Printer Friendly

Who becomes an "at-risk" student? The predictive value of a kindergarten screening battery.

In recent years, a major thrust in conceptualizing special education services has shifted from a direct-services approach to a prereferral-consultation model (Carter & Sugai, 1989; Graden, Casey, & Christenson, 1985). This model has multiple goals. These include (a) the reduction of inappropriate special education referrals or placements, (b) more efficient use of educational resources and personnel, (c) identification of strategies that are successful in "the mainstream," and (d) the promotion of collaboration between regular and special educators.

Carter and Sugai (1989) found that most state education agencies require or recommend the use of a prereferral model. Although the concept of prereferral may be expressed in varying ways in different school administrative units, two approaches predominate: a consultative model, using the expertise of one or more specialists, or a team model, involving a school-based problem-solving team (Pugach & Johnson, 1989). Implicit in this manner of conceptualizing and implementing a prereferral system is the belief that a problem must be dealt with either through "intervention assistance" (Graden, 1989) or through the more formal special education referral process.

Concurrently, researchers and practitioners have become concerned with the problem of an increasing number of today's children considered to be "at risk" of school failure (Davis & McCaul, 1990). Changing demographics indicate that the numbers of homeless children, children living in poverty, and children whose primary language is other than English are all expected to increase. Estimates currently place 20 million school-aged children at risk (Catterall & Cota-Robles, 1988). Levin (1985) estimated that one third of America's school children are educationally at risk.

There are indications that at-risk students may be identified early in their school careers. For example, Camper (1989) reported that students' grades and absenteeism in the 3rd grade could predict dropout behavior with 87% accuracy. Speece and Cooper (1990) found that students considered at risk by their first-grade teachers may demonstrate distinct profiles on measures of intelligence, achievement, and social behavior. Fedoruk and Norman (1991) emphasized a more ecological approach, arguing that different classroom environments may lead to different types of students becoming at risk. At any rate, several authors (Hahn, Danzberger, & Lefkowiwtz, 1987; Slavin, 1989; Slavin & Madden, 1989) have argued that efforts at early intervention are essential in any effort to seriously attack the problem of at-risk children.

Both the studies concerning at-risk students and those concerning prereferral suggest that the most meaningful model of intervention would be one that would increase successful school experiences and "intervene" (i.e., appropriately program) before serious educational problems develop. As our school district began to implement a consultative model of prereferral prior to evaluation for special education needs, we noted a trend away from referrals to special education. In the 1987-88 academic year, there were 29 new referrals for evaluation. In the 1989-90 school year, there were only 17 such referrals; this represents a 40% reduction in new referrals. The apparent success of the prereferral process led us to consider whether certain approaches to early identification of high-risk students would then allow even earlier intervention, thus promoting a greater degree of school success for a greater number of children.

This study focused, therefore, on two research questions:

1. How do we effectively identify young children who are potentially at risk?

2. How effective is one nationally validated screening battery for such identification purposes?


To determine whether students who exhibited later educational problems could have been identified at the time of the kindergarten screening, we employed screening results and other longitudinal information (e.g., free or reduced-price lunch status, scores of the Iowa Test of Basic Skills, etc.). The kindergarten screening method was a nationally validated project entitled Early Prevention of School Failure (EPSF) (Werner, 1985). Although many such screening instruments are available (see Meisels & Provence, 1989 for a listing of early childhood screening instruments), the school district found this instrument useful in assessing students for possible early intervention. The stated goal of EPSF is to identify the developmental skills and learning styles of 4- to 6-year-old children to implement a developmentally appropriate curriculum for each child.

Several researchers have emphasized the necessity of this linkage between screening batteries and instruction (Barclay, 1983; Judy, 1986; Ysseldyke, Thurlow, & O'Sullivan, 1987). The EPSF also combines both teacher rating scales and formal tests, two essential components of screening batteries (Lindsay & Wedell, 1982). Parker and Ciechalski (1990) argued that the EPSF skill-building activities and the collaborative approach to both assessment and follow-up led to gains in student achievement that were detectable as early as 1st grade. Similarly, Betz (1990) found that groups using EPSF consistently demonstrated greater gains than did control groups on standardized achievement tests.

The study took place in a rural school district that provides educational programs for about 1,300 students in five towns, with kindergartens in four schools. In addition to the EPSF, tests of speech, vision, and heating are administered to every kindergarten student in September and May of each school year. The screening team consists of a school nurse, a psychological examiner, a primary level teacher, a speech-language pathologist, an EPSF paraprofessional, and a kindergarten teacher. Each kindergarten teacher works with her own students. The speech-language pathologist assigned to the particular school serves on its team. All other team members are constant throughout the district.

Results dictate the manner in which each child moves through the associated developmental curriculum. Posttesting in May provides information about a student's growth in each modality. A management system that tracks every student's mastery of the sequential skills within the curriculum currently allows the program to be tailored to the individual needs of the student both in kindergarten and in 1st grade, if necessary.



Since 1985, the district has routinely compiled EPSF pretest and posttest results for each student entering kindergarten. For every student, basic background information was compiled, as well as retention, special education referral, and special education placement status.

We used data from the kindergarten classes of 1985-86 and 1986-87 because these students, at the time of the study, had been in school long enough to allow an examination of retention and special education referral status. Students for whom longitudinal data through 3rd grade were not available were eliminated from most analyses. The final sample consisted of 161 students, 95 boys (59%) and 66 girls (41%).


The EPSF battery of tests administered include: (a) an adaptation of the Goodenough Draw-A-Person (DAP), (b) the Motor Activity Scale (MAS), (c) the EPSF Preschool Language Scale (PLS), (d) the Peabody Picture Vocabulary Test-Revised (PPVT-R), and (e) the Developmental Test of Visual-Motor Integration (VMI). The tests cover a broad range of developmental skills, a common consideration both in selecting screening instruments and in developing more accurate at-risk indexes (Meisels & Provence, 1989). The following are EPSF components:

* Draw-A-Person. This EPSF measure is an adapted version of the Goodenough Draw-A-Man Test. Students' fine motor and visual skills are ranked according to their degree of control over the drawing process and the degree of detail included in their drawing. The ranking may be used to derive a developmental age.

* Revised Motor Activity Scale (MAS). The MAS assesses fine and gross motor skills through a sequence of activities. Areas of assessment include body imagery and spatial orientation. manual dexterity, and body control. Ratings are determined by successful completion of the steps in each activity and by the evaluator's observations of the child's performance. Interrater reliability for the period of this study was .90.

* EPSF Preschool Language Scale (PLS). The PLS is designed to assess integrarive and conceptual areas of language development. The PLS is a criterion-referenced instrument that is closely related to the EPSF skills checklist and the EPSF classroom activities.

* Peabody Picture Vocabulary Test-Revised (PPVT-R). The PPVT-R is a standardized test of receptive language skills, a widely recognized prerequisite to success in schools. The PPVT-R test manual reports split-half reliability coefficients of .70 to .84 for the age groups involved in this study. They also report a median value of .82 for alternative forms reliability and a median value of .78 for test-retest reliability. Concurrent validity with the Stanford-Binet Vocabulary Subtest is reported as .72 (median value) and as .62 (median value) with the Stanford-Binet Intelligence Scale (American Guidance Services, 1981).

* Developmental Test of Visual-Motor Integration (VMI). The VMI assesses a child's visual-motor skills by asking the child to copy various geometric forms. The VMI manual provides scoring criteria and converts scores to age equivalents. The VMI test manual reports an interrater reliability coefficient of .93, a test-retest coefficient of .63 (7-month period), and a split-half reliability coefficient of .78. Concurrent validity as measured by correlations with school readiness tests "have averaged about .50" (Berry, 1982).

Scores on these measures are evaluated by the EPSF team and converted to ratings in each of the six modalities: (a) receptive language, (b) expressive language, (c) auditory skills, (d) visual skills, (e) fine motor skills, and (f) gross motor skills. Each child's skills in each of the modalities are described as being in one of the following ranges as they relate to his or her chronological age:

1. Considerable need(2 years or more below age expectancy).

2. Moderate need (approximately 1 year below age expectancy).

3. Average.

4. Moderate strength (1 year above age expectancy.)

5. Considerable strength (2 years or more above age expectancy).

The ratings are based on specific criteria, and combinations of scores are applied consistently by the screening team across individuals and schools; thus, interrater reliability is maintained. The reading segment of the Iowa Test of Basic Skills (ITBS) was employed to assess second-grade reading achievement.


Descriptive statistics were computed for (a) student age, (b) socioeconomic status (SES) as measured by free or reduced-price lunch status, (c) gender, (d) EPSF pretest modality and total scores, (e) EPSF posttest modality and total scores, (f) retention status, (g) special education referral status, and (h) special education placement status. Cross-tabular analyses were performed to investigate differential retention, referral, and placement patterns. T tests were employed to examine differences on EPSF and reading scores.

EPSF subtest ratings were entered into discriminant analyses equations to determine their power in predicting which students would be retained, referred to special education, or placed in special education programs. Because basic background factors are often identified as influencing retention, as well as referral and placement of students, a second set of analyses was also performed. In these discriminant equations, EPSF subtest ratings, age, gender, and SES were entered.

In attempting to further refine a "profile" of these students at risk, we employed two additional approaches. First, stepwise discriminant analysis isolated the EPSF subtests that were the most powerful predictors of which students were retained, referred, or placed in special education. Second, we compared the EPSF scores of students whose retention, referral, or placement status was correctly predicted by the discriminant model with those whose status was incorrectly predicted (see Walker, Stieber, Ramsey, & O'Neill, 1990, for a similar approach). Because we were interested in specific comparisons between groups of students (e.g., correctly predicted retained vs. incorrectly predicted retained) and not in overall main effects, groups were compared with t tests, not analysis of variance. To guard against errors of inference based on multiple comparisons, we employed the Bonferroni procedure (as outlined by Hafner, Ingels, Schneider, Stevenson, & Owings, 1990), in which the alpha level (e-g., .05) is divided by the number of pairwise comparisons. In this case, six comparisons were possible [k= 4; k(k-1)/2], and results with a probability of less than .008 were considered to be significant.

For the stepwise discriminant analyses, we entered all EPSF subtests into the discriminant equation. Stepwise analysis retained those predictors with F values greater than one, and these were then entered into another stepwise discriminant analysis. Those that emerged for each category of student status are described; these predictors provide a framework for identifying those students at high risk of being retained, referred, or placed.

Finally, we employed two levels of regression analyses to assess the value of EPSF subtests in predicting later reading achievement. The ratings for the pretest EPSF subtests were entered into a regression equation as independent variables with spring and fail reading scores as the dependent measures. Then, the background factors of age, gender, and SES were added to the regression equations. Thus, the predictive effect of the EPSF subtests could be determined when basic background factors were statistically controlled.


Of the 161 subjects from the 1985-1987 kindergarten classes, a majority (59%) were boys. The subjects were also relatively low in SES, with approximately 65 % receiving free or reduced-price lunch (see Table 1).

For the total district, approximately 17% of the students were retained, with retention patterns varying somewhat from school to school. This percentage appears to be relatively high, at least according to a recent national estimate that indicated an 11% retention rate (Marion & Coladarci, 1990). Most students (14) were retained in the first grade. Some subjects (roughly 8%) had been retained in kindergarten previously. Also, a majority (70.4%) were boys. Approximately 14% of students had been referred for special education services by the 3rd grade; however, only 8% of students were actually placed. (In Maine, approximately 13.1% of the total school population receives special services; nationally, this figure is approximately 11.4% [Kierstead & Gray-Hanc, 1990]. Nationally, approximately 72% of students referred for special education are actually placed [Algozzine, Christenson, & Ysseldyke, 1982].) In the district, 17 (73.9%) of those referred were boys, and 6 (26.1%) were girls. Of those actually placed, 8 (61.5%) were boys and 5 (38.5%) were girls.

As is evident from Table 1, teacher ratings on pretest EPSF subtests were relatively constant, having a mean of approximately 3.00 and a standard deviation of approximately 1.00. This indicates that roughly 68% of the subjects were classified as in the average range scoring between moderate strength and moderate need. Values for the posttest were slightly higher, with a total mean of 20.45 as compared to the total pretest mean of 17.69, indicating that teachers did perceive pupil growth in EPSF skill areas during the students' kindergarten year. The mean scores for fail and spring reading achievement were comparable; but, rather surprisingly, fall reading scores were 1.22 points higher. Although not evident from Table 1, a correlation of .72 was found between the fall and spring reading results.

Results indicated that several significant differences existed between those retained and those students not retained. Differences were found in all EPSF modalities (with the exception of the gross motor modality), total EPSF scores and fall reading. No differences were found between groups on spring reading achievement. Students who were referred to special education had significantly lower EPSF (except in the gross motor modality on the pretest) and reading scores than those not referred. Significant differences also existed between students who were placed in special education and those who received no special education services. Differences were found on all pretest and posttest EPSF scores except for the gross motor modality on the pretest. Students who were placed in special education had significantly lower fall and spring reading test scores (see Table 2).

An examination of the data contained in Table 3 shows that approximately 83% of subjects were correctly classified on their retention status on the basis of their EPSF pretest scores. When age, gender, and SES were included as predictors, the predictive value of the model rose to 89%, but the accuracy of predicting those who were retained actually declined. EPSF subtests correctly predicted 73% of the retained group (Table 3) while only 54% were correctly predicted when age, sex, and SES were included in the model. The predictive value was thus enhanced for the students who were not retained. Tests for statistical significance indicated that all EPSF modalities, except gross motor, contributed to the predictive power of the model in the first level of analysis. The power of the model in predicting retention status must be viewed with caution, however. One could obtain an 83% successful prediction rate simply by guessing that the entire group would not be retained (the percentage of the students in the sample who were not retained); in addition, the model correctly predicted the retention status of 19 (73%) of the retained students.

Similarly, EPSF scores on the pretest were 82% accurate in predicting which students were referred to special education (Table 3). When age, gender, and SES were added to the predictive model, approximately 91% of cases were classified correctly; but accuracy in correctly identifying referred students decreased from 70% using just EPSF subtests (Table 3) to 65% in the expanded model. Again, because 85 % of students in the sample were not referred, one could obtain an 85% prediction rate by guessing that all members of the sample were not referred. In the first level of discriminant analysis, the model was only 65% accurate in predicting the status of the referred students. All EPSF modalities except gross motor contributed significantly to group prediction.

In terms of placement in special education, EPSF modality scores were accurate predictors for approximately 80% of cases (Table 3). Predictive value increased to roughly 94% when age, gender, and SES were added to the model; but as with the analyses for retention and referral, the enhanced model was less effective (54% vs. 69%) in predicting the status of the 13 subjects placed in special education. These students placed in special education accounted for only 8% of the sample, so one could obtain a 92% prediction rate by guessing that none of the students in the sample would be placed. In the first level of analysis, tests for statistical significance indicated that all EPSF modalities, except gross motor, contributed to the predictive power of the model.

In the stepwise discriminant analysis on retention status, EPSF subtest scores on expressive language were the most powerful predictors, followed by scores in the fine motor modality. For referral to special education, EPSF scores in the fine motor modality were the most powerful predictors, followed by scores in the auditory and receptive language modalities. Stepwise discriminant analysis results for placement in special education indicated that EPSF subtest scores on the fine motor modality were the most powerful predictors. Auditory and expressive language modalities were also significant predictors.

Results of the analyses on EPSF scores for predicted versus actual groups are shown in Table 4. Scores tended to fall along a continuum; students who were retained, referred, or placed and whose status was correctly predicted, had the lowest scores. (This pattern, however, was not evident with the gross motor scores.) We conducted multiple comparisons between groups, in an attempt to better understand misclassifications, but our analyses focused on students who had been retained, referred, or placed.

For retained students, we found significant differences in the modalities of fine motor and expressive language between those correctly and incorrectly classified. Similarly, referred and placed students who were correctly classified differed significantly from those who were incorrectly classified in the expressive language and fine motor modalities. These results suggest that a more intensive evaluation of these modalities may enhance the predictive power of the district's screening battery.

We used multiple regression to determine which EPSF modalities might be predictive of later reading achievement. The auditory modality was a significant predictor of fall reading even when background factors of age, gender, and SES were statistically controlled. Table 5 shows that the auditory and fine motor modalities were significant predictors of spring reading achievement. When age, gender, and SES were added to the multiple-regression model, gender and ratings on fine motor ability emerged as significant predictors of spring reading achievement scores.

These results indicate that differences in ratings on auditory processing may actually be explained by differences in age, gender, or SES. The fine motor modality still appears to be an important predictor of spring reading ability, however.


Several issues emerge from our attempt to identify students who are at potential risk for educational failure. A review of the literature on retention suggests that retention may have dubious effects in increasing achievement but, on the average, tends to decrease student self-esteem and increase dropout behavior (Marion & Coladarci, 1990). It may be, therefore, that retention failed to address the actual learning problems of the students. Instead, an earlier focus on classroom adaptations and consideration of referral for special education may have been more effective.

Results indicated that students who were retained, referred, or placed differed from their peers along expected dimensions. These students tended to be lower in SES, have lower EPSF scores, and have lower reading achievement scores. Surprisingly, however, retained students did not have significantly lower spring reading scores than those who were not retained.

An examination of the predictors of retention, referral, and placement status indicated that EPSF scores had some predictive value. Nevertheless, the discriminant equation models were not appreciably better than chance in specifically determining student retention, referral, or placement status. Models were far better at predicting which students would not be retained, referred, or placed. These results are consistent with earlier research on the predictive value of screening instruments (Lindsay & Wedell, 1982). Two EPSF variables emerged from stepwise discriminant analyses as the most powerful and consistent predictors of students' status. These variables were the fine motor and auditory modalities.

Kindergarten students who received low ratings in these modalities, therefore, could be considered as potentially at risk. Because significant differences were found between students retained, referred, and placed versus those who were not, these results may have considerable practical value. Though retention and special education placement are, in a way, forms of intervention, it is the school district's goal to intervene by providing more appropriate programming at this formative stage of a student's education. The introduction of the EPSF curriculum, which is then tailored to a child's needs, as a result of the EPSF modality profile, is an important beginning step in this direction.

The results of this study suggest the need to closely examine the areas most significantly associated with later indicators of risk. For example, the fine motor modality was consistently found to be a strong predictor of retention, referral, or placement in special education. Although this result does not indicate that all intervention must be directed at improving fine motor ability, its predictive value may direct us to modifying instruction to a greater extent than is done at present. If we develop a successful prereferral plan for children identified as at risk as they enter school, we should see a resulting decrease in school failure and ineffective interventions, such as inappropriate retention. The EPSF profile must be considered carefully in conjunction with other at-risk indicators, such as poor family support systems; low levels of parent education; and lack of, or inadequate, preschool experiences (Davis & McCaul, 1990).

The predictive value of EPSF subtests, although clearly limited, may be of use by the district. To be practicable, however, they must be used in conjunction with other data before students are automatically considered to be at risk. Clearly, a variety of factors related to the school, family support network, and the child must be considered in identifying children at high risk of educational failure. If the district continues to keep careful records on students, then statistical analyses may become increasingly sophisticated and useful in providing a "red flag"; but schools and teachers must appraise a child's individual circumstances in actually identifying a particular child as "at risk."

Certainly, many questions remain concerning student retention: Does appropriate retention actually enhance student achievement? Are retentions reflective of student social maturity or teacher expectations rather than actual student achievement? Do retention practices reflect self-fulfilling negative expectations on the part of teachers? These questions were beyond the scope of the present study, but they need to be addressed.

The results of this study suggest that some difficulties can be at least partially predicted by preschool screening results. What interventions may then change the student's school experience? What classroom modifications are necessary? How do student-assistance teams help in this process? At what point is a referral for special education appropriate?

An ecological approach is necessary to determine which students are truly at risk. Low scores on certain EPSF modalities (particularly the fine motor modality) are somewhat predictive of later difficulties in school, but must not be used in isolation in determining the status of "at risk." The identification of students at risk during the kindergarten and 1st-grade years entails the danger of creating self-fulfilling prophecies by teachers, parents, and school administrators. Each student's situation must be carefully considered, and the student-assistance-team model may be helpful in this process. EPSF screening results, if interpreted with proper caution, may be useful in targeting students who may be at risk of later school failure.



Algozzine, B,, Christenson, S., & Ysseldyke, J.E. (1982). Probabilities associated with the referral to placement process. Teacher Education and Special Education, 5, 19-23.

American Guidance Services. (1981). Peabody Picture Vocabulary Test-Revised Manual for Form L and M. Circle Pines, MN: Author.

Barclay, L.K. (1983). Developmental screening of early childhood skills. Elementary School Guidance and Counseling, 18, 94-100.

Berry, K.E. (1982). The revised administration, scoring, and teaching manual for the Developmented Test of Visual - Motor Integration. Cleveland, OH: Modem Curriculum Press.

Betz, P. (1990). Evaluation of student impact and program fidelity. Peotone, IL: Early Prevention of School Failure.

Camper, J. (1989, August 14). Early signs can pinpoint dropouts, study finds. Chicago Tribune, p. 7.

Carter, J., & Sugai, G. (1989). Survey on prereferral practices: Responses from state departments of education. Exceptional Children, 55, 298-302.

Carterall, J., & Cota-Robles, E. (1988). The educationally at-risk: What the numbers mean. In Accelerating the education of at-risk students [Summary]. Conference on Accelerating the Education of At-Risk Students (pp. 6-7). Stanford, CA: Center for Educational Research, Stanford University.

Davis, W.E., & McCaul, E. (1990), At-risk children and youth: A crisis in our schools and society. Orono, ME: College of Education, University of Maine.

Fedoruk, G.K., & Norman, C.A. ( 1991 ). Kindergarten screening predictive inaccuracy: First-grade teacher variability. Exceptional Children, 57, 258-263.

Graden, J.L. (1989). Redefining "prereferral" intervention as intervention assistance: Collaboration between general and special education. Exceptional Children, 56, 227-231,

Graden, J.L. Casey, A., & Christenson, S.L. (1985). Implementing a prereferral intervention system: Part 1. The model. Exceptional Children, 51, 377-384.

Hafner, A., Ingels, S., Schneider, B., Stevenson, D., & Owings, J.A. (1990). National educational longitudinal study of 1988: A profile of the American eighth grader. (Report No. NCES 90-458). Washington, DC: National Center for Education Statistics.

Hahn, A., Danzberger, J., & Lefkowitz, B. (1987). Dropouts in America: Enough is known for action. Washington, DC: Institute for Educational Leadership.

Judy, J. (1986). Early screening is essential for educational accountability: Response to Salzer and to Shepard and Smith. Educational Leadership, 44, 8788.

Kierstead, J.T., & Gray-Hanc, D.M. (1990). Maine special education data summary report: 1988-89. Augusta, ME: Maine Department of Education.

Levin, H.M. (1985). The educationally disadvantaged: A national crisis. (The State Youth Initiatives Project, Working Paper #6). Philadelphia: Public/Private Ventures.

Lindsay, G.A., & Wedell, K. (1982), The early identification of educationally "at risk" children revisited. Journal of Learning Disabilities, 15, 212-2 17.

Marion, S.F., & Coladarci, T,(1990). A longitudinal study of grade retention and its effects*. Paper presented at the annual meeting of the American Educational Research Association, Boston.

Meisels, S.J., & Provence, S. (1989). Screening and assessment: Guidelines for identifying young disabled and developmentally vulnerable children and their families. (National Early Childhood Technical Assistance System and National Center for Clinical Infant Programs report, ISBN 0-943657-15-6.)

Washington, DC: National Center for Clinical Infant Programs.

Parker, L.D., & Ciechalski, J.C. (1990). Kindergarten screening and remediation strategies with implications for school counselors. Elementary School Guidance and Counseling, 25, 116-122.

Pugach, M.C., & Johnson, L.J. (1989). Prereferral intervention: Progress, problems, and challenges. Exceptional Children, 56, 217-226.

Slavin, R.E. (1989). Students at risk of school failure: The problem and its dimensions. In R. Slavin, N. Karweit, & N. Madden (Eds.), Effective programs for students at risk (pp. 3-19). Boston: Allyn & Bacon. Slavin, R.E., & Madden, N.A. (1989). What works for students at risk: A research synthesis. Educational Leadership, 46, 14-20.

Speece, D. L. & Cooper, D.H. (1990). Ontogeny of school failure: Classification of first-grade children. American Educational Research Journal, 27, 119-140.

Walker, H.M., Stieber, S., Ramsey, E., & O'Neill. R. E. (1990). School behavioral profiles of arrested versus nonarrested adolescents. Exceptionality: A Research Journal, 1, 249-266.

Werner, L. (1985). Early prevention of school failure training and resource handbook. Peotone, IL: EPSF Nationally Validated Project.

Ysseldyke, J.E., Thurlow, M.C., & O'Sullivan, P.J. (1987). The impact of screening and referral practices in early childhood special education: Policy considerations and research directions. Journal of Special Education, 21, 85-96.


MARYANN ROTH (CEC ME Federation) is a school psychologist at Maine School Administrative District No. 64, East Corinth, Maine; EDWARD MCCAUL (CEC #192) is a Senior Program Associate at the National Association of State Directors of Special Education in Alexandria, Virginia; and KAROLDENE BARNES (CEC ME Federation) is Director of Student and Staff Services at Maine School Administrative District No. 64 in East Corinth, Maine.

We thank Leonard Ney and William E. Davis for their unwavering guidance and support during the course of this study. We also thank Anne Levasseur for her excellent and diligent work in the preparation of this article.

Manuscript received April 1990; revision accepted October 1991.
COPYRIGHT 1993 Council for Exceptional Children
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Roth, Maryann; McCaul, Edward; Barnes, Karoldene
Publication:Exceptional Children
Date:Feb 1, 1993
Previous Article:Transition from school to adulthood: case studies of adults with learning disabilities who dropped out of school.
Next Article:"I've counted Jon": transformational experiences of teachers educating students with disabilities.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters