USING TECHNOLOGY-BASED TEACHER SUPPORT TOOLS TO SCALE UP THE CLASSWIDE PEER TUTORING PROGRAM
Mary Abbott Charles R. Greenwood
Jay Buzhardt Yolanda Tapia
Juniper Gardens Children’s Project, University of Kansas, Kansas City, Kansas, USA
The purpose of this paper is to describe the role of computer and infor-mation technology in scaling up research-validated instructional strategies like ClassWide Peer Tutoring (CWPT). Recently, implemen-tation software, the CWPT Learning Management System, web and e-mail communications, and interactive multimedia resources have been developed to support teachers’ and schools’ use of CWPT. Prelimi-nary findings suggest that this technology improves access and use of CWPT, and that when local school factors of implemenatation are in place, improvement in student academic scores replicate past CWPT results. Results suggest that the initial and ongoing professional development, the strength of local school leadership, and a building faculty’s ability to work with information technology are key elements to widespread implementation and accelerated student learning. The implications of such are discussed.
This work is supported by a Steppingstones of Technology Innovation—Phase 3 (H327A000038) grant from the Office of Special Education Programs (OSEP), US. Depart-ment of Education. Opinions expressed herein are those of the authors and do not represent the position of the U.S. Department of Education. Additional support was provided by the Kansas Mental Retardation and Developmental Disabilities Research Center (HD 02538).
Address correspondence to Dr. Mary Abbott, Juniper Gardens Children’s Project, 650 Minnesota Avenue, 2nd floor, Kansas City, KS 66101. E-mail: [email protected]
ISSN: 1057-3569 print=1521-0693 online DOI: 10.1080/10573560500203525
INTRODUCTION
Among its various provisions, the ‘‘No Child Left Behind Act’’ (NCLB) (U.S. Department of Education, 2002) requires that schools employ instructional practices supported by rigorous evaluations of effectiveness and the NCLB provides consequences for schools whose students fail to achieve these standards. As a result, a move toward large-scale implementation of research-validated instructional prac-tices has become of heightened interest for many school districts. Large-scale implementation, however, is not easily accomplished. Elmore (1996) envisioned four main components for large-scale program implementation:
1. create strong professional development structures that include an external system of evaluation of good teaching practice, 2. create organizational structures small enough to intensify and
focus on systematic change,
3. establish an intentional process for reproducing large-scale change, and
4. establish structures that promote the learning of new practices and incentive systems that support them.
Professional development (PD) questions then arise regarding how to create these strong organizational structures to promote tea-cher access, learning, and the implementation of research-validated strategies. How is an effective practice best taught to teachers? How is it best migrated to and replicated by others at reasonable cost? How can PD be made to support and motivate classroom- and school-wide implementation? The most advanced PD methods for this purpose have included partnerships between schools and uni-versity researchers or the creation of a school-district system of tea-chers as researtea-chers, responsible for translating research-validated strategies into local practice (Boudah et al., 2001). PD forms like these may or may not be available due to costs and because the developers of research-validated practices have not sought to promote use of their practice beyond initial research studies (Greenwood & Abbott, 2001).
However, the use of computer and information technology (CIT) could be an efficient way to increase awareness, access, and implementation support either alone or as a hybrid with PD models. CIT overcomes the need for an increasingly larger staff required to provide PD in different locations for an increasingly large population
of users. CIT also addresses the difficulty of accessing information for decision making and the materials and tools needed for PD and guiding implementation. The purpose of this paper is to describe the use of CIT as an integral component in implementing the long-distance use of the Class Wide Peer Tutoring—Learning Management System (CWPT—LMS) (Greenwood et al., 2001b) in schools.
BACKGROUND
CWPT is a reciprocal, peer-mediated instructional strategy in which members of the same classroom tutor one another using the regular classroom curriculum. Unlike teacher-mediated instruction, CWPT makes it possible for students to experience one-on-one, pupil–tutor dyads, relatively immediate error correction, and fast-paced instruc-tion with multiple opportunities to respond. Students experience both the teacher and learner roles in a classroom, with students with dis-abilities included for instruction, simultaneously providing opportu-nities for peer social interaction (Delquadri et al., 1986; Mortweet et al., 1999; Sideridis et al., 1997; Utley et al., 1997).
