AN EXPLORATORY INVESTIGATION OF THE EFFECTS OF CLASS SIZE AND SCHEDULING RELATED TO ACHIEVEMENT AND MOTIVATIQNAL OUTCGMES Thesis for the Deane of Ed. D. MICHIGAN STATE UNIVERSITY Clarence Murray W imam: I962 IIII II IIIII IIIIIIIII 0671984 II IIIIIIIZIWIII This is to certify that the thesis entitled AN EXPLORATORY INVESTIGATION OF THE EFFECTS OF CLASS SIZE AND SCHEDULING RELATED TO ACHIEVEMENT AND MOTIVATIONAL gTCOMES present 6 g C larence Murray; Williams has been accepted towards fulfillment of the requirements for Ed.D. degree in Education QWZ I/méW fimflw/ Major profess?I Da! 0-169 AN EXPLORATORY INVESTIGATION OF THE EFFECTS OF CLASS SIZE AND SCHEDULING RELATED TO ACHIEVEMENT AND MOTIVATIONAL OUTCOMES BY CLARENCE MURRAY WILLIAMS A THESIS Submitted to Michigan58tate University in partial fulfillment of the requirements for the degree of DOCTOR OF EDUCATION College of Education 1962 @2rq43 If [’9’ 1'93 TABLE OF CONTENTS Chapter Page Table of Contents ii List of Tables iv Acknowledgments vi Abstract vii I. THE PROBLEM . . . ._. . . . . . . . . . . . . . . . . . . 1 Purpose of the Study . . . . . . . . . . . . . . . . . 1 Need for the Study 1 Statement of the Problem 2 General Hypothesis 3 Overview of the Study 3 II. REVIEW OF THE LITERATURE . . . . . . . . . . . . . . . . 5 III. PROCEDURES . . . .'. . .5. . . . . . . . . . . . . . . . 16 Experimental and Control Schools . . . . . . . l6 Non-Standard and Standard Size and Scheduling . . . . 18 Other Experimental Conditions . . . . . . . . . . . . l8 Rationale of the Observation Schedule . . . . . . . . 22 Instrumentation . . . . . . . . . . . . . . . . . . . 27 Achievement . . . . . . . . . . . . . . . . . . . 28 Intelligence . . . . . . . . . . . . . . . . . . . 29 Motivation . . . . . . . . . . . . . . . . . . . 29 Self-Initiation . . . . . . . . . . . . . . . . . 30 Specific Hypotheses . . . . . . . . . . . . . . . . . 31 Statistical Models . . . . . . . . . . . . . . . . . 32 Data Analysis . . . . . . . . .'. . . . . . . . . . . 33 Summary . . . . . . . . . . . . . . . . . . . . . . . 34 IV. ANALYSIS OF DATA . . . . . . . . . . . . . . . . . . . . 37 Part One-Achievement . . . . . . . . . . . . . . . . . 37 Part TwoéMotivation . . . . . . . . . . . . . . . . . 47 Summary . . . . . . . . . . . . . . . . . . . . . . 53 Part One . . . . . . . . . . . . . . . . . . . . . 53 Part Two . . . . . . . . . . . . . . . . . . . . . 53 ii Page V. SUMMARY, DISCUSSION, AND RECOMMENDATIONS . . . . . . . . 54 sumry O O O O O O O O O O O O O O O O I O I O O O O 54 Discussion . . . . . . . . . . . . . . . . . . . . . 55 Recommendations . . . . . . . . . . . . . . . . . . 58 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . 62 APPENDIX NO. 1, NON-STANDARD AND STANDARD CLASS SIZE AND SCHEDULING I O O O C I O I O O O O O O O O O O O 64 APPENDIX NO. 2, OBSERVER'S CHECK SHEET . . . . . . . . . . . . . 68 APPENDIX NO. 3, SELF-INITIATED PROJECTS QUESTIONNAIRE . . . . . 71 iii TABLE 3. 1 LIST OF TABLES Frequency Comparisons of Observations of Pace of Instruction Related to Subject Materials and Students in Large and Small Groups in the Experimental School . . . . . . . . Frequency Comparisons of Observations of Activities Considered to be Articulatory or Connecting one Instruction Period to Another . . . . . . . . Frequency Comparisons of Observations of Certain Categories of Classroom Activities in Large and Small Classes . . . . . . . . . . . . . . . . . Frequency Comparisons of Observations of Categories of Teaching-Learning Activities and Control of Teaching Means and Analysis of Covariance for English Controlled for I. Q. . . . . . . . . . . . . . Means and Analysis of Covariance for English Controlled for Pre-Achievement Means and Analysis of Covariance for English Controlled for Motivation (WRL) ‘Means and Analysis of Covariance for Physics Controlled for I. Q. . . . . . . . . . . Means and Analysis of Covariance for Physics Controlled for Pre-Achievement Means and Analysis of Covariance for Physics Controlled for Motivation (URL) . . . Means and Analysis of Covariance for Chemistry Controlled for I. Q. . . . . . . . . Means and Analysis of Covariance for Chemistry Controlled for Pre-Achievement . . . . . Means and Analysis of Covariance for Chemistry Controlled for Motivation (WRL) iv Page 24 25 26 26 38 39 40 41 42 43 44 45 46 TABLE Page 4.10 Means and Analysis of Covariance of Pre- and Post-Motivation (WRL) for English Population . . . . . . . 48 4.11 Means and Analysis of Covariance of Pre- and Post-Motivation (URL) for Physics Population . . . . . . . 49 4.12 Means and Analysis of Covariance of Pre- and Post-Motivation (URL) for Chemistry Population . . . . . . 50 4.13 Means and Analysis of Covariance of Pre- and Post-Motivation (WRL) for Combined Population . . . . . . 51 4.14 Experimental and Control SIPQ Means, Variance, and "t" test . . . . . . . . . . . . . . . . . .I. . . . . 52 ACKNOHLEDGMENTS To write a thesis a person needs patience and persistance. He also needs advisers, colleagues, friends, and family with patience and persistence. Without the constant support of the following individuals, some of whom played more than one role at various times, this study could not have been completed. To Dr. William W. Farquhar, the writer's major advisor, colleague, friend, and chairman of his committee. To the other members of the committee, Dr. Bernard Corman, Dr. John D. Krumboltz, and Dr. S. Howard Bartley. To Dr. William Fullagar and Dr. Charles Adair for providing judicious motivation. And to his wife and children whose faith, encouragement, and willingness to spare him the time provided the inspiration necessary to complete the study. vi AN EXPLORATORY INVESTIGATION OF THE EFFECTS OF CLASS SIZE AND SCHEDULING RELATED TO ACHIEVEMENT AND MOTIVATIONAL OUTCOMES BY Clarence Murray Williams AN ABSTRACT OF A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree Of DOCTOR OF EDUCATION College of Education 1962 Approved ABSTRACT AN EXPLORATORY INVESTIGATION OF THE EFFECTS OF CLASS SIZE AND SCHEDULING RELATED TO ACHIEVEMENT AND MOTIVATIONAL OUTCOMES BY Clarence Murray Williams The problem in this exploratory investigation was to examine the effects of different class sizes and scheduling on the educational development of students in high school physics, chemistry, and senior English. Educational development outcomes for the study were measured in, (1) subject matter and skill achievement, and (2) motivation. In the experimental school, the class sizes ranged from approxi- mately 60 to 100 students, meeting twice a week in double periods interspersed with one small class (6-24 students) "seminar" and laboratory periods (for appropriate subjects) for each student. In the control school, class size was standard of approximately 30 students and scheduled five times a week in 50 minute periods with necessary and appropriate laboratory periods for the subjects. The control school was selected for the investigation by members Of the state education department. The main variable on which seleCtion was based was completed years Of high school and college of the parents of the students included in the investigation in both schools. Other conditions such as suburban location and size of the school were approximated as closely as possible. vii viii The general hypothesis was that there would be no differences in either achievement or motivation outcomes as result of instruction under different class sizes and scheduling; Nine specific Statistical null hypotheses were tested on achievement outcomes using analysis of covariance to control, in each Of the three subject matter areas, for intelligence, pre-achievement, and motivation. Furthermore, five specific statistical null hypotheses were tested on motivational outcomes. Analysis of covariance was used to control, in each of the three subjectareas, pre-motivation differences. A fourth motivation outcome analysis was made of the entire experimental sample compared to the control sample using analysis of covariance to adjust for pre-motiva- tion. In a fifth instance, an instrument designed to detect self- initiation of instruction-related projects, provided scores for a "t" test of means between the total experimental sample and total control sample. The achievement instruments used for the study were those provided by the state education department. For chemistry, a state education score was available in biology as a pre-achievement measure. For physics, a math 10 test score was available as a pre-achievement measure.. The junior English examination served as both a pre- and post-instrument because the senior examination was discontinued recently. For pre- and post-motivation scores a new scale, the Word Rating List which measures "Academic Self-Concept" was used. This scale was developed by Farquhar and associates in an on-going investigation sponsored by the U.S. Office ix . l of Education and Michigan State University. Another new instrument, 2 the Self-Initiated Projects Questionnaire is under development by the writer and was used as a measure of self-initiation of instruction- related projects. These two instruments are correlated .15. Significant differences in statistically adjusted achievement outcomes were obtained in all three specific tests in senior English in favor of the experimental conditions. In physics, the adjusted achieve- ment outcomes were not significantly different and in chemistry, the adjusted outcomes in achievement favored the control conditions. No differences were found in motivation within each subject but the analysis of covariance on the entire experimental sample compared to the control sample was significantly different in favor of the experimental condi- tions. No difference was obtained between the entire experimental sample and the control sample on self-initiation. The major conclusion of this exploratory investigation is that class size, as a variable, affects the teaching and learning situation. Due to certain concessions necessary to experiment in a field setting and necesSary assumptions regarding achievement measures, it was impossible to separate sufficiently the full effects of class size on 1Farquhar, William W., "A Comprehensive Study of the Motivational Factors Underlying Achievement of Eleventh Grade High School Students," Research Project N. 846 (8458) in cooperation with the U.S. Office of Education, 1959. Initiation Related to Instruction, in preparation, 1962. instruction and learning and motivation outcomes. However, awareness of the importance to teaching of manipulation of class size (with the attendant schedule shifts) was increased. CHAPTER I THE PROBLEM Purpose of the Study 1, 2 Writers in education advocate that class size and scneduling be changed so that better use of teaching staff, student time, and scnool facilities might be obtained. This study was designed to compare some relevant outcomes of instruction under two different class size and scheduling arrangements. Need for the Study How to gain the effective use of teacher-time has concerned educa- tors for some time. Because the profession of education is still faced with a shOrtage of qualified teachers, and more and more students are finisning high school, the efficient use of teacher-time has become even more critical. Furthermore, the number of students continuing their educational experience increases year by year. The increasing use of technology in classrooms makes it possible for teachers to instruct different sized groups using such learning aids as movies, slides, film strips, over-head and opaque projectors, 1Arthur D. Morse, Schools_gf Tomorrow-Today. (Garden City, New York: Doubleday and Company, 1960). 2Lloyd J. Trump and Dorsey Baynham, Focus on Change - Guide to Better Schools. (Chicago: Rand, McNally and Company, 1961). -1- -2- television, and programmed learning. Small group instruction and well prepared assignments also help in that they place more responsibility on the student and so prepare him for advanced study. In summary, the need for the present study is based on the following factors: 1) a proportionately smaller number of qualified teachers are available each year, 2) an increasing number of students are finishing high school and continuing their education, and 3) a better understanding of the relationships between teacher aid-technology and instruction is emerging. These factors demand controlled empirical investigations of manipula- tions of class sizes (with the attendant snifts in scheduling) and their effects on teaching and learning outcomes. Because all of the variables are not evident at this time, an exploratory study is necessary. Statement of the Problem The problem in this exploratory investigation was to examine the effects of different class size and scheduling on the educational development of students in high school physics, chemistry, and senior English. Educational development outcOmes for the study were measured in, 1) subject matter and skill achievement, and (2) motivation. -3- General Hypothesis The basic assumption of the study is that achievement outcomes, as they are commonly measured, will not be affected significantly by manipulation of class size and scheduling. If is further assumed that motivational outcomes, such as "Academic Self-Concept" as found in the Word Rating List3 and self-initiation of individual projects related to instruction4 would not be affected by varying class size and scheduling. The general hypothesis, therefore, is that there will be no differences in either achievement or motivation outcomes as a result of instruction under different class sizes and scheduling. Overview of the Study The remainder of the study is separated into four chapters. In chapter two is found a review of the pertinent literature related to class size and scheduling. The procedures used to select the experi- mental control schools, how large and small class size and different scheduling affected instructiohal procedures, the instruments used for data collection, and the statistical models are described in chapter 3William W. Farquhar, A Comprehensive Study of the Motivational Factors Underlying Achievement 2f Eleventh Grade High School Students, Research Project No. 846 (8458); Supported by the U.S. Office of Educa- tion, in cooperation with Michigan State University, 1959. 4Clarence M. Williams, The Development 2£.E Measure Of Self-Initia- tion Related £2 Instruction, in preparation, University of—Rochester, 1962. -4- three. The data analysis is found in chapter four. The summary, conclusions, discussion, and recommendations of the study are included in chapter five. CHAPTER 11 REVIEW OF THE LITERATURE Most of the literature on the effects of class size and scheduling on instructional outcomes is spread over a number of years and is con- founded with ancillary variables such as single-subject-matter analysis, visual aids, or teacher-centered versus student-centered methods. Most investigators conclude that class size, as a variable, has negligi- 1. ble effects on outcomes of instruction . Meanwhile, visionaries are suggesting that, "the future secondary school will be organized around large-group instruction, individual study, and small group discussion. The majority of reports on class size effects are not experimental in that a problem is defined, hypotheses developed, a methodology for data collection and analysis detailed, and warranted conclusions based on probabilities drawn. Only a few directly related studies are available which meet the minimum requirements of being scientific inquiry. Other reports using class size as a subsidiary variable slightly extend the body of knowledge in this area. 1C.M. Fleming, "Class Size as a Variable in the Teaching Situation." Educational Research, February, 1958, pp. 35-48. 2Wallace B. Nelson, "An Experiment in Class Size in the Teaching of Elementary Economics." Educational Research, October, 1959, pp. 330-341. 3 J. Lloyd Trump, Images 9f the Future -‘A New Approach £2 the Secondary School. Urbana: University Of Illinois, Commission on the Experimental Study of the Utilization of the Staff in the Secondary School, 1959, 46 p. -6- The major emphasis of this review will be on the directly related research with passing reference given to a few pertinent studies which treat class size as a by-product. An ongoing study is described by Trump as "teachers (in Snyder, Texas) are experimenting with sections consisting of seventy-five students each. Each teacher teaches three such sections, utilizing many audio-visual aids, and then has three periods a day free for " To date, no other follow-up could be found planning and conferencesh. on the experiment in large-group instruction in Snyder, Texas. Another author reported that "students do not suffer by being in large classes...(and)...teachers varied in their success in accordance with their ability to endure the strain of a number of large classes...5" but does not document or present evidence of a systematic study of these factors. In addition, several "experiments" and investigations are alluded to without citation._ A comprehensive study of large-group instruction at the university level where achievement in large groups in several subject areas was 6 controlled by achievement in small classes was conducted in Miami . 4J. Lloyd Trump, New Horizons for Secondary School Teachers. Urbana, Illinois: University of Illinois, Commission on the Experimental Study of the Utilization of the Staff in the Secondary Schools, 1956, 35 p. SDora V. Smith, "Vital Factors in the Present Situation in Class Size." English Journal, Vol. 22, 1933, pp. 366-74. 6Lawrence Siegel, F.G. Macomber, and James F. Adams, "The Effective- ness of Large~Group Instruction at the University Level." Harvard Educational Review, Vol. 29, 1959, pp. 216-226. Achievement scores in general were not adversely affected by large group instruction. A confounding variable for which no control was provided was the use of television as a media in the large group instruc- tion. The authors report for controls that the course content, the instructors, and the final examination were constants in the experiment. Teaching effectiveness is equated with objective test scores and final grades in another study of class size in economics7. The author reports that large and small group students were matched for major subject, student classification, sex, and the remaining differences eliminated by analysis of covariance. Unfortunately, the original source of this information is not available for detailed study. When English literature is the subject material, learning is reported as enhanced when taught in large groups and compared to learning in small groupsa. This finding is said to be true for "general reading also." In addition, "students preferred large class and the teacher ' Again, no mention is made in this article of found them more active.’ any kind of control or attempt to systematize the investigation. This specific article is being used in a number of summaries and reviews on achievement in large classes versus small classes as favorable evidence for large group instruction. 7Wallace B. Nelson, "An Experiment in Class Size in the Teaching of Elementary Economics." Educational Research, Vol. 40, October, 1959, pp. 330-341. 8Dora V. Smith, Class Size 33 High School English: Methods and Results. University of Minnesota, 1931. Another improper use of research information is found in the doctoral dissertation of Bittickg. He concludes, "students from large high schools where classes tend to be large achieve as much in the language arts as do students from small high schools where classes tend to be small." The same logic could be used in favor of small classes. Large lecture sections make for increased motivation of lecturers to engage in more careful preparation and may have provided for greater stimulation for excellence in presentation is the conclusion of another reporterlo. The statistical comparison of the efficiency of the large lecture section and the small recitation section was not analysis of covariance and did not provide control over differences when assign- ment to large or small class was not random. The Critical Ratio between large and small class means obtained by an unidentified weighting of quizzes and examinations was not significant. A project entitled SUPRAD conducted under the auspices of Harvard University is described by Morsell. He reports that "large group, small group, and individual tutoring are all in the context of best 9Edsel F. Bittick, Differentials 1g Collegg Success A; The University ‘9: Texas gf_Students From Large and Small Texas High Schools. Doctor's Thesis, University of Texas, 1956. 10Richard W. Husband, "A Statistical Comparison of the Efficiency of Large Lecture vs Small Recitation Sections upon Achievement in General Psychology." Journal 3; Psychology, Vol. 31, 1951, pp. 297-3OQ. 11 ‘ Arthur D. Morse, Schools 9f Tomorrow-Today. (New York: Doubleday and Company, 1960). -9- use of talents of the group of teachers in SUPRAD project." No research evidence is supplied to support this contention. No further information about what is being done specifically is supplied. Goodlad, in writing on "Classroom Organization," for the 1299 Encyclgpedia 3; Educational Research, concludes, "... Class size, like other problems of classroom organization, cannot be satisfactorily studied apart from the problems of curriculum and instruction tied up with itlz." While this seems a reasonable conclusion and one which says that class size still needs to be investigated along with other problems, another conclusion is reported by Fleming as he completes a review of the literature on class size, "There have been many investigations; but with few exceptions...under typical conditions, class size in itself appears to be an unimportant factor. The benefits of small classes, though commonly taken for granted by theorists, are as yet largely undemonstrated in the pages of accredited research reports. This conclusion has been reached at every level from infant- room to university lecture-theatre. It has been formulated in relation to many subjects; and it is supported both by test results and by assessment of various types ." 12John I. Goodlad, "Classroom Organization," Encyclopedia 2; Educational Research. (Ed. by Chester w. Harris. New York: The Macmillan Company, 1960). 13C.M. Fleming, "Class Size as a Variable in the Teaching Situation." Educational Research, February, 1958, pp. 35-48. -10- These two points of view are too contradictory to leave unconsidered. The latter appears to be built on uncritical acceptance of many reports of studies of class size from pre-school to university applications. The controversy demands empirical testing. An interesting study of the effects of large group instruction is 14 ‘ discussed by Dranes . He first states that there are no critical differences between mean achievement and attitudes of large and small group classes of the same subject. The results, however, favored the small classes. The study was then rerun adjusting the teaching method so that the instructors could know the students better, plan the semes- ter work more carefully, and provide more opportunity for discussion with individuals. In the next semester, the large group did better than the large group of the previous semester. The conclusion drawn was that method and class size are inseparably related. 15 , There is a suggestion in a study by Entwisle that for certain purposes it is better to instruct students in large groups. If the purpose of instruction is to widen the scope of knowledges to which a student is exposed then Entwisle's statement, "merely calling attention to a given body of subject matter (administering attensity) fostered learning irrespective of the means by which attention was directed," has some merit. There are reaSons why instruction should be directed 14David Daniel Dranes, "A Study of Class Size and Instructional Methods," Doctoral Dissertation, The University of Wisconsin, 1957. 15Doris R. Entwisle, "Attensity: Factors of Specific Set in School Learning." Harvard Educational Review, Vol. 31, Winter, 1961, pp. 84-101. ' ’ -11- to large groups which might be different from those developed for small groups. Support for this notion comes from a summary of teaching methods in science education. The reviewer states that if the aim is to produce learning of the informational type, demonstration methods (in lectures) . . . . 16 are as effective as indiVidual laboratory work . In a study in which the election of advanced courses in psychology . . .. 17 . was used as one criterion, McKeacnie reports that the more autocratic recitation method as compared to discussion and tutorial methods proved not only to produce superior performance on the final examination but also to produce greater interest in psychology as measured by the election of advanced courses in psychology. . . . . . 18 In another study of higher education, Giffin and Bowers conclude that the "mass-lecture" method of teaching can be employed without significantly diminishing the amount of learning. They used "several" control groups and experimental groups of fourteen students each in a course, Fundamentals of Speech} in which the aim was to bring about improvement in speaking skill. The total number of students was 297. The experimental condition was one "mass-lecture" per week by the 16Harry A. Cunningham, "Lecture Demonstration Versus Individual Laboratory Method in Science Teaching--A Summary." Science Education, Vol. 30, 1946, pp. 70-82. l7 Wilbert J. McKeachie, "Students, Groups, and Teaching Methods." American Psychologist, Vol. 13, 1958, pp. 580-584. 18Kim Giffin and John Waite Bowers, "An Experimental Study of the Use of Lectures to Large Groups of Students in Teaching the Fundamentals of Speech." The Journal of Educational Research, Vol. 55, No. 8, May, 1962, pp. 383-385. ' -12- department head with all students in experimental classes attending and then separation into small groups of fourteen students each for two discussion periods per week. The control groups met in classes of fourteen each for three periods per week. The presentation of the substance of the fundamentals of speech was during the “mass-lecture" for the experimental group. For the control group, the fundamentals were presented when needed or at opportune times between the short (two to six minute) speeches of students. In order to control experi- mental variables the same graduate assistants were used as instructors, all classes met in the mornings of the same days and met in the same or comparable classrooms. The assignment of students to the different sections at enrollment time was assumed to be random. An oral speaking test and an objective test were administered pre- and post- the course. The oral test was rated (one to ten) by three judges and the mean scores used for analysis. The two form ob- jective test consisted of fiftyJTive items with no estimates of reliabi- lity or validity. The model for data analysis was a two by two cell arrangement with a "t" test between pre- and post-results for both groups and between both groups pre- and post for both the oral and the objective tests . experimental control pre ”t" test post "t” test "t" Ht” -13- The major criticism of this study is the use of different instructors to present the substance of the course to the experimental and control groups. The differential effects of the department head's presentation in the ”mass-lecture" might have been lessened had he guest-lectured in each of the control classes on the same topics. Other criticisms are, (l) the assumption of random assignment to sections by normal enrollment procedures, (2) the lack of estimates of reliability or validity on the tests, and (3) the method of data analysis. For the possible non-random assignment of students to experimental and control sections and the method of data analysis the statistical method, analysis of covariance, would have improved the design. Means are available for obtaining estimates of the reliability of the tests. Other conclusions might have been made. For example, Giffin and Bowers might have concluded that large group instruction does not significantly diminish the amount of learning when the lecturing is done by an exper- ienced lecturer. Or, that graduate assistants can supply the necessary course substance to students when they have small classes to deal with. These kinds of tentative conclusions perhaps should have been made because it would have shown the possibilities for serendipity in research. As an example of how school personnel are beginning to manipulate the schedule in order to improve teaching-learning the following description of a schedule change plan is given. In a junior high school, . . . . l9 . the baSic unit of time has been reduced to twenty-two minutes. This 19M.H. Robb, "Modular Scheduling at Euclid Central." National Association gf_Seconda£y School Principal's Bulletin, Vol. 46, February, 1962, pp. 66-69. -14- unit or "module" is considered to be sufficient for some classes like foreign language instruction in the seventh and eighth grades and in certain remedial sections. Other classes like the "core" classes in social studies might have as many as five modules. The teachers can use the time in the assigned modules any way they wish, according to the subject, assignments, and individual student's need. As of the writing of the report, the author states that now school personnel are considering "sliding classes." If one teacher needs more than the planned time, she takes it and gives it up later in exchange to the teacher from whom it was originally taken. Summary There is little empirical research available on class size as a variable which effects the outcomes of instruction. When reports relating to class size are found, other variables such as a single subject matter, visual aids, or teacher-centric versus student—centric methods are confounded with it. Several reports in which claims about the effects of class size are presented are not well conceived in research design, and/or described in such a way as to mask clear understanding. Some of the reports allude to findings not presented and others refer to "experiments" not described or cited. A controversial issue arises when comparing two different reviews of the literature. One reviewer states that class size is inextricably tied up with problems of curriculum and instruction and needs further study while another reviewer concludes that class size is an unimportant J -15- factor in teaching-learning situations. One recent study was reviewed in detail. The design did not account for teacher differences between the large group and small group instruc- tion, did not include estimates of reliability or validity for the instruments used, and the analysis of data did not provide for non- random assignment of students to the two treatments. A possibility for other tentative conclusions such as, an experienced teacher can lecture to all the students in a "mass-lecture" and then graduate assistants can monitor speeches as a method to expose all students to the experienced teacher for the substance of the course or that graduate assistants can teach the substance of the course in small group situations was suggested. . i No research was located on scheduling as a variable. Some recent articles refer to "experimentation" in scheduling and one was reviewed so that an idea of what is being considered and applied by school per- sonnel might be gained. In this article, the scheduling unit or "module" is described as twenty-two minutes in length. Numbers of modules vary from one to five depending on the kind of class and its purpose. The author also describes the next procedure, that of adopting "sliding classes.“ If a teacher needs more time to finish a lesson or assign- ment, she takes it and releases the students when she is finished. She "repays" the time to the class from which it was taken at a later date. CHAPTER III PROCEDURES In order to understand the setting in which an experimental manipulation of class size and scheduling was possible, it is necessary to describe, in the procedures section, the designation and selection of an experimental and a control school, define the experimental con- ditions, and show preliminary evidence that the experimental conditions affected the teaching-learning situation. Finally, in this section, the data instruments are described and the schedule for the collection and. analysis of data is reported. Experimental and Control Schools The experimental school-was a suburban New York school where the administrators and staff of the secondary unit were interested in manipulating class size in English IV, physics, and chemistry. In co- operation with the State Department of Education in Albany, New York, a control school was selected. Members of the education research division of the State Department made the selection using data already available I I O I o 1 o 0 from their continuing study of education quality. The chief variable on which selection was based was the number of years of high school and 1William D. Firman and associates, Procedures i3 School Quality Evaluation, mimeographed report, publisaedhby the DiviSion of Research, State Education Department, Albany, New York, 1961. -16- -17- college completed by the parents of the students. This variable is one which has been identified as a stable indicator of achievement in a particular school. The data was collected for the parents of students in the three experimental classes and sent to Albany. Selection of a school of similar parent backgrounds and as comparable as possible with respect to size and type of location (both scnools were in suburban areas) was then made. The contgol school staff's cooperation in admin- istering the testing schedule was secured. A coordinator of the project was named in the education department's research division. His function was to serve as an intermediary to facilitate communication between the experimental and control schools. At no time, however, did the personnel in the two schools know the other's identity and location. Because the relinquishing of complete direction over both treatments in the experiment was necessary in order to get complete separation of the experimental and control situations this cumbersome technique for selection and data collection was accepted. It constitutes a limitation on the study in that some data was lost. The inordinate length of time necessary for communication to be effective meant, in most cases, that fill-in and replacement data were not available due to the press of year-end school time committments and the students dispersing. -18- Non-Standard and Standard Class Size and Scheduling Non-standard class size and scheduling2 refers to two large group (with 60 to 90 students) instruction periods of double length per week interspersed with small group for seminars, independent study, and laboratory experiences.‘ Standard class size and scheduling refers to classes of approximately 30 students in regular length periods of from 45 to 55 minutes meeting five times a week with the laboratory periods appropriate to the subject. The total time in minutes per week was the same. Other Experimental Conditions It is necessary at this point to explain more fully three aspects of the study, objectives for the three courses, English IV, physics, and chemistry, their relation to the examinations provided by the state, and the necessary and complete separation of the experimental and control schools with the attendant assumptions. The objectives for Regent's courses in New York State are stated in general terms. For English IV, they are3: Reading and literature 1. To develop skill in reading (with concentration, speed, comprehension, organization, and recall). 2See Appendix No. l for a copy of the class schedule in the experi- mental school compared to traditional scheduling. 3Personal communication, the principal of the experimental school. -19- 2. To develop skill in the use of the tools of reading, including the card catalog, reference books, the dictionary, the book index, and table of contents. 3. To form a permanent reading habit based upon a love of reading. 4. To become acquainted with the reading field including books and current periodicals. 5. To develop skill in making discriminating choices of reading materials. Expression 1. To develop skill in expressing thought in clear, correct, courteous, interesting, and forceful oral English. 2. To develop skill in taking part in public discussion within the rules and courtesies of parliamentary procedures. 3. To develop skill in expressing thought in clear, correct, courteous, interesting, and forceful written English. For physics: "The objectives of the course in physics should extend far beyond a minimal comprehension of the basic facts and prin- ciples outlined in these syllabuses. The appreciation of the scientific method, the ability and willingness to change bClleS and )pinions after careful weigning of new evidence, -20- and the development of the habit of critical thinking are the intangible but most important outcomes of the study of this science 4" The aims of the physics course "... should extend far beyond acquisition of knowledge about utilitarian applications of physics principles ... should be the practice laboratory (to) ... reinforce the so-called 'scientific method' ... afford a glimpse of the orderliness of the patterns ... of the universe part played by math as an expression of this systematic or- 5 ganization should be brought home to the ... student " For chemistry: "The objectives of the course in chemistry should extend far beyond a minimal comprehension of the basic facts and princi- ples outlined in these syllabuses. The appreciation of the scientific method, the ability and willingness to change beliefs and opinions after careful weighing of new evidence, and the development of the habit of critical thinking are the intangible but most important outcomes of the study of this science " 4Chemistry and Physics: Ag Outline gf the Scope 9f the Content and Related Understandings of the Courses 2f Study. Pub. by Secondary Curriculum Development, New York State Education Department, 1957, p. 7. 5Physics Handbook. Publication of Bureau of Secondary Curriculum Development, New York State Education Department, 1956. 6Chemistry and Physics: Ag Outline ... p. 7. -21- The Rengent's examinations in New York State are revised each year by a committee of teachers from various schools throughout the state working with test specialists of the education department. These tests are constructed to fit the broad objectives contained in the state syllabi. In essence, the items in these tests probably could be said to be the operationally defined objectives of the particular Regent's course Insofar as this is true, then, the Regent's courses English IV, physics, and chemistry as taught in the experimental school and the control school are similar. They are similar in content, extent and breadth of coverage, the kinds of experiences needed to establish and reinforce the particular skills, and in the source materials used. This is a broad but necessary assumption for the exploratory study. Because the Regent's syllab} and examinations are accepted by . , scnool personnel desiring to offer Regent's credit to students and because of state-wide teacher participation in examination preparation it is assumed, then, for this exploration, that the only major variable unaccounted for is the teacher. This limitation was accepted in order to achieve separation of the experimental and control conditions and so that a preliminary study of achievement and motivational outcomes could be made. Also, it was thought that the non-standard class size and scheduling might affect instruction and prove a source of contamination to the standard classes. 7This year's examinations (1962) were not available at this writing. -22- An observation schedule was developed to permit study of the effects of non-standard class size and scheduling on instruction. It was thought this procedure would provide information which might prove helpful in explaining results. Pre-experimental discussion and planning with the teachers revealed that a sufficient description of the method and findings of the observations should be reported to increase their understanding of some possible effects of non-standard size and scheduling on teaching. Rationale of the Observation Schedule J Three weeks in late winteriwere selected as being a period of time relatively free of radical distraction in which to conduct a study of the effects of non-standard class size on teacher method in the experi- mental school. The mid-year activities were over and sufficient time remained in the school year before the press of closing activities. Furthermore, the three weeks constituted a period which best approximated a sample of the total forty weeks of instruction. Three observers were chosen who were doctoral candidates in the College of Education, University of Rochester. Each spent several periods of observations in the experimental school as part of his training. All had had three or more years of teaching experience. The observation schedule for each observer was set to be twelve Complete class periods randomly selected out of the possible forty-five to sixty-one periods available in the non-standard size classes. Two COnditions in the selection of observation periods were made. The first was an equal division of observations of large and small group -23- classes; the second was a ten per cent overlap of observer time in the schedule. The latter condition was made to gain some logical estimate of reliability of observations. In addition, on inspection, there appeared to be no logical differences when observations overlapped. The teachers being observed were asked to check the observer's 8... . . .. reports , indicate discrepanCies and make adjustments pertaining to instructional procedures, sequence, or content. In only a few cases did teachers change some aspect of an observation. None of the changes proved to be substantial or logically significant. In the following summary all overlapping observations were deleted: 8See Appendix No. 2 for copy of Observer's Check Sheet. -24- The data included in Tables 3.1, 3.2, 3.3, and 3.4 serve to characterize the kind of instruction being practiced in the experimental large and small classes. Table 3.1 Frequency Comparisons of Observations of Pace of Instruction related to Subject Materials and Students in Large and Small Groups in the Experimental School Classification of Observed Sessions* Pace 9f Instruction: Large Group Small Group Very Fast 0 2 Fast 4 4 Moderate 9 9 Slow 3 0 Pace: for Subject Materials: Quite Appropriate 8 7 Appropriate 7 6 Somewhat Appropriate l 2 Inappropriate 0 0 Pace: for Students Quite Appropriate 6 8 Appropriate 6 6 Somewhat Appropriate 4 l Inappropriate 0 O * Large groups were two periods and small groups were one period in length. From Table 3.1 can be observed that in three out of sixteen observations the pace was rated Slow for large group instruction whereas in two of fifteen observations of small class instruction the pace was rated Very Fast. The points of most difference suggests a -25- possibility that speed of instruction is not adequately taken into account as a factor when teachers plan activities. Table 3.2 Frequency Comparisons of Observations of Activities Considered to be Articulatory or Connecting one Instruction Period or Topic to Another Classification of Observed Activities References to other Largg Group Small Group periods of instruction (within same subject) 0 6 Introductory or Preparatory Remarks 10 8 Teacher Summary as Closing Activity 3 1 Student Summary at Close of Session 0 0 -26- Table 3 . 3 Frequency Comparisons of Observations of Certain Categories of Classroom Activities in Large and Small Classes Classification of Observed Activities Assignment Activities: Large Group Small Group Giving 9 0 Clarifying 4 0 Checking 4 O Use of Resources and Aids: Textbook 2 1 Local "expert" 0 l Over-head Projector 3 0 Chalkboard 6 0 Movies 3 O Table 3.4 Frequency Comparisons of Observations of Categories of Teaching-Learning Activities and Control of Learning Classification of Observed Sessions Dominant Teaching- Large Group Small Group Learning Activity: Lecture 15 1 Discussion 0 13 Classification of Observed Activities Problem-Solving: Large Group Small Group by Group 0 7 by Individuals 1 7 Critical Thinking Practice 0 11 Individual Review 0 1 Classification of Observed Sessions Control of Learning Large Group Small Group Activities: Teacher-centered 15 5 Student-centered l 10 -27- The data in Tables 3.2, 3.3, and 3.4 indicate that the "largeness" or size of a class influence the activities of the teacher and student within the classroom. In Tables 3.2 and 3.3, it can be seen that teachers behaved differently in terms of articulatory and assignment activities in classes of different sizes. More significantly, the tendency toward teacher-centered instruction noted in nearly all of the large group classes and in one-third of the small classes indicates the need for investigating the teaching-learning variables which arise when class size and scheduling is manipulated. The significant question still remains, do these reactions and adjustments made to different class sizes affect instruction and achievement? Instrumentation There are several factors which limit the time available for teaching and learning in the modern high school. In addition to the careful scheduling, there is the press of desire for knowledge, the competition of students for marks, and the increasing rigor of college admission policies. School personnel are wary, justifiably or not, of the experimenter who wishes to take time from the schedule to administer tests. Therefore, as much use as possible had to be made of instruments which already had a place in the testing schedule of the school. Even though the staff of the control scnool expressed an interest in the Study, some guidance and administrative resistance had to be overcome. The forced choice of instrumentation therefore places a limitation on the study. -28- Achievement The achievement instruments employed in the study were the New York State Regent's tests for the subjects, English III, Physics, and Chemistry. Because most schools follow state recommendations for sequence of courses, and use state syllabi and tests, a Regent's Math 10 score was available in both schools as a pre-achievement measure for physics. A Regent's Biology score was available as a pre-achievement measure for chemistry. These data were used because it was thought they might prove helpful in determining the usefulness of the control school selection technique and in further exploration of their importance as predicting variables. It would, of course, have been desirable to have pre-achievement measures in the appropriate subjects but this was not possible. In the 1961-62 school year, the English IV Regent's test was cancelled. In lieu of any more satisfactory substitute, the Regent's English III test was re-administered in both schools. This test, then, served as both a pre- and post-achievement measure. The test-retest reliability of Regent's examinations is reported in past years as .93 for English III in 1959, .97 for Chemistry in 1957, and .96 for Physics in 1957. Reliability data on the current administrations are not available at the time of this writing. Regent's Test Development Office, Personal communication. -29- Intelligence The Otis Quick-Scoring Mental Ability Test,5 form Gamma, was administered in the fall as a measure of intelligence. The inter-form (test - retest) coefficient of correlation is reported to be .86. The validity coefficients range from .20 to .69 when correlated with grade p01nt average. Motivation The Word Rating List of the Michigan State Motivation Scales developed by Farquhar6 and associates was used as a measure of motivation. This scale is a measure of "Academic Self-Concept." The WRL has a reported reliability coefficient of :93.7 A validity coefficient of .43 is reported8 when WRL is correlated with GPA on a stratified proportionate sample drawn from nine high schools in cities of various size in Michigan. A detailed description of the scale along with its rationale for 5A.S. Otis, Manual for the Otis Self-Administering Tests gf Mental Ability. (Yonkers, N.Y.: World Book Co., 1922). 