Bette Chambers
Centre for the Study of Learning & Performance
Concordia University
Abstract
Researchers increasingly recognize that literacy is the most important academic skill taught in elementary school. If children do not have this skill by grade three, their chances of dropping out of school before finishing secondary school are very high. By literacy we not only mean the ability to read and write but computer literacy as well. Individuals joining the workforce need to be both proficient readers and computer literate as well.
This research builds on the study that we have been conducting on Success for All (SFA) a school restructuring program designed at Johns Hopkins University to improve the achievement of at-risk children. This research will focus on the design and evaluation of different methods of tutoring for children who experience problems learning to read, including computer-assisted tutoring (Reading CAT).
One-to-one tutoring has been identified as an extremely effective form of instruction, because tutors are able to respond to the individual child's needs, adjust their instruction so that it is in the tutee’s zone of proximal development, and provide instant feedback on progress. However, the costs of providing certified teachers as trained tutors are large and appear beyond the fiscal means of most Canadian schools. Furthermore, even trained tutors appear challenged to deliver remediation as prescribed by the model.
We propose to develop a computer-assisted tutoring programme, (the Reading CAT or RC) that will track and monitor the tutee using cumulative summary statistics of the session, which are useful for the teacher, for assisting a cross-age tutor, for feedback to parents, for further diagnostic purposes, and for further individual program adjustments. RC will be developed on-site to ensure a dovetail between the pedagogical strategies in place and the computer programme being developed. To meet the need for software that can be used flexibly within the reading curriculum, RC will be developed as a computer-assisted tutorial template for reading, which can be loaded with stories, vocabulary, and activities from diverse cultures, for different reading-level abilities and grades. This template will be used to support the tutoring component of the SFA programme to improve the reading achievement of disadvantaged children. Most importantly, RC should provide strategic assistance to students who require remediation in reading instruction.
The principal method of inquiry will be quasi-experimentation and descriptive/ correlational studies augmented by case studies. The classroom research sites will be available because the members of the team are actively engaged in training educators to implement SFA. We will employ standardized measures of reading achievement and aptitude, observation checklists, and other process measures.
The cuurent proposal represents the evolution and elaboration of research proposed in the original, successful EVNET application for funding as part of Theme 3D: Evaluating Computer-assisted literacy learning. Due to budgetary and other constraints, no funds have yet been made available within this theme area for the project. Consequently, I request that this proposal be considered for funding as a priority in recognition of my ongoing membership in the network. Since the project is already part of EVNET, this proposal needs to be understood in light of the pertinent section of the original application which is italicized below.
We will also run a comparative evaluation with the Success for All (SFA) programme in Montreal, a prevention and early intervention programme in literacy development. The SFA programme focuses on the promotion of school achievement, beginning at the earliest grades. It is designed to meet the needs of children at-risk for academic difficulties. The programme includes the following elements: a) an intensive, innovative reading programme b) one-on-one tutoring, c) cooperative learning, d) eight week assessments, e) preschool and kindergarten programmes, f) family support teams, g) a school facilitator, and h) staff support teams.
One component that has, thus far, been excluded from the SFA programme is the use of computer technology. Due to diminishing resources allocated for education, the tutoring component is often not implemented as it was originally designed. The high cost of providing one-on-one tutoring by certified teachers to all students who experience difficulty learning how to read is simply prohibitive for many school boards. Computer-based tutoring can make tutoring affordable, thereby exposing disadvantaged children to computers so that their futures will not be further hindered by limited access to this technology. But is the more cost effective computer-based tutoring as effective for disadvantaged children as the traditional one-on-one approach? We plan to examine this question by comparing the effectiveness of computer-based tutoring sessions with the one-on-one tutoring provided by certified teachers that the programme presently prescribes. In a prototypical study of children at-risk, students will be exposed to several forms of tutoring including cross-age peer tutoring, adult tutoring, individual on line tutoring, and CSCL tutoring,. Software programmes such as Wiggle Works will be adapted to the various tutoring conditions used in the SFA programme.
To conform to the application announcement, the body of the proposal contains the following components: objective, literature, ongoing, innovation, theory, methodology, dissemination, training, and schedule. In addition and following the body, I have included the following components in order to speak briefly and directly to the criteria used for adjudication, namely: integration, network value-added, significance, training, partnerships, and practical applications. Although not specifically requested, I have also included a brief budgetary request and my CV.
This proposal directly confronts issues of serious concern in Canadian society, namely school failure and the related problems of poor school attendance, low self-esteem, weak motivation, behaviour problems, and parents with low expectations for their children. Together these factors are significant influences on the school dropout rate which remains shockingly high throughout Canada, exceeding 50% in certain urban areas (Statistics Canada, 1996). Researchers increasingly recognize that literacy is the most important academic skill taught in elementary school; if children do not have this skill by grade three, their chances of dropping out of school before finishing secondary school are very high (Slavin et al., 1992). When they first encounter difficulties learning to read, children benefit more from immediate one-to-one tutoring rather than waiting for their learning problems to be diagnosed by a specialist, and only then receiving remedial instruction, all the while falling further behind (Wasik, & Slavin, 1994).
Individuals joining the workforce need to be both proficient readers and computer literate as well. Overall, the literature suggests that: a) there is an important place for computer technology in the modern curriculum; b) children enjoy and can effectively use computers as learning tools; c) computer technology can increase at-risk readers' appreciation of literature; d) the computer should be systematically included and evaluated for its strengths in the areas of reading exposure, reading practice, reading diagnosis; and e) the locus of software development must be shifted from commercial publishers to a coalition of teachers, developers and researchers in order to produce, integrate, and evaluate effective educational software (Williams & Ogletree, 1992).
Consequently, we propose to conduct a needs assessment of both tutors and tutees to examine the tutoring experience, to critically evaluate existing software, to design Reading CAT software for use in the SFA programme, to alpha and beta test it, and to evaluate the effectiveness of the programme against other models of tutoring: a) certified teacher tutoring alone, b) cross-age tutoring alone (e.g., grade five or six students serving as tutors for grade one and two students), c) paraprofessional tutoring alone. The core design will be a use of software (Reading CAT, no software) by type of tutor (certified teacher, cross-age tutor, paraprofessional tutor) factorial. We predict that students who use Reading CAT tutoring will achieve more than students who are not tutored with the aid of the software. Reading Cat students will be more motivated to learn, have higher reading achievement, and be more computer literate than students tutored without benefit of the software. Furthermore, we hypothesize that the effectiveness of Reading Cat will be modified by tutor type: students who use Reading Cat with certified teachers will learn substantially more than students who are tutored by cross-age tutors with the Reading CAT or paraprofessionals with the Reading CAT.
One-to-one tutoring has been identified as an extremely effective form of instruction, because tutors are able to respond to the individual child's needs, adjust their instruction so that it is in the tutee’s zone of proximal development, and provide instant feedback on progress (Torgesen, 1997; Vygotsky, 1978; Wasik, & Slavin, 1994). In the SFA programme, children who experience problems keeping up with their reading group are immediately assigned a certified teacher who tutors them individually for twenty minutes a day for eight weeks in order to help them catch up to their reading group. The tutoring begins before the child’s motivation deteriorates due to long-term repeated failure (Michael, 1990).
The twenty minute tutoring time is divided into several components consisting of: a one minute drill activity that incorporates positive personal feedback; five minutes of reading a familiar story to build on success; seven minutes working on the current story with modeling, coaching, subtle hints, questions, challenges, and instruction tailored to each child; and a short time for writing to build on the child’s comprehension and writing skills.
In their review of five different tutoring programmes, Wasik and Slavin (1994) found that the programmes that employed certified teachers as tutors led to superior gains than programmes that used paraprofessionals. However, the costs of providing certified teachers as trained tutors are large and appear beyond the fiscal means of most Canadian schools. Furthermore, even trained tutors appear challenged to deliver remediation as prescribed by the SFA model. In our pilot observations of tutoring sessions we found that tutors often spend most of the time drilling students and little or no time having the students write. Consequently, both tutor availability and skill at implementation are concerns in SFA.
Cross-age tutoring may be a viable alternative. Research on cross-age tutoring (Fogarty & Wang, 1982; Cohen, Kulik, & Kulik; 1982) suggests that academic gains can be made by both tutor and tutee, especially when: a) lessons are properly structured; b) tutoring is sufficiently long for the benefits to take hold; and c) tutors are trained in appropriate interpersonal skills and supervised by teachers.
The potential benefits of the computer in the reading curriculum are: a) individualized learning (particularly in reading practice for at-risk readers), b) the allure of the computer as a motivating learning medium, c) cost-effective and quality-controlled experiences, d) instant feedback for the young reader, e) cumulative summary statistics at the end of program sessions that can track children's progress, and, f) its ability to support the growth of the learner while accommodating diversity `patiently' (Schulz, 1993; Soloway et al., 1997). In addition, students who drop out of school often lack the employability skills required for a successful work career, including facility with computers (Conference Board of Canada, 1992). Computer-assisted tutoring would expose disadvantaged children to computers so that their futures will not be hindered further by limited access to this technology.
Unfortunately, the actual initiatives to promote the use of new technologies in the classroom over the past two decades have not been as successful as anticipated (Stoks, 1993). The reasons range from a lack of computer literacy on the part of educators, hindering them from integrating technology adeptly into the curriculum (Haugland & Wright, 1997), through commercially-developed courseware that does not dovetail with educational theory or classroom practice (Reinking, 1989), to a lack of hardware and software resources sufficient to meet the needs of students and teachers (Schulz, 1993).
The limited research on computer-assisted tutoring in reading (e.g., Baron & Abrami, 1992a, 1992b) shows some positive effects with instructional designs based on behavioral applications of programmed instruction (e.g., individualized, self-paced, immediate, corrective feedback) but which can also lead to surface processing and learner isolation. More recent designs have been urged which emphasize constructivist principles including the contextualization of learning and increased interactivity among learners. For example, Abrami and his colleagues (Abrami & Bures, 1996; Abrami & Chambers, 1996; Bures, Abrami & Amundsen; under review; Savard et al., 1995) have suggested ways of incorporating small group instructional techniques of positive interdependence and individual accountability into computer-assisted instruction.
Chambers and her colleagues describe the challenges of implementing Success for All in Canadian schools (Chambers, Abrami, Massue, under review). The major components of SFA are that the students are regrouped for reading instruction, cooperative learning, eight-week assessments, and a school facilitator. Because Canada does not have the same level of support for high poverty schools that they have in the U.S., the SFA schools do not have certified teacher tutors, full day kindergarten, or the level of community involvement in the family support teams.
To validate the programme, we compared the reading achievement of 543 at-risk elementary students from 1 experimental and 3 control schools over 2 years, as measured by standardized reading measures. After one year, students participating in the Success for All Program performed significantly better than control students on a variety of standardized reading measures (Chambers, Abrami, Massue, & Morrison, in press). After the second year in the programme, reading scores were significantly higher for these students (Chambers, Abrami, & Morrison, accepted for publication).
One of the major obstacles that are encountered by schools in their implementation of SFA is that when children experience difficulty learning to read, tutoring is limited in availability and variable in quality (Morrison, Aslin, Chambers, & Abrami, 1998). Consequently, we propose to improve upon the SFA programme in by creating and evaluating different methods of tutoring that will increase opportunities for access to remediation and enhance the effectiveness of tutoring.
We propose to develop a computer-assisted tutoring programme, (the Reading CAT) that will track and monitor the tutee using cumulative summary statistics of the session, which are useful for the teacher, for assisting a cross-age tutor, for feedback to parents, for further diagnostic purposes, and for further individual program adjustments. The Reading CAT will be developed on-site to ensure a dovetail between the pedagogical strategies in place and the computer programme being developed. To meet the need for software that can be used flexibly within the reading curriculum, the Reading CAT will be developed as a computer-assisted tutorial template for reading, which can be loaded with stories, vocabulary, and activities from diverse cultures, for different reading-level abilities and grades (Morrison, et al., 1998). This template will be used to support the tutoring component of the SFA programme to improve the reading achievement of at-risk children.
One factor that stands out clearly in the literature is that computers have been shown to increase student motivation in most software applications regardless of content matter or grade level. (Wright & Shade, 1995). A Reading CAT should help to remove some of the negative connotations attached to the need for tutoring by providing a child identified as requiring tutoring with an interactive, interesting, and supportive learning environment. Tutorial software can present an infinitely patient, quality controlled, non-judgmental expert technological face to the child, who controls its functions and directs his or her learning with the click of a mouse.
Most importantly, the Reading CAT should provide strategic assistance to students who require remediation in reading instruction. The Reading CAT should also provide savings of time for teachers and in training tutors, quality control of sessions, self-esteem benefits for children, as well as the potential benefits of using the Reading CAT where trained reading diagnosticians or trained tutors are not available and individual cumulative summary session statistics are saved and sent to locations where such experts can assess sessions and make program design recommendations.
Undergirding reading instruction today are evolving theoretical and pedagogical principles that combine constructivist, especially Vygotskian, concepts about how children learn through social interaction and behavioral principles of systematic instruction with immediate feedback. Clearly, there is a need for innovative technological assistance in education, for software that meets classroom needs, that provides new approaches to improve teaching, and that supports theoretical developments in reading instruction. However, most software packages are commercially produced outside of the school setting as "turnkey" [used-as-delivered] programs directed not by research but by marketplace sales (Ritter, 1993; Stoks, 1993). They often surmise what the educational setting might be, include no teacher or student input, do not include cumulative session record-keeping options, and are not developed with an eye to developmentally-appropriate practice (Bredekamp, 1992). Without these ingredients, it is difficult for teachers to integrate software effectively into the teaching strategies currently in place.
The principal method of inquiry will be quasi-experimentation (Cook & Campbell, 1976) and descriptive/correlational studies (Pedhazur, 1982) augmented by case studies (Yin, 1994). The classroom research sites will be available because the members of the team are actively engaged in training educators to implement SFA.
We will conduct a needs assessment of both tutors and tutees to examine the tutoring experience, to critically evaluate existing software, to design Reading CAT software for use in the SFA programme, to alpha and beta test it. We will evaluate the effectiveness of the programme against other models of tutoring: a) certified teacher tutoring alone, b) cross-age tutoring alone (e.g., grade five or six students serving as tutors for grade one and two students), c) paraprofessional tutoring alone. The core design will be a use of software (Reading CAT, no software) by type of tutor (certified teacher, cross-age tutor, paraprofessional tutor) factorial.
The research team will augment the quasi-experimental control of variables with statistical control. Statistical analyses will rely heavily on multiple linear regression with dummy coding of categorical predictors. Case study methodology will be used for in-depth analyses of learning and motivational processes.
Measures: We will employ standardized measures of language development (e.g. Test of Language Development), reading achievement (e.g., Durrell Analysis of Reading Difficulty), aptitude (e.g., Peabody Picture Vocabulary Test), motivation (e.g., SAGE, Abrami & Kouros, 1998) and computer literacy. We will also develop and refine specific measures to evaluate the processes involved in the tutoring programmes. We will conduct observations of the tutoring sessions to obtain information about the processes involved.
We will communicate our research findings to EvNet researchers through publishing them in the EvNet Electronic Working Papers Series and presenting the findings at EvNet annual meetings.
We will communicate our research findings to other researchers through journal publications (e.g., American Educational Research Journal, Early Childhood Research Quarterly, Canadian Journal of Education) and conference presentations (e.g., American Educational Research Association, National Association for the Education of Young Children, Canadian Society for the Study of Education).
Additionally, we believe that in an applied field such as education, it is important to disseminate our findings to practitioners in the field. We accomplish this via: a) our workshops and inservice training programmes, b) resource centre and clearinghouse for information on instructional practices, c) our multi-media and print-based instructional materials, d) presentations to educators at professional conferences (e.g., Provincial Association for Protestant Teachers) and e) through training provided for SFA tutors.
This project will involve the collaboration of: a) faculty, graduate students, and staff members of the Centre for the Study of Learning and Performance of Concordia University; (who will serve as the primary researchers and conduct some of the training), b) administrators, teachers, tutors and students of the English Montreal School Board (who will implement the programme, participate in the planning and conducting of the research); and c) faculty and staff from the Center for Research on the Education of Students Placed at Risk (CRESPAR) of Johns Hopkins University (who will serve as consultants to the team).
A team approach also enhances the training and involvement of graduate students. For example, in this research, both faculty and students will have opportunities to engage in the active exchange of ideas, improving the research through both elaborated explanations (Webb, 1989) and through collaborative responsibilities such as positive interdependence and individual accountability (Abrami et al., 1995). The faculty members of this research team have a sound, lengthy, record of working collaboratively both with each other and with others in the field
Summer 1999 We will identify potential schools to participate in the pilot testing of the Reading CAT (RC) software and the cross-age tutoring programme. We will continue developing the programme.
Fall1999 We will train potential schools and tutors in the programmes.
Winter 2000. We will complete, revise, and refine the programme based on pilots and feedback.
Spring 2000 We will implement the different types of tutoring programmes in SFA elementary schools, designate control groups for evaluation and pretest children. We will observe and complete rating scales of the implementations, conduct implementation checks, and code and record observations of experimental and control student progress. We will administer outcome assessments and collect permanent record data on programme and control children.
Summer 2000 We will analyze data and write reports, train any new teaching and research team members, beta-test reading programme stories in the RC, make alterations as necessary according to feedback, and finish loading the stories into the trial set for field testing.
2000 - 2001 We will revise all programme elements in light of experiences and findings and continue field testing, conduct quasi-experiments on programmes, prepare reports on findings, and disseminate findings to academic and educational communities.
Integration: I will continue to collaborate actively with my colleagues and fellow EvNet members affiliated with the Centre for the Study of Learning and Performance, Concordia University. In addition, once the project is funded within EvNet, I am fully prepared to share any research developments with my EvNet colleagues in Theme 3D: Evaluating computer-assisted literacy and others who share my interests within the network. I would be pleased if I was able to conduct the investigation at a variety of sites across the country and with input from others. Otherwise, I have the support of the faculty and staff of CRESPAR at Johns Hopkins University. Finally, I hope to share what is learned among EvNeters through the toolkit, electronic publication, annual meetings, and otherwise.
Network Value-added: In the original EvNet application, the project(s) listed under Theme 3D were described as …"an integral part of EvNet's focus on the effectiveness of various means of delivering education and training. It focuses directly on literacy as the key determinant of academic and workplace success." I share the belief that the tools and techniques my colleagues and I develop to use and evaluate technology as a means to encourage literacy skills is of fundamental importance to everyone and will provide real value-added to the network.
Significance: The importance of research on an essential skill like literacy cannot be overestimated. This research promises to illuminate the means by which I can promote the ability to read among youngsters using technology as well as provide methods for evaluating the mechanisms of effective computer-assisted tutoring.
Training: See section entitled Training in the body of the proposal. Note that training will also affect the expertise of teachers and other school partners with whom we will work.
Partnership: I continue to enjoy an active partnership with the English Montreal School Board (formerly the Protestant School Board of Greater Montreal). This partnership has allowed the research programme on Success for All to expand within the Board beyond what was described in the original EvNet application. Furthermore, the programme enjoys a collaborative link with schools in Northern Vermont and I have been approached by school boards in Ontario and British Columbia about getting involved.
Practical applications: Developing and testing the RC software prototype should be of direct use and benefit to young children struggling to read and the teachers who facilitate their learning.
Abrami, P.C., & Bures, W. M. (1996). Computer-supported collaborative learning and distance education. American Journal of Distance Education, 10, 37-42.
Abrami, P.C., & Chambers, B. (1996). Research on cooperative learning and achievement: Comments on Slavin. Contemporary Educational Psychology. 21, 70-79.
Abrami, P.C., Chambers, B., Poulsen, C., De Simone, C., d'Apollonia, S., Howden, J. (1995). Classroom connections: Understanding and using cooperative learning. Toronto, Ontario: Harcourt Brace.
Baron, L.J. & Abrami, P.C. (1992a). Microcomputer learning with a tutorial program--Manipulating group size and exposure time. Journal of Computing in Childhood Education. 3, 231-245.
Baron, L.J. & Abrami, P.C. (1992b). The effects of group size and exposure time on microcomputer learning. Computers in Human Behaviour, 8, 353-365.
Bures, E.M., Amundsen, C. & Abrami, P.C. (under review). Student motivation to learn via computer conferencing. Submitted to Research in Higher Education.
Chambers, B., Abrami, P.C., & Massue, F. (under review). Success for Canadians too: Implementing an American school restructuring programme in a Canadian context. Education Today.
Chambers, B., Abrami, P.C., Massue, F., & Morrison, S. (in press). An ounce of prevention in schools: A prevention and early intervention programme for children at risk. Canadian Journal of Research in Education.
Chambers, B., Abrami, P.C., & Morrison, S. (accepted for publication). Can Success for All succeed in Canada? Educational Research and Evaluation.
Morrison, S., Aslin, L., Chambers, B., & Abrami, P. (1998). Reading CAT please: A case study in progress. Paper presented at the annual conference of the EvNet, Montreal, Quebec.
Savard, M., Mitchell, S., Abrami, P.C. & Corso, M. (1995). Learning together at a distance. Canadian Journal of Educational Communication, 24, 115-128..
Slavin, R. E., Madden, N. A., Karweit, N. L., Dolan, L. J., & Wasik, B. A. (1992). Success For All: A relentless approach to prevention and early intervention in elementary schools. Arlington, VA: Educational Research Service.
Statistics Canada. (1996). Reading the future. Ottawa, ON: Canadian Government.
Vygotsky, L. S., (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press.
Wasik, B. A., & Slavin, R. E. (1994). Preventing early reading failure with one-to-one tutoring: A review of five programs. In R. E. Slavin, N. L. Karweit, & B. A. Wasik, (Eds). Preventing early school failure: Research, policy, and practice. Needham Heights, MA: Allyn & Bacon.