CWPT is supported by more than two decades of experimental research (Greenwood, 1997; Greenwood et al., 2002). Effect sizes across a number of randomized trials using CWPT have consistently fallen in the ‘‘medium’’ to ‘‘large’’ range (i.e., 0.3 to 0.8 ES; Cohen, 1988) on indicators of achievement and improvement in classroom engagement and social behavior (see Table 1). One twelve-year longi-tudinal study demonstrated that students who used CWPT experi-enced significantly higher growth in grades 1 to 4 (Greenwood et al., 2001b); continued higher growth in achievement into middle school, with fewer special education services (LD, EMR, BD, Conduct Disorders, ADHD) (Greenwood et al., 1993c); and produced significantly fewer students who dropped out of school prior to gradu-ation from high school (Greenwood & Delquadri, 1995). Other forms of CWPT, such as Peer Assisted Learning Strategies (PALS) based on CWPT procedures, have proven similarly effective in younger and older students with and without disabilities (see Table 1).
In summary, CWPT is a research-validated strategy that acceler-ates the academic outcomes of students with a wide range of abilities. Further, findings have established that when teacher fidelity is improved using feedback and consultation, so too is student learning (Arreaga-Mayer, 1998; Delquadri et al., 1990; Greenwood et al., 1992). However, if CWPT is to be widely adopted by large numbers of schools=districts in the country, PD issues of long-distance training and teacher fidelity must be addressed.
Table 1. Methodologi cally rigorous evidence (randomiz ed trials) supportin g the efficacy of CWPT C itation Des cription Indic ator Effect size Gre enw ood, Delqua dri, & Hall (1989) y Prospe ctiv e, longitudinal, randomized CW PT trial, 1st– 4th g rades (N ¼ 416) R ead ing ac hie vemen t La nguage ac hi ev emen t 0.57 0.60 Arit hmeti c ac hie vemen t 0.37 Acad emic eng agement 1.41 Gre enw ood (1991) y Multiy ear beh avioral trajectories, Acad emic eng agement 0.63 1st– 3th g rades (N ¼ 115) T ask man agem ent 0.61 Inap propriat e beha vior 0.83 Gre enw ood et al. (1 993c) y F ollo w-u p at se venth gr ade R ead ing ac hie vemen t 0.39 (N ¼ 303) La nguage ac hi ev emen t 0.35 Arit hmeti c ac hie vemen t 0.57 Soc ial studies ac hie vemen t 0.39 Scien ce ac hie vemen t 0.48 R educ tion in SPE D ser vice s 0.54 " Les s restrictiv eness ser vic es 0.73 " Gre enw ood & Delqua dri (1995) y F ollo w-u p at tw elfth gr ade (N ¼ 231) R educ tion in sc hool dropout 0.66 " Ma thes, Ho w ard , Allen , & Fuc hs (1998) (N ¼ 96) R and omized P ALS trial, gr ade K W oodcoc k w ord ID W oodcoc k w ord attack 0.70 0.78 W oodcoc k comp rehension 0.27 CBM (lo w ac hie vers) .03–1.35 Fuc hs, Fuc hs, T hompso n et al. (2001) (N ¼ 379) R and omized P ALS trial, gr ade K Segm entatio n, b lend ing Alp habetic s 0.45–2.1 0.02–1.96 Fuc hs, Fuc hs, Phi llips, et al. (1995) (N ¼ 120) R and omized P ALS trial, gr ades 2–4 Ma th ac hie vement 0.34 Fuc hs, Fuc hs, M athes, & Simmo ns (1997) (N ¼ 120) R and omized P ALS trial, gr ades 2–6 R ead ing ac hie vemen t 0.22–0.56 Not e . Ef fec ts size s are Cohen ’s d ; "¼ ef fect size calcula ted from chi -square as w . y These pee r-re vie w ed pub lications repor t th e lon gitudinal ac hie vemen t, beha vio r and life ev ent ef fects of a sin g le CWPT trial. 50
THE USE OF COMPUTER AND INFORMATION TECHNOLOGY (CIT) WITH CWPT
To address these issues, subsequent work led to a computer-assisted learning management system. The CWPT Learning Management System (CWPT—LMS) is a software application that provides teachers with a convenient way to create and deliver peer-tutored content, monitor student participation, and assess student perform-ance and progress.
In an early project, an interactive multimedia CD was developed that contains text, photos, and video illustrations of CWPT that could be viewed any time at a teacher’s convenience (Greenwood et al., 1993a). A second effort assembled experts’ knowledge about CWPT implementation problems and solutions and created a ‘‘rules-based advisor’’ system for teachers. Based on student progress, the weekly advisor component provided suggestions for improving CWPT implementation (Greenwood et al., 1993b). For example, the program advises the teacher when the tutoring material may be too difficult or too easy.
Finally, to promote high quality implementation, an electronic record-keeping system was developed to help teachers manage CWPT data (Greenwood et al., 2001a). The system also permitted teachers to randomly or manually form weekly peer-tutoring part-ners and teams, time the duration of daily tutoring sessions, enter tutoring point scores, and record weekly pre–post scores of mastery or fluency in the actual sequence of real-time classroom implemen-tation of CWPT. Other tools in the system allowed teachers to eas-ily produce graphs of student and group progress and use this information to make instructional decisions. These graphs were also used as reports for parents during parent–teacher conferences (see Fig. 1). These CIT components were joined with the previously developed CWPT instruction technologies to form the CWPT— LMS. During the process of revising and creating CIT materials, an advisory board of teachers, administrators, parents, and research staff involved in former CWPT projects reviewed materials and made suggestions. Improvements were made based on advisory board suggestions.
A BLUEPRINT FOR REACHING SCALE WITH FIDELITY The CWPT—LMS program was only part of what was needed to prepare for large-scale implementation. Also needed was a plan or blueprint for reaching scale. Fig. 2 shows the finalized CWPT—LMS
blueprint that has four components: core, support, replication, and information technology. Core components are procedures focused on designing and integrating CWPT into classroom instructional applications (e.g., second grade spelling) to yield a comprehensive curriculum. This component contains information about skills to be taught and the unique peer-mediated aspects of CWPT that
Figure 1. CWPT—LMS graph of mean class spelling accuracy over weeks for one teacher with high implementation fidelity.
Figure 2. CWPT—LMS scaling up blueprint.
complement conventional, teacher-led instruction (e.g., Elmore, 1996; Utley et al., 1997).
Support components are products and tools that make CWPT usable by classroom teachers. For example, the CWPT—LMS includes information on the peer-tutoring process, curricula, student progress monitoring, and computer-assisted analysis of implemen-tation. These tools support and guide weekly planning, daily implementation, and instructional decision-making to improve indi-vidual and group learning in CWPT (e.g., Carnine, 1997).
Replication components are procedures that support migrating CWPT to diverse, new, local school programs using a trainer-of-trainer model (see Fig. 3). At the top of this model is a national or state consultant (NSC), a person with the ability to recruit and involve schools. The consultant recruits at least two schools for involvement by working with local schools and by recruiting a site coordinator (SC) to act as the lead trainer and coordinator of CWPT—LMS implementation. The SC works directly with a building facilitator or principal, who in turn works with teachers, who in turn work with students and parents. These components involve elements of marketing and public awareness (see, for example, http:==www. kansaskidshealth.org=archived detail.asp?ProgramID¼4), combined with leadership training and professional development (e.g., annual CWPT Training Institutes) in an effort to identify and recruit new sites. As in other professions, the model of competing in a market-place, ‘‘supply and demand’’ environment has become a reality
in educational large-scale replication efforts. The CWPT—LMS blueprint fits this model. On the supply side, the CWPT—LMS blueprint is an effort to increase the supply of products and prac-tices based on rigorous empirical evidence. On the demand side, federal policies and regulations (e.g., No Child Left Behind Act of 2002) have boosted the ‘‘market value’’ of evidence-based practices. These mandates and policies contain economic consequences for failing schools and provide incentives for use of effective practices. Gladwell (2000) argued that in a market economy, the spread of social phenomena, like fads, can be explained by the concepts of epi-demic and contagiousness. According to Gladwell, ‘‘exposing people to phenomena infects them so that they will do the same as others’’ (p. 6), and ‘‘small individual changes eventually lead to large group effects.’’ The tipping point, he argued, is where ‘‘an idea, trend, or social behavior crosses the threshold, tips, and spreads like wild fire.’’ Market forces with powerful implications for reaching scale are addressed in the CWPT—LMS blueprint. Computer and information technology play a key role in this scaling-up plan.
