Altering Teacher and Pupil Behaviorwith Mastery Teaching1

James R. OkeyScience EducationIndiana University

Bloomington, Indiana 47401

Bloom (1968) has hypothesized that most pupils can master thecontent of the courses they study and that it is the task of teachersto make this occur. Mastery, according to Bloom, means achievementin a course at a level that is regarded as "A" work. Bloom is suggestingthat most pupils can do top quality work, not just a few; it is thesuccessful teacher that can make increased mastery happen. Toincrease mastery of subject matter, teachers need to raise their

expectations for pupils and to change some of their teaching practices.There is some evidence in support of Bloom’s hypothesis that pupil

achievement can be dramatically altered. A number of these studiesare described by Block (1971). The classroom strategy that seemsto increase pupil mastery is relatively straightforward: objectives arespecified, tests for the objectives are prepared, pupils are instructed,diagnostic tests are given, and pupils restudy those objectives thatthey fail. This study-test-restudy cycle is repeated as needed in aneffort to help all pupils achieve the objectives. When possible,alternative learning materials are provided so that each pupil mayfind suitable instruction.A critical component of mastery teaching is frequent diagnostic

testing. Instead of testing at the end of a unit, say every two orthree weeks, pupils are tested as soon as they finish a short bit ofinstruction. This means that pupils take short tests at frequent intervals.The purpose of tests in a mastery strategy, however, is to locatelearning deficiencies, not to give grades. Thus, in mastery teachingtests serve a diagnostic function.The major purpose of this study was to teach teachers to use mastery

teaching and to determine what effect this would have on pupilachievement. Teachers were instructed in the strategy and providedwith materials to implement it. The criterion or dependent variablewas pupil achievement when teachers put the mastery strategy intopractice. A second purpose of the study was to find out if the attitudesof teachers toward testing and diagnostic teaching could be altered

1. This article is based on a paper presented at the National Association for Research in Science Teaching

(NARST) Annual Meeting, Detroit, March, 1973. The study reported in this paper was supported by the National

Center for the Development of Training Materials in Teacher Education at Indiana University under a grant

from the U.S. Office of Education.

530

Altering Teacher and Pupil Behavior 531

by an instructional program that teaches the frequent use of diagnostictests.

PROCEDURE

Teachers enrolled in a graduate science methods class (n = 21)were the subjects for the study. All but three were employed asteachers in pre-school through eighth grade classrooms. The four menand 17 women in the group had teaching experience ranging fromno teaching to about 20 years in the classroom.As a requirement of the course, each student studied a four hour,

multimedia, self-instructional module called Teaching for Mastery(Okey and Ciesia, 1972) that is designed to help teachers implementBloom’s mastery teaching strategy. Objectives, practice exercises,feedback on exercises, self-tests, and answers are given for eachof the six sections in the Teaching for Mastery (TFM) materials.A total of 22 objectives are stated for the materials that cover suchtopics as sequencing objectives, constructing and administering diag-nostic tests, and selecting alternative instruction for unsuccessfulpupils. About two hours of class time were devoted to independentstudy of the TFM program with the remainder done outside of class.The TFM module is designed to help each teacher implement the

five step mastery teaching plan shown in Figure 1. To use the masteryplan in teaching, specific performance objectives are developed andtests for each objective prepared. Pupils are then instructed usingany methods or materials the teacher chooses. Diagnostic tests aregiven following instruction; this is done frequently, perhaps each dayor two. Pupils are then given additional instruction if their diagnostictests show that they have not achieved the objective.

Specify DevelopTeach usingTest achievementReteach andperformance diagnosticany preferredof objectivesretest untilobjectives tests forproceduresusing diagnosticpupils achieve

objectivestestsobjectives

FIG. 1. A five step mastery teaching strategy.

Each of the 21 students in the class took a 24 item attitude measurebefore and after studying the TFM program. Responses were madeon a five point Likert scale to statements like these: Tests aid studentsin learning, Students could administer and score their own tests, orTesting takes more time than its worth. The measure was designedto assess the attitude of a teacher toward tests, grades, and diagnosticteaching.

532 School Science and Mathematics

After instruction in the TFM program, each employed class member(n = 18) implemented mastery teaching in either a science or mathe-matics class. Three different experimental designs were used formeasuring the effects of mastery teaching and are described elsewhere(Okey and Ciesia, 1973) 2 The results from the subgroup of thirdand fourth grade teachers (n = 5) using a Posttest-Only Control GroupDesign (Campbell and Stanley, 1963) will be reported here.Each of the teachers split their class by assigning alternate names

from an alphabetized class list to two groups. Then, for a two weekunit, one group was taught using the mastery teaching strategy. Theother group was taught the same two week unit with the same objectiveswithout using a mastery strategy. This meant that the teacher gavefrequent diagnostic tests to one group and attempted to remedy anypupil errors and did not do this with the other group. While theteacher taught one group the other half of the class was out of theroom.The 20 objectives provided the teachers for the two week unit

were all related to the topic of fractions. The objectives coveredsuch skills as shading a fractional portion of a figure, writing fractions,selecting the larger of two fractions, writing equivalent fractions, andindicating the fractional portion of a set of figures.

In addition to receiving the objectives, each teacher received copiesof instructional materials to hand out to their pupils to help learnthe objectives, 3 diagnostic test items for each objective, and additionalinstructional materials related to each objective to use for remedialpurposes if needed. Although each teacher received all materials,only pupils in the experimental group from each class used thediagnostic tests and remedial materials.A posttest on the objectives was also given to each teacher. They

administered it at the end of the two week unit to learners in bothgroups. The tests from all pupils were collected and scored by theinvestigator.

ANALYSIS AND RESULTS3

The results obtained by the five teachers using mastery and non-mastery techniques are shown in Table 1. Achievement of pupilson the 20 objectives favored the mastery group for each of the teachers.Even though only one of the teachers produced significantly different

performance using the mastery procedure (using p < .05 as the criterion)

2. The first and second grade teachers from the class used a Time Series Design and the fifth through eighthgrade teachers used an Equivalent Time Samples Design. The details of both designs are described in Campbelland Stanley (1963).

3. The assistance of Jerome L. Ciesia and Martin L. Goodson in acquiring and processing data is gratefullyacknowledged.

Altering Teacher and Pupil Behavior 533

TABLE 1: ACHIEVEMENT BY MASTERY AND NON-MASTERY PUPILSON 20 OBJECTIVES

ExperimentalControlSignificanceTeacher GroupGroupLevel

n X SDn XSD t

A1312.83.261310.23.302.02 .05B1314.42.981212.62.961.51 .10C138.62.77127.42.871.06 .15D1514.52.811513.52.701.00 .20E1215.72.371215.22.750.48 .35

the direction of the results consistently favored the mastery pupils.Fisher (1938) suggests that for cases in which repeated tests are madeof the same hypothesis that a single test of the significance of theresult be made by calculating the likelihood of obtaining a set ofprobabilities each in the same direction. Following the method ofanalysis suggested by Fisher (1938, p. 105) produced a value of \2= 19.7 (p < .05, df = 10). The interpretation of this result is thatthe probability of obtaining the series of outcomes each in the favoreddirection is l^ss than 5 in 100.The 24 items on the teacher attitude pre- and posttest were scored

by assigning values to the five point Likert scale; responses thatfavored the use of diagnostic teaching were given a value of fiveand unfavorable responses a one. Two items were dropped from theanalysis when agreement could not be reached on a desired response.Scores were then obtained for 20 of the teachers (one teacher misseda test) by summing their responses on the 22 items. In one casepre- and posttest scores were identical; in the remaining 19 casesa more positive attitude toward tests, grades, and diagnostic teachingwas found following study of the TFM program. The difference inattitude scores was highly significant (X^ = 92.1, SD = 8.26; X^== 81.1, SD = 6.83; t == 7.3; p < .001, df = 19).The interpretation of the attitude data is that feelings teachers have

toward testing can be altered by showing them ways to use testsfor the benefit of students. If mastery teaching procedures are togain wide use, this will be necessary. These results, however, shouldbe interpreted with some caution. At this time the attitude test hasbeen found to be only marginally reliable (r = .58) using a test-reteststrategy on another group of 18 students in a similar graduate methodsclass.

DISCUSSION

A critical test of teacher training materials is whether they leadto increased pupil achievement. In this study, the effect of teachers

534 School Science and Mathematics

studying and using skills designed to promote pupil achievement wasfound to do that. With a minimum of instruction for teachers (lessthan five hours), pupil achievement could be altered. Teachers wereable to produce somewhat higher achievement when they used themastery teaching skills than when they did not. Although achievementresults of pupils in the mastery sections were not dramatically differentfrom non-mastery classes, they were consistently higher.