6William W. Farquhar, A Comprehensive Study 2: the Motivational Factors Underlying Achievement 2f Eleventh Grade High School Students, Research Project No. 846 (8458); Supported by the U.S. Office of Educa- tion, in cooperation with Michigan State University, 1959. 7 William W. Farquhar, Personal communication. 8William W. Farquhar, 5 Comprehensive Study 2f the Motivational Factors Underlying Achievement 9f Eleventh Grade High School Students, Research Project No. 846 (8458); Supported by the U.S. Office of Educa- tion, in cooperation with Michigan State University, Preliminary Report, 1961. -30- development can be found in Payne's doctoral dissertation. Self-Initiation A Self-Initiated Projects Questionnaire of thirty-two items designed to detect variable amounts of self-initiated projects such as designing and building a model, discussing class-learned concepts with adults outside the school situation, and reading an instruction-related book or novel has been under development as part of another study at the 10 University of Rochester. Two different scales, "I Do" and "I Would Like To," have yielded data from students in secondary schools indicating the possibilities of added dimensions to present motivation instruments. The "I Do" scale was chosen as a measure of self-initiation in the present study because of its possible orthogonal relationship to the WRL. The coefficient of correlation between the ”I Do" scale of SIPQ and WRL is .15 on a sample of 83 students. The coefficient of reliabi- lity of SIPQ "I Do" scale as determined by Hoyt's analysis of variance method11 is .70. 9David Allen Payne, A Dimension Analysis gf the Academic Self-Concepts 9f Eleventh Grade Under- and Overachieving Students, Doctoral Disserta- tion, 1961, (Ann Arbor, Michigan: University Microfilms, Inc.). 10 Initiation Related £9 Instruction, in preparation. 11C.J. Hoyt, "Test Reliability Estimated by Analysis of Variance,” Psychometrika, Vol. 6, 1941, pp. 153-160. into relating tion. -31- Specific Hypotheses Tne general null hypothesis described in chapter one is separated two sets of hypotheses. Set A is composed of those hypotheses to achievement and Set M consists of those related to motiva- Hypotheses Al, A2, and A3: There will be no differences in acnievement outcomes between the experimental and control conditions in English when statistically con- trolled for (1) intelligence differences, or (2) pre- achievement differences, or (3) pre-motivation differences. Hypotheses A4, A5, and A6: There will be no differences in achievement outcomes between the experimental and control conditions in Physics when statistically con- trolled for (1) intelligence differences, or (2) pre- achievement differences, or (3) pre-motivation differences. Hypotheses A7, A8, and A9: There will be no differences in achievement outcomes between the experimental and control conditions in Chemistry when statistically con- trolled for (1) intelligence differences, or (2) pre- achievement differences, or (3) pre-motivation differences. Hypotheses M1, M2, M3: There will be no differences in motivation outcomes as measured by the WRL between the experimental and control conditions in (1) English IV, or (2) Physics, or (3) Chemistry when pre-motivation -32- differences are statistically controlled, or in Hypothesis M4: All three subjects combined when combined pre-motivation differences are statistically controlled. Hypothesis M5: There will be no differences in motiva- tion outcomes between the experimental and control conditions as measured by the SIPQ. The level of confidence for the rejection of the null hypotheses is .05. Statistical Models As a method for controlling possible differences between the experimental and control groups in ability, past achievement, and motivation where there was not random assignment of subjects, the analysis . 12 . . of covariance model was chosen. This type of analySLS allows for statistical control of differences which might exist in the two groups of the sample before the experimental condition or treatment is begun. Because the two groups were in two different schools, no random assign- ment of students was possible. Therefore, for the I.Q., achievement 12Allen L. Edwards, Experimental Design in Psychological Research, (New York City: Rinehard and Company, Inc., 1950). -33- scores, and the WRL scale, the following two by two model is applicable as a design: Experi- mental Control School School Pre-test (X) I.Q., Achievement and Motivation WRL Post-test (Y) Achievement and WRL Analysis of Covariance For the SIPQ scores (which were collected only at the end of the . 13 school-year) the "t" test of differences between means was chosen. Data Analysis Available in the university computing center library was a covariance program for the IBM 709. Originally, an attempt was made to modify the program for the 7070, however, after encountering much difficulty, it was decided an original 7070 covariance program ought to be developed. The program which the center devised and validated tests the covariance assumptions of linearity and homogeneity of regression. It tables the output data with analysis of variance, means, analysis of covariance, and computes F on the mean squares obtained from the adjusted sums of squares and adjusted degrees of freedom. 13Ibid. -34- Summary The study in an experimental school where there were staff members who wished to explore the instructional outcomes of different class sizes and scheduling necessitated the selection of a control school. The State Department of Education in New York chose the control school in which the parent population in the secondary unit matched that of the experimental school in educational background or years of high school and college completed. The experimental variable was non-standard size and scheduling which refers to two large instruction groups with 60 to 90 students meeting twice a week for double periods interspersed with small groups for seminars, independent study, and laboratories. The control school used standard size and scheduling or 30 students meeting in single periods five times a week. The objectives for the three Regent's courses, English IV, physics, and chemistry were examined and found to be abstract. They are used by all teachers in New York State who teach the Regent's subjects. They are also used by selected teachers from throughout the state WHO with test specialists in the state education department revise the examinations each year. It was assumed that teachers teaching these subjects and using the common objectives and examinations would be similar in their use of teaching method and materials. It was also assumed that the only major variable unaccounted for in the exploratory study was the use of different teachers. -35- An observation rationale was developed for the experimental classes to better understand how non-standard size and scheduling might effect instruction. Major points of difference between large and small group instruction were found in the pggg of instruction, the appropriate- gggg of the pace for the subject material and students, the argiculation of topics and materials between different classes on same subject, the giving of assignments, the use of resources and aids, in the kinds of teaching-learning activities, and in whether the control of the class was teacher or student-centered. The instruments selected for the study consist of Regent's examinations for achievement; the Otis Intelligence test; the Word Rating List; and the Self-Initiated Projects Questionnaire. Each of these is described and, for each, estimated reliability and other infor- mation reported. Two sets of hypotheses were made. Set A is comprised of nine different null hypotheses about achievement outcomes. Three were made for each of the three subject areas. In each subject area the first null hypothesis is concerned with differences statistically controlled for intelligence scores. In the second, the statistical adjustment is for pre-achievement, and in the third, for pre-motivation. In set M are five null hypotheses. The first, second, and third were made in the three subject areas English IV, Physics, and Chemistry, and in each, post-score differences are adjusted for pre-scores. The fourth M hypothesis was made regarding motivation differences over the combined population of the three subject areas. And the last null hypothesis is ~36- again over the combined population but with a different measure of motivation. The statistical model chosen for the two by two design of pre- and post-achievement and motivation in the experimental and control schools was the analysis of covariance. This model allows for statistical control of prior differences in the two groups since there was no matching or random selection of students possible. A "t" test statistic was chosen to test the significance of mean differences obtained from data collected only post the instruction period. The data were analyzed in the university computing center using an original analysis of covariance program for the IBM 7070. The ,program tested the assumption of linearity and homogeneity of regression. CHAPTER IV ANALYSIS OF DATA Chapter four is divided into two parts. Those data pertaining to achievement in the three subjects, English IV, Physics, and Chemistry are included in the first part. In the second part, data on two different and orthogonal dimensions of motivation are presented. Part One Achievement Hypotheses A1 through A9 afh concerned with the expectancy of finding no significant differences in achievement between two different treatments of class size and scheduling. Analysis of covariance was used as a statistical test of such differences. I Three tables for each subject are provided. In each table is presented means and analysis of covariance for the dependent or achieve- ment variable (Y) and a different independent variable (X). Also included are the adjusted sums of squares, the appropriate adjusted degrees of freedom (df), and mean squares for the experimental-and control conditions. F is shown with level of significance where appropriate. -37- ~38- Analysis.gf English I! Scores English achievement in relation to the treatment effects was analyzed adjusted for intelligence, pre—achievement, and pre-motivation. English IE: A1 Ho: 1K1 fi/Hh, Adjusted for I.Q. Means and analysis of covariance for achievement in English adjusted for I.Q. scores are shown in Table 4.1. Means and Analysis of Covariance for English Controlled for I.Q.. U (Re-test) N f I.Q. Means (X) English III (Y) Experimental 63 ' 116.16 82.24 Control 52 116.08 78.00 Total 115 116.12 80.32 Source Adjusted SS (Y) Adj. df . Mean Squares Between 500.48 1 500.48 Within 5319.53 112 47.50 Total 5820.01 113 F1, 112 = 10.5374 p < .01 The data in Table 4.1 indicate that achievement in English, when adjusted for intelligence is better in the experimental school than in the control school. The obtained F of 10.5374 with the appropriate degrees of freedom 1 and 112 is significant beyond the one percent level of confidence. Hypothesis A1, therefore, is rejected. -39- English IV: A2 HO: a;l =]12, Adjusted for pre-achievement. Means and analysis of covariance for acnievement in English adjusted for pre- achievement are presented in Table 4.2 Table 4.2 Means and Analysis of Covariance for English Controlled for Pre-Achievement N Pre-Ach Means (X) English III (Y) Experimental 63 80.81 82.24 Control 52 78.25 78.00 Total 115 79.65 80.32 Source Adjusted SS (Y) Adj. df Mean Squares Between 171.22 1 171.22 Within 4138.68 112 36.95 Total 4309.90 113 F1’ 112 = 4.6334 p < .05 f Achievement in English, ainndicated in Table 4.2, when controlled for pre-achievement differences is higher for the experimental school over the control school. The F with 1 and 112 degrees of freedom of 4.6334 is significant beyond the five percent level of confidence and is cause for rejection of Hypothesis A2 of no difference. -40- English IV: A.1 HQ: ”FA, Adjusted for pre-motivation. In Table 4.3 is presented means and analysis of covariance of English achievement p J adjusted for motivation differences obtained on the WOrd Rating List. Table 4.3 Means and Analysis of Covariance for English Controlled for Motivation (WRL) N Motivation Means (X) Experimental 63 30.16 Control 52 32.87 Total 115 31.38 Source Adjusted SS (Y) Adj. df Between, 749.37 1 Within 6460.78 112 Total 7210.15 113 F1, 112 I 12.9906 p (1 .01 English III (Y) 82.24 78.00 80.32 Mean Squares 749.37 57.69 The data presented in Table 4.3 show that achievement in English when adjusted for motivation differences is higher in the experimental school. hypothesis A3 of no difference is rejected by the obtained F with one and 112 degrees of freedom of 12.9906 which is the one percent level of confidence. -41- Analysis 2f Physics Scores Physics achievement in relation to the treatment effects was analyzed adjusted for intelligence, pre-achievement, and pre-motivation. Physics: A4 Ho: flfflmAdjusted for I.Q. Means and analysis of covariance for achievement in Physics adjusted for I.Q. scores are shown in Table 4.4. Table 4.4 Means and Analysis,of Covariance for Physics Controlled for I.Q. N I.Q. Means (X) Physics (Y) Experimental 32 119.62 79.97 Control 58 119.96. 82.55 Total 90 119.84 81.63 Source Adjusted SS (Y) Adj. df Mean Squares Between 117.53 1 . 117.53 Within 8301.19 87 95.42 Total 8418.72 88 F1, 87 = 1.2318 not significant (ns) Achievement in Physics, when controlled statistically for differences in I.Q. scores, is very similar in the two treatments. The obtained F with l and 87 degrees of freedom of 1.2318 is not significant. There- fore, Hypothesis A4 of no difference is accepted. The unadjusted mean difference in achievement of approximately two and a half points is in favor of the standard class size and scheduling treatment. -42- Physics: Ali Ho: fl,=/g,Adjusted for pre-achievement. In Table 4.5 is presented Physics achievement means and analysis of covariance adjusted for pre-achievement. Table 4.5 Means and Analysis of Covariance for Physics Achievement Controlled for Pre-Achievement N Pre-Ach Means (x) Physics (Y) Experimental 32 81.94 79.97 Control 58 82.69 82.55 Total 90 82.42 81.63 Source Adjusted SS (Y) Adj. df Mean Squares Between 104.78 1 104.78 Within 7775.22 87 89.37 Total 7880.00 88 F1, 87 = 1.1725 ns Physics achievement, when controlled statistically for differences in pre-achievement in Math 10, is not very different between the two treatments of class size and scheduling; The analysis of covariance presented in Table 4.5 provides an F with one and 87 degrees of freedom of 1.1725. Minor Hypothesis E of no difference is accepted. -43- Physics: A6 HO: /K.fi/«;,Adjusted for pre-motivation. Physics achievement differences analyzed and adjusted for pre-motivation are shown in Table 4.6. W Means and Analysis of Covariance for Physics Achievement Controlled for Motivation N Experimental 32 Control 58 Total 90 Source Adjusted SS (Y) Between 299.56 Within 9200.19 Total 9299.76 Motivation Means (X) 35.78 ,32.l7 33.46 Adj. df 1 .87 88 F1, 87 = 2.8328 ns Physics (Y) 79.97 82.55 81.63 Mean Squares 299.56 105.75 The F value of 2.8328 is insignificant and Hypothesis A6 of no difference in achievement between the two treatments is accepted. -44- Analysis 22 Chemistry Scores Physics achievement in relation to the treatment effects was analyzed adjusted for intelligence, pre-achievement, and pre-motivation. Chemistry: A7 H0: [(1 =fl2, Adjusted for I.Q. Results for achievement in Chemistry controlled for differences in I.Q. scores are shown in Table 4.7. Table 4.7 Means and Analysis of Covariance for Chemistry Achievement Controlled ‘ for Intelligence N I.Q. Means (X) Chemistry (Y) Experimental 41 119.78 79.86 Control 40 115.80 87.75 Total 81 117.81 83.75 Source Adjusted SS (Y) Adj. df Mean Squares Between 1787.85 1 1787.85 Within 10584.25 ; 78 135.70 Total 12372.10 79 F1, 78 = 13.1755 p < .01 When Chemistry achievement is adjusted for I.Q. score differences, the mean difference is significant. The standard class size and scheduling treatment is favored in the difference and Hypothesis A7 is rejected. -45- Chemistry: A8 HO: [(1 “2, Adjusted for pre-achievement. In Table 4.8 is presented mean scores on Chemistry achievement statisti- cally adjusted for pre-achievement scores. Table 4.8 Means and Analysis of Covariance for Chemistry Achievement Controlled for PreeAchievement.in Biology N Pre-Ach Means (X) Chemistry (Y) Experimental 41 85.95 79.85 Control 40 81.15 87.75 Total 81 83.75 83.75 Source Adjusted SS (Y) Adj. df Mean Squares Between 2807.87 1 2807.87 Within 6069.08 78 77.81 Total 8876.95 79 F1, 78 = 36.0869 p < .01 In Table 4.8 is presented mean scores on Chemistry achievement statistically adjusted for pre-achievement scores. A significant difference in favor of the control conditions is again obtained. Hypothesis A8 of no difference is rejected. -45- Chemistry: .A9 HO'./“l §;42,-Adjusted for pre-motivation. Mean scores on Chemistry achievement adjusted for pre-motivation differences are shown in Table 4.9. Table 4.9 Means and Analysis of Covariance for Chemistry Achievement Controlled for Motivation N Motivation Means (X) Chemistry (Y) Experimental 41 34.29 79.85 Control 40 32.07 87.75 Total 81 33.20 83.75 Source Adjusted SS (Y) Adj. df Mean Squares Between 1677.14 1 1677.14 Within‘ 9086.75 78 116.50 Total 10763.90 79 F1, 78 = 14.3965 p < .01 Chemistry achievement adjusted for motivation differences between the two treatment groups is significantly higher for the control treatment. Therefore, Hypothesis A9 of no difference is rejected. -47- Part Two Motivation Hypotheses M1 through M.4 are null hypotheses concerned with the finding of no significant differences in motivation between the two different treatments of class size and scheduling. The statistic, analysis of covariance was used as a means of adjusting final scores obtained by administering the Word Rating List for scores obtained on the same instrument before the treatments. The next three tables, 4.10 through 4.12, contain separate covariance analyses on pre- and post-motivation scores for the p0pu1a- tion in each of the three subjects, English IV, Physics, and Chemistry. In Table 4.13 is presented an analysis of covariance of combined pre- and post-WRL data from all three of the subject populations. The last table in this chapter contains a comparison of mean scores obtained on the Self-Initiated Projects Questionnaire from the combined experi- mental and control groups with a "t" test of the significance of the mean difference. -48- Analysis 2f_Motivation (WRL) Scores Motivation _i_1_1_ English: Ml H0: [‘1 =M2, Adjusted for pre-motivation. In Table 4.10 is data on post-motivation scores adjusted for pre-motiva- tion differences. Table 4.10 Means and Analysis of Covariance of Pre- and Post-Motivation (WRL) for English Population . N Pre-Motivation (X) Post-Motivation (Y) Experimental 63 30.16 33.38 Control 52 32.86 32.06 Total 115 31.38 33.69 Source Adjusted SS (Y) Adj. df Mean Squares Between 34.98 1 34.98 Within 5056.92 112 . 45.15 Total 5091.90 113 F1, 112 - 0.7747 ns Post-motivation mean scores adjusted for pre-motivation scores were not significantly different as presented in Table 4.10. The non-standard treatment did not significantly effect motivation as measured by the WRL in English IV. Hypothesis A10 of no difference is accepted. -49- Motivation in Physics: M2 HO: J11 =fl2, Adjusted for pre-motivation. Analysis of covariance of pre- and post-motivation scores obtained from the population of students in both schools in physics is presented in Table 4.11. Table 4.11 Means and Analysis of Covariance of Pre- and Post4Motivation (WRL) for Physics Population N Pre-Motivation (X) Post-Motivation (Y) Experimental 32 l 35.78 36.97 Control 58 32.17 33.17 Total 90 . 33.46 34.52 Source Adjusted SS (Y) Adj. df Mean Squares Between 38.11 1 38.11 Within 4253.44 87 48.89 Total 4291.55 88 F1, 87 = 0.7795 ns The adjusted post-motivation scores yield an F of insufficient value to be significant. Hypothesis All of no difference is therefore accepted. -50.. Motivation i3 Chemistry: M3 HO: lkfi f/fiz, Adjusted for pre-motivation. For the Chemistry population, when post-motivation means are adjusted for pre-motivation differences, as shown in Table 4.12. Table 4.12 Means and Analysis of Covariance of Pre- and Post—Motivation (WRL) for Chemistry Population N Pre-Motivation (X) Post-Motivation (Y) Experimental 41 34.29 36.66 Contro1 40 32.07 32.80 Total 81 33.20 34.75 Source Adjusted SS (Y) Adj. df Mean Squares Between 114.46 1 114.46 ‘Within 2734.55 78 35.06 Total 2849.01 79 F1, 78 = 3.2647 ns Analysis of covariance produces an insignificant F. Hypothesis A12 is accepted. -51.. Motivation i3 Combined Subjects Population: M.4 HO'.Z%1 f/Ké, Adjusted for pre-motivation. For all three subjects population combined, analysis of covariance of pre- and post-motivation scores obtained over both treatments as shown in Table 4.13 provide an F of sufficient amount to be significant. Hypothesis A13 is rejected. Table 4.13 Means and Analysis of Covariance of Pre- and Post-Motivation (WRL) for Combined Population N Pre-Motivation (X) Post-Motivation (Y) Experimental 136 32.73 35.21 Control 150 32.39 33.38 Total 286 32.55 34.25 Source Adjusted SS (Y) Adj. df Mean Squares Between 183.95 1 183.95 'Within 12067.88 283 42.64 Total 12251.83 284 F 283=4.3137 p < .05 1, -52- Motivation (SIPQ) _1_._r_1' Combined Subjects Population: M5 H0:fl1 =_'/«2, “t" test on post-motivation. Using another measure of motivation, the Self-Initiated Projects Questionnaire, data obtained from all students in both treatments is analyzed by a "t" test of significance as shown in Table 4.14. Egble 4.14 Experimental and Control SIPQ Means, Variance, and "t" test 2 Means N S (variance) Experimental 82.297 138 391.688 Control 84.42 99 148.959 Standard error of difference - 2.08 t a 84.420 - 82.297 2.08 = 1.021 ns In this measure, the higher the score, the lower is self-initiation. While the trend is in favor of the experimental treatment, the "t" of 1.021 proved to be not significant. Therefore, Hypothesis A14 of no difference is accepted. -53- Part One Summary Achievement, as measured by Regent's examinations in all three subjects, English IV, Physics, and Chemistry, was adjusted statistically for independent measures of intelligence, pre-achievement, and motivation. The F values obtained through covariance techniques indicated that: (1) mean differences favored the experimental or non-standard treatment in English IV in all three adjustments, (2) means in Physics were not significantly different for all three adjustments, and (3) mean differences favored the control or standard treatment in Chemistry for all three adjustments. Part Two Summary For the separate subjects, English IV, Physics, and Chemistry, moti- vation scores of the Word Rating List from the experimental and control groups were analyzed using covariance techniques. .No significant differences were obtained for the individual subject p0pulations. How- ever, when the combined subject populations were analyzed the F was of sufficient amount to be significant in favor of the experimental treatment. Therefore, null hypotheses M1 through M3 were accepted and M4 was rejected. The combined subject population means on the Self—Initiated Projects Questionnaire were tested with a "t" test and the difference was not significant. Null hypothesisM5 was accepted. CHAPTER V SUMMARY, DISCUSSION, AND RECOMMENDATIONS Summary The main problem of this exploratory study was to examine the effects of non-standard class size and scheduling on achievement and motivation outcomes of instruction in three high school subjects, English IV, physics, and chemistry. An experimental and control school were established for data collection under the two conditions, non-standard and standard class size and scheduling. The effects of large class size and reduced scnedu- ling on instruction were also considered. Regent's examinations in the three subject areas were used for l achievement measures. Motivation data consisted of the Word Rating List of Farquhar and associates and the Self-Initiated Projects Questionnaire2 The analysis of covariance was selected as a test of all the comparisons where pre- and post-data were obtained. The "t" test was used for a comparison of means obtained from the Self-Initiated Projects Question- naire scores. For English IV achievement, the null hypotheses were rejected in favor of the experimental conditions. The null hypotheses were accepted for physics achievement. For chemistry achievement, the null hypotheses were A 1WilliamW. Farquhar, A Comprehensive Study pf the Motivational lfiactors Underlying Achievement pf Eleventh Grade High School Students, Research Project No. 846 (8458); Supported by the U.S. Office of Educa- tion in cooperation with Michigan State University, 1959. 2Clarence M. Williams, The Development pf 2_Measure‘gf Self-Initia- jgion Related £3 Instruction, in preparation, 1962. -54- -55- rejected in favor of the control conditions. In a comparison of motivation differences between the two conditions in each of the subject matter areas, pre- and post-WRL scores were tested with analysis of covariance. All null hypotheses were accepted. When the motivation data for the three subject matter areas were combined, a pre- and post-analysis of covariance provided a significant difference in favor of the experimental conditions of large class size and reduced scheduling. Therefore, the null hypothesis was rejected. In a similar test, combining scores for the three subject matter areas, a "t" test of means of SIPQ scores between the two conditions proved to be not significant and the null hypothesis was accepted. Discussion The major conclusion of this exploratory investigation is that class size, as a variable, affects the teaching and learning situation. Due to certain concessions necessary to experiment in a field setting and necessary assumptions regarding achievement measures, it was impossible to separate sufficiently the full effects of class size on instruction and learning and motivation outcomes. However, awareness of the impor- tance to teaching of manipulations of class size (with the attendant scuedule shifts) was increased. The differences between teachers could account for the findings of better achievement in English in the experimental school, same achieve- ment level in physics, and better achievement in tne control school chemistry. Differences of this kind could be called the "between" ~56- teacher differences. Between teacher differences could reflect several other variables like, (1) a difference in capacity, (2) a difference in knowledge, understanding, and skill in the particular subject, (3) a difference in knowledge, understanding, and skill in teaching, (4) a difference in amount of experience, (5) a difference in kind of experience, or (6) a difference in willingness to work as hard as necessary to insure that every student learns as much as he can. Differences of another kind might be called the "within" teacher differences. Within teacher differ- ences might reflect other variables such as, (1) capacity to adapt method of teacuing to the exigencies of a particular situation, or (2) tolerance to stress and strain. If a teacher's ability to cause students to learn so much knowledge, understanding, and skill in a particular subject is the sum of all of these and perhaps other possible kinds of variables, then just controlling, in future experimentation, for the gross, obser- vable teacher differences or balancing them out might not be sufficient. However, the practical controlling by matching or, preferably, by balancing is to be desired in the immediate next steps in empirical investigation of teacher effects on learning, whether it be in large, small, or standard size groups. The differences between the subject matter areas could be the reason why the results were so disparate. Perhaps an advanced high school English class for seniors is one in which the objectives are rather broad and flexible. Details like punctuation, grammar, or proper choice of word may not need to be of great concern. Therefore, the students can -57- be concerned with ideas and concepts at a level where communication is more certain. A course in high school physics may have both specific and broad objectives. Because it is the only course of its kind in most high school schedules and not one of a sequence of courses, in all probability it is necessary to make it extremely comprehensive. Similarly, high school chemistry must also be considered a comprehensive course. It is probable that applications of knowledges and skills gained in chemistry are much fewer and more difficult to arrange than those in physics and senior English. Large group instruction might have made senior English a more interesting and vital class to the students and made physics and chemistry difficult to comprehend sufficiently. Because this exploration did not provide, in its design, tests of sufficient warranty of the assumptions made about the similarity of objectives, materials, and teaching of Regent's courses, nothing can be said of their real effects. However, iall of these kinds of variables need to be considered in future empirical testing of the effects of non-standard size and scheduling versus standard size and scheduling on achievement and motivation. The initial differences in intelligence, achievement, and WRL scores in the student populations were slight. Although the effects of these differences on the dependent variable were statistically controlled by the analysis of covariance, they probably would not have proven different from chance had they been tested by the design. However, students do differ in ability, achievement, motivation, and in other ways. If practical experimentation is desired on the effects of large group-small -58- group versus standard size group instruction, these differences should be taken into account. For example, students of high ability may be able to learn sufficiently well under more class size conditions than students of low ability. Recommendations In the present exploration, an attack on the problem of class size was attempted. The results obtained would indicate that considerable more sophistication about the important variables needs to be gained before more productive experiments can be designed. The following are recommendations for design variables if class size differences are to be studied: (1) §i§g. If practical information regarding the most efficient size of a class is needed then what is presently being practiced and what might be desired ought to be considered. Students learn by themselves, in small groups, in medium size groups, and in large groups. 30, perhaps these four different sizes of learning units should be studied: one student, two to 14, 15 to 39, and 40 to ___. (The units are arbitrary and can be altered if necessary). (2) Subject. Probably the best delineation of a subject could be made if it were to be especially constructed or artificial. A unit on a number system to a base other than the common ones might be devised or, if this is (3) -59- impossible, something like the geography of a non- existent island might be devised. This unit ought to be Snort so that explicit objectives, as operational as possible, could be developed; so that a comprehensive objective test could be constructed; so that teachers might be willing to try to learn it and teach it; and so that it could be accomplished in a reasonable time with students in a school setting. In other words, the unit ought to be: of sufficient scope and depth so that differential learning would obtain; sufficiently easy so that teachers gggld teach it; easily and completely evaluated, and; short. Because scheduling has been mentioned previously as a variable, it should be noted that for the next steps in investigation of size of class, it may be left out. The initial unit should be short enough to accomplish in one session. Teacher. Because of the possibility of many kinds of teacher differences (as discussed earlier) affecting learning outcomes, tne teacher as a variable should be limited as much as possible. Perhaps a start could be made with one teacher teaching the selected subject unit in both ascending and descending order of size units averaging the learning accomplishment for like size units. Analysis of variance (with random assignment of .(4) -60- students) would be an appropriate statistical method. The next step might be to have several teachers study the subject unit and take the comprehensive test. Different levels of teacher learning might then be arbitrarily set so that partial answers to some of the questions posed in the earlier discussion might be obtained. The effects of age, sex, training, and experience might become evident in these different levels when they are applied to the experimental subject- 1 size teaching task. Other variables. Random assignment of students would take care of student differences at any particular level or set of levels but the kind of student ought to be considered carefully before the subject unit is constructed. The most serious consideration is that the stipulated re5ponses to the stimuli of the subject unit be in the response repertoires of the students. That is, if there are physical responses like drawing a certain kind of figure or sequences of figures, the physical development of the student needs to be con- sidered. If there are verbal responses then there should not have to be any blocking because of previous experience or achievement of the students. The class- room setting should be constant over the size units -61- and the time of day and week ought to be as closely approximated over the size units as possible. The directions and explanations given tne teachers and students ought to be carefully constructed and administered in a similar fashion to all of the participators. It might even be possible to give teachers variable, specified units of time to plan for the presentation of the subject unit. In other words, the next steps ought to include consideration of all of the foregoing variables and every attempt should be made to approximate learning laboratory conditions as closely as possible. BIBLIOGRAPHY Bittick, Edsel F., Differentials 23 College Success 33 the University 2f Texas gf Students From Large and Small Texas High Schools. Doctor's Thesis, University of Texas, 1956. Cunningham, Harry A. "Lecture Demonstration Versus Individual Laboratory Method in Science Teaching -- A Summary. " Science Education. Vol. 30, 1946, pp. 70- 82. Dranes, David Daniel, "A Study of Class Size and Instructional Methods," Doctoral Dissertation, The University of Wisconsin, 1957. Edwards, Allen L., Experimental Design 32 Psychological Research, (New York City: Rinehard and Company, Inc., 1950). Entwisle, Doris R., "Attensity: Factors of Specific Set in School Learning." Harvard Educational Review. Vol. 31, Winter, 1961, pp. 84-101. Farquhar, William‘w. , A Comprehensive Study of the Motivational Factors Underlying Achievement of Eleventh Grade High School Students, Research Project No. 846 (8458); Supported by the U. S. Office of Education, in cooperation with Michigan State University, 1959. Firman, William D. and associates, Procedures ig_School Quality Education, mimeographed report, published by the Division of Research, State Education;Department, Albany, New York, 1961. Fleming, C. M., "Class Size as 3 Variable in the Teaching Situation." Educational Research, February, 1958, pp. 35-48. Giffin, Kim, and John Waite Bowers, "An Experimental Study of the Use of Lectures to Large Groups of Students in Teaching the Fundamentals of Speech. " The Journal of Educational Research, Vol. 55, No. 8, May, 1962, pp. 383- 385. Goodlad, John I., "Classroom Organization." Encyclopedia 9: Educational Research. (Ed. by Chester W. Harris. New York: The Macmillan Company, 1960). Hoyt, C. J., "Test Reliability Estimated by Analysis of Variance.” Psychometrika, Vol. 6, 1941, pp. 153-160. Husband, Richard W., "A Statistical Comparison of the Efficiency of Large Lecture vs Small Recitation Sections Upon Achievement in General Psychology." Journal 2f Psychology, Vol. 31, 1951, pp. 297-300. -62- -63- McKeachie, Wilbert J., "Students, Groups, and Teaching Methods." American Psychologist, Vol. 13, 1958, pp. 580-584. Morse, Arthur D., Schools 2f Tomorrow - Today. (Garden City, New York: Doubleday and Co., 1960). Nelson, Wallace B., "An Experiment in Class Size in the Teaching of Elementary Economics." Educational Research, October, 1959, pp. 330-341. Otis, A. 8., Manual for the Otis Self-Administering Tests 2; Mental Ability. (Yonkers, N.Y.: World Book Co., 1922). Payne, David Allen, A Dimension Analysis 2; the Academic Self-Concepts pf Eleventh Grade Under- and Overachieving Students, Doctoral Dissertation, 1961, (Ann Arbor, Michigan: University Microfilms, Inc.). Robb, M. H., "Modular Scheduling at Euclid Central." National Association ‘9: Secondary School Principal's Bulletin, Vol. 46, February, 1962, pp. 66-69. Siegel, Lawrence, F. G. Macomber, and James F. Adams, "The Effectiveness of Large-Group Instruction at the University Level."' Harvard Educational Review, Vol. 29, 1959, pp. 216-226. Smith, Dora V., Class Size 13 High School English: Methods and Results. University of Minnesota, 1931. Smith, Dora V., "Vital Factors in the Present Situation in Class Size." English Journal, Vol. 22, 1933, pp. 366-374. Trump, J. Lloyd, Images gf the Future -‘A New Approach £9 the Secondary School. Urbana: University of Illinois, Commission on the Experimental Study of the Utilization of the Staff in the Secondary School, 1959, 46 p. Trump, J. Lloyd, New Horizons for Secondary School Teachers. Urbana, Illinois: University of Illinois, Commission on the Experimental Study of the Utilization of the Staff in the Secondary School, 1956, 35 p. Trump, J. Lloyd, and Dorsey Baynham, Focus on Change - Guide to Better Schools. (Chicago: Rand, McNally and“ Co. , 1961). Williams, Clarence'M., The Development gf_g Measure gf Self-Initiation Related £2 Instruction, in preparation, 1962. APPENDIX NO. 1 EXAMPLES OF SCHEDULE DIFFERENTIATION -64- H.,. ..1..!. a. . .N. .r ....r an mochdrw mHMmawmw H aucz bvvoamwx H nosuugamnd mmuwoam : a 2 .2. w H Hub macmmbam N 5.0 macmouam x N Hum: HM we”. co avian wow 00 3.35.3 . Mum m meEEm mmszgm mgztwm mlwm macmmuam mnwm macmmnam mIHw magmas won 5. awn—3mm mow km Random no”. hm 3.95am mcuuwmamuémw H. 5. Hanna A. psawdwmcmw madman." 00:93:08 :2“: ammo—emu. mgwum $6 “$3.. Manon—H5305“ w. Hummumummud macaw 3.38%.. w. mmawupdm apnoea agocmsodw mormaswo Ammo: madness mosmmcwmm 050m m town we macaw on. mun—by b. 5% Honda”. u. H.333 mum $553. 326 won ammofidcm mum. gaoummamba macaw. 9. wmowuuoomw womoasm mod mvmowmw mummm we do? mums mam Hmumm magnum. q. osdmdmummbm 385.8 coouwm madMHdem no Hmumm mwozw £5591... muwwwofigw Huang an 3.4.... 33333 ownmmmm. manning : a a .3 m. EH53 Hm H mu macamaam won. #013 Bataan $30 when mafia MES mx>3%bm ow mommqum UmwwmwmzaHbeHoz Zoslmdmsmmum ullllx mmflHomm 3 a z 65 w wwwmwom H 00 madmmuam H: 00 masdmuam Ha wwmeom bmdowmdouw M NM Hmumm muofiu NM Hmumm macaw co Bpusaom I Hsmdudoawou won Humaudodwos mos muocv ow HOIHM oo Swasdmm co sHuswmm mfiuuHmamsamH H. oarma mocdHo umuHoa Hmuoumdouwmm museum «unocmw :mow.m marmach. H5Mowamdwou" m. dmchHw mHmwowmua Humanuonoum H5 Hmcoum¢0dw. flamma domoUHsmv. u. bmdoumwoew mHmo uuoqwmmm HHBm mow msDHH muofiu HsdeSO¢Hou. b. mucouummmamsa om Huamvozaoua mdfiQfitmOHmuom uncuooam. .mflmsmmum z a 2 as w wvwmwom H Mmlwo mnnmmudm: mmsm muss when mean bouhw Buucdmm m H. can mHamHm umuHoa Hudonmnouw moam aHam acuHsm dam tmow AcmcmHHw mmao Humdnconowv MNwafim ow momwddbm waewnmaanewmm zoulmdmnmuua 33.an x .H. z a: m. It. Qw mdsamuum H5 Hmnmm am assesses Ha Hmumm oumaHmauw HmdonaoHH .w macaw HsmauconHou xxx macaw Humeusoauou xxx 00 suuaemm 1 05.5.5. ~ as. 8 E338 as. 8 Random 93:? a. 5.5 mauUHmeouamH H. canoe mocde umHHom Hmconmdoawmm muummm aruocmw somw.m maeraHm. Humouemdwouu n. dmanHw duwmmumua Hsmnucodoum H: Hmuoumwouw Adams ammowwumv. w. reconnaouw mHmo bandpass «Ham won memHH macaw HumduaoaHou. b. msoocummmswud ow Hummumsamud macawrmowmuom ”seamedm. m mean 3 a a as - a... bOIHm aHueamm I O..- l ..OII' 1" armsemduw. “ H W nmlwo macaoudm mean # mmao mesa mean #1 M H. cam mHumHm umnwoo Hmwoumaonw moan «Ham chHsm wan amok. AdmsmHHw mmsm Hamausoaouv Faster schedule set-up for large group instruction including those in the experiment, other large groups and alternating classes. Periods M '1‘ W TH F ! - X45 English 12 Physics I English 12 Physics Cit. Ed. 12 9O “1% 110 60 no 60 '70 students minutes L students students students students 45 min. seminars throughout week th 11 peaking emina a 1 (Regular ' lass size) . Science 1’ boratcrieg / ’/ 1‘ . / 3 " Chemistry World Chemistry orld Seminars in 45 History story History with 125 60 125 60 subjects Vt students students students students minutes A [.5 roblenf Americ n oblafls rid an to“ min, in in 81.301? emocracfy emocrac y ience (regular {1 class a 12?) students 3 t I_ 1! . APPENDIX NO. 2 OBSERVER'S CHECK SHEET -68.. PITTSFORD CENTRAL SCHOOL OBSERVER'S GUIDE SHEET Varying Size Group Instruction Program Observer Observation # Room # Date Time Number of Students General organization of group (circle predominant nature of group) Teacher leader, student leader, discussion, panel, independent work, sub—groups, other (specify) If organization changes during class, note here COMMENTS l. How would you judge the pace of instruction? very fast fast moderate slow very slow How appropriate was the pace with respect to , materials? Students? a) quite appropriate , b) appropriate c) somewhat appropriate d) inappropriate 2. What references were made (specifically or by implication) to other portions of the Varying Size Group Program? 3. From what resources were student contributions drawn? (example - library, home, research, large group presentation) 4. What approximate percentage of time of the seminar is teacher oriented? student oriented? 5. The following procedure seemed unusual: -2- TYPES OF ACTIVITY (order in which activity happened and time spent) (Items left blank indicate no evidence during this observation) I Introductory & transitional III Related activities outside of activities this meeting Preparatory remarks Assignments Teacher summary giving periodic clarifying '_—_ at close Checking _—— Student summary Research written library use oral other references Other (describe) Individual projects :::: References to other type instruction in II Kinds of teaching—learning this course (large/ activities small group) Lecture Discussion IV Resources used between teacher/ Records, tapes class Slides within class Local "experts'I -——. Student panel Film Demonstration Introduction Dimulation showing role playing follow-up by entire group Buzz groups Show of hands Instruction in note—taking library use other skill (List ) Review Use of blackboard Solving preframed problems by group individually Drill Cautionary remarks re ideas re mechanics Practice of critical tthking questioning of subject material by students by teacher(Answer eliciting) evaluation of ideas problem sol/ing Testing Interpreting test APPENDIX NO. 3 SELF-INITIATED PROJECTS QUESTIONNAIRE -71.. SELF-INITIATED PROJECTS QUESTIONNAIRE 0n the following pages you will find a list of possible outside school activities which you might choose, or would like to choose because you find them interesting and enjoyable. For each activity indicated, you should select the one term that best describes how often you g2_this activity. For this same activity you must also select the one term that best describes fig! giggg you would like £9,g__ this same outside school activity. The five possible descriptions for each after-school activity that you do or would like to do are: l - Very often (v.o.) 2 - Often (0.) 3 - Sometimes (5.) h - Seldom (sel.) 5 - Never (n.) The numbers representing each description are listed beside each outside activity. To describe how often you do each activity, circle one of the num- bers at the left of the activity. Then, to describe how often you would like t d this same activity, circle one of the numbers at the right of the activity. sel. #' 3' £' 3' $‘ t' -P ..t‘ Read a novel Discuss 3 book I have read with someone Watch a sports event Make models of autos, airplanes,etc. Draw pictures about what I see Play on a school sports team Visit a public library (Unassigned) Observe people doing some activity Ask people questions about what they are doing Write short stories or poems Participate in group discussions about school See a movie Discuss politics with someone Try to solve math puzzles, cross- word puzzles, etc. Read a newspaper Play a musical instrument Visit the library to obtain a novel to read Write a letter to a friend Play a Sport with a friend or a group Sing with a vocal group I would like to: v.0. l l O. 2 2 s. 3 3 sel. L, L, #' $' £' $‘ t' 3' 3‘ J? n. \DU‘IU‘U‘IUT U1U'I -_a‘---“ . ...-...-.. sel. #4??? J? 4? Read about famous people in history Try simple science experiments Read an interesting magazine article Play challenging games such as checkers, chess, charades and twenty questions Participate in a school dramatic activity Listen to music by myself Work on a scrapbook which contains material I am studying in school Read a foreign language Attend a play Read a book review Enter a contest - puzzle, writing, photography, etc. Do work in photography Work on a collection - insects, rocks, stamps, etc. Build scientific equipment - ham radio, telescope, etc. Read a scientific journal lead a literary journal Discuss a theory with friends I would like to: v.0. l 0. 2 2 S. 3 3 sel. I. I, if??? 43' J? UWU'I U1 . - H.... .....-h. .. ..- -- . ... — - ”.1- . ._.- n. .,. - ..-”-.. . .... “._.. ¢.~. r ’ l i , c . c c . a , l‘ t i s 1 . I ~ . . , . . i , i i ‘ l . ‘ i n o ‘ f . . . l i ' I‘ . , i ‘ l i . , ~ ' f . t i - _ . u. . - u v V 1 I . i r s ‘ . . v . . ‘ , ',- - N i v . . , . r 1 I l 0’ . .. ‘ ‘ 1‘ v. , I . A 1" i . ‘ ,i‘ . . p \ — ' I _ p . , ,J i . ‘I M _J ( - l ' _' - . . . ' q ."\ , ‘ n -. - _ v ' ~ , . . J . .. . ' . ‘ ' ' | c . . - ' -‘ . ' i ' ‘ . ' ‘ ‘ T I \ f . l J ' I . A . ' 1 4 . I . \ ‘ ii . . . i , ‘ - - . l . ‘ t - . m - . .. ‘ . l ‘ I ' ’- I T . ')