Computer and information technology (CIT) components support world-wide electronic communications and immediate access to CWPT—LMS information (e.g., web site., document information, procedures, e-mail, and databases). Teachers can install the two-CD CWPT—LMS set on any classroom desktop computer and link in a number of ways to a set of CWPT resources. In the Help menu, teachers can access an electronic copy of the CWPT—LMS teacher’s manual, the CWPT—LMS web site, or the Interactive multimedia CD among other tools (see Fig. 4). For example, from the web site links below, teachers can take a CWPT—LMS tour or engage in tutorials showing how it works:
1. http:==www.lsi.ku.edu=jgprojects=cwptlms=html2002=Virtual Tour=virtual_tour.htm
2. http:==www.lsi.ku.edu=jgprojects=cwptlms=html2002=Flash% 20Folder=Introtutorial.swf
3. http:==www.lsi.ku.edu=jgprojects=cwptlms=html2002=Flash% 20Folder=ProgSuppTutorial_v6.swf
CIT components make CWPT readily accessible to teachers and administrators. CIT also establishes a central database for the receipt of individual student progress and other valuable indicators (e.g., school and teacher implementation) that are of interest to program evaluation and empirical research on scaling up. The computer and
information technology offered within the CWPT—LMS has the potential to aid in streamlining PD because it can facilitate fas-ter access, provide efficient communication, support individual implementation, assist data collection and analysis, and provide problem solving advice.
For example, e-mail and the Internet create speedy communication among various project participants within and across sites, negating barriers of geography. Personal CWPT software for teachers sup-ports weekly planning and daily implementation. Adapting CWPT instruction to fit group and individual learner needs becomes faster and more accurate because students’ learning results are captured in software or web forms and displayed immediately for use in decision making. Weekly progress data may also be transported over the web directly to a centralized database for centralized analysis, evaluation, and research on scaling up (Bullock & Ory, 2000).
In the current project, we have tested the CWPT—LMS program in a two-year implementation study. Below, we address the following research questions based on the first year’s implementation effort.
1. What was the level of successful CWPT—LMS implementation and research participation, and what local contexts appeared to moderate it?
2. Where implementation was ‘‘on protocol,’’ what spelling out-comes were achieved by students?
METHOD Participants
Using the steps enumerated in the replication components (see Fig. 3), we recruited Consultants (NSCs) by contacting a list of CWPT-experienced faculty around the country, posting an announcement
on theCouncil for Exceptional Children’s website, conducting
presen-tations, and disseminating by word of mouth. Four NSCs recruited one or two site coordinators (SCs) and five implementation schools in four states. SCs were a mix of principals, graduate students, and teachers willing to assume this role (Abbott et al., 2003).
Schools and teachers agreeing to participate committed themselves to having Internet access in each classroom with a computer to run the LMS software, providing time for professional development, implementing the CWPT—LMS in spelling or reading, transmitting data back to researchers electronically, and including at least one student with a disability in every classroom. Participating K–5 elemen-tary schools ranged from 14 to 73 teachers, with student populations ranging from 160 to 745. Schools were public and private=parochial located in urban cities in Florida and Maryland and rural settings in Nebraska and Mississippi. The researchers=developers were located in Kansas City.
In summary, we sought to draw together our past experiences implementing research-validated practices in schools (Abbott et al., 1999) and Elmore’s ideas concerning a strong PD structure. This structure included an external teacher system of evaluation within local organizational structures small enough to intensify and focus on systematic change using the trainer-of-trainer model (see Fig. 4). In this approach, the NSC supported the SC, who established teacher access to and use of the CWPT—LMS and who trained and supported building personnel and monitored their progress.
Training National Site Consultants (NSC) and Site Coordinators (SC) In prior CWPT projects, the researchers=developers had provided teacher training and used onsite implementation visits, direct teacher consultation, and observational data collection. In this project, the NSC, SC, and local school staff took over these duties in the context of the CWPT—LMS blueprint. The researchers assisted the NSC and
SC by phone and email, but had limited problem-solving contacts with teachers or local tech staff. The training of district personnel on how to train local teachers became paramount to successful large-scale implementation.
We conducted a three-day leadership institute in Kansas City to prepare NSCs and SCs with the skills, media, and materials needed to implement the CWPT—LMS at their locality. The training taught participants how to implement CWPT, use the CWPT—LMS pro-gram, recruit teachers, negotiate contracts with local schools, deliver technical assistance, and monitor, report, and evaluate site progress. Prior to attending the Summer Institute, participants received a copy of the CWPT—LMS software and resource CDs. Participants were asked to install the software on a laptop computer and complete a short LMS assignment. In the institute, we trained the participants using the same instructional format that we wanted them to use in fall teacher in-services. The first two morning sessions included a Power-Point presentation covering relevant content. Afternoon sessions were spent engaged in hands-on computer applications of the mor-ning’s lessons. The last day included a conversation about adminis-trative and research issues.
We created two PowerPoint presentations, organized according to key statements concerning CWPT. Key principles were supported by a detailed explanation of important information to convey during teacher training. Additionally, a web simulation CD was created that allowed participants to view most components of the website without having an Internet connection. Information about accessing web forms and data transmission was covered.
Teachers’ Training, Software, and Materials
Teachers’ implementation materials included the CWPT—LMS two-CD set, containing the LMS software and the multimedia resource CD. The resource CD provided supplementary instruction and training that a teacher or any person interested in the CWPT— LMS program could access anytime. The supplementary resource CD included the CWPT manual (see Fig. 2), the CWPT—LMS Gallery of illustrative videos and pictures detailing procedures of the CWPT–LMS program, and a tutorial that navigated through the major LMS components. The tutorial began with how to launch CWPT—LMS and continued through the creation of a class roster, data entry, graphs creation, accessing the Advisor system, and finally the FTP procedure for data transmission. These training and resource materials provided a layering of information from initial training through PD to improve existing practice. Upon returning
to their localities, each consultant–coordinator team prepared a schedule for implementing teacher training and CWPT—LMS implementation across the 2000–01 school year.
CWPT—LMS Implementation
We used five criteria to determine the level of each teacher’s implemen-tation of the CWPT—LMS. Implemenimplemen-tation was measured as the percentage of registered teachers at each school who had completed:
1. registration at the website, thereby transmitting a confirmatory web form
2. administration and transmission of target student CBM assess-ments
3. implementation of the CWPT instructional process
4. CWPT—LMS setup, use, and transmission of weekly student progress data
5. fidelity of implementation checklists returned for each teacher. These five tasks could have been initially completed as early as October, but in most cases took longer to complete.
Measures of Student Progress in the Spelling Curriculum
Classroom teachers collected weekly pre- and post-test measures of student progress in the spelling curriculum and analyzed these data using the CWPT—LMS program. Teachers created, administered, and graded a test of spelling words tutored each week and created a set of words to be tutored in the upcoming week (Greenwood & Hou, 2001). Teachers entered the data into the CWPT—LMS and transmitted them each Friday via the Internet for storage in the central database.
RESULTS
What Was the Level of Implementation?
Results indicated that 57% of participating teachers implemented all five key tasks (see Table 2). In Schools 1 and 2, 100% of teachers implemented all tasks. In Schools 3 and 4, 45%and 42%of teachers implemented all five tasks, respectively. No implementation occurred in School 5. Across the five individual implementation, implemen-tation ranged from 80%to 40%. Eighty percent of the teachers regis-tered (Task 1), 70% completed CBM evaluations (Task 2), and 60%
conducted CWPT (Task 3). Tasks 4 and 5 were completed least often, 42%for Task 4 (implemented the CWPT—LMS) and 40% for Task 5 (completed teacher fidelity checks).
Successful implementation hinged on three key factors: support of administration, fall training, and the ability to overcome technology challenges. Schools 1 and 2 (full implementation) both had adminis-trators who strongly supported the program, provided at least adequate fall teacher training, and worked closely with project staff to overcome any technology challenges. School 3 (low implementation) had good administrative support but poor training and technology problem-solving abilities. In School 4 (low implementation), the training was sufficient, but administrative and technology support was weak. Low teacher implementation occurred. School 5 (no implementation) had low administrative support, no training, and no technology problem-solving abilities.
Students’ Spelling Outcomes for Those Teachers Who Implemented CWPT-LMS
This next analysis was based on seventeen teachers reporting spelling data for a total of 337 students who collectively contributed 4,369 weekly CWPT data records. Weekly mean trends showed variation in pretest spelling accuracy ranging from 1 to 32 weeks (M¼10.3 weeks) of CWPT.
Table 2. Schools/teachers meeting on implementation steps, overall and by site coordinator
Site
Implementation criterion (%)
Overall
School coordinator 1 2 3 4 5 mean (%)
1 Principal 100 100 100 100 100 100 2 Doc student 100 100 100 100 100 100 3 Principal 100 100 25 0 0 45 4 Doc student 100 0 100 12 0 42 5 Teacher 0 0 0 0 0 0 Mean Steps 80 60 65 53 40 57 Implementation criterion.
1. Teachers registered at the web site and transmitted a confirmation report via the web. 2. Target student CBM data was administered and sent to JGCP.
3. Teachers implemented the CWPT strategy.
4. Teachers used the CWPT-LMS (setup classroom program, entered data, and sent data to JGCP).
CWPT substantially impacted weekly post-test spelling accuracy. Pretest scores were in the 40–50% range, but all classroom groups increased to within the 80–100% range, with minor exception. Over-all, the pretest mean calculated at Week 7 was 47.2%correct (pretest slope over weeks¼0.16% correct), increasing to a mean of 93.4% correct (post-test slope over weeks¼ ".02%) following a week of CWPT. The overall mean gain was 47.2% correct (slope over weeks
¼ "0.21%). The increasing pretest linear slope (0.16% per week) indicated that the pretest words became easier; also evident was a slightly decreasing trend in weekly mean gain ("0.21%).
DISCUSSION
Results indicated that when the use of CWPT—LMS was ‘on proto-col’ in local schools—with local SC training and monitoring and tea-cher implementation—students’ weekly spelling accuracy gains were consistently large, replicating prior research (e.g., Greenwood et al., 2001a). Results also confirmed that scalability success requires tools and expertise in areas beyond just a proven, research-validated strat-egy. Specifically, we found that the factors described by Elmore (1996), including a strong PD structure (Summer Institute, fall tea-cher training, ongoing PD), an organization small enough to focus on systematic change (trained SCs training local teachers), an inten-tional process for reproducing the change (CWPT—LMS blueprint), and an incentive system that supports change (improved student achievement), were all essential in successful implementation.
How is an effective practice best taught to teachers on a large-scale basis? First year results clearly illustrated the critical role that local school leadership played in training teachers to implement the CWPT—LMS program. Reaching scale requires local school leaders to set the expectation of full implementation, ensure sufficient training, and monitor the daily details of implementation. Findings favored having a school administrator take responsibility to ensure that both training and implementation occurred ‘‘on protocol.’’ In the most successful schools, PD continued throughout the year in the form of individual consultation and mentoring as needed.
How are research-validated strategies and programs more easily replicated at reasonable cost using technology? In a PD model of trai-ners (see Fig. 2), training local school administration staff to train teachers, initial training that emphasized computer and information technology (CIT), strongly enhanced successful implementation. CIT required relatively little local staff to conduct in-services,
consult, observe, and problem solve with teachers. Findings also sup-ported the detailed involvement of the local individual responsible for and capable of serving and maintaining the local school’s technology systems and negotiating the local networks infrastructure.
Resolving CIT challenges with ongoing PD were also important. During training and throughout the school year, the quick resolution of technology issues was crucial to establishing and sustaining success-ful implementation. For example, for School 1’s fall training, the administrator planned to demonstrate the CWPT—LMS software on an older, school-centralized computer that could not correctly load the software. Fortunately, the administrator communicated with the CWPT—LMS project programmer several days before the actual training. The programmer provided the needed programming patch to fix the problem before the fall training, and CWPT—LMS was quickly implemented in early fall. Even though the multimedia resources were valuable training tools for individualized PD, some tea-chers reported occasional hardware problems on individual compu-ters. To resolve this and other local computer specific challenges, we compiled a troubleshooting list of potential challenges and fixes and incorporated this information into the web site and future training.
The technology expertise of teachers at the local level cannot be underestimated. Pope and Golub (2000) reported that technology-experienced teachers who also knew how to use software applications in their teaching managed to create beneficial learning environments for their students. Similarly, in the CWPT—LMS program, teachers experienced in using spreadsheets, word processing software, and other communication software easily generalized to using the CWPT—LMS. Teachers without CIT background knowledge required additional training and support, and in some cases, even this was not enough for them to succeed. The time for this additional training was often difficult to work into school schedules, and the lack of teacher and administrator expertise affected implementation (Buzhardt et al., in press).
How can PD be made to support and motivate classroom- and school-wide implementation? Initial results confirm prior research, that widescale implementation within a school district begins at the highest district levels. In school districts where strong administrative support began at the district level and continued down to the local school administrative level, strong implementation occurred. In areas with weak district support, implementation was non-existent or less successful. For example, at the beginning of the project, School 3 administrator strongly supported program implementation and facili-tated good training efforts. It was first thought that implementation
was proceeding slowly because of technical difficulties. However, once those issues were resolved, implementation continued to stag-nate. We later learned that, although the CWPT—LMS program fit within the philosophy of the school district, strict implementation of the new mandated reading program took first priority. With ongoing scheduling and training conflicts, the administrator was unable to provide the needed support for complete CWPT—LMS program implementation. Thus, only 25% of the teachers included CWPT instruction, and none of the teachers incorporated the LMS software portion of the program.
Improving the scalability of CWPT at a low cost using CIT tools has not been easy, but first year results show its feasibility. Classroom implementation occurred in four of five implementation schools, and student achievement in these schools was comparable to past CWPT studies. Our experiences with two decades of CWPT implementation suggest that this degree of scaling up success at these costs could not be possible without use of CIT. Our blueprint for scaling up the CWPT—LMS program fits well within Elmore’s vision for large-scale implementation of strong professional development structures that are locally controlled to create sufficient intensity to ensure sys-tematic change. Our belief is that technology enhances that blueprint because it facilitates the process for reproducing systematic change and creates the incentive of academic student improvement to main-tain implementation.
REFERENCES
Abbott, M., Buzhardt, J., & Greenwood, C. R. (2003). Scaling up ClassWide Peer Tutoring: Effects of a blueprint on use, implementation, and student learning. Paper presented at the Pacific Coast Research Conference, LaJolla, CA. Abbott, M., Walton, C., Tapia, Y., & Greenwood, C. R. (1999). Research to
practice: A blueprint for closing the gap in local schools.Exceptional Children, 65(3), 339–352.
Arreaga-Mayer, C. (1998). Increasing active student responding and improving aca-demic performance through Class-Wide Peer Tutoring.Intervention in School and Clinic,34(2), 89–94.
Boudah, D. J., Logan, K. R., & Greenwood, C. R. (2001). The research to practice projects: Lessons learned about changing teacher practice. Teacher Education and Special Education,24(4), 290–303.
Bullock, C. & Ory, J. (2000). Evaluating instructional technology implementation in a higher education environment.American Journal of Evaluation,21(3), 315–328. Buzhardt, J., Greenwood, C. R., Abbott, M., & Tapia, Y. (in press). Research on scaling up effective instructional intervention practice: The need for a formative
measure of the rate of implementation.Journal Educational Technology Research and Development.
Carnine, D. (1997). Bridging the research-to-practice gap. Exceptional Children, 63(4), 513–521.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: Erlbaum.
Delquadri, J., Greenwood, C. R., Whorton, D., Carta, J., & Hall, R. V. (1986). ClassWide Peer Tutoring.Exceptional Children,52, 535–542.
Delquadri, J. C., Elliott, M., & Hughes, V. (1990).The effects of an administrative model for school principals on the effectiveness and maintenance of the reading ClassWide Peer Tutoring program. Kansas City, KS: Juniper Gardens Children’s Project, University of Kansas.
Elmore, R. F. (1996). Getting to scale with good educational practice.Harvard Edu-cational Review,66(1), 1–27.
Fuchs, D., Fuchs, L. S., Mathes, P. G., & Simmons, D. C. (1997). Peer-assisted learning strategies: Making classrooms more responsive to diversity.American Educational Research Journal,34, 174–206.
Fuchs, D., Fuchs, L. S., Thompson, A., Svenson, E., Yen, L., Al Otaiba, S., Yang, N., & Nyman, K. (2001). Peer-assisted learning strategies: Extensions downward into kindergarten=first grade and upward into high school.Remedial and Special Education,22(1), 15–21.
Fuchs, L. S., Fuchs, D., Phillips, N. B., Hamlett, C. L., & Karns, K. (1995). Acquis-tion and transfer effects in class-wide peer-assisted learning strategies in math-ematics for students with varying learning histories. School Psychology Review, 24, 620–640.
Gladwell, M. (2000).The tipping point: How little things can make a big difference. Boston: Little, Brown and Company.
Greenwood, C. R. (1991). Longitudinal analysis of time engagement and academic achievement in at-risk and non-risk students.Exceptional Children,57, 521–535. Greenwood, C. R. (1997). ClassWide Peer Tutoring.Behavior and Social Issues,7,
11–18.
Greenwood, C. R. & Abbott, M. (2001). The research to practice gap in special edu-cation.Teacher Education and Special Education,24(4), 276–289.
Greenwood, C. R., Arreaga-Mayer, C., Utley, C. A., Gavin, K., & Terry, B. J. (2001a). ClassWide Peer Tutoring learning management system: Applications with elementary-level English language learners.Remedial and Special Education, 22(1), 34–47.
Greenwood, C. R. & Delquadri, J. (1995). ClassWide Peer Tutoring and the preven-tion of school failure.Preventing School Failure,39(4), 21–25.
Greenwood, C. R., Delquadri, J., & Bulgren, J. (1993a). Current challenges to beha-vioral technology in the reform of schooling: Large-scale high-quality implemen-tation and sustained use of effective educational practices. Education and Treatment of Children,16, 401–440.
Greenwood, C. R., Delquadri, J., & Hall, R. V. (1989). Longitudinal effects of class-wide peer tutoring.Journal of Educational Psychology,81, 371–383.
Greenwood, C. R., Finney, R., Terry, B., Arreaga-Mayer, C., Carta, J. J., Delquadri, J., et al. (1993b). Monitoring improving and maintaining quality implementation of the ClassWide Peer Tutoring program using behavioral and computer tech-nology.Education and Treatment of Children,16, 19–47.
Greenwood, C. R., Hou, S., Delquadri, J., Terry, B. J., & Arreaga-Mayer, C. (2001b). ClassWide Peer Tutoring program: A learning management system. In J. Woodward & L. Cuban (Eds.), Technology, curriculum, and professional development: Adapting schools to meet the needs of students with disabilities (pp. 61–86). Thousand Oaks, CA: Corwin.
Greenwood, C. R. & Hou, S. (2001).The ClassWide Peer Tutoring learning manage-ment system (CWPT-LMS): Manual for teachers. Kansas City, KS: Juniper Gardens Children’s Project, University of Kansas.
Greenwood, C. R., Maheady, L., & Delquadri, J. (2002). ClassWide Peer Tutoring. In M. R. Shinn, H. M. Walker, & G. Stoner (Eds.),Interventions for achievement and behavior problems(pp. 611–649). 2nd ed. Washington, DC: National Associ-ation for School Psychologists (NASP).
Greenwood, C. R., Terry, B. J., Arreaga-Mayer, C., & Finney, R. (1992). The Class-Wide Peer Tutoring program: Implementation factors moderating students’ achievement. Special issue: The education crisis: Issues, perspectives, solutions. Journal for Applied Behavior Anaylsis,25(1), 101–116.
Greenwood, C. R., Terry, B. J., Utley, C. A., Montagna, D., & Walker, D. (1993c). Acheivement, placement, and services: Middle school benefits of ClassWide Peer Tutoring used at the elementary school. School Psychology Review, 22(3), 497–516.
Mathes, P. G., Howard, J. K., Allen, S. H., & Fuchs, D. (1998). Peer-assisted learn-ing strategies for first-grade readers: Respondlearn-ing to the needs of diverse learners. Reading Research Quartely,33, 62–94.
Mortweet, S. L., Utley, C. A., Walker, D., Dawson, H. L., Delquadri, J. C., Reddy, S. S., et al. (1999). ClassWide Peer Tutoring: Teaching students with mild mental retardation in inclusive classrooms.Exceptional Children,65(4), 425–436. Pope, C. A. & Golub, J. N. (2000). Preparing tomorrow’s English language arts
tea-chers today: Principles and practices for infusing technology.Contemporary Issues in Technology and Teacher Education,1(1), 89–97.
Sideridis, G. D., Utley, C. A., Greenwood, C. R., Delquadri, J. C., Dawson, H., Palmer, P., et al. (1997). Classwide peer tutoring: Effects on the spelling perform-ance of students with mild disabilties and their typical peers in an integrated instructional setting.Journal of Behavioral Education,7(4), 435–462.
U.S. Department of Education. (2002).The No Child Left Behind Act of 2001. P. L. 107–110.
Utley, C. A., Mortweet, S. L., & Greenwood, C. R. (1997). Peer-mediated instruc-tion and interveninstruc-tions.Focus on Exceptional Children,29(5), 1–23.
