Exceptional ChildrenVol 72. No. 3. pp. 3! 7-331.©2006 Council far Exceptional Children.

The Sustainability of InclusiveSchool Reform

PAUL T. SINDELAR

DEIRDRE K. SHEARER

DIANE YENDOL-HOPPEY

TODD W. LIEBERT

University of Florida

ABSTRACT: FoT ovev u dccude. University of Florida researchers worked with middle schools in a

Urge urban and suburban south Florida district, as they developed and then worked to sustain

inclusive reform. One middle school Socrates, was notably successful, having built its inclusion

model on a foundation of previous reform and a school culture characterized by shared decision

making, collaboration, and teaming. For 4 years, we studied Socrates and the sustainability of its

program. Inclusion was not sustained; our analysis of teacher and administrator interviews

revealed three primary factors that help explain why: leadership change, teacher turnover, andstate

and district assessment policy change. Reduced support for the program, a by-product of the

primary factors, also contributed to the lack of sustainability.

The idea of including studentswith disabilities in general educa-tion classrooms is nearly 30 yearsold. In 1975, the Individualswith Disabilities Education Act

(IDEA) introduced the concept of insttuctingstudents in the least restrictive environment(LRE), and for many students with disabilities,the LRE is a general education classroom. Adecade later. Will's (1986) call for shared respon-sibility in educating students with disabilities setschools and researchers on a quest for successfulmodels of inclusion. During the 1990s, we wereable to follow one school's transition from tradi-tional special education to a teacher-developedinclusion program. This transformation took

place with the assistance of University of Floridaresearchers and a federal grant. Two years afterthe original project ended, we returned toSocrates Middle School (SMS) to study the sus-tainability of those reforms.

S U S T A I N A B I L I T Y R E S E A R C H

Despite the growing body of knowledge aboutschool reform and special education practices, re-searchers know httle about the extent to whichinnovations are sustained over time and what fac-tors influence their sustainability. In short, empir-ical research on sustainability factors is limited(Gersten, Chard, & Baker, 2000).

Exceptional Children

Research on the sustainability of innovationhas focused on both classtoom-specific (e.g., read-ing strategies) and schoolwide reforms (e.g., mag-net programs); (Gersten et al., 2000). Schoolwidereforms affect the structure and day-to-day opera-tion of a school, and research on them has beenlimited to general education reforms (Florian,2000; Huberman & Miles, 1984). By contrast,research on the sustainability of classroom-specificreform includes several special education studies(Gersten et al.; Klingner, Arguelles, Hughes, &Vaughn, 2001; Vaughn, Klingner, & Hughes,2000).

Despite the growing body of knowledge

about school reform and special education

practices, researchers know little about the

extent to which innovations are sustained

over time and what factors influence their

sustainability.

In studies of classroom reforms, researchershave identified three main factors related to sus-tainability: district and state policy, leadership,and teaching/classroom factors. Districts thatshow sttong commitment to a reform recognizeschools for adopting new practices and take mea-sures to ensure that principals follow through.Both actions have been linked to sustained use ofreforms (Klingner et al., 2001; Vaughn et al.,2000). In contrast, innovations stand less chanceof survival when districts are not committed tothem (Klingner et al., 2001) or when districts in-vest heavily in high-stakes assessments (Furney,Hasazi, Clark/Keefe, & Hartnett, 2003) and im-proving student outcomes on [hem (Klingner etal., 2001). A second factor affecting the sustain-ability of classtoom-specific innovation is schoolleadership (Klingner et al., 2001). Schools atwhich principals devote time to the developmentof an innovation are more likely to have teacherscommitted to its practice. Further, districts thatprocedurally rotate principals may have more dif-ficulty sustaining a classroom-specific strategythan schools where principals are retained(Klingner et al., 2001). The third factor is teach-ers' acceptance of the practice. Successful adop-

tion of innovative practices occurred when it wasconsistent with teachers' beliefs or teaching style(Klingner et al., 2001; Vaughn et al., 2000), whenthe practice helped the most difficult-to-teachstudents, and when teachers received supportivetraining (Gersten et al., 2000). Because learningnovel teaching practices places new demands onteachers, they will not sustain their use of innova-tive practices unless they see benefits for students(Gersten et al.; Klingner et al., 2001).

In research on the sustainability of school-wide reform, policy and leadership also play arole. In addition, school culture and factors re-lated to the innovation itself are likely to affectschoolwide reform. For the most part, district andstate policy and ptincipal leadership have thesame impact on schoolwide reform as on class-room-specific reforms (Florian, 2000; Huberman& Miles, 1984), but Huberman and Miles alsoemphasized the role of teacher leaders. Theynoted that a group of teacher leaders—whomthey call enforcers—often provide resources andencouragement to other teachers for adopting anew practice. Enfotcers understand the innova-tion and are invested in its continued use. Huber-man and Miles found that enforcers were oftenmotivated by the opportunities for advancementthat a new reform created in the district (Huber-man & Miles).

A third factor in the sustainability of school-wide reform is school culture: Schools with sharedvision and cultures of communication and shareddecision making, and schools that involve teach-ers in the design of an innovation, are more likelyto sustain innovations (Florian, 2000; Huberman& Miles, 1984). Another important element ofschool culture is teacher mobility. Huberman andMiles described two waves, one when an innova-tive practice commences and one later. The sec-ond wave, which is likely to have the most impacton sustainability, results from teachers whodemonstrate success with the innovation and ad-vance to higher positions. Successful teachers werefrequently enforcers in the Huberman and Milessense; when they left the classroom, support forand communication about the innovation waned(Huberman & Miles).

The final factor derived from the nature ofthe innovation itself. In their study of variouslarge-scale, longitudinal innovations, Huberman

3 1 8 Spring 2006

and Miles (1984) found that innovations thatwere smaller in scope and that placed fewer de-mands on teachers were more likely to take rootand be sustained. By contrast, innovations thatcreated too many demands or were too complexto understand did not have the success of moremanageable programs. Furthermore, innovationsthat required too many changes in the currentfunctioning of the school were less successful thanmore proscribed innovations (Huberman &Miles).

The inclusion of students with disabilities ingeneral education is one such complex and de-manding reform. Inclusion is often misunderstoodand sometimes resisted by teachers, and it is notalways fully understood or supported by schooladministrators. The Individuals with DisabilitiesEducation Act (1997) stipulates that students withdisabilities be educated in the least restrictive envi-ronment hut also requires that districts provide acontinuum of placement options. Thus, states anddistricts have some latitude with regard to IDEAimplementation, and as a consequence special ed-ucation practice varies dramatically from districtto district and state to state. Inclusion, in short,would seem to be a challenging schooiwide reformto establish and implement, and its sustainabilitywould seem difficult to achieve. No previous re-search on the sustainability of inclusive reform isavailable in the literature.

PURPOSE OF THE STUDY

I he purpose of the current study was to addressthis gap in knowledge about factors leading to thesustainability of a schooiwide special educationreform. For a decade, faculty from the Universityof Florida (UF) collaborated with teachers andadministrators at Socrates Middle School (SMS),first assisting with the design and implementationof an inclusion program, then focusing on thesustainability of the reform. Six years into the col-laboration, the researchers removed themselvesfrom rhe role of participant observers to assumean observational research stance that focused onrhe documentation of the sustainability effort. IntJiis article, we report results of a qualitative anal-ysis of teacher, administrator, and staff interviewsconcerning the factors that led to the rise and fall

of the school's inclusive reform. In doing so, weaddress two questions: Wouid inclusive reform besustained beyond the life of the initial university-school collaboration.' and What factors influencedsustainability?

METHOD

We addressed these questions of sustaining inclu-sive reform using a case study approach. Ethno-graphic case studies produce what Stake (1980)has called "naturalistic generalizations" (p. 69).Naturalistic generalizations are useful for evalua-tors because they identify hidden variables andproduce hypotheses researchers can later verifyusing quantitative strategies (Ashton & Webb,1986; Sherman & Webb, 1988; Smith, 1978).Case study research, as Fetterman (1983) has said,"with its close attention to details . . . can identifycausal features and causal linkages that may beoverlooked or misinterpreted on the basis of cor-relational analysis of survey data, or predeter-mined observational category systems" (p. 21).Naturalistic generalizations also are useful to prac-titioners who want practical guidance on programimplementation in similar settings. Detailed de-scriptions of program activities, identification ofbarriers, clarification of significant variables, sur-facing of taken-for-granted assumptions, andanalyses of causal program processes are useful tothose working in the setting being studied or oth-ers like it.

In our case studies, Bronfenbrenner's (1979)ecological framework served as a lens for our anal-ysis. We used rewriting (Stake, 1980), coding(PfafFenberger, 1988), and the constant compara-tive method of data analysis (Strauss, 1987) inour layered case study analysis of interview data.These activities appear linear, but the process wasrecursive and occurred throughout data collection(Spradley, 1980; Taylor & Bogdan, !984; Webb& Glesne, 1992). Rewriting involved transcribingindividual interviews (Spradley, 1979); coding in-volved attaching category names to basic units offield-generated data (Strauss & Corhin, 1990;Taylor &C Bogdan).

Initially rwo members of the research team,who were not involved in the study design or datacollection, independently coded the data, in the

Exceptional Children 919

process producing 47 codes. They discussed simi-larities among the first set of codes and collapsedthem into 16 codes. After discarding 3 codes thatdid not generate adequate support from the data,two additional members joined the tesearch teamfor the next stage of analysis. This group clusteredthe remaining 13 codes into 4 themes.

In an effort to enhance credibility, the au-thots engaged in multiple researcher triangula-tion. Researcher triangulation (Patton, 2002)involved multiple investigators in the analysis,three of whom were not involved as participantobservers during the Project RISES (Restructur-ing for the Inclusion of Special Education Stu-dents) and had little stake in or knowledge of thedevelopment of the SMS program. Additionally,the lead author's statement of researcher bias isprovided to make known any preconceived no-tions regarding the importance of sustainabiliry:

When we began the sustainability study, Ibelieved that teachers at Socrates MiddleSchool bad developed a first-rate inclusionprogram and tbat structures—supportiveleadership, collaborative culture, bigh-qualityprofessional development, and shared deci-sion making—were in place to sustain it. Ihad come to admire the SMS teachers andtheir commitment to students with disabili-ties, but seldom agreed completely witb tbedecisions they made about the inclusion pro-gram. As a result, entering into tbe sustain-ability study, I was curious about the shapethe program bad taken and wbetber SMSstakebolders retained tbeir commitment toit. 1 felt less investment in the particulars ofthe program; I knew its flaws and under-stood tbat I would have little opportunity tocontribute substantially to tbe process of im-proving it. Yet, in tbis regard, little badcbanged; it was tbe teacbers who shaped andrefined tbe original program, not RISES re-searchers.

The final four themes created a conceptualframework that organized and explained the vari-ables associated with successful or unsuccessfulschool restructuring for inclusion (Pfaffenberger,1988; Spradley, 1980; Strauss & Corbin, 1990).During these discussions, we paid special atten-tion to the micro- and macro-politics of educa-tion (Ball, 1987) and variables at all levels of theeducational system (Bronfenbrenner, 1979). Dur-

TABLE 1

Data Collection

School Year

1998-99

1999-00

2000-01

2001-02

2002-03

Interviews

Teacbers

Key Players

Administrators

Key Players

TeacbersAdministratorsKey Players

Teacbers

Key Players

Quantity

457

55

30118

30

11

ing tbis final analysis, we included key informantsfrom every level by member checking themes.

DATA COLLECTION AND PROCEDURE

From 1998 to 2002, we conducted individual in-terviews with 95 teachers and 16 administrators.Individual interviews generally lasted 30 to 50minutes, Tahle 1 describes the chronology of theinterviews; the term "key players" refers to stake-holders who contributed substantially to the de-velopment of the inclusion program duringRISES. All interviews were held in private, out ofeyesight and earshot of other school personnel.All data were typed into laptop computers ortaped and edited later. Because we did not typefast enough to capture every word, our contetn-poraneous transcriptions were not verbatim. Weleft some information out and compressed someideas into shorter statements; however, we believewe fairly represented stakeholders' comments andopinions. Because teachers and administrators atSMS had taken part in many research projects,they were familiar with the interview process. Ingeneral, most stakeholders were friendly, open,and cooperative. Their observations typically werethoughtful and candid.

B A C K G R O U N D

SOCRATES MIDDLE SCHOOL

Socrates Middle School is a large middle school ina large urban and suburban district in southeast

3 2 O Spring 2006

Florida. Like many schools in South Florida, itserves a fast changing, culturally diverse studentpopulation. From 1990 to 1996, the student pop-ulation at SMS grew from 1,200 to over 2,000students, and the faculty expanded from 60 toover 100 teachers. In the 1996-97 school year, anew middle school opened nearhy causing SMSto lose 700 students and 30 faculty. However, be-cause the district was growing so rapidly, enroll-ment at SMS recovered. By the 2001-2002 schoolyear, 1,765 students attended SMS.

To meet the challenge posed by populationchanges, the faculty and staff at SMS initiatedand sustained several reforms including shared de-cision making (SDM), a district-sponsored initia-tive, and the Coalition of Essential Schools(CES). In 1992, SMS collaborated with the Uni-versity of Florida (UF) on Project RISES, a U.S.Department of Education grant with a focus onincluding students with disabilities in the processof systemic reform. In the following section, webriefly describe the evolution of inclusion duringRISES; a more complete account can be found inKiigote, Griffin, Sindelar, and Wehb (2001,2002).

INCLUSION AT SMS

SMS moved toward inclusion gradually. In thelate 1980s, teachers first developed a model theycalled "ESE (Exceptional Student Education)with EASE," in which students with disabilitieswere taught by a single team of teachers at eachgrade level. ESE with EASE offered two place-ment options: general education classrooms withsupport as needed or self-contained ESE classestaught hy teams of content area and ESE teachers.Students who made the transition from ESE togeneral education classes were taught by the samecontent teachers, making the move less difficult.

The ESE with EASE program was considereda success by the faculty. Although teachers sawgrowth in special education students and theirclassmates, teachers and principals were not satis-fied with what they had created. The model stillsegregated special education students. Putting allstudents with disabilities on one team concen-trated theit numbers in inclusion classes, andsuch concentrations sometimes frustrated teach-ers, students, and parents.

Program development continued apace. Fac-ulty and staff, parents, district and state adminis-trators, and university researchers discussedlimitations of the ESE with EASE model andways to improve it. SMS made significantprogress toward implementing a broader vision ofinclusion during a series of retreats facilitated bythe UF research team, in 1994 and 1995. As facil-itators, we helped plan the retreats, worked tomake sure all stakeholders were heard, raisedquestions, reported on relevant research findingswhen asked, kept minutes, and reported thoseminutes to the faculty. The gtoup used severalprinciples to guide program improvement, amongthem (a) natural proportions, (b) general classplacement, (c) heterogeneous grouping, and (d)multidisciplinary intervention.

The movement toward inclusion was inter-rupted briefly when the principal was transferredduring the 1994-95 school year. However, thenew principal shared the faculty's commitment todemocratic governance, academic excellence,school improvement, professional development,CES principles, and inclusion. She quickly en-tered the school's ongoing conversation about in-clusion and supported the faculty's effort to solvethe problems they identified in the ESE withEASE Program.

The 1995-96 school year began with thenew teacher-developed inclusion model in place.All at-risk students were assigned proportionatelyto teams. Teams were assisted by co-teachers, andco-teachers helped all students who needed spe-cial attention, regardless of their classification. Sixfull-time co-teachers each worked with two orthree teams, while two half-time co-teachersworked with one team each. The exact role of theco-teachers was not prescribed by the planninggroup and evolved differently on diffetent teams.Most co-teachers worked in different ways withdifferent teams. The program was implementedin this form through the 1996-97 school year.

SUCCESS AND LIMITATIONS DURING

PROJECT RISES

The focus of our RISES research was on the cul-ture of the school and the organizational processesin place that allowed teachers to participate mean-ingfully in decision making. The project was

Exceptional Children S2 I

funded in a competition designed to supportstudies of efforts to include students with disabili-ties in the process of systemic teform. As a result,the effectiveness of the SMS inclusion programvi-as not the primary focus of our wotk, and ourdata relate primarily to how the school organizeditself to develop and implement the inclusionprogram. We collected no student achievementdata, but observed in classrooms and conductedfocus groups with teachers and students.

In focus groups and interviews, we askedteachers to identify practices that contributed totheit ability to include students with disahilitiessuccessfully, and, in response, they discussed the-matic units, cooperative learning, peer tutoring,explicit instruction, and small group and one-to-one instructional arrangements (Kilgore et al.,2001, 2002). Teachers learned and taught one an-other such techniques as split-page note taking,mnemonic strategies, and organizational strate-gies. Furthermore, SMS teachers consistentlydemonstrated high-quality gtoup instructionalpractices, which they supplemented with smallgroup and individualized remediation, as needed.

Despite enthusiasm for the inclusion pro-gram and other indicators of its success, problemsemerged (Kilgore et al., 2001, 2002). Some teach-ers felt they did not get enough help from theirco-teacher, and, in fact, the quality of implemen-tation varied from team to team (and sometimesfrom teacher to teacher within a team). Further-more, not all teachers were confident about theirability to accommodate students with disabilitiessuccessfully. (Math teachers tended to be morediscouraged with the performance of includedstudents than other teachers.) Teachers also wereconcerned about grading and the fairness of usingdiffetent criteria to evaluate different students.They were reluctant to adjust their evaluationmethods but disliked giving included studentspoor grades. Finally, in many cases, teachers wetecomfortable including students with less signifi-cant disabilities—students with learning disabili-ties, for example—but were not prepared forstudents with mental or emotional disabilities.

In spite of these problems. Project RISES wasconsidered a success when its implementation wascompleted in 1996. In 1998, UF tesearchers re-turned to SMS to study the extent to whichRISES reforms had been sustained. This second

project was called SIR, an acronym for SustainingInclusive Reform. In the interim, much hadchanged, both at the school and in the policycontext in which the school operated.

F I N D I N G S

STATE AND DISTRICT CHANGES

During the 1990s, Florida developed a statewidesystem of school accountability, initiated by thedevelopment of standards—the Sunshine StateStandards—that provided benchmarks in reading,mathematics, and writing. To assess studentprogress, the state developed and validated theFlorida Comprehensive Achievement Test, orFCAT, first administered in 1997. In the follow-ing year, Florida instituted the "A"̂ Accountabil-ity' system for measuring a school s progresstoward achieving the standards. School gradeswere first assigned in the summer of 1999 usingFCAT scores, the percentage of students tested,and the percentage of students making gains onthe FCAT Students with disabilities (except thosewith speech impairments) were not included inthe calculation of school gtades.

The Florida School Recognition Programprovided financial rewards and greater autonomyto schools that sustained high performance ordemonstrated exemplary improvement due to in-novation and effort. Schools receiving "A" gtadesand schools improving at least one grade were eli-gible for school recognition awards, which couldbe used for performance bonuses for faculty andstaff ot for equipment and materials. School advi-sory council members and school staff deter-mined how these funds would be used. Thedistrict and state placed great emphasis on im-proving test scores and school grades. The districtdeveloped curricula (keyed to the standards) andscripted teaching methods for each subject andgrade level. In addition, it announced it wouldsupport only "laser-focused" professional develop-ment that specifically addressed teachers' needsvis-a-vis assigned subject areas and approvedteaching strategies.

CHANGES AT SMS

Leadership. Before we returned to SMS, inthe spring of 1997, the SMS principal learned

Spring 2006

that she was to be transferred. Her replacement,the third SMS principal we worked with, hadbeen principal of a nearby elementary school buthad no middle school experience. She faced thedifficult challenge of fitting into a school withstrong traditions and a faculty with strong opin-ions, She diffeted from het predecessors in beingless familiar with and committed to either CES orinclusion principles.

Scbool Demographics. As the 1998-99 schoolyear began, SMS's student population increasedfrom 1,376 to 1,582 (see Table 2), which led toan increase in the number of teachers and staff. Ina departure from previous practice, new teacherswere not chosen because of their understanding ofand commitment to inclusion (or CES), and fewhad experience in inclusive or CES schools.Moreover, SMS continued to grow throughoutthe project. As a result of redistricting, the popu-lation of the school grew to 1,725 in 2000-01,and the number of teachers increased commensu-tately.

At the same time, the inclusion support staffwas reduced. In 2000—01 (see Table 2), for exam-ple, there was the same number of support staff asin 1996, when SMS served 500 fewer students.The number of full-time co-teachers had droppedfrom six to three from 1996 to 1999, and al-though there were still two part-time co-tcachers,guidance counselors and assistant principals nolonger served as co-teachers. Furthermore, onlytwo of the eight co-teachers from the PtojectRISES eta were still working at SMS.

Program Changes. Initially, SMS held to itsinclusion ideal, but significant events during the1998-1999 school year presaged change. For onething, co-teachers' responsibilities were expanded.In addition to their usual duties (collaboratingwith colleagues, adapting assessments, modifyingcurriculum, offering small group and one-on-oncinstruction, and completing individualized educa-tion programs and other paperwork), co-teachetswere asked to substitute teach and were assignedregular lunchroom duty.

SMS also began to provide services for stu-dents who performed poorly on the FCAT. Con-cern about meeting these students' academicneeds led to the establishment of a basic skills in-struction program, known as the Ditect Insttuc-tion (DI) Lab. The DI Lab served roughly 1 40

students found eligible on the basis of perfor-mance on screening tests. They participated in thelab 2 days a week for reading and 4 days a weekfor math, depending on their needs.

To pay for the DI Lab instructor, SMS in-vested less in co-teaching. Figure 1 shows the de-clining number of co-teachers (by total FTE)from 1995 to 2001; the decline in co-teachersfrom the original cohort was even more precipi-tous. These data show that SMS's investment inthe inclusion program decreased markedly overthese years. It was reduced first in 1997-98, whentotal FTE dropped from 7 to 5.5. By 1998-99,with the school's investment down to 5.0 FTE,full-time co-teachers worked with three or moteteams, and individual teams had assistance avail-able to them for no more than a day and a half aweek. In 2002, when the project ended, SMS em-ployed only 3.5 FTE co-teachets.

Between 1998 and 2002, SMS's A"*" Account-ability grade fluctuated between A and B (seeTable 2). After earning an A in 1998-1999, SMSadded a computer-based reading lab to its array ofstudent pullout services. Most students attendingthe Read 180 Lab, as it was called, scored betweenthe 20th and 40th petcentiies on FCAT reading,and most were not special education students.This group was targeted for intervention becauseSMS stakeholders believed that they were mostlikely to improve their test performance substan-tially; students who scored below the 20th per-centile apparently wete deemed poor investmentsfor remediation. Nonetheless, in large measuredue to the stringent criteria for sustaining Agrades, SMS fell to B in 1999-2000.

After a second B in 2001-02, SMS broughtits grade back up to A. During these 2 schoolyears, enrollment grew substantially, and the per-centage of students on free and reduced lunchpeaked (albeit at 16%). The percentage of stu-dents with disabilities also peaked in 2001-2002.Although the faculty grew by 10 positions from1999-2000 to 2001-2002, the number of fUll-time co-teachers dropped to three, and the num-ber of part-time co-teachers dropped to one. Moststudents with mild disabilities were included ingeneral education classes, but a self-contained classand tesource room were added (both serving pri-marily students with behavioral disorders), and

Exceptional Children SS3

•« 1 ^^

v^

Ov

o

q\6

ooo oo

324 Spring 2006

F I G U R E 1

Co-Teacher Turnover

DTotal FTE for Co-Teaching OOriginal Team

6 -i

5 -

4

3 -

2 -

1 -

iiT

II

I

IIii [

1995-96 1996-97 1997-98 1998-99 1999-2000 2000-01

both the DI and Read 180 Labs continued to op-erate.

During this time period, our observations ofteachers led us to conclude that the quality ofgroup instruction remained strong, and experi-enced teachers continued to use strategies theyhad learned previously. On the other hand, thereseemed to be less opportunity to share ideas andexpertise, and professional development focusedexclusively on FCAT preparation. Teachersseemed less collegial, and schoolwide activity gaveway to concentrating on the work of teams. Thus,although the quality of instruction remainedstrong, tbere seemed to be less potential at SMSto sustain its high-quality instruction.

ANALYSIS OF CHANGES

The purpose of this longitudinal qualitative studywas to identify factors affecting the sustainabilityof inclusion reform at SMS. Analyses indicatedthat the school underwent dramatic changes,moving from an inclusion program in the mid1990s to a special education program character-ized by a menu of pullout services and self-con-

tained placements. Our analysis suggests thatchanges in leadership, teacher turnover, and ashift in district and state priorities led to thesechanges. Reduced resources, a by-product of theseprimary factors, played a secondary role.

Shifting Leadership Priorities. Given keenpublic interest in school grades and the link be-tween funding and FCAT performance, the dis-trict's continued focus on accountability and testscores came as a surprise to no one. SMS's firsttwo principals were both effective at shieldingteachers from district pressures; however, teachersfelt the new principal at SMS was less effective:

Whenever the area office requests anything,che prtncipal stops everything else to fulfillthe request. (L. J.)

Our past principals have never said no to us.They always said "no" [ro] downtown, andwe would figure something out. Now it is ev-erything comes from downtown. (M, S.)

In defense ot the new principal, one teacher ac-knowledged the power of district pressure.

Exceptional Children 325

[The principal] has a pile of things to sharewith us ftom the district. They want moredocumentation, [Now] too much time spentin telling us how to teach and not enough inletting us do it. (P. B.)

The new principal also placed less emphasis onCES principles tban the previous principals had.Tbe teachers also began to sense less support forinclusion.

My first two principals carried che torch andinspired us to use the coalition principles tohelp kids. Now, there is no talk of purpose,no use of coalition principles and language tocreate a community educational agenda.There is no sense of teaching as a commongood in the service of the community,(M. L,)

She [current principal] will keep our schoolout of the red. She cares very much for thehottom line, but she may end up holding thehottom line by herself In the past, teacherswere emotionally in the black, rhe budgetwas in the red; today, teachers are in the redemotionally, and the budget is in rhe black.(W.J.)

Shifiing District and State Priorities. In con-trast to CES evaluation principles, which empha-size "unanxious expectation," state mandatedreform seemed coercive. Teachers felt that the dis-trict was pressuring schools to spend their timeand money on preparing students for the FCAT.Because SMS was historically a high-performingschool, there was added pressure to produce highstudent achievement and good school grades.Teachers put it this way: "I think that account-ability that the state puts on us . . . inhibits usfrom having time to do those kinds of things weused to do at SMS." (K. S.)

Despite the added pressure, many teachers atSMS reported that they did not change how theytaught and had no intention of doing so to meettbe needs of the FCAT: "I was already teachingthe thinking skills, the reading and writing in thecontext of the curriculum. I did not need tochange anything." (B. P.)

However, in practice, most teachers adoptedstate-developed practice materials. We observedmost of these activities in math, where manyteachers began lessons with FCAT practice prob-

lems. The activities themselves were organizedand afforded students direct practice on skills onwhich they would be assessed. Most teachers care-fully articulated FCAT practice activities to tbelessons they planned independently. All were ac-tive and directive during these activities, some-times more so than during the rest of the lesson.

In respect to content I don't teach to FCAT,However, I do spend 25 minutes warm-upperiod where I do FCAT content. (B, N,)

I did some more simulations and prompts;other than that, I did the same things I havealways done. (T. J.)

"In the past, teachers were emotionally in

the black, the budget was in the red; today,

teachers are in the red emotionally, and the

budget is in the black. "

Teachers on strong, longstanding reams—almostexclusively four-person, seventh- and eighth-gradeteams—were least likely to report changes in theirapproach to teaching. The quality of instructionon these teams had been and continued to be out-standing, and the practices they commonly em-ployed supported inclusion. In addition, strongteams occasionally shielded telatively inexperi-enced team members from ECAT pressures.

Our ream does very little FCAT drill. We letthe kids see what it looks like but we don'tspend time practicing for rhe test. What weare doing is working on specific skills, likeevidence pracrice in reading and social stud-ies, that has affected what we teach. We areconstantly asking ourselves, are we teachinghigher order thinking? (B, R.)

On the other hand, teachers on weak or newlyconstituted teams seemed to be more heavily in-fluenced by tbe FCAT pressure. Some of themtold us:

Absolutely, no doubt, the strongest influenceon my teaching is knowing that there will bean FCAT Test. It has become an ail-encom-passing goal. I was one of those people whobecame very stressed by it. I think I sacrificedsome of the creativity and warmth that Iwould have used. I got very businesslike. I

Spring 2006

may have made some of the kids stressed andtense. (B. M.)

There is no time for innovation; you have to[each to the FCAT test to some degree. Inno-vation takes trial and error, reflection, adjust-ment, and renewed effort. We don't havetime for trying new things, and trying newthings is nor valued here. It is all about get-ting the kids to score better on the test, notgetting them to learn things in depth or in anew way. (P. B.)

Thus, although FCAT pressures had a pro-found impact on SMS teachets' sense of commonpurpose and collegialiry, it did not necessarily di-minish the quality of teaching or hamper teachets'efforts to accommodate students with disabilities.Competent teachers, particularly those on strong,longstanding teams, were unlikely to change, andteachers who did change wound up adding an ex-plicit teaching activity to their lessons—a changefrom which students with disabilities were likelyto benefit.

Teacher Turnover. During Project RISES,SMS's teacher ranks were rife with enforcers.These key individuals worked closely with admin-istrators to develop the inclusion program andwere heavily invested in its success. However, be-tween 1996 and 1998, as the number of staff in-creased, tbe number of staff with knowledge ofthe history of inclusion diminished. The depar-ture of enforcers also impacted the inclusion pro-gram. Figure I shows that only one of the originalseven SMS co-teachers remained in 2001-2002.Furthermore, the turnover rate at SMS during the1990s averaged 21.3%; average turnover in sub-urban xScbools bas been estimated at roughly 15%(Ingersoil, 2001).

Stakeholders were well aware of thesechanges. They told us: "It is real hard to sustain[reform] if the principal doesn't stay there, espe-cially when she takes her faculty with her." (B. L.)"I think it is a level of general accepted maximthat it [inclusion program] was better and all thegood people left." (K. D.)

The turnover also diluted SMS s commit-ment to the inclusion philosophy. With key play-ers leaving and new teachers entering SMS, therewas less understanding and less acceptance of theideal.

1 have been dealing with inclusion for a longtime. When it doesn't work for some stu-dents there needs co be something in place totake care of them. I agree with the inclusionphilosophy, but it doesn't work for every-body. (C. R.)

I think that students that are behind forwhatever reason should get what theymissed. I shouldn't be teaching seventh grademath to kids that don't know fourth grademath yet. They can't do algehra when theycan't multiply, h is hard as a teacher to teachthat kind of a class. I think that everyoneloses. I feel really bad for the good students.(S. L.)

It also became cleat that not all teachers un-derstood what was meant by inclusion. We weretold: "We have a total inclusion program; severelybandicapped students are in a self-containedclass." (B. P.) "We have full inclusion, so we haveall kinds of students in the toom. There are somepull out programs, and we have a self-containedclass." (F. J.) "[We ate a] full inclusion school. Wedon't have separate, self-contained classes. It is anexcellent program. There are some separateclasses." (N. K.) "We have full inclusion exceptfor kids who get pulled out for needs. I guess thatis not really full inclusion, but that is what teach-ers call it." (F R )

The failure to socialize new teachers to SMSideals may have resulted from dilution of theschool culture. For example, less emphasis wasplaced on collegiality and collaboration. In re-sponse to a question about the status of shareddecision making, teachers told us:

No, it [SDM] is not as important. Theredoes not seem to be as much considerationof teacher voices; not as mtich support forcreative programs. There are some problemswith communication after decisions aremade; the results don't always get clearlycommunicated to the faculty. (E J.)

I am not even sure what it is any more. Iknow what a decision Is and what shared is.The joke around here is they will make a de-cision and share it with us. (G. K.)

Commitment to CES principles also waned. Newteachers in particular did not understand theprinciples or how to apply them in theit class-

Exceptional Children

rooms. Furthermore, the successful teamingmodel set up at SMS suffered as a result ofturnover and lack of consistent team membership.

We were given the essential principles to putin our room, but we have never really dis-cussed the coalition and the school's involve-ment with ir. No one really oriented us to it.I have not been to any coalition activity thisyear. (C. T.)

We only talk about the coalition in the newteacher program monthly meeting, not at theschool. I don't know anything about the waythe coalition works at the school. I have notattended any coalition activity yet. (L. B.)

Again [Teaming] not as much as it used tobe, and there has been a tremendousturnover of teaching staff. And if it is notyout planning or your lunch you dont get tomeet the new people. There are not as manysocial activities as there used to be, except bythe same little cliques. So, the new peopledon't show up. (H. S.)

Reduced Support for the Inclusion Program.

The SMS inclusion program was never without

flaws. RISES teachers bemoaned the lack of ade-

quate support and expressed uncertainty about

their ability to include students with disabilities

successfully.

From 1999-2001, teachers gtew even more

frustrated with reductions in co-teacher support.

The changing tenor of their comments—ftom

"She is a Superwoman!" (L. M., 1999) to "She is

spread too thin to be useful." (M. L., 2000)—re-

flected this frustration.

The Special Education Department is not fa-miliar with these kids. . . . Even at IEP meet-ings the Special Education person might notbe able to pick a kid out of a lineup. She'dcome in and ask me to point out kids she'ssupposed to be seeing. (B. M., 2001}

Our students with disabilities are pulled outof class and then put in special classes; thereis no support facilitation. (B. R., 2001)

Only one teacher ever told us she received ade-

quate support from her co-teacher. Resources for

the inclusion program had been depleted, and it

took a toll on the faculty.

It's ridiculous the number of teams they (co-teachers) are supposed to service. This sub-bing thing is ridiculous; there's a shortage ofsubs, and they are pulled away from theirteams, and their credibility is shot. A lot ofteams have not bought into inclusion be-cause they don't get the help they need andcan't depend on their co-teacher. (G. M.)

Furthermore, resources once invested in inclusionwere redirected to programs for students who didnot score well on statewide assessments.

We were in a parent conference yesterday andwe were asked: "Do we have a guidance per-son?" Before we had guidance at each level.Now we don't have adequate support for ourinclusion kids or our regular kids. (R. S.)

There are things that we need to be doing asco-teachers that we can no longer do. TheSOS lab, for example, where kids couldcome for help. Programs have been droppedthis year because there is not enough sup-port. (R. S.)

One teacher summed things up this way:

Putting kids that were not emotionally or in-tellectually equipped to handle the work intothe regular classes and not having the re-sources to help them be successful, later ondown the road . . . it wasn't good for them.When you have a small number of studentswith disabilities combined with supportteachers to help them be successful, inclusionworks, and under those conditions I supportit fully. It does not work for all students, inall settings. You have to honestly evaluate theconditions and the students. (V. J.)

D I S C U S S I O N

In our research on the sustainability of reform atSocrates Middle School, three primary assertionshelped us explain why inclusion was not sus-tained: leadership change, teacher turnover, andstate and district policy change. An additional fac-tor—reduced support for the program—also con-tributed to the lack of sustainability. In thissection, we discuss these fectors, how they interre-late, and how our understanding of these fmdingscompares with principles established in previousresearch. We also consider the implications of ourfmdings for school practice.

Spring 2006

Three primary assertions helped us explain

why inclusion was not sustained: leadership

change, teacher turnover, and state and

district policy change.

LEADERSHIP CHANGE

SMS changed principals twice, once during theinitial project and a second titne in 1997, a yearbefore the second project began. Previous litera-ture on special education classroom reform sug-gests that leadership change may underminesustainability, particularly when a new principaldoes not devote time to the innovation (Klingner,Vaughn, Hughes, & Arguelles, 1999). Our find-ings about schootwide reform were consistentwith this assertion. Because rhe first new principalwas committed to her predecessor's reform agendaand was successful at keeping teachers workingtowards the inclusive vision, that change had littleimpact on the program. By contrast, the secondnew principal seemed more committed to recentdistrict initiatives related to student assessmentsand school grades. She expressed support forCES, inclusion, and shared decision making, thebackbones of SMS reform, but she understoodthat her school—and her own performance as itsleader—would be judged primarily on how wellSMS students scored on the FCAT.

In their study of classroom specific reform,Klingner et al. (2001) determined that procedu-rally rotating principals added ro the difficulty ofsustaining reform. To this generalization, we wouldadd that changes in school leadership may affectschoolwide reform in different ways, dependingupon the principal's affinity for and commitmentto an established schoolwide reform agenda.

TEACHER TURNOVER

Teachers transferred out of SMS at a surprisinglyhigh rate, even before the second principalchange. In a sense, all of the principals broughton new teachers whose beliefs were aligned withthe prevailing reform philosophy. Thus, duringthe tenure of the third principal, teachers werehired with less attention paid to their knowledgeof inclusion and their commitment to the co-teaching model than had been the case before we

returned to SMS. It is not surprising that thisprocess diluted faculty commitment to the origi-nal reform or that, as a group, the SMS facultygrew less knowledgeable about including studentssuccessfully—and less enthusiastic about trying.

Teacher turnover had a second deleterious ef-fect on the inclusion program. Many of the teach-ers who left were the enforcers {Huberman &Miles, 1984) who had helped to create and estab-lish inclusion at SMS. Huberman and Milesnoted that enforcers used reform expertise todemonstrate their leadership and move to admin-istrative positions, and they considered such up-ward mobility a problem inherent in reform. Ourfindings at SMS were consistent with these asser-tions; several enforcers devoted to the inclusionprogram left: with the first two principals and fol-lowed them to new schools, often in leadershiproles. Others left with the arrival of the secondnew principal. If other teachers were beinggroomed for leadership roles—for the inclusionprogram or otherwise, it was never apparent to usor to the teachers we interviewed. Apparently, atSMS, teacher leadership was a victim of highteacher turnover and the loss of focus that camewith new and competing reform.

CHANGE IN STATE AND DISTRIGT POLICY

Long before the passage of the No Child Left Be-hind Act in 2001, the state of Florida initiated aprogram of high-stakes assessments and account-ability in which schools were graded, largely onthe basis of FCAT performance. Previous work onthe sustainability of both classroom specific andschoolwide reforms (Huberman & Miles, 1984;Klingner et al., 2001) emphasized the importanceof the match between a reform and rhe policycontext in which it is implemented. Furney et al.(2003) concluded that high-stakes assessment wasa poor policy context for classroom-specific re-forms; our findings suggest that high-stakes as-sessment also proved to be a poor context forinclusion, a schoolwide reform. The third princi-pal began at SMS a year before school grades werefirst awarded, as pressure to perform was beingratcheted up. She made clear to her teachers theimportance of improving FCAT performance. Allteachers felt pressure; they were being held tohigh accountability standards but lacked the re-

Exceptional Children

sources to meet all students' needs. They did nothave adequate co-teacher support, social supportsfor students, or training ro deal with studentswith diverse needs. Programs like the DI Lab wereadded that focused on low-performing—but notnecessarily special education—students. Thetewas less communication and teaming as teachersbegan to sacrifice creativity to the demands of testpreparation.

Research on classroom reform has found thatteachers are mote likely to use new practices ifthey see improvements in students, especiallyhard-to-teach students {Klingner, et al., 1999;Vaughn, Klingner, & Hughes, 2004). At SMS,district accountability poHcy forced teachers to re-define student improvement as performance onstandardized tests rather than performance onmore sensitive academic assessments (e.g., curricu-lum-based measures) or measures of social growth.Redefming success on the basis of academic testperformance obscured the benefits of inclusion,particularly for students with disabilities and otherstudents with learning difficulties, and thus under-mined the sustainability of the reform.

IMPLICATIONS FOR SCHOOL PRACTICE

Our findings make clear that change in schoolleadership can sustain reform—or drive new re-form, depending on the new leaders commitmentto a particular agenda. At SMS, we witnessed anexample of both. The first principal change hadlittle impact on the program because hoth thefirst and second principals were committed to in-clusion and the broader reform agenda on whichit was built. The second principal change coin-cided with the emergence of a new reformagenda—high-stakes assessment and school ac-countability. As a result, inclusion and high-stakesaccountability were pitted against one another,with the new principal thrust into the role ofchampioning a new agenda. Her circumstancewas unenviable: Marching orders in hand, she en-tered an environment where most teachers re-mained committed to inclusive reform.

Given state and federal pressure to imple-ment high-stakes assessments and judge schoolperformance, district policy to improve FCATscores overwhelmed its commitment to inclusivereform—and that priority was communicated in

no uncertain terms ro principals. In fairness, thereis no way to determine how the previous princi-pals would have dealt with such pressure. Further-more, it does not seem likely that the secondchange was intended to subvert inclusion at SMS.After all, the second principal was reassigned—tothe largest high school in the district, a prestigeposition, and the new principal had administra-tive experience at a nearby feeder school. In alarge district, principal change is a gamble; some-times it works to sustain reform, and sometimes itdoes not. That such a change would render SMSvulnerable to the powerful pressure of high-stakesassessment and the need to improve test perfor-mance, in retrospect, seems obvious.

SUMMARY AND IMPLICATIONS

In sum, we found three major factors than con-tributed to the demise of the inclusion program atSMS: changes in leadership, shifting district/statepolicy, and teacher turnover. These factors also ledto diminished philosophical and financial com-mitment to the reform. In turn, lack of resourcesstressed an already fragile program structure. Thisstudy provides further support for the importanceof strong principal leadership, proper teachingtraining, and adequate resources in maintainingreform. It adds to the literature by demonstratingthese same factors are important in schoolwidespecial education reform.

R E F E R E N C E S

Ashton, P. T , & Webb, R. B. (1986). Making a differ-ence: Teachers' sense of efficacy and student achievement.New York: Longman.

Ball, S. J. (1987). The micro-politics of the school: To-

wards a theory of school organization. London: Methuen.

Bronfenbrenner, U. (1979). The ecology of human devel-opment: Experiments by nature and design. Cambridge,MA: Harvard University Press.

Education of Al! Handicapped Children Act. (1975).Pub. L. No. 94-142.

Fetterman, D. M. (198.3). Ethnography in educationalresearch: 'I he dynamics of diffusion. Educational Re-searcher. 11(1), 17-22.

Florian, J. (2000). Sustaining education refhrm: Influen-

tial factors. (Report No. EA 031023). Aurora, CO: Mid

Continent Research for Education and Learning. (ERICDocument Reproduction Service No. ED 453 583)

3 3 OSpring 2006

Furney, K. S., Hasazi S. B., Clark/Keefe, K., & Hart-nett, J. (2003). A longitudinal analysis of shifting pol-icy landscapes in special and general education reform.Exceptional Children, 7(9,81-94.

Gersten, R., Chard, D., Baker, S. (2000). Factors en-hancing sustained use of research-based instructionalpTACiices. Journal of Learning Disabilities, 33, 445—457.

Huberman, A. M., & Miles, A. B. (1984). Innovationup close. How school improvement works. New York:Plenum.

Individuals with Disabilities Education Act of 1997.(1997). Pub. L No. 105-17,611 etseq,

Ingcrsoll, R. (2001). Teacher turnover and teachershortages: An organizational analysis. American Educa-tional Research Journal, 38, 499-534.

Kilgore, K., Griffin, C. C , Sindelar, P. T, & Webb, R.B. (2001). Restructuring for inclusion: A Story of mid-dle school inclusion (Part I). The Middle School Journal,35(2), 44-51.

Kilgore, K., Griffin, C. C , Sindelar, P. T, & Webb, R.B. (2002), Restructuring for inclusion: Changingteaching practices (Part II). The Middle School Journal,33(3), 7-13.

Klingner, J. K., Arguelles, M. E., Hughes, M. T, &Vaughn, S. (2001). Examining the school wide"spread" of research-based practices. Learning DisabilityQuarterly, 24, 221-254.

Klingner, J. K.. Vaughn, S., Hughes, M. T, & Ar-guelles, M. E. (1999). Sustaining research-based prac-tices in reading. Remedial and Special Education, 20(5},263-274.

No Child Left Behind Act: Reauthorization of the Ele-mentary and Secondary Education Act. (2001). Pub. L.No. 107-110, §2102(4).

Patton, M. Q. (2002). Qualitative evaluation and re-

search methods (3rd ed.). Thousand Oaks, CA: S^e.

Pfaffenberger, B. (1988). Microcomputer applications in

qualitative research. Newbury Park, CA: Sage.

Sherman, R. R., & Webb, R. B. (Eds.). (1988). Quali-tative research in education: Focus and methods. London;Falmer.

Smith, L. M. (1978). An evolving logic of participantparticipation in educational ethnography and other casestudies. In L Shulman (Ed.), Review of Research in Edu-cation, (Vol. 6, pp. 316-377). Ithaca, IL: Peacock Press.

Spradley, J. P. (1979). The ethnographic interview. NewYork: Holt, Rinehart, & Winston.

Spradley, J. P. {1980). Participant observation. NewYork: Holt, Rinehart, & Winston.

Stake, R. (1980) The case study method in social in-quiry. In H. Simons (Ed.), Towards a science of the sin-gular (pp. 62-73). Norwich, England: Centre forApplied Research in Education.

Strauss, A., & Corbin, J. (1990). Basics of qualitative re-search methods: The search for meaning. Newbury Park,CA: Sage.

Strauss, A. L. (1987). Qualitative analysis for social sci-entists. New York: Cambridge University Press.

Taylor, S. J., & Bogdan, R. (1984). Introduction to qual-itative research methods. New York: John Wiley & Sons.

Vaughn, S., Klingner, J. & Hughes, M. (2000). Sus-tainability of research-based practices. Exceptional Chil-dren, 66, 163-171.

Vaughn, S., Klingner, J. K., & Hughes, M. T (2004).Sustainability of research-based practices: Itnplicationsfor students witb disabilities. In A. M. Sorrells, H. J.Rieth, & P T. Sindelar (Eds.), Critical issues in special ed-ucation: Access, diversity, accountability {pp, 135-153).Boston: Allyn &C Bacon.

Webb, R., &c Glesne, C. (1992). Teaching qualitativeresearch. In M. D. LeCompt, W. L. Mlllroy, &C J.Preissle (Eds.), The Handbook of qualitative research ineducation (pp. 771-813). San Diego, CA: AcadetnicPress.

Will, M. C. (1986). Educating children with learningproblems: A shared responsibility. Exceptional Children,52, 4\ 1-415.

ABOUT THE AUTHORS

PAUL T. SINDELAR (CEC FL Federation

#1024), Professor, Department of Special Educa-tion; DEIRDRE K. SHEARER, Doctoral Candi-date, Department of Educational Psychology;DIANE YENDOL-HOPPEY, Assistant Ptofessor,School of Teaching and Learning; and TODD W.

LIEBERT, Research Assistant, Department ofCounselor Education, University of Florida,Gainesville.

Address all correspondence to Paul Sindelar, De-partment of Special Education, G-315 NotmanHall, 605 SW 13th Street, Gainesville, FL 32611-7050 (e-mail: pts@coe.ufl.edu).

Manuscript received October 2004; acceptedApril 2005.

Exceptional Children