Experimental studies measuring the effect on pupils of teachersengaging in certain behaviors are few in number (cf. Rosenshine andFurst, 1971). More often, studies are reported on the use of a skill(e.g., whether teachers ask high level questions or make supportingstatements to pupils following responses). Of course we should beconcerned about the use of particular teaching skills, but we shouldalso conduct studies that measure the effect on pupil achievementof the use of skills by teachers.Because of a desire to develop effective competency or performance

based teacher education programs, a good deal of attention will haveto be given to validation of teaching skills. In order to have confidencein the skills that teachers are expected to learn, evidence is neededthat use of the skills makes a difference in classrooms. To a greaterextent, training programs for teachers and classroom practice shouldbe based on skills with demonstrated classroom power.The study reported in this paper is not a pure case in which a

set of teaching skills is validated. There already was evidence (e.g.,the studies reported in Block) that use of a mastery strategy couldhave an effect. Instead the study constitutes the partial validationof a set of training materials. It was shown that teachers could learnto use mastery teaching by studying the TFM materials and that useof the skills had a desirable effect.Even though the benefits to pupils of mastery teaching can be

demonstrated, wide use of the strategy may not result. The extraburden on the teacher is considerable: objectives have to be writtenor selected, diagnostic test items must be prepared and administered,records of individual progress by each pupil on each objective mustbe kept, alternative instruction for some pupils must be devised, andfrequent decisions must be made about individual pupils as the teacherworks with them to achieve mastery. The task for the teacher isformidable. A quote from a book by Piaget (1971) is appropriate.

Generally speaking, the more we try to improve our schools, the heavier theteacher’s task becomes; and the better our teaching methods, the more difficultthey are to apply, (p. 123)

Altering Teacher and Pupil Behavior 535

REFERENCES

BLOCK, J. (Ed.) Mastery learning: Theory and practice. New York: Holt, Rinehartand Winston, Inc., 1971.

BLOOM, B. Learning for mastery. Evaluation comment. May 1968, 1, Center for theStudy of Evaluation, University of California, Los Angeles.

CAMPBELL, D. and STANLEY, J. Experimental and quasi-experimental designs for research.Chicago: Rand McNally and Co., 1963.

FISHER, R. A. Statistical methods for research workers. Edinburgh: Oliver and Boyd,1938.

OKEY, J. R. and CIESLA, J. L. Designs for the evaluation of teacher training materials.AV Communication Review, 1973, 21, 299-310.

OKEY, J. R. and CIESLA, J. L. Teaching for mastery. Bloomington: National Centerfor the Development of Training Materials in Teacher Education, Indiana University,1972.

PIAGET, J. Science of education and the psychology of the child. New York? The VikingPress, 1971.

ROSENSHINE, B. and FURST, N. Research in teacher performance criteria. In B. 0.Smith (Ed.) Research in teacher education. Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1971.

PUBLIC SMOKING BAN NEEDED?

Is smoking a hazard to the health of the people around the smoker? Shouldsmoking be allowed only in private among consenting adults? These are morethan rhetorical questions. In the seven years since warning labels first appearedon cigarette packages, a major change has occurred in the biopolitics ofsmoking. As evidence has grown that cigarette smoke is harmful to nonsmokers,first steps have been taken toward treating smoking as a form of air pollution.Airlines divide their cabins into smoking and non-smoking sections and theyhave long since abandoned the practice of giving away cigarettes. The stateof Arizona has prohibited smoking on public conveyances and places ofassembly.The heavy use of cigarettes and the role of tobacco in public revenues

and in employment may suggest that protective action against smoking hasgone about as far as it can go without arousing popular resistance. On theother hand, three factors continue to work toward restricting use of cigarettes.They are:�An increase in the incidence of lung cancer and emphysema.�The concept that each individual has a right to a healthful environment,

or at least a court-enforceable right against hazards to his health inflictedupon him by other people.�The proposition that ill-health unnecessarily incurred because of individual

indulgence or indifference places a burden on all society, and the healthhazarding behavior of the individual becomes a matter of public concern.The indications point toward the treatment of cigarette smoking as a taxable

private vice, tolerated among consenting adults but prohibited from inflictionupon the nonsmoking public. "Under such circumstances the glamor of thecigarette could sink to the level of snuff."