"Pull Out" in Compensatory Education

"PULL OUT" IN COMPENSATORY EDUCATION
Gene V Glass
Mary Lee Smith
Laboratory of Educational Research
University of Colorado

Paper Prepared for the Office of the Commissioner
U.S. Office of Education, November 1977

ACKNOWLEDGMENTS
We were asked to address the problem that forms the subject of this paper by Marshall Smith and Fritz Edelstein of the Office of the Commissioner, U.S. Office of Education. We have no idea whether or not they had opinions about the "pull out" issue at the outset. If they did, they hid them from us. They encouraged us in every way to examine the question freely and render an independent opinion. That requires commendable nerve, even when the stakes are only $3,000.

We gathered a large number of documents and interviewed many people in just a few weeks. It would not have bean possible to produce this report without the voluntary cooperation of dozens of persons who consented on a moment's notice to be interviewed, or dig up old data, or mail us their reports and papers. The persons who gave us interviews are acknowledged in the Appendix. The staff of System Development Corporation, particularly Ralph Hoepfner and Clarence Bradford, were helpful in performing new data analyses and explaining old ones. Len Cahen of the Far West Laboratory for Educational Research and Development shared several unpublished papers with us. Bob Stonehill of the U.S. Office of Education found reports for us that we wouldn't have obtained otherwise.

SUMMARY
This report examines the research on "pull out," a method or type of school organization for remedial teaching of Title I eligible pupils. Four major issues addressed are: (1) the educational benefits of pulling students out of the daily routine to provide them with compensatory education services; (2) the impact of such action on students; (3) whether a child is better served if he remains in the classroom all day; and (4) alternatives to "pull out" available for providing compensatory assistance to educationally disadvantaged children. Other related issues examined are: the prevalence of "pull out" programs, the benefits or losses resulting from "pull out" programs, teacher contact with and attitudes toward pulled out pupils, financial costs of "pull out" programs, and the potential contribution of "pull out" programs to cultural separatism, racial segregation, or even racism. It is concluded that despite the near universality of pulling Title I eligible pupils out of regular classrooms for compensatory instruction, the procedure has neither academic nor social benefits, may be detrimental, and is used mostly to satisfy Title I regulations. Alternatives to "pull out" are recommended.

"Pull Out" is a method or type of school organization for remedial teaching of Title I eligible pupils. With this plan, Title I eligible pupils are pulled out of regular classes containing both eligible and non-eligible pupils and sent to a different room to receive instruction from a remedial specialist teacher. "Pull Out" has emerged as a prominent feature of compensatory education in the past few years, and it now concerns policy-makers, researchers, and educators alike. This report was written in response to a request from the Office of the Commissioner of Education to examine the research on "pull out." In the course of preparing this opinion, we interviewed about thirty persons in schools, state education agencies, the federal government, universities, and teacher organizations; in addition, we read and, in some instances, reanalyzed data from approximately 150 documents.

The Incidence and Context of "Pull Out"
Roughly 75% of compensatory education pupils receive remedial reading instruction in the "pull out" setting; the comparable figures for mathematics and language are 45% and 41%, respectively. When these figures are corrected to eliminate pupils in 100% Title I eligible classrooms who do not need to be "pulled out," the "pull out" rates in all other classrooms rise to 84% for reading, 54% for mathematics, and 50% for language arts. When one considers further that pupils might be "pulled out" for one of these subjects and not the other, it is plausible to say that in classes not 100% "Title I eligible" the practice of "pull out" for compensatory teaching is nearly universal.

"Pull out" is probably more prevalent in small and medium-sized districts than in large, urban schools. "Pulled out" and "mainstreamed" pupils compensatory education pupils taught in regular classes do not differ in their academic performance before remedial teaching; thus, "pull out" seems not to be prescribed differentially for pupils with varying remedial needs.

The amount of the entire instructional day spent in the "pull out" setting rose from around 5% in 1973-1974 to around 9% in 1974-1975. The percentage of time does not vary by subject taught (reading vs. mathematics) or by grade level (elementary vs. secondary). Although time in the "pull out" setting is a small part of the total instructional day, at Grade 1 it constitutes almost half of the instructional time funded by TitleI. At the elementary school level, one-fifth of the "pulled out" pupils miss regular classroom instruction in the subject for which they are removed from the regular class (i.e., they are "pulled out" of regular reading to receive remedial reading). One-fourth miss social studies; one-seventh miss science. One-third miss no academic subject at all since they are pulled out during study periods in the regular class. (By some chop logic we do not understand, supplanting is not supplanting at all if one supplants science and social studies.)

The "pulled out" pupil has three chances in four of receiving remedial instruction from a remedial subject matter specialist. (The comparable chances for a mainstreamed compensatory education pupil are only one in three.) However, the "specialist" teachers receive very little training for their job (less than ten hours in any one year on the average) and they receive virtually no extra pay (less than 5% more than regular teachers). These data seem to indicate that remedial specialists are distinguished from regular teachers neither by more intensive training nor by the pay they receive. The most cynical assessment of their role and contribution would be that remedial, specialist teachers are merely rechristened regular classroom teachers -- the motive for so designating them being, perhaps, the need to comply with certain Title I regulations.

Finally,"pull out" programs appear to be roughly twice as expensive per pupil as mainstream compensatory programs, probably because of much smaller class size in the former than the latter.

The Effects of "Pull Out"

Experimental evidence is skimpy on the effects of the "pull out" technique per se on pupils' academic progress. A study recently published by the National Institute of Education (September 1977) alleges to show beneficial effects of "pull out" at certain grade levels and in certain subjects and detrimental effects elsewhere. We have examined the data and find little support for these conclusions. The academic gains made by "pulled out and "mainstreamed" compensatory education pupils in the NIE data are virtually identical, differing overall grades and subjects by less than one-quarter month in grade equivalent units. Perhaps a better database for assessing the effects of "pull out" exists in the data files of the evaluation of the Emergency School Aid Act (ESAA) conducted by Systems Development Corporation. There one finds a consistent negative relationship between the percentage of time pupils spend in the "pull out" setting and their math and reading achievement. This relationship was consistent across all grades and subjects; it held true for samples numbering about 10,000 pupils in total. Moreover, the relationship persisted even after more than a dozen background variables were controlled statistically.

A vast body of empirical research on instructional methods and organization is pertinent to the "pull out" problem because the phenomena investigated share various features with the "pull out" technique. Such related topics include the following: (a) ability grouping, (b) mainstreaming the handicapped, (c) racial desegregation, (d) labeling pupils with consequent changes in teachers' expectations of them, and (e) peer tutoring.

Our synopses of the research evidence on these topics are as follows: (a) The research on ability grouping is inconsistent, uninformative, and a battleground for various social ideologies; it was not helpful to us in forming opinions about "pull out." (b) The research on mainstreaming the handicapped was exceedingly skimpy, but the findings of three studies point toward the benefits of integrating EMH, EMR, or emotionally disturbed pupils into the regular classroom. (c) Nearly all research on racial desegregation fails to trace racial mixing at levels lower than the school building. Using the Coleman data, McPartland (1969) assessed the effects on verbal achievement of black pupils having predominantly black classmates instead of white classmates. Although the effect diminished as more background variables were partialed out, the effect of racial segregation at the classroom level was negative in first analyses and never appeared beneficial regardless of how many variables were statistically controlled. (d) Research into the effects of labeling pupils on teachers' behavior toward them and expectations of them proved to be most pertinent and startling. Labeling a pupil "mentally retarded," "intellectually slow," or "academically weak," reduces his academic achievement by one-quarter standard deviation below that of comparable pupils not so labeled. Furthermore, teachers' attention and support for pupils invidiously labeled are reduced by one-third standard deviation below those for comparable unlabeled pupils; and teachers' judgment of labeled pupils' success, motivation, social competence, etc. is reduced by nearly one-half standard deviation. These findings from more than forty experiments indicate that the effects of labeling pupils are large and worrisome. (e) Finally, research on peer tutoring, which presumably could occur less often in the "pull out" programs. Pupils pulled out of regular classrooms would have to receive remarkably effective compensatory programs to offset the potential risks incurred. In our opinion, the "pulled out" pupil is placed in moderate jeopardy of being dysfunctionally labeled, of missing opportunities for peer tutoring and role modeling, and of being segregated from pupils of different ethnic groups.

Historical and Political Context of "Pull Out"

We believe that the "pull out" problem was created by the ESEA Title I regulations and the manner in which they have been interpreted and enforced. To quote one state education department official, "'Pull out' exists for one reason only; because the 'locals' are afraid Big Brother will catch them in a 'supplanting' violation." The practice of pulling Title I eligible pupils out of regular classrooms so that a "specialist" teacher could give them instruction in a separate classroom did not grow out of professional judgment about curriculum or instruction. The history is complex, but nearly any disinterested reading of it leads to the same conclusion: "pull out" is an artifice created by schools at the urging of USOE's minions in state education departments to satisfy regulations concerning "supplementing, not supplanting" and "excess costs." The regulations themselves reflect a philosophy that seems seldom to have been seriously challenged. They are enforced with a nearly obsessive concern that "noneligible" pupils might receive Title I services. Yet the argument can be made that even pupils performing at grade level and above are educationally deprived by merit of attending a school with large concentrations of poor (since such schools attract less qualified teachers, have poorer opportunities for peer tutoring, etc.).

One finds virtually no support for the "pull out" concept among educators or their professional organizations. Teachers worry that pulling pupils out of class creates discontinuities in their schooling and makes coordination of teaching difficult. Others worry that the regular classroom teachers will feel less responsible for pupils whose needs are presumably being met somewhere else by a specialist teacher. The National Education Association regards "pull out" as a minor issue and will merely watch its evolution, being concerned only with keeping pupil-to-teacher ratios low. The "pull out" problem seems to be no one's major concern. But it may well be one of those quiet, inconspicuous matters that count heavily in ways seldom clearly seen.

Conclusions, Observations, and Recommendations

Our work has led us to the following conclusions and observations about the "pull out" technique and several recommendations for dealing with the problems it raises.

1. Pulling Title I eligible pupils out of regular classrooms for compensatory instruction is virtually universal.
2. The "pull out" procedure per se has no clear academic or social benefits and may, in fact, be detrimental to pupils' progress and adjustment to school.
3. The "pull out" procedure is used by schools more to satisfy Title I regulations than because it is judged by teachers to be a sensible and beneficial plan.

We wish to bring the following recommendations to the attention of those persons at all levels who administer Title I programs and who will influence the evolution of compensatory education:

1. The Title I regulations, which now reflect an overweening concern with targeting funds on "eligible" pupils, should be examined.New considerations should be given to the needs of all pupils in poor schools and the integrity of total school programs.
2. Instructional strategies should be designed that would eliminate the invidious labeling of compensatory education pupils and their segregation from classes of "regular" pupils.
3. Teachers, administrators and other persons connected with Title I programs should be informed of the findings of research on the "pull out" method and associated phenomena. 4.Methods should be devised of counteracting the possibly detrimental effects of "pull out" where educators choose to use it or have no reasonable alternatives. Such methods could include means for coordinating instruction across two sites and techniques of teacher observation that lessen the possibility that "pulled out" pupils will be unconsciously neglected in regular classes.

BIBLIOGRAPHY

Bailey, S.K. and Mosher, E.K. (1968). ESEA: The Office of Education Administers a Law . Syracuse, N.Y.: Syracuse University Press.

Billett, R.O. (1932). The administration and supervision of homogeneous grouping. Columbus:The Ohio State University Press.

Borg, W.R. (1966). Ability Grouping in the Public Schools, (2nd ed.). Madison, Wis.: Dembar Educational Research Services, Inc.

Campbell, E.Q. (1965). Structural effects and interpersonal relationships. American Journal of Sociology, 71, 284-289.

Cornell, E.L. (1936). Effects of ability grouping determinable from published studies. In G.M. Whipple (Ed.),The Ability Grouping of Pupils. Yearbook of Nattional Social Studies Education, Part I. Bloomington, Ill.: Public School Publishing Co., pp. 289-304.

Coulson, J.E., et al. (1977). The Third Year of Emergency School Aid Act (ESAA) Implementation. Santa Monica, Calif.: Systems Development Corporation.

Dienemann, Paul F., Donald L. Flynn, and Nabeel Al-Salam. (1974). An Evaluation of the Cost Effectiveness of Alternative Compensatory Reading Programs. Bethesda, MD: RMC Research Corporation.

Eash, M. (1961). Grouping: What have we learned?. Educational leadership, 18 429.

Ekstrom, R. (1961). Experimental studies of homogeneous grouping: A criticalreview. School Review, 69, 216-226.

Esposito, D. (1973). Homogeneous and heterogeneous ability grouping: Principal findings and implications for evaluating and designing more effective educational environments. Review of Educational Research, 43 (2), 163-179.

Findley, W.G. and M.M12\Byran.Ability Grouping: 1970,Status Impact and Alter- natives.Athens, Georgia:University.of Georgia, Centerfor.EducationImprovement, 1971. \

Flynn, D.L, Hass, A.E., & Al-Salam, N.A. (1976). An Evaluation of the Cost Effectiveness of Alternative Compensatory Reading Programs. Bethesda, Maryland: RMC Research Corporation.

Gampel, D.H., Gottlieb, J., and Harrison, R.H. (1974). Comparison of classroom behavior of special-class EMR, integrated EMR, low IQ, and nonretarded children. American Journal of Mental Deficiency, 79, 16-21.

Glass, Gene V, et al. (1970). Data Analysis of the 1968-1969 Survey of Compensatory Education. University of Colorado.

Glass, G.V (1977). Integrating findings: The meta-analysis of research. Review of Research in Education, (in press).

Goldberg, M.L., Passow, A.H., and Justman, J. (1966). The Effects of Ability Grouping. New York: Teachers College Press, Columbia University, 1966.

Goodman, H., Gottlieb, J., and Harrison, R.H.Social acceptance of EMRs integrated into a nongraded elementary school.American Journal of MentalDeficiency, 1972, 76, 412-413.

Kiesling, H.J. Input and Output in California Compensatory Education Projects. Santa Monica, Calif.:Rand Corporation, 1971.

McDill, E.L., Meyers, E.D., and Rigsby, L.C. Sources of Educational Climates in High Schools. Washington, D.C.: U.S. Office of Education Cooperative,Research Report No. 1999, 1966.

McLaughlin, M. W. Evaluation and Reform: The Elementary and Secondary EducationAct of 1965, Title I. Cambridge, Mass.: Ballinger Publishing Co., 1975.

McPartland, J.M. (1969). The relative influence of school desegregation and of classroom desegregation on the academic achievement of ninth-grade Negro students. Journal of Social Issues, 25, 93-102.

Miller, W.S. and Otto, H.J. (1930). Analysis of experimental studies in homogeneous grouping. Journal of Educational Research, 21, 95-102.

National Institute of Education. The Effects of Services on Student Development. Washington, D.C.: The National Institute of Education, U.S. Department of Health, Education and Welfare, September 1977.

Rock, R.T., Jr.A critical study of current practices in ability grouping.Education Research Bulletin.Catholic University of America, Nos. 5 and 6, 1929.

Smith, M.L. Meta-analysis of teacher expectation research. Unpublished paper. Boulder, Colorado: Laboratory of Educational Research, University ofColorado, 1977.

Smith, M.S. "Equality of educational opportunity: The basic findings reconsidered." In Equality of Educational Opportunity, edited by F. Mosteller and D.P. Moynihan. NY: Random House, Inc., 1972.

Tuckman, B.W. and Bierman, M. Beyond Pygmalion: Ability group reassignment and its effects. Paper presented at the Annual Meeting of the American Educational Research Association, New York, 1971.

Vacc, N.A. A study of emotionally disturbed children in regular and special.classes. Exceptional Children, 1968, 197-204.

Appendix

PERSONS INTERVIEWED ON THE "PULL OUT" ISSUE

Dr. Richard Cortright National Education Association
Dr. George Cronk New York Department of Education
Dr. Joy Frechtling National Institute of Education
Dr. Gerald. Freeborn New York Department of Education
Dr. John Garrett Denver Public Schools
Dr. David Gordon California Department of Education
Dr. Susan S. Hartley Northwest Missouri State College
Office of Senator Floyd Haskell Denver, Colorado
Dr. Ralph Hoepfner Systems Development CorpOration
Ms. Linda Jones Colorado Department of Education
Dr. Martin kaufman BEH, U.S. Office of Education
Dr. Michael Kean Philadelphia Public Schools
Dr. Bernard McKenna National Education Association
Dr. Richard Mallory National Education Association
Dr. Robert Mendro Dallas Public Schools
Dr. Lynn Morris Center for the Study of ':valuation, UCLA
Dr. Iris Rothberg National Institute of Education
Ms. Ann Rutherford Denver Public Schools
Dr. Robert Stonehill OPBE, Office of Education
Dr. Gary Toothaker Superintendent of Rifle Public Schools
Dr. Bruce W. Tuckman Rutgers University
Dr. Jean Wellch Systems DevelopMent Corporation
Dr. David E. Wiley CEMREL