APPLICATIONS OF A COMPUTER SIMULATION MODEL TO LOGISTICAL DECISIONS IN A UNIVERSITY Thesis for the Degree of Ph. D MICHIGAN STATE UNIVERSITY GEORGE MERRILL VAN DUSEN 1969 This is to certify that the thesis entitled Applications of a Computer Simulation Model To Logistical Decisions in a University presented by GEORGE MERRILL VAN DUSEN has been accepted towards fulfillment of the requirements for Ph.D. degree in Educational Adminis- tration _ SWW Major professor Dz,he May 14, 1969 0-169 ABSTRACT APPLICATIONS OF A COMPUTER SIMULATION MODEL TO LOGISTICAL DECISIONS IN A UNIVERSITY by George Merrill Van Dusen A means was sought in this study of demonstrating how a specific model might be employed to describe the operations of a university so that educational administrators can become aware of the potential of a systems approach as an aid in rational decision- making. It was the purpose of this study to: (I) describe in non- technical language a systems model and an implementation of the model using data descriptive of Michigan State University as devel- oped by a research group in the College of Engineