xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam a22 u 4500
controlfield tag 008 c20059999azu 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E11-00472
Educational policy analysis archives.
n Vol. 13, no. 50 (December 14, 2005).
Tempe, Ariz. :
b Arizona State University ;
Tampa, Fla. :
University of South Florida.
c December 14, 2005
Charter school type matters when examining funding and facilities : evidence from California / Cathy Krop [and] Ron Zimmer.
Arizona State University.
University of South Florida.
t Education Policy Analysis Archives (EPAA)
xml version 1.0 encoding UTF-8 standalone no
mods:mods xmlns:mods http:www.loc.govmodsv3 xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govmodsv3mods-3-1.xsd
mods:relatedItem type host
mods:identifier issn 1068-2341mods:part
mods:detail volume mods:number 13issue 50series Year mods:caption 20052005Month December12Day 1414mods:originInfo mods:dateIssued iso8601 2005-12-14
Readers are free to copy display, and distribute this article, as long as the work is attributed to the author(s) and Education Policy Analysis Archives, it is distributed for noncommercial purposes only, and no alte ration or transformation is made in the work. More details of this Creative Commons license are available at http:/ /creativecommons.org/licen ses/by-nc-nd/2.5/. All other uses must be approved by the author(s) or EPAA EPAA is published jointly by the Colleges of Education at Arizona State University and the Universi ty of South Florida. Articles are indexed by H.W. Wilson & Co. Send commentary to Casey Cobb (c firstname.lastname@example.org) and errata notes to Sherman Dorn (epaa-editor@s hermandorn.com). EDUCATION POLICY ANALYSIS ARCHIVES A peer-reviewed scholarly journal Editor: Sherman Dorn College of Education University of South Florida Volume 13 Number 50 December 14, 2005 ISSN 1068Â–2341 Charter School Type Matters When Examining Funding and Facilities: Evidence From California1 Cathy Krop Ron Zimmer RAND Corporation Citation: Krop, C., & Zimmer, R. (2005). Ch arter school type matters when examining funding and facilities: Evidence from California. Education Policy Analysis Archives, 13 (50). Retrieved [date] from h ttp://epaa.asu.edu/epaa/v13n50/. Abstract Currently, charter schools represent one of the fastest growing movements of educational reform The first charter school opened in 1992 and there are now over 3,400 charter schools nationwide. Despite this growth, we are only beginning to learn about the performance and operation of these schools. This article adds to our knowledge of ch arter schools both by examining the finances of charter schools in California, which has more charter students than any other state, and by highlighting their fiscal challenges. Usin g survey data of Ca lifornia charter and conventional public schools, the results sug gest that the degree charter schools are struggling with resources and facilitie s depends upon ch arter school type. Keywords: school choi ce; school finance. 1 This paper was funded by the Calif ornia Legislative AnalystÂ’s Office.
Education Policy Analysis Archives Vol. 13 No. 50 2 Introduction In 1992, California became the second state to enact legislation that created charter schools. Charter schools are publicly funded schools of c hoice that operate autonomously outside the direct control of conventional school districts. Instead, they operate under the authority of quasi-contracts, or charters, granted generally by a public body. These schools are designed to provide greater educational choice to families, reduce bureaucr atic constraints on educators, and provide competitive pressure to induce improvement in conventional public schools while remaining publicly accountable and having wide ranging appr oaches to educating students. In most states, charter schools can be converted from conventional public schools or can be started from scratch, known as Â“start-upÂ” schools. Some charter schools are as small as a half-dozen students while others can be as large as 10,000 students. Some charter sc hools are part of larger for-profit or non-profit Educational Management Organization while others run independently of any larger organization. Much of how charter schools operate is a function of in dividual state charter laws. In total, 40 states plus the District of Columbia have charter laws wi th nearly one million students attending over 3,400 charter schools, and nearly a fourth of thos e students attending charter schools in California (Center for Education Reform, 2005). As charter laws have spread across states, so me have voiced opposition to this growth. One concern is that charter schools drain public sc hools of much needed resources leaving school districts in desperate fiscal conditions. Critics ar gue that ultimately the loss of revenue created by transferring student to charter schools forces many districts to close schools, layoff teachers, and generally feel a financial pinch in the overa ll operation of schools. Meanwhile, charter school proponents argue that charter school s are not playing on a level playing field financially and note that charter schools need additional revenues to fully educate students (Finn, Mammo, & Vanourek, 2000). For example, one key source of inequity be tween public school and charter school funding is often the lack of facility funding for charter schools (RPP International, 2000; Sugarman, 2002; Thomas B. Fordham Institute, 2005). In addition, other factors, such as a lack of participation in specific state and federal funding programs, may also affect charter school funding. The question of charter school finances is an important one, particularly given the rapid growth in charter schools across the nation. Califor nia has been at the forefront of the charter movement and presents an interesting site for exam ining charter school funding. In 1992, California was the second state to pass charter school legislat ion and nowhere is the growth in charter schools more apparent than in California. In addition, char ter school financing in California is influenced by the extensive public school finance reforms instituted over time. CaliforniaÂ’s current public school finance system evolved through a combination of court decisions, legislative actions, voter-approved initiatives, and government regulation. The transformation began in 1971 when the California Supreme Court ruled in Serrano v. Priest that differences in property tax revenue per pupil across districts could not be related to differences in the property wealth of those districts. Since that ru ling, California has moved to equalize revenues among local school districts. Proposition 13, passed by California voters in 1978, helped shape the new system by imposing a 1% limit on general-pu rpose property tax rates. California quickly moved from a system in which each school district deter mined its own revenue through local property taxes to a system in which school revenues are controlle d at the state level, with school districts heavily dependent on state aid for support. This effort to create greater equity within Ca lifornia raises questions of whether charter schools are being equitably funded. Against this backdrop, we examine charter school funding in California. We begin by discussing existing charter school research, much of which has focused on
Charter School Type Matters When Examining Funding and Facilities 3 student achievement effects of charter schools with little attention given to charter school finances. We then describe some of the general characteristics and trends of CaliforniaÂ’ s charter schools, their funding, and the research methodology used to ex amine charter school finances. Finally, we evaluate charter school funding sources and the provision of charter school facilities. The analysis sheds light on the fiscal challenges of charter schools. Literature While the debate over charter schools has spu rred research, most of it has focused on student achievement and the racial/ethnic integra tion effect of charter schools with mixed results. Much of the current student achievement literat ure uses school-level or cross sectional analysis, which masks changes over time in the schoolÂ’s popu lation of students and variation of performance across different subjects and grades and cannot fa ctor out the various non-school forces at work (Carnoy, Jackson, Mishell, & Rothstein, 2005; Greene, Forster, & Winters, 2003; Hoxby, 2004; Miron, Nelson, & Risley, 2002; Nelson, Rosenberg & Van Meter, 2004; Ro gosa, 2003). The best student achievement research currently uses longitud inally-linked student-level data, which provides the ability to track students over time and creates a mechanism for controlling for differences among student who choose to go to charter schools and th ose who do not. Currently, there is a handful of such studies focusing on individual states, incl uding Arizona (Solmon, Paark, & Garcia, 1999), California (Zimmer et al, 2003), Flor ida (Sass, 2005), North Carolina (Bifulco & Ladd, in press), and Texas (Booker Gilpatric, Gronberg, & Jansen, 20 04; Gronberg & Jansen, 2001; Hanuskek, Kain, & Rivkin., 2002). These studies have not created a consensus but genera lly have suggested that charter schools have either small positive or negative effect s, which vary by state. For our current research, the most relevant research is of California. Zimme r et al. found that charter schoolsÂ’ performance is on par with conventional public schools and that the performance of charter schools varies by charter type.2 While student achievement analyses have recei ved the bulk of attention, there is also a growing literature that has examined the effect ch arter schools have on racial/ethnic integration of students. Although a number of studies have ex amined the racial/ethnic representativeness of charter schools, most of these studies compared snapshots of charter-school enrollments with the average enrollments of their surrounding districts and states (Frankenberg & Lee, 2003; Miron & Nelson, 2002; Powell, Blackorby, Marsh, Finnega n, & Anderson, 1997; RPP International, 2000; Zimmer et al., 2003). This approach, while provid ing some important insights, cannot determine whether individual students are moving from hetergenous to homogenous schools or vice versa. One study that has used student-level data (Biful co & Ladd, 2005) examined migration patterns of North Carolina students of differ ent race/ethnicity as they choose to go to charter schools. The study suggests that black students are more likely to go to charter schools with higher concentration of black students than their exiting school. Thes e results raise some concern that charter schools may be creating greater segregation among studen ts and suggest a closer examination of these schools. Despite the growth of the research on stud ent achievement and integration, the actual operation, including the critical issue of finances, has not received much attention. Two reports that provide some initial insights into the revenues and expenditures of charter schools relative to public school districts are Miron and Nelson (2002) and Mi ron, Nelson, and Risley (2002). In the first of 2 Some of these results are also pr esented in Buddin and Zimmer (2005).
Education Policy Analysis Archives Vol. 13 No. 50 4 these two reports, Miron and Nels on (2002) compare instructional expe nditures of Michigan charter schools and school districts within the state and fo und that charter schools spend more money than conventional public schools on administration. Th is analysis, however, has the obvious drawback that school districts have a different organizationa l and fiscal structure than an individual school, which affects costs. As part of that study, the au thors also conduct in-depth case studies of four schools that provide important insights, including that charter schools focus their attention on less costly students. But because of the limited sample, the results are not generalizable to a statewide or nationwide population of charter sc hools. Miron, Nelson, and Risl ey (2002) examine charter schools in Pennsylvania using data provided by Standard and Poors and found that charter schools receive $750 less per pupil than their host districts, a deficit for which charter schools partially compensate by collecting private resources. A third report by the American Federation of Teachers (Nelson, 2000) examines state financing mechanisms and simulates revenue allocated to charter schools based on the types of students enrolled in different types of charter scho ols. The analysis assumes that charter schools take advantage of programs available to them. The repor t shows substantial differences in the allocation of resources among charter schools based on state policies and student makeup of these schools. More recently, the Thomas B. Fordham Foundation (2005) released a report that examines charter school finances over 16 states and the Di strict of Columbia, which collectively enroll 84% of charter school students nationwide. For their analysis, the authors requested data from states and districts and in some cases received only partial information. However, their examination is the broadest to date and provides some important insi ghts. Overall, the study found that charter schools receive $1,801 less per pupil than conventional pu blic schools. In California, the disparity was greater, with charter schools r eceiving $2,223 less per pupil. The authors argued that these differences are driven primarily by the lack of fa cility funding, but other factors may contribute to the shortfall. We argue one such factor is the lack of participation in state and federal categorical programs. In this study, we build on the previous research but are able to make comparisons between charter and conventional schools through surveys to find out both their level of participation in government funding programs and their level of priv ate support, as well as to analyze the challenges of acquiring facilities. California Charter Schools Below, we describe some general characteristics and trends of charter schools in California. One of the major distinctions within CaliforniaÂ’s charter schools are between charter schools started from scratch and charter schools converted from c onventional public schools. Figure 1 suggests that the majority of recently implemen ted charter schools are start-ups and accounted for 70% of all charter schools by the 2001Â–02 school year.
Charter School Type Matters When Examining Funding and Facilities 5 260 222 163 102 73 54 41 32 14 6 112 110 102 91 60 49 48 36 29 90 50 100 150 200 250 300 1993199419951996199719981999200020012002 DateNumber of Schools Start-up schools Conversion schools Figure 1 Number of Conversion and StartUp Schools by Year. (Sour ce: 2001Â–02 California Basic Educational Data Sy stem, or CBEDS) Breaking the figures down by grade span, Figure 2 shows that 72% of all elementary charter students are in conversion schools with the remainin g 28% in start-up schools. In contrast, 46% of all secondary students are in conversion schools, with the remaining 54% in start-up schools. 46 72 54 280 20 40 60 80 100 Elementary SecondaryPercent of Students Conversion Schools Start-up Schools Figure 2 Percent of Charter Students in Conversion and Startup Schools by Grade Levels. (Source: 2002 California Department of Educati on Statewide Student-Level Data) As Figure 3 suggests, the distribution of enrollment sizes varies by school type. Start-up charter schools tend to be much smaller than c onversion charters or conventional public schools, while conversion schools more closely mimic the si ze distribution of conventional public schools.
Education Policy Analysis Archives Vol. 13 No. 50 6 The differences between start-up and conversion sc hools in terms of funding is explored in our analysis. 12 6 35 29 18 9 13 3030 19 30 20 36 6 8 0 5 10 15 20 25 30 35 40 Less than 100 100 to 199200 to 599600 to 999Greater than 1000Percentage Conventional public schools Conversion schools Start-up schools Figure 3 Differences in School Size by Sc hool Type. (Source: 2002 CBEDS) Charter-School Funding A basic precept of charter schools across all states is that money should follow the students. However, the formula varies from state to state. Generally, across states, there are four bases for determining how much money should follow a studen t to the charter school: per-pupil revenue of a district, per-pupil expenditures of a district, perpupil statewide average expenditures, or a per-pupil district budget formula (Nelson, 2000). The per-pup il revenue of a district approach relies primarily upon the taxable revenues of the district coupled with the characteristics of students that the school serves, including low-income and special needs stud ents. In contrast, in the per-pupil expenditure approach, charter schools receive the same level of av erage per-pupil expenditures as the rest of the district, which assumes the district and any particul ar school serve similar students. The third option, statewide average expenditures, is usually a stra ightforward funding system in which the schools receive a flat per-pupil rate. Finally, a fourth appr oach leaves the funding decision up to the district that charters the school and often results in a negotiation process between the district and the school. In California, public school funding in Calif ornia is provided by the State and funneled through the districts, but the system has evolved. Under the initial charter school funding model, funding was modeled on the system used to fund conventional public schools. Charter schools received funding through two means: state revenu e limit funding which is general purpose money allocated based on average daily attendance (ADA) at a school; and categorical aid which is generally more restricted state and federa l funding for particular students or programs and based on application and eligibility. The per-pupil revenue limits that are set by the state each year are
Charter School Type Matters When Examining Funding and Facilities 7 essentially equalized across all districts but vary ba sed on the level and size of the district. Categorical aid, on the other hand, is allocated to serve specific purposes and/or specific populations of students. The California courts do not require categorical aid to be evenly distributed. Most categorical aid is accompanied by conditions for its use, in contrast to revenue limit funding. Currently, about one-third of school funding is earmarked by the state for about 70 specific categorical aid programs. Over time, revenue limit fu nding has declined as a percentage of total state KÂ–12 while the proportion of funding devoted to categorical aid has grown. While revenue limit funding has become more equally distributed, it ha s also become a smaller fraction of total funding (Carroll, Krop, Arkes, Morrison, & Flanagan, 2004). One concern regarding this initial model for fu nding charter schools was that charter schools may have trouble applying for and administering in dividual categorical aid programs. Because many charter schools are started from scratch and are relatively small in size, they may not have the experienced administrative staff who knew about the various programs or be willing or able to go through the challenges of applying for and mainta ining the programs. Therefore, there was concern that charter schools are not receiving appropria te aid due to a lack of categorical funds. In 1999, the legislature passed AB544, requiring the California Department of Education to propose a funding model for charter schools that wo uld provide operational funding equal to total funding received by a school district serving a si milar pupil population. In addition, the model was to provide charter school funding in a simple manner [Ed Code 47630]. Out of this mandate, a charter school block grant system was created.3 The charter school block grant funding model contains two parts, a general purpose entitlement in lieu of revenue limit funding and a block grant in lieu of some categorical funding. Both block grants are provided on an ADA basis calculated separately for each of four grade ranges (KÂ–3, 4Â–6, 7Â–8, and 9Â–12). The general-purpose block grant is based on comparable revenue limit funding so that conventional public schools and charter schools receive similar per-pupil general purpose monies. General purpose funding rates per student in ADA for the 2001Â–02 school year were $4,421, $4,478, $4,600, and $5,341 for grades KÂ–3, 4Â–6, 7Â–8, and 9Â–12, respectively. The funds are unrestricted and may be used for any school purpose. The categorical block grant is provided in lieu of funding for many state categorical programs. Therefore, a charter school is not elig ible for separate funding for any state program included in the categorical block grant. Charter schools must apply separately for categorical programs not included in the block grant. Charter schools are exempt from the program requirements of the individual state categorical prog rams included in the block grant calculation. The federal government does not allow charter schools any flexibility on the use of federal funds, so charter schools must continue to apply for and fully comply with the conditions of federal programs. Like the general-purpose block grant, the funds pr ovided in the categorical block grant may be used for any purpose determined by the charter school. Ma ny of the largest sta te categorical programs and all federal categorical programs fall outside of the state categorical block grant and are applied for separately, as is discussed in greater detail later in the paper. 3 The new funding model was established by AB 115 in the 1999Â–2000 state budget Â“education trailer bill.Â” All new charter schoolsÂ—those assigned number s after June 1, 1999Â—are funded under the Charter School Funding Model. Charters that were previously assigned numbers were allowed to continue to operate through the 2001Â–02 school year under a district apportionment or be fu nded through the Charter School Funding Model. The exception to this is district-wide charter schools, which retain the option to be funded under the revenue limit model or the Charter School Funding Model.
Education Policy Analysis Archives Vol. 13 No. 50 8 At the option of the school, charter schools may receive their funding through local or direct allocation of funds. Under local funding, the charter school has its funds deposited in the appropriate fund or account of the authorizing lo cal education agency. Under direct funding, the charter school has its funds deposited in the appr opriate fund or account of the charter school. The decision on whether to be locally or directly fund ed does not affect the amount of block grant funding provided to the charter school. In general, local funding tends to be a more popular option with conversion schools and those that rely on a di strict for fiscal services. In addition, some districts prefer the local option if they intend to he lp charter schools with cash flow because of the sense of control or recourse by handling the sta te dollars and passing them through. Direct funding tends to be a more popular option with start-up ch arter schools, possibly giving them a greater sense of control as the money goes directly to them. The decision to be locally or direct funded ca n affect the level of support provided to apply for and manage the categorical programs outside of the block grant. Charter schools that choose to be funded locally must apply through the approving local educational agency for categorical programs that are not included in the block gr ant, unless legislation for individual programs specifically allows charter schools to apply separate ly. Locally funded charter schools can work with their approving local educational agencies to ensure they are included in those agenciesÂ’ applications for programs for which the charter schools are eligible and in which they choose to participate. In particular, conversion charter schools that are char tered by their local districts already have the mechanisms in place to be included in the sponsoring districtsÂ’ applications for categorical aid programs. In contrast, start-up charter schools do not have a history of participation in categorical aid programs and are more likely to be direct-funded. A charter school that is direct-funded must apply individually for state or federal funds not includ ed in the block grant. The school may not be included in the application or eligibility of the authorizing entity for any categorical programs. An election to receive funding directly applies to all funding the charter receives, including other state and federal categorical aid. Method To examine the finances of charter schools in California, we use both secondary and primary data sets. The secondary data set includes the CaliforniaÂ’s Basic Education Data System (CBEDS) and JÂ–200 data. CBEDS data are compiled by the California Department of Education (CDE) and include school enrollment, free and reduced-price lunch program participation, racial and ethnic breakdowns, and other student characteristics. JÂ–20 0 data, also compiled by CDE, include detailed revenue and expenditure data for each school dist rict across the state, data which provide a good base for determining the type and size of revenues school districts receive and how they spend their money. The primary data sources are based on a surv ey of the universe of charter schools and a matched sample of conventional schools, and a supplemental financial survey of charter schools. The surveys of charter schools an d the matched sample of conv entional schools use consistent questions as much as possible to allow comparisons. A third supplemental survey was administered
Charter School Type Matters When Examining Funding and Facilities 9 to probe charter schools with more detailed questions of expenditures and revenues.4 Each of these surveys was administered in the spring of 2002. Survey Sample To identify the universe of California charter schools, we began with a list from merging the California charter school office publicly available da ta with the charter scho ols listed in the 2000Â– 2001 CBEDS data. Schools were eligible if th ey opened before September 15, 2001 and were operating as of February 2002. In total, 357 sc hools met these requirements. We then contacted the individual schools and their respective chartering authorizers to verify the data in our initial list. We made changes in our database to reflect updated information we received during these interviews, including adding any schools that were not in our original list. Twenty schools were added to our sample this way, while 25 were eliminated. Of the 25 schools that were dropped, nine had never existed,5 five schools responded they were not charter schools (three told us that were no longer charters and two said they were not Â“public chartersÂ” ), two others were ineligible for our sample because they had not opened, and nine had either closed or had their charter revoked. Thus, the final sample included 352 charter schools. One limitation to this method is the small possibility that a charter school was not included in our sample be cause it was not in either the California charter office data or the CBEDS data. Crucial in our analysis are the matches we cr eated for the charter schools. In the past, researchers have generally found that charter sc hools disproportionally serve low-income and highminority students (Finnigan, et al., 2004; Gill, Timpane, Ross, & Brewer, 2001; Zimmer, et al., 2003), a fact that may cause schools to have different cost and governance structures. To avoid confounding differences associated with school type with differences related to students served, we matched charter and non-charter schools by an estimated propensity score (Rosenbaum & Rubin 1983). The propensity score is the probability that a school with a given set of characteristics is a charter school as opposed to a conventional public school. These propensity scores can then be used to match charter schools to non-charter school s by finding those that have similar propensity scores. To carry out the propensity match, we used a fo ur-step process. First, we stratified charter schools into eight categories (elementary schools, middle schools, high schools, county schools, continuation schools, juvenile hall schools, speci al education schools, and alternative education schools) used by CDE to designate school types for all public schools.6 Roughly 60 charter schools 4 In a pilot of our surveys, we found that charter schools are more capable of answering detailed financial questions than are conventional public schools, primarily because the district handles more of the financial responsibilities for conventional public school s. Thus, we created a one-page supplemental survey of financial questions answered only by charter schools. 5 Through our investigation, we could find no evidence that the charter schools ever opened. 6 Some charter schools had grade ranges that intersected multiple strata (e.g., and kindergarten through grade 12 school intersects the elementary, middle and high school strata). In these cases, the charter schools were included in each catego ry and matched to a traditional public school for each category. Due to the small sample of county, continuation, juvenile hall, special education, and alternative education schools, a propensity match was not used in these cases. Instead, if demographic data were available for these schools, the schools were matched based on the criteria of getting schools within 10% of racial characteristics of the charter schools. In many cases, de mographic characteristics were not available for these schools and schools were matched to a traditional public school of the same school type within the district or the closest district.
Education Policy Analysis Archives Vol. 13 No. 50 10 were new in the 2001Â–2002 school year and were not included in the 2000Â–2001 CBEDS data, and thus, could not be matched to public schools. Second, within grade range strata, we fit a logistic regression model to predict designation (1=char ter; 0=conventional public) as a function of aggregate school characteristics, including percenta ge ethnicity (percentage White, percentage Blacks, percentage Asian, and percentage Hispanic), pupil socioeconomic status (percentage free-andreduced lunch),7 and percentage English language learners. Using these characteristics, predicted values for charter school i and conventional public school j were created ( pi and pj). Then, the distance between these schools ( dij) are estimated as the absolute value of the difference between their propensity scores, dij = | p pj |. We calculated the distance between each charter school and every conventional public school. Fourth, we matche d to each charter school a conventional school that minimizes dij over all California conventional public schools j. As part of the matching process, we allowed a conventional public school to be matched to multiple charter schools due to budget and time co nstraints. While the propensity scores do not create perfect matches, they do create a sample of conventional public schools with characteristics that closely resemble those of charter schools. Ta ble 1 displays the characteristics of the matched elementary, middle, and high schools for charter and conventional public schools. Table 1: Match School Ethnic/Racial and Lo w-English Proficient Breakdown 8 Schools Percent White Percent Black Percent Hispanic Percent Asians Percent Others Percent LEP Elementary schools Charter schools 48.5 14.9 27.8 2.7 6.1 15.6 Match public schools 51.5 13.3 27.7 2.9 4.6 17.1 Middle schools Charter schools 51.8 11.7 23.8 2.3 10.4 9.4 Match public schools 54.3 13.8 22.5 4.0 5.4 10.6 High schools Charter schools 52.9 9.6 26.4 4.0 7.1 10.0 Match public schools 53.2 5.3 28.8 6.8 5.9 10.2 Source: 2001Â–02 CBEDS Data Response Rates Not all of the charter or matched conventi onal public schools responded to our survey. Table 2 highlights the number surveyed, the number of respondents, and the percentage response 7 It was later discovered that ma ny charter schools do not participate in free-and-reduced lunch programs. Since the original propensity match included percentage free and reduced lunch, the final sample had to be weighted to account for this bias. 8 We only matched conventional public schools to charter schools for those schools we had demographic information.
Charter School Type Matters When Examining Funding and Facilities 11 rate for each sample. As highlighted in the tabl e, our response rates were nearly 75% for both charter and conventional public schools, and 56% for the charter supplemental survey. Table 2 Response Rate Survey Sample Respondents Response Rate Charter school survey 352 257 73% Charter school supplemental survey 352 200 56% Conventional public school survey 245 184 75% To adjust for differentia l response rates among and across charter and conventional schools, which may create bias in our results if type s of charter schools are underrepresented, we weighted the data so that the sample of charte r schools reflected the population of charter schools in the state, and co nventional public school re sults were weighted to ensure comparability with the full sample of conven tional public schools created through the propensity match. To weight the data for no n-response, we used a lo gistic regression that predicts whether the school res ponds or not based on demographic characteristics of a school, including percentage racial/ethnic breakdowns, percentage free-and-reduced lunch (including a dummy variable for whet her the school participated in th e free-and-reduced lunch program), and percentage language proficient (Little & Rubin, 1987). In this approach, the universe of charter and conventional public sc hools is included in a single data set. Using the coefficient estimates from the regression, we plug each sch ools characteristics to gain a predicted probably (p) of responding. This analysis weight for each charter and conventional public school that responded is the odds of responding p/(1-p) as describe d by Hirano, Imbens and Ridder (2000). Table 3 displays the characteristics of the sample after weighting. Table 3: Student Characteristics of Weighted Sample School Type Percent White Percent Black Percent Hispanic Percent Asians Percent Others Percent LEP Charter schools 50.1 13.5 26.8 2.8 6.8 14.8 Conventional public schools 47.4 11.6 30.7 3.9 6.4 17.7 Source: 2001Â–02 CBEDS data The data produced from the surveys serves as the foundation for our analyses in the rest of the paper. In some cases, we comp are the responses of charter and conventional schools. In other cases, we examine differences among charter schools by analyzin g the difference in responses between schools that are conver ted from conventional public schools, or conversion charter schools, and schools started from scra tch, or start-up charter schools. Limitations Before we proceed, we should also mention some possible drawbacks of our analysis. First and foremost, we are relying upon self-reported information through a survey of charter-school
Education Policy Analysis Archives Vol. 13 No. 50 12 principals, whose responses may have some erro rs. In addition, other states use different mechanisms of funding charter schools, and this cu rrent analysis may have limited implications for those schools. In addition, while we did have a hi gh response rate, and did weight for non-response, we did not have 100% response rate, which also ma y create small errors in our percentages and averages. Finally, because special education is a major educational cost, we would have liked to match charter and conventional public schools ba sed on the percentage of special education students. However, we were unsure whether charter and conventional public schools consistently classify special education students. In particular we were not sure if the percentage of special education students were accurately reported for charter schools. Therefore, we did not include it in our matching procedure. This may create some limitations in the comparison between charter and conventional public schools. Results Participation in Catego rical Funding Programs In California, the charter school categorical bloc k grant is intended to simplify the process of obtaining and maintaining state categorical aid for charter schools as well as to allow charter schools freedom in the use of the funds. The block grant model is desirable for charter schools both because the cash is more likely to be received in a time ly manner and because charter schools can avoid the applications and reporting involved in obtaining fu nds from a large number of categorical programs (Sugarman, 2002). Charter school operators are ofte n unsophisticated in completing the forms and carrying out procedural activities that have take n districts years to master. In addition, charter schools may not have the economies of scale to operate categorical pr ograms on their own. Approximately 30 state categorical programs are currently included in the charter school categorical block grant. It is important to note that several of the largest state categorical aid programs and all federal categorical aid programsÂ—including KÂ– 3 Class Size Reduction, Transportation, Special Education, and Title 1 funding for disadvantaged pu pilsÂ—fall outside of the categorical block grant and require charter schools to apply separately and to adhere to the statues and regulations that govern the programs. Also, the categorical block grant rates have declined over time largely due to the removal of several state programs from the bl ock grant and to the expiration of or reduction in programs that previously contri buted to the block grant. Participation or lack of participation in ca tegorical aid programs can have significant financial effects on schools. Charter school and conventional public school finances may be relatively equal per pupil based on the general purpos e block grant. But, as noted earlier, categorical funds are not required by the California courts to be equally distributed. Over time, categorical aid has become a larger share of total school fund ingÂ—currently about one-third. Through the charter school and conventional public school surveys, we soug ht to address a number of questions related to charter school participation in categorical programs outside of the block grant First, do charter schools participate in programs outside of the block grant? If charter schools choose not to participate in these programs, why? Specifically, for nine relatively large state and federal categorical aid programs outside of the block grant, we asked charter school and conventional public school principals whether they are currently receiving funding, have an application pending, are ineligible to apply, are not applying but eligible, or donÂ’t know whether eligible or not. To adjust for multiple related questions within the
Charter School Type Matters When Examining Funding and Facilities 13 survey, we used a Bonferroni correction to offset th e chance of incorrectly reporting statistical differences. Table 4 shows the percentage of principals w ho reported Â“currently receiving fundingÂ” for start-up charter schools, conversion charter school s and matched conventional public schools for each individual categorical prog ram. Start-up charter schools have statistically significant lower participation rates for every categorical progra m compared with matched conventional public schools. By contrast, conversion charter schools in general have similar participation rates in categorical programs as matched conventional pub lic schools and, in some cases, have higher participation rates (but those differences are not statistically significant). Table 4 Percent of Schools Currently Rece iving Funding from Va rious Categorical Aid Programs, Separately for Conversion Charter Schools, Start-Up Charte r Schools, and Matched Conventional Public Schools 9 Conventional Public Schools Conversion Charter SchoolsStart-up Charter Schools Categorical Program N % N % p N % p KÂ–3 class size reduction 167 66 69 72 .42 176 40 < .001* Pupil transportation 172 54 70 55 .85 170 4 < .001* Public School Accountability Act 169 48 66 47 .90 167 10 < .001* Special education funding 179 94 70 83 .03 175 67 < .001* Title 1 funding 173 64 70 73 .19 175 34 < .001* Staff development buyout days 179 88 68 89 .96 174 47 < .001* Child nutrition programs 177 82 69 77 .39 180 34 < .001* Supplemental instructional program 171 59 67 54 .59 172 16 < .001* Desegregation program 167 10 66 28 .003* 169 3 .026 indicates charter school percentages that are statistically different from matched conventional public school percentages at .05 level. 9 In Table 4, we report p values as an example of how we did our analysis. In later tables, we do not report the p values. Instead, we provide an for when the comparison is si gnificantly different.
Education Policy Analysis Archives Vol. 13 No. 50 14 To explore in greater detail the lower particip ation rates of start-up schools, we examined whether schools are Â“eligible but not applyingÂ” and whether schools Â“donÂ’t know whether they are eligible or not.Â” Tables 5 and 6, respectively, s how the percentage of start-up charter school, conversion charter school and conventional public school principals who reported their schools were Â“eligible but not applyingÂ” or Â“donÂ’t know wh ether eligible or notÂ” to specific categorical programs outside of the categorical block grant. Table 5 Percentage of Charter Schools Â“Eligible for Categorical Aid Funding but Not ApplyingÂ” by Matched Conventional Public Schools, Convers ion Charter Schools and Start-Up Charter Schools Conventional Public Schools Conversion Charter Schools Start-up Charter Schools Categorical Program N Percent N Percent N Percent KÂ–3 class size reduction 167 2 69 2 176 3 Pupil transportation 172 5 70 7 170 8 Public School Accountability Act 169 3 66 3 167 8 Special education funding 179 2 70 3 175 6 Title 1 funding 173 2 70 0 175 23* Staff development buyout days 179 3 68 0 174 7 Child nutrition programs 177 4 69 8 180 24* Supplemental instructional program 171 1 67 1 172 12* Desegregation program 167 4 66 3 169 3 indicates charter school percentages that are statistically different from matched conventional public school percentages at .05 level. In general, no more than 5% of conventional public school principals responded that their schools are Â“eligible but not applyingÂ” to the indivi dual categorical programs. Similar results are seen for the conversion charter schools. Start-up charte r schools are generally more likely than either conversion or conventional public schools to be Â“e ligible but not applyingÂ” to the individual aid programs. In particular, there are large differences for the Child Nutrition programs and Title 1. Both programs provide, on average, relatively large per-pupil funding to participating schools. Participation in Child Nutrition programs can pose a problem to charter school s, particularly those that do not have a sponsoring district willing to include them in their programs. Many charter schools have neither economies of scale nor administrative resources to support a Child Nutrition program on their own (Zimmer et al., 2003). Similarly Title 1 is a federal aid program with extensive statues and regulations that govern the program as requirements of receipt of funding. Those charter schools without links to a district chartering authori ty willing to include them in the districtÂ’s Title 1 program likely cannot participate alone. When asked in the charter school survey to agree or disagree with the statement Â“our school has given up pursui ng certain categorical funds because they are too complex,Â” about 25% of conversion charter schools st rongly agreed or agreed and 48% of start-up charter schools strongly agreed or agreed. Turning to the issue of knowledge about categ orical programs, Table 6 again shows that conversion charter schools and conventional public schools are generally similar in the percentages that Â“donÂ’t know whether eligible or notÂ” to the various categorical aid pr ograms. Again, start-up charter schools show considerably larger percen tages than either conventional public schools or
Charter School Type Matters When Examining Funding and Facilities 15 conversion charter schools, which may suggest that these principals are relatively inexperienced in applying for categorical programs and ulti mately leads to less money for the schools. Added together, those who responded they are Â“eligible but not applyingÂ” and Â“donÂ’t know whether eligible or notÂ” result in considerably lowe r participation in categorical aid programs outside of the block grant for start-up charter schools than conversion charter schools or conventional public schools. This suggests that the removal or exclusion of programs from the block grant has a sizable effect on start-up charter schools in particular. Table 6 Percentage of Charter Schools Â“DonÂ’t Know Whet her Eligible or Not Â” by Matched Conventional Public Schools, Conversion Charter Schools and Start-Up Charter Schools Categorical Program Conventional Public Schools Conversion Charter Schools Start-up Charter Schools N Percent N Percent N Percent KÂ–3 class size reduction 167 2 69 7 176 3 Pupil transportation 172 26 70 20 170 35 Public School Accountability Act 169 38 66 38 167 60* Special education funding 179 1 70 7 175 11* Title 1 funding 173 3 70 7 175 15* Staff development buyout days 179 5 68 9 174 21* Child nutrition programs 177 6 69 9 180 9 Supplemental instructional program 171 35 67 40 172 46 Desegregation program 167 53 66 42 169 62 indicates charter school percentages that are statistically different from matched conventional public school percentages at .05 level. The results from this section suggest that star t-up charter schools have a lower participation rate in categorical programs, which implies that these schools are receiving less public revenue. A portion of the lower participation rates can be ex plained by the fact that some start-up charter schools do not know whether they are eligible for these categorical programs. In other cases, the start-up schools know that they are eligible for the program but do not participate for what are likely a variety of reasons, including the administrative complexity of participation, the lack of economies of scale necessary for participation and a lack of fiscal administrative experience by charter school principals. It should be noted that statewide data do not exist on how much individual schools receive from individual categorical aid programs, and stat ewide data are collected at the district level. In addition, as categorical aid is generally distributed at the district level with the district providing services with the money, schools generally do not know how much they receive from categorical aid programs. Therefore, we cannot put dollar amounts to the lower participation in these large categorical aid programs. However, our survey resul ts suggest that charter schools received about
Education Policy Analysis Archives Vol. 13 No. 50 16 $6,500 per pupil in 2001Â–02 compared to $8,000 for conventional public schools and that this difference may be partially explained by differen ces in participation in categorical programs.10 Private Donations to Charter Schools Private donations may play a unique role in ch arter schools. Especially for start-ups, charter schools face a number of costs ranging from books and materials to facility needs. In addition, some charter schools have distinct educational focuses th at may be used to identify and attract donors. Previous research (Miron, Nelson, & Risley, 2002) has suggested that private donations can be a mechanism charter schools use to make up for insufficient public funding.11 Because there is little systematic public data quantifying how much mon ey or in-kind services schools receive from private donors, we asked both conventional and charter school principals how much private funding their schools received for the 2001Â–02 school year. Table 7 shows the average private dollars gi ven to conventional and charter school types. The table suggests that conventional public and conversion charter schools receive similar amounts of private giving per pupil and start-up charter schools receive significantly more. However, these results are skewed by a few start-up schools that received a large share of the total donations. In fact, the median value of per-pupil donations are $0 for conversion schools and $3.86 for start-up schools. The high average per-pupil donation for star t-ups is partially driven by three schools that received over $10,000 of per-pupil donations for the 2001Â–02 school year. Taking these out, the average per-pupil donation for startup schools is $293, which is still significantly more than conversion charter or conventional public schools. In addition, about 17% of start-up schools received a donation of more than $500 per pupil 20 01Â–02. So, even after taken into account some of the outliers, start-up schools appear to receive la rger support from private sources than conversion charter or conventional public schools. The higher level of private support is likely in part due to start-up schoolsÂ’ having greater initial expenses and facility needs. Conversion school s generally have facilities, supplies and materials to begin instruction and so might not have as great a need for private support, particularly in the early years. In addition, start-up schools may seek private donations to fill some of the gap in categorical funding described in the previous section. Ideally, we would know the extent to which private funding merely helps to compensate for the fa ct that start-ups need to pay for facilities while conversions do not versus as well as the extent to which private funding is used for other purposes such as teacher salaries, curriculum, or other instructional activities. In addition, we do not currently know if private funding consists of one-time gifts or ongoing contributions. Nevertheless, it is safe to assume that these private contributions are gene rally not making up for the average shortfall of approximately $1,500 per charter school. 10 Charter school per-pupil revenue is derived fr om our survey of charter schools in the 2001Â–02 school year. Conventional public school revenue per pupil is derived from the National Center for Education Statistics website, http://nces.ed.gov/quicktables/Detail.asp?Key=760. 11 Public schools call on a variety of private givers to provide a spectrum of goods and services. Recent research suggests public school s have increasingly sought private support (both financial and in-kind) in recent years (Brunner & Sonstelie, 1997; Zimme r, Krop, Kaganoff, Ross, & Brewer, 2001). However, private financial contributions still acco unt for a relatively small share of to tal resources for the vast majority of public schools.
Charter School Type Matters When Examining Funding and Facilities 17 Table 7 Private Funding to Conventional Public and Charter Schools School type N Average Dollar Value Conventional public 184 $83 Conversion charter school 63 $56 Start-up Charter School 153 $576* Charter school percentages are statistically different from m atched conventional public school percentages at .05 level. Charter School Expenditures Having examined California charter school part icipation in revenue programs, we now turn to charter school expenditures. Several expenditur e-related questions issues of how charter schools spend their resources, how spending differs among types of charter schools, and how charter school expenditures differ from conventional public schools. These questions are difficult to answer given current data sources. First, there are no systematic state data collected in California, or in most other states, at the school level on expenditures. Instead, data are collected and reported at the district level. Even if there were systematic, reliable charter school expenditure data at the school level, these data could only be compared to public school distri ct averages. Further, it is also difficult to collect expenditure data for individual charter schools. For example, some locally funded charter schools rely on a district to pay for some large expenditures while others do not. Data reported by the charter schools may or may not include expenditures that are assumed by the district, and charter schools may not be able to accurately report suc h expenditures. In addition, charter school expenditures are influenced by large capital expe nditures in a given year. Without a detailed cost studyÂ—one that correctly apportions overhead, administration, and personnel to the Â“rightÂ” schoolsÂ—it is difficult to document and compare school expenditures. Due to the lack of systematic charter school or conventional public school finance data collected by the state, we addressed questions about charter school expenditures through a supplemental survey as described earlier. In addition to other items, we asked charter schools to report their total expenditures, teacher salary and benefit expenditures, and other staff salary and benefit expenditures for the 2001Â–02 school year. These numbers should be interpreted with caution given the limitations discussed above. Charter schools as a whole reported an average total expenditure per st udent of $6,204 for the 2001Â–02 school year.12 This appears to be lower than the reported statewide average per-pupil expenditures. There are various sources of averag e total expenditures per pupil for California schools, and although the exact number differs depending on what is included, the reported statewide averages tend to be closer to $6,500 per pupil.13 12 The mean expenditures throughout this section are influenced by several high outliers, often due to the inclusion of capital expenditures in total expendit ures. In the cases of extreme outliers, we called the survey respondents to confirm the reported numbers. The median total expenditure reported b`y charter schools is $5,408 for the 2001Â–02 school year. The standard deviations for total expenditures per pupil are as follows: all charter schools, $4,984; s tart-ups, $4,658; conversions, $6,084. 13 For example, the National Education Associatio n (NEA) reports statewide average expenditure per ADA in California for 2000Â–01 to be $6,837. Th e California Department of Education reports the statewide average ex pense of education per unit of ADA to be $6,360 for the 2000Â–01 school year.
Education Policy Analysis Archives Vol. 13 No. 50 18 In addition, we looked at how per-pupil tota l expenditures, teacher salary and benefit expenditures, and other per-pupil staff salary and benefit expenditures differ among different types of charter schools. Charter schools report per-pup il teacher salary and benefit expenditures of $2,841 and other per-pupil staff salary and benefi t expenditures of $1,075. Table 8 documents these expenditures for start-up an d conversion charter schools. Table 8 Charter School Expendit ures Per Pupil, 2001Â–02 Type of Charter School Total Expenditures Per Pupil Teacher Salary and Benefit Expenditures Per Pupil Other Staff Salary and Benefit Expenditures Per Pupil All Charter Schools $6,204 $2,841 $1,075 Start-Up $6,168 $2,729 $1,006 Conversion $6,366 $3,237 $1,340 The table suggests that, on average, start-up charter schools spend less overall per pupil as well as less per pupil on teacher salaries and benefits and other staff salaries and benefits compared with conversion charter schools. In addition, star t-up charter schools, on average, allocate about 60% of their total expenditures to teacher and othe r staff salaries and benefits compared with about 72% for conversion charter schools. This may be due to start-up charter schoolsÂ’ need to allocate larger shares of their expenditures to such items as facilities and start-up costs.14 The results from examining statewide average per-pupil expenditures suggest that charter schools as a whole may have lower per-pupil expe nditures than conventional public schools. An accurate estimate of the difference in per-pupil spending between charter schools and conventional public schools is difficult to secure given current data sources. With that said, a possible inference from the charter school survey data, which would need to be affirmed through a systematic collection of detailed conventional public school and charter school costs, is that charter schools, and particularly start-up charter schools, receive lower revenues from categorical programs and spend less per pupil than conventional public schools. Facilities Acquiring and funding school facilities has been a stumbling block for charter schools as a whole, across states (Finn, Mammo, & Vanourek, 2000; Powell, et al., 1997; RPP International, 2000; Sugarman, 2002;). Charter schools do not have access to similar revenu e sources for facilities as conventional public school districts. Conventiona l public school districts pay for facilities by issuing bonds, an avenue unava ilable to many charter schools.15 In addition, charter schools pay for 14 Sugarman (2002) suggests that start-up schools of ten have to redirect perhaps 20% or more of their core funding to pay for space. 15 In addition, charter schools in California often do not have access to state or district bond monies or other capital resources for school improvements or building of new fa cilities. Unless the charter provides that its facilities must comply with the Field Act, charter schools are exempt from the Act. Often bond monies or other state or federal facility monies ar e dependent on Field Act comp liance. Conversion charter schools generally comply with the Field Act while start-up charter schools may not.
Charter School Type Matters When Examining Funding and Facilities 19 facility expenses that conventional public school s do not. These expenses may include rent on facilities, utilities, maintenance, and off-site storag e facilities. Finally, charter schools often find it difficult to find suitable facilities and face landlo rds who are cautious about leasing facilities to new entities and to charters that are granted for only a few years. Until the 2003Â–04 school year, charter schools in California largely had to find their own facilities. Conversion charter schools often already ha d facilities that they had been occupying as a conventional public school. Start-up charter schools generally had to acquire facilities. The charter law in effect until the 2003Â–04 school year stated that a school district in which a charter school operated (which is not necessarily the approvin g district) shall permit a charter school to use facilities cost-free when not being used by the dist rict for instructional or administrative purposes, unless historically used for rental purposes (ref. E. C. 47614). Some districts in which charter schools were operating were already overcrowded and had no unused facilities. And, the media suggested that a number of districts with unused facilitie s were not complying with the law (Space Crunch, 2004). To examine charter school facility issues, we asked charter school principals how they arrange for facilities and whether they are strugglin g with financing capital expenditures. As Table 9 shows, charter schools appear to use multiple mean s to arrange for facilities. Among charter school principals, 42% report that their facilities are provid ed by a district, free or at a nominal costÂ—with most of these being conversion charter schools. In addition, 30% of charter schools lease their facilities from a commercial sourceÂ—with most of these being start-up charter schools. About 12% of charter schools used two different means to provide facilities (e.g., leased from a commercial source and donated by sponsors) and about 2% of charter schools used more than two means to provide facilities. Table 9 Acquisition of School Facilities, by Conversion and Start-Up Charter Schools16 Charter Schools Conversion Charter Schools Start-up Charter Schools Categorical Program N Percent N Percent N Percent Leased from Commercial Site 255 30 703 185 40 Provided by District, Free or Nominal Cost 255 42 7091 185 23 Leased at, or Near Market Price from District 255 9 703 185 12 Privately Rented or Owned 255 18 703 185 24 Donated by Sponsors Other than District 255 3 701 185 4 Obtained through Another Arrangement 255 13 705 185 16 In addition, we asked charter school principals whether they were struggling with financing charter school capital expenditures. Among the ch arter school principals, 62 % strongly agreed or 16 We asked charter schools to check all that apply. Therefore, adding across all response categories for start-up charter schools, for example, will result in more than 100 percent.
Education Policy Analysis Archives Vol. 13 No. 50 20 agreed with the statement Â“Our school is strugglin g with financing capital expendituresÂ” (with 68% of start-up charter school principals and 46% of conversion charter school principals agreeing).17 In response to charter school facility obstacles, the provision of charter school facilities in California changed in 2003Â–04. Proposition 39 an d Senate Bill 740 requires districts to provide facilities for eligible charter schools and allows re imbursements of facility costs for schools in lowincome areas. Approved by voters in November 2000, Proposition 39 took effect in November 2003 for most districts. Proposition 39 directs school districts to provide facilities for charter schools who have an in-district ADA of 80 students or more. The charter school does not need to be currently located within the district, nor does th e charter have to have been granted by the district where the eligible students live. The district is only required to provide space for the in-district students. The law states that facilities must be Â“reasonably equivalentÂ” to facilities which the students would otherwise attend in non-charter sc hools in that district. Districts may charge the charter school an amount equivalent to what the district spends per student on facilities from their general fund. Senate Bill 740 is another legislative measure de signed to alleviate some of the facilities burden on charter schools. This measure was implem ented for the first time in the 2002Â–03 school year. The legislation created a small charter facilit ies aid program for schools in low-income areas. Eligible schools receive a cash reimbursement after the close of the fiscal year. The law currently allows schools in which more than 70% of the charter school students are eligible for free or reduced lunch to be eligible for this funding. The charter school and chartering authority res ponses presented in this section provide a baseline for changes in charter school facility rule s and regulations. Future research will need to examine how chartering authorities respond to an d how charter schools are affected by Proposition 39 and Senate Bill 740. Conclusions The results from our analysis indicate that any fiscal challenges charter schools are experiencing are most likely experienced by start-up rather than conversion charter schools and that these challenges result in part from lower partic ipation in categorical programs and from facility needs. The disparity in participation may ultimatel y lead to disparities in funding among charter schools and between charter schools and conventi onal public schools. Given CaliforniaÂ’s long history to create equity in funding across schools, any resulting disparities are reason for concern. However, because the state is making the categorical programs available to charter schools, it is not entirely clear how it affects the stateÂ’s legislat ive requirement to provide equitable funding. Our results further suggest that start-up charter school s may be relying more on private sources of funds than conversion charter schools or conventional pub lic schools. Additional research is needed to determine the extent to which the private funds merely help to compensate for start-up and facility costs versus the extent to which private funds are used for other purposes. Finally, our results also suggest that CaliforniaÂ’s current focus on providin g greater facility support for charter schools is warranted. Together, these results suggest that policymakers need to be particularly conscious of how funding models affect start-up charter schools, esp ecially given the recent growth of these schools. 17 We also looked at the charter school responses based on when the charter was granted, but the responses were consistent acro ss different lengths of time si nce the charter was granted.
Charter School Type Matters When Examining Funding and Facilities 21 In addition, charter schools and policymakers w ill need to develop innovative solutions that encourage start-up charter schools to participate in categorical programs for which they are eligible to receive funding. For example, to the extent th at individual charter schools cannot support child nutrition programs, avenues need to be opened for the schools to be included in district child nutrition programs or to form networks with othe r charter schools. The state may also want to provide technical assistance to ch arter schools in accessing and filling out appropriate forms for categorical programs or create networks of charter schools that can facilitate information sharing. While much of the charter school research has focused on student achievement, this research supports the call for additional research to develop innovative approaches to charter school finances in general and start-up charter school finances in particular. References Bifulco, R. & Ladd, H. (2005). Race and charter schools: Ev idence from North Carolina. Unpublished manuscript at the University of Connecticut. Bifulco, R. & Ladd, H. (in press). The impact of charter schools on student achievement: Evidence from North Carolina. Education Finance Policy. Booker, K., Gilpatric, S., Gronbe rg, T.J., & Jansen, D.W. (2004). Charter school performance in Texas. College Station, TX: Priv ate Enterprise Research Center, Texas A&M University. Brunner, E. & Sonstelie, J. (1997). Coping with Serrano : Voluntary contribut ions to CaliforniaÂ’s local public schools Paper presented at 89th Annual Conference on Taxation, National Tax Association, October. Buddin. R. & Zimmer, R.W. (2005). A closer look at charter school student achievement. Journal of Policy Analys is and Management, 24 (2), 351Â–372. Carroll, S. J., Krop, C., Arkes, J., Morrison, P., & Flanagan, A.. (2005). CaliforniaÂ’s KÂ–12 public schools: how are they doing? Santa Monica, CA: RA ND Corporation, MG 186. Center for Educati on Reform. (2005). Quick facts about charter schools Retrieved December 12, 2005, from http://www.edreform.c om/_upload/ncsw-numbers.pdf. Cornoy, M., Jackson, R., Mishell, L., & Rothstein, R. (2005). The charter school dust-up: Examining the evidence on enrollment and achievement. Washington, DC: Economic Policy Institute. Finn, C., B. Mammo, & Vanourek, G. (2000). Charter schools in acti on: Renewing public educaiton. Washington, DC: Hudson Institute. Prin ceton, NJ: Princeto n University Press. Finnigan, K., Adelman, N., Ande rson, L., Cotton, L., Donnelly, M.B., & Price, T. (2004). Evaluation of the public charte r schools program: Final report U.S. Department of Education, 2004Â–08.
Education Policy Analysis Archives Vol. 13 No. 50 22 Frankenberg, E. & Lee, C. (2003). Charter sch ools and race: A lost opportunity for integrated education. Education Policy Analysis Archives, 11 (32). Retrieved December 12, 2005, from http://epaa.asu.edu/epaa/v11n32/. Gill. B. P., Timpane, M. Ross, K. E., & Brewer, D. J. (2001). Rhetoric versus reality: What we know and what we need to know about vouchers and charter schools. Santa Monica, CA: RAND, MRÂ–1118. Greene, J.P, Forster, G., & Winters, M.A.. (2003). Apples to apples: An evaluation of charter schools serving general student populations. New York, NY: Manh attan Institute, Education Working Paper, (July 1). Gronberg, T.J., & Ja nsen, D.W.. (2001). Navigating newly chartered waters: An analysis of Texas charter school performance San Antonio and Aust in, TX: Texas Public Policy Foundation, Retrieved December 12, 200 5, from http://www.texaspolic y.com/pdf/2001-05-17-educnewly.pdf. Hanushek, E. A., Kain, J. F. & Rivkin, S. G. (2002). The impact of charter schools on academic achievement Unpublished Manuscript, Hoover Institution. Hirano, K., Imbens, G., & Ridder, G. (2003). Efficient estimation of average treatment effects using the estimated propensity score. Econometrica, 71 1161-1190. Hoxby, C.M. (2004). Achievement in charter schools and regular pu blic schools in the United States: Understanding the differences. Retrieved December 12, 2005 from http://post.economics.harvard.edu/faculty/hoxby/papers/hoxbycharter_dec.pdf. Little, R. J. A., & Rubin, D.B. (1987). Statistical analysis with missing data New York: Wiley. Miron, G. N., & Nelson, C. (2002). WhatÂ’s public about charter schools? Thousand Oaks, CA: Corwin Press, Inc. Miron, G., N., Nelson, C ., & Risley, J. (2002). Strengthening PennsylvaniaÂ’s charter school reform: Finding from the stat ewide evaluation and discussion of relevant policy issues Kalamazoo, MI: The Evaluation Center, Western Michigan University. Nelson, F.H, Rosenberg, B, & Van Meter, N. (2004). Charter school achiev ement on the 2003 National Assessment of Educational Progress Washington, D.C.: Amer ican Federation of Teachers. Nelson, F.H. (2000). Venturesome capital: State charter school finance systems. Washington, DC: American Federation of Teachers Educat ional Foundation. Retrieved December 12, 2005, from http://www.aft.org/topics/cha rters/downloads/vent uresomefull.pdf. Powell, J., Blackorby, J., Marsh, j., Finnegan, K., & Anderson, L. (1997). Evaluation of charter school effectiveness. Menlo Park, CA: SR I International. Rogosa, D. (2003). Student progress in California charter sc hools, 1999Â–2002 Unpublished manuscript, Stanfo rd University.
Charter School Type Matters When Examining Funding and Facilities 23 Rosenbuam, P. R. & Rubin, D.B.. (1983). Th e central role of the propensity score in observational studies for causal effects. Biometricia, 70 41Â–45. RPP International. (2000). The state of charter schools Washington, DC: U.S. Department of Education. Sass, T.R. (2005). Charter schools and studen t achievement in Florida. Unpublished Manuscript Florida State University. Solmon, L., Paark, K., & Garcia, D. (1999). Does charter school attendan ce improve test scores? The Arizona results Phoenix, AZ: Goldwater Instit ute Center for Market Based Education. Space Crunch for Charter Sc hools. (2004, December 20). Los Angeles Times Retrieved December 20, 2004, from http://www.latimes.com/news/local/la-mecharter20dec20,1,2827267.story. Sugarman, S.D. (2002). Charter school funding issues. Education Policy Analysis Archives, 10 (34). Retrieved December 12, 2005, from http://epaa.as u.edu/epaa/v10n34/. Thomas B. Fordham Foundation. (2005). Charter school funding: InequityÂ’s next frontier Retrieved September 25, 2005, from http://www.edexcellen ce.net/institute/c harterfinance/. Zimmer, R., Buddin, R., Chau, D., Daley, G., G ill, B., Guarino, C., Hamilton, L., Krop, C., McCaffrey, D., Sandler, M ., & Brewer, D. (2003). Charter school operations and performance: Evidence from California Paper MRÂ–1700. Santa Monica, CA: RAND Corporation. Zimmer, R., Krop, C., Kaganoff, T. Ross, K. & Brewer, D. (2001). Private giving to public schools and district s in Los Angeles Co unty: A pilot study Paper MRÂ–1429Â–EDU. Santa Moncia, CA: RAND.
Education Policy Analysis Archives Vol. 13 No. 50 24 About the Authors Cathy Krop Ron Zimmer RAND Corporation Email: email@example.com Dr. Cathy Krop is a policy analyst at RAND where sh e has done extensive work on the allocation and use of education dollars and on designing and co sting options fo r alternative systems of school finance in su pport of education reform effort s. Dr. Krop has also estimated the costs of class size reduction policies and K12 enrollment growth. In addition, Dr. Krop led a multi-state evaluation of a National Science Foundation program to im prove math and science instruction. Before joinin g RAND, Dr. Cathy Krop worked on education policy at the Congressional Budget Office and was a project manager at the Evaluation and Training Institute. Dr. Ron Zimmer is an economist at RAND focusi ng his research on educational finance, peer interactions, No Child Left Behind, and school ch oice. Dr. Zimmer recently led a team of researchers that examined the operatio n and performance of Cali fornia charter schools and is currently on a large project team evaluati ng the student achievement impact of No Child Left Behind in seve ral major urban districts ac ross the U.S. Previously Dr. Zimmer has worked on projects evaluating Edison schools, the mo vement toward greater state funding, and the effects of school tracking.
Charter School Type Matters When Examining Funding and Facilities 25 EDUCATION POLICY ANALYSIS ARCHIVES http://epaa.asu.edu Editor: Sherman Dorn, University of South Florida Production Assistant: Chris Murre ll, Arizona State University General questions about ap propriateness of topics or particular articles may be addressed to the Editor, Sherman Dorn, firstname.lastname@example.org. Editorial Board Michael W. Apple University of Wisconsin David C. Berliner Arizona State University Greg Camilli Rutgers University Casey Cobb University of Connecticut Linda Darling-Hammond Stanford University Mark E. Fetler California Commission on Teacher Credentialing Gustavo E. Fischman Arizona State Univeristy Richard Garlikov Birmingham, Alabama Gene V Glass Arizona State University Thomas F. Green Syracuse University Aimee Howley Ohio University Craig B. Howley Appalachia Educational Laboratory William Hunter University of Ontario Institute of Technology Patricia Fey Jarvis Seattle, Washington Daniel Kalls Ume University Benjamin Levin University of Manitoba Thomas Mauhs-Pugh Green Mountain College Les McLean University of Toronto Heinrich Mintrop University of California, Berkeley Michele Moses Arizona State University Anthony G. Rud Jr. Purdue University Michael Scriven Western Michigan University Terrence G. Wiley Arizona State University John Willinsky University of British Columbia
Education Policy Analysis Archives Vol. 13 No. 50 26 EDUCATION POLICY ANALYSIS ARCHIVES English-language Graduate -Student Editorial Board Noga Admon New York University Jessica Allen University of Colorado Cheryl Aman University of British Columbia Anne Black University of Connecticut Marisa Cannata Michigan State University Chad d'Entremont Teachers College Columbia University Carol Da Silva Harvard University Tara Donahue Michigan State University Camille Farrington University of Illinois Chicago Chris Frey Indiana University Amy Garrett Dikkers University of Minnesota Misty Ginicola Yale University Jake Gross Indiana University Hee Kyung Hong Loyola University Chicago Jennifer Lloyd University of British Columbia Heather Lord Yale University Shereeza Mohammed Florida Atlantic University Ben Superfine University of Michigan John Weathers University of Pennsylvania Kyo Yamashiro University of California Los Angeles
Charter School Type Matters When Examining Funding and Facilities 27 Archivos Analticos de Polticas Educativas Associate Editors Gustavo E. Fischman & Pablo Gentili Arizona State University & Universidade do Estado do Rio de Janeiro Founding Associate Editor for Spanish Language (1998Â—2003) Roberto Rodrguez Gmez Editorial Board Hugo Aboites Universidad Autnoma Metropolitana-Xochimilco Adrin Acosta Universidad de Guadalajara Mxico Claudio Almonacid Avila Universidad Metropolitana de Ciencias de la Educacin, Chile Dalila Andrade de Oliveira Universidade Federal de Minas Gerais, Belo Horizonte, Brasil Alejandra Birgin Ministerio de Educacin, Argentina Teresa Bracho Centro de Investigacin y Docencia Econmica-CIDE Alejandro Canales Universidad Nacional Autnoma de Mxico Ursula Casanova Arizona State University, Tempe, Arizona Sigfredo Chiroque Instituto de Pedagoga Popular, Per Erwin Epstein Loyola University, Chicago, Illinois Mariano Fernndez Enguita Universidad de Salamanca. Espaa Gaudncio Frigotto Universidade Estadual do Rio de Janeiro, Brasil Rollin Kent Universidad Autnoma de Puebla. Puebla, Mxico Walter Kohan Universidade Estadual do Rio de Janeiro, Brasil Roberto Leher Universidade Estadual do Rio de Janeiro, Brasil Daniel C. Levy University at Albany, SUNY, Albany, New York Nilma Limo Gomes Universidade Federal de Minas Gerais, Belo Horizonte Pia Lindquist Wong California State University, Sacramento, California Mara Loreto Egaa Programa Interdisciplinario de Investigacin en Educacin Mariano Narodowski Universidad To rcuato Di Tella, Argentina Iolanda de Oliveira Universidade Federal Fluminense, Brasil Grover Pango Foro Latinoamericano de Polticas Educativas, Per Vanilda Paiva Universidade Estadual Do Rio De Janeiro, Brasil Miguel Pereira Catedratico Un iversidad de Granada, Espaa Angel Ignacio Prez Gmez Universidad de Mlaga Mnica Pini Universidad Nacional de San Martin, Argentina Romualdo Portella do Oliveira Universidade de So Paulo Diana Rhoten Social Science Research Council, New York, New York Jos Gimeno Sacristn Universidad de Valencia, Espaa Daniel Schugurensky Ontario Institute for Studies in Education, Canada Susan Street Centro de Investigaciones y Estudios Superiores en Antropologia Social Occidente, Guadalajara, Mxico Nelly P. Stromquist University of Southern California, Los Angeles, California Daniel Suarez Laboratorio de Politicas Publicas-Universidad de Buenos Aires, Argentina Antonio Teodoro Universidade Lusfona Lisboa, Carlos A. Torres UCLA Jurjo Torres Santom Universidad de la Corua, Espaa