors utilized available standardized testing data from the years 2001–2008 and provide a strong longitudinal picture of charter school student performance compared with virtual student matches of traditional public school students. Their student sample from Louisiana included 34,479 charter school students with an 85% virtual student match. Their outcome variable of interest was academic growth on standardized tests in reading and math for this amount of time. The authors utilized ordinary least squares (OLS) regression and found that Louisiana charter school students, across all types of charters, as a whole outperform their traditional public school counterparts in both reading and math. They found that African Americans (.13 pts.) and Hispanics (.09 pts.) in charters do significantly better than traditional public school students in math. Students in poverty also do better than their counterparts in public schools in reading (.05 pts) and math (.04 pts). Special education students are not significantly outperforming their counterparts. Retained students do worse when in charter schools than traditional public school students. These relative growth measures are seen only in students enrolled for longer periods of time in charters. This study only examined charters that were operational up to the year 2008. Most of the forced conversion charters in the state became operational that year, leaving this study with only one year of data for most conversion charters. Thus the results of this study cannot be generalized to forced conversion charters in Louisiana.

30 CRUTCHFIELD

METHODS

This exploratory research study aimed to examine the nature of the differences in the enrollment of African American students and low income students. The central research questions is:

1. Do race and socioeconomic status vary as a function of charter school type? A secondary research question addresses the nature of differences in enrollment of African American students and low income students according to charter admission policies. The second research question is:

2. Do race and socioeconomic status vary as a function of charter school admission practice?

Sample

All of the charter schools in the state of Louisiana with demographic information on students were included in this study. As of the Fall 2013 school year, the total population of charters is currently 120, but for each independent and dependent variable, there were a total of 88 schools that had all the data were included in the sample. Others were opened as recently as 2012 and 2013 and did not yet have student data available. All data were downloaded from publicly available records housed online at the Louisiana Department of Education.

Independent Variables

“Charter type” was the main independent variable under investigation in this study. There are five charter types in Louisiana as outlined in Table 1.

The second independent variable was “admission practices.” Charter type was linked to ad- mission practice as outlined in Table 1. All charters in Louisiana, with the exception of Type 5 charters, are allowed to admit students based on whether they meet particular criteria as it fits with the school’s mission and purpose.

Dependent Variables

Race is designated by the ethnic group to which a student belongs. The majority of the public school students in Louisiana are African American, including those in the Louisiana Recovery School District (LDOE, 2013). This study aims to identify whether the educational plight of the African American charter students in Louisiana is similar to the pattern of the rest of the nation. As discussed in this article, minorities experience far higher enrollment in charters than their white counterparts (National Center for Education Statistics, 2004). “Free and reduced-fee lunch status” is a variable often cited in school effects literature. It is a national measure of poverty that correlates to low scores and outcomes for students (McKinsey & Company, 2009). The majority (90–100%) of African American students in Louisiana’s public schools receive free and/or reduced-fee lunch (LDOE, 2013). The percentage of racial minority students in the school

WHAT DOES CHARTER SCHOOL MEAN TO YOU? 31

TABLE 2 Descriptive Statistics by Charter Type

Free/Reduced Lunch African American Enrollment

Charter Type Mean SD Mean SD

1 70.0 22.1 86.6 12.4 2 63.4 23.6 61.9 35.1 3 63.1 27.0 76.9 22.0 4 73.5 3.4 72.0 41.2 5 84.5 11.8 95.9 9.9

is linked to achievement. Those schools with higher racial minority percentages perform lower than those with lower minority percentages (Carey, 2005).

Data Quality

All of the data were screened for missing data, outliers, and normality, using scatterplots and histograms to examine the shape of the distribution. Using the Doornik Hansen statistic, which is an alternative to the Shapiro-Wilk test (Doornik & Hansen, 2008), results demonstrated that there was both univariate and multivariate normality distributed within groups, but not across groups, which is expected in Mulivariate Analysis of Variance (MANOVA). The univariate outliers were first sought within each group and then separately. For multivariate outliers, the mahalanobis distance (a measure that accounts for the variance and covariance of variables) was first computed, and then scores were screened. There were minimal missing data, although 32 schools did not have current demographic data on the outcomes. These 32 schools were, therefore, excluded from the analysis. This data was missing because the schools were recently opened and did not have current information.

Table 2 lists descriptive summary characteristics of the sample of charter schools on the outcome variables. Type 5 charters have the largest means for both outcomes (84.5%; 95.9%) with small standard deviations.

Inferential Statistics

MANOVA is particularly suited for testing differences between group means on dependent variables (Stevens, 2002). It is an extension of Analysis of Variance (ANOVA) and can assess the main effects on more than one outcome. ANOVA is used to analyze differences in group means with one independent variable. The current study sought to determine whether means on enrollment of African American students and students who qualify for free and reduced-fee lunch are significantly different from zero. MANOVA also controls for potential inflation of Type 1 error seen in ANOVA.

32 CRUTCHFIELD

RESULTS

A one way MANOVA yielded a significant main effect for charter type and for admission practices: Pillai’s trace 0.40 df 8.0 F(8, 190) = 6.48 p < .000 (Stevens, 2002). Pillai’s trace calculates the amount of variation in enrollment of African American and low income students accounted for by the greatest separation of charter types. Thus, we concluded that charter enrollment of African American and low income students was significantly dependent on which type of charter these students attended. This result indicates a difference in race and socioeconomic status by charter type. Both research questions can be answered with a significant result: Race and socioeconomic status do vary as a function of charter type, with charter Types 2 and 5 showing a higher enrollment of African American and low socioeconomic status students than other charters. Also, these results demonstrate that race and socioeconomic status do vary as a function of charter school admission practice.

Given the significance of the overall test, the univariate main effects were examined for socioeconomic status and race. Significant univariate effects for charter type were revealed for race, F(4) = 7.42, p < .000, power = 1.00 and socioeconomic status, F(4) = 7.42, p < .000, power = 1.0.

Pairwise comparisons were conducted for significant charter types. Type 5 charters had a 14-points higher score on the outcomes than Type 1 charters; a 20-point higher score than Type 2 charters; a 21-point higher score than Type 3 charters; and a 10-point higher score on the outcomes than Type 4 charters. This result clearly demonstrates the strength of the difference according to Type 5 charters. Type 5 charters served 14% higher African American and low income students than any other charter subtypes. Thus, Type 5 charters stand out as characteristically different than other Louisiana charters on these dependent variables and reflect the potential for wide variation among similar types of charters nationally (Frankenberg, Siegel-Hawley, & Wang, 2010).

DISCUSSION

The goal of this research was to examine whether race and SES vary as a function of charter type and/or as a function of charter school admission practice in Louisiana charters. Using all of the charter schools with publicly available data on the variables of interest, a one-way MANOVA yielded a significant effect of charter type/admission practice on both outcomes. Given these findings I can conclude that the type of charter influences enrollment patterns. Patterns of enrollment and resegregation of racial minority students have been at the forefront of charter literature since the inception of charter schools on the national level. Charter school resegregation has been highlighted as a civil rights issue within the larger context of the United States’ public schools’ pattern of segregation. Not only are the majority of public schools segregated, but charter schools also have high segregation rates. It is well documented in the literature that schools with higher concentrations of low income, racial minority students face negative outcomes for students that range from less access to mainstream education to higher dropout rates; issues that charters are said to be able to combat (Frankenberg, Siegel-Hawley, & Wang, 2010). This research goes beyond previous research to isolate these high concentration patterns by type of charter. Not only do these results shed light on charter type and the influence this has on enrollment, but also on admission practices and enrollment patterns. Because the type of charter is a direct predictor of

WHAT DOES CHARTER SCHOOL MEAN TO YOU? 33

admission practices in Louisiana, these results also demonstrate that charters without the ability to engage in selective enrollment have a higher amount of African American students and low income students.

CONCLUSION AND RECOMMENDATIONS

The Recovery School District has opened three new Type 5 charters this school year alone and promises to continue. In 2011, representatives from New Jersey, Oklahoma, and Tennessee came to Louisiana to receive briefings and participate in workshops given by the Louisiana Department of Education (LDOE), including RSD staff (Hill & Murphy, 2011). The work- shops stressed the success that Louisiana has seen in closing and transforming low performing schools through the RSD. Indeed, Hill and Murphy (2011) assert that the Louisiana RSD is “a vital asset to the state” (p. 2). The course of the New Orleans style takeover model is not only being spread to Indiana, Michigan, and Tennessee, but also to places such as Haiti, Chile, and Venezuela by the former RSD superintendent, Paul Vallas (Buras, 2012). With the spread of the RSD model, there is an imperative that the LDOE conduct sound research to demon- strate the ability of state takeover to improve student achievement and decrease the risk of dropout.

While it is informative to know that Louisiana has an even higher enrollment of racial minority and low income students than other states, it also is valuable to know how this enrollment breaks down by charter type and admission practice. Only in Type 2 and Type 5 charters is this over- enrollment an issue. When discussing any outcomes regarding Louisiana charter schools, charter type must be delineated.

Both proponents and opponents of choice agree that enrollment in charters is a key issue in the charter debate. Further research on outcomes of achievement should be delineated by charter type/admission practice. The California Charter School Association has made a distinction between its start-ups and voluntary conversions in terms of enrollment and achievement and can, therefore, make more pointed policy decisions (Zimmer & Buddin, 2005). Charter policy that attempts to address segregation, privilege, and access to schools of choice must consider the specific types of charter schools to target.

Given that both enrollment of African American students and low socioeconomic students can be linked to the specific charter type and admission practice, forced conversions and the Louisiana Subtype 5, I have two final recommendations. First, policy makers and education re- searchers must consider the economic, social, and racial impact of charters according to their distinguishing characteristics rather than their overarching similarities. For example, rather than reporting performance results for charter schools in Louisiana as one monolithic group, documen- tation of outcomes, such as graduation rates and standardized test scores, should be chronicled according to charter type. This could reveal differences that may exist similar to the findings in this study. Next, due to the persistence of the educational equality gap, it is imperative that all aspects of new reforms, including charters and their subtypes, are included in the discussion so that a greater equality divide can be avoided. Although there are many factors that contribute to the vast differences and inequality in charter student outcomes addressed in the literature, space needs to be made for the consideration of how charter types exacerbate or improve these

34 CRUTCHFIELD

differences/challenges. As charter schools increase in number across the United States, evidence from the literature on the influence of charter type must also increase.

REFERENCES

Acts 1995, No. 192, §1, eff. June 14, 1995. Allen, J., & Consoletti, A. (Eds.). (2010). Annual survey of America’s charter schools. Washington, DC: Center for

Education Reform. Retrieved from https://www.edreform.com/wp-content/uploads/2011/09/CER Charter Survey 2010.pdf

Arkin, M., & Kowal, J. (2005) Reopening as a charter school. Washington, DC: Center for Comprehensive Education Reform. Retrieved from http://www.centerforcsri.org/pubs/restructuring/KnowledgeIssues2Chartering.pdf

Buras, K. (2012). Review of the Louisiana Recovery School District: Lessons for the buckeye state. Boulder, CO: National Education Policy Center. Retrieved from http://nepc.colorado.edu/files/TTR-NOLAOhio-Fordham 0.pdf

Carey, K. (2005). The funding gap 2004: Many states still shortchange low-income and minority students. Washington, DC: Education Trust.

Center for Research on Education Outcomes. (2009). Multiple choice: Charter school performance in 16 states. Standard, CA: Author. Retrieved from http://credo.stanford.edu/reports/MULTIPLE CHOICE CREDO.pdf

Cowen Institute. (2011). Charter schools in Louisiana. Retrieved from http://www.coweninstitute.com/wp-content/ uploads/2010/03/1-CharterSchoolsinLA 000.pdf

Doornik, J. A., & Hansen, H. (2008). An omnibus test for univariate and multivariate normality. Oxford Bulletin of Economics and Statistics, 70, 927–939.

Elementary and Secondary Education Act. (1965). P.L. 89–10. Frankenberg, E., & Lee, C. (2003). Charter schools and race: A lost opportunity for integrated education. Education

Policy Analysis Archives, 11. Retrieved from http://epaa.asu.edu/epaa/v11n32/ Frankenberg, E., Siegel-Hawley, G., & Wang, J. (2010). Choice without equity: Charter school segregation and the need

for civil rights standards. Los Angeles, CA: The Civil Rights Project. Retrieved from http://www.schoolinfosystem. org/pdf/2010/02/CRP-Choices-Without-Equity-report.pdf

Hill, P., & Murphy, P. J. (2011). On Recovery School Districts and stronger state education agencies: Lessons from Louisiana. Seattle, WA: Center on Reinventing Public Education. Retrieved from http://www.crpe.org/sites/default/ files/WP States Recovery Jun11 0.pdf

Lacireno-Paquet, N. (2006). Charter school enrollments in context: An exploration of organization and policy influences. Peabody Journal of Education, 81(1), 79–102.

Louisiana Department of Education. (2013). Charter schools. Retrieved from http://www.louisianabelieves.com/schools/ charter-schools

Lubienski, C. & Weitzel, P. (2010). The charter school experiment: Expectations, evidence, and implications. Cambridge, MA: Harvard Education Publishing Group.

McKinsey & Company. (2009). The economic impact of the achievement gap in America’s schools. Washington, DC: Author. Retrieved from http://mckinseyonsociety.com/downloads/reports/Education/achievement gap report. pdf

Miron, G., Urschel, J. L., Mathis, W. J., & Tornquist, E. (2010). Schools without diversity: Education management organizations, charter schools and the demographic stratification of the american school system. Boulder, CO: Education and the Public Interest Center & Education Policy Research Unit. Retrieved from http://epicpolicy.org/ publication/schools-without-diversity

National Center for Education Statistics. (2004). The condition of education. Washington, DC: Institute of Education Sciences. Retrieved from http://nces.ed.gov/pubs2004/2004077.pdf

Recovery School District. (2013, October 24). New Orleans share of Louisiana failing schools falls from 44 percent in 2005 to only 8.5 in 2013. Retrieved from http://rsdla.net/apps/news/show news.jsp?REC ID=287113&id=0

Renzulli, L. A. (2006). District segregation, race legislation, and black enrollment in charter schools. Social Science Quarterly, 87(3), 618–637.

Rosenbaum, S. (2005). A renewed IDEA and the need for more ardent advocacy. Human Rights, 32(4), 3–4, 6–7. Shanker, A. (1988). Restructuring our schools. Peabody Journal of Education, 65(3), 88–100.

WHAT DOES CHARTER SCHOOL MEAN TO YOU? 35

Smith, N. (2012). The Louisiana Recovery School District: Lessons for the buckeye state. Columbus, OH: The Thomas Fordham Institute. Retrieved from http://files.eric.ed.gov/fulltext/ED528943.pdf

Steiner, L. (2005). State takeovers of individual schools. Washington, DC: Learning Points Associates. Stevens, J. P. (2002). Applied multivariate statistics for the social sciences (4th ed.). Mahwah, NJ: Lawrence Erlbaum

Associates. Ziebarth, T. M. (2004). Closing low performing schools and re-opening them as charter schools: The role of the state.

Denver, CO: Education Commission of the States. Zimmer, R., & Buddin, R. (2005). Student achievement in charter schools: A complex picture. Journal of Policy Analysis

and Management, 24(2), 351–371. Zimmer, R., Gill, B., Booker, K., Lavertu, S., Sass, T. R., & Witte, J. (2009). Charter schools in eight states: Effects

on achievement, attainment, integration, and competition. Santa Monica, CA: RAND Corporation. Retrieved from http://www.rand.org/content/dam/rand/pubs/monographs/2009/RAND MG869.pdf

Jandel Crutchfield is a visiting assistant professor at the University of Mississippi’s Depart- ment of Social Work where she engages in education reform research. She has many years of professional experience as a school social worker in traditional public schools, charter schools, and state takeover schools.

Copyright of Equity & Excellence in Education is the property of Routledge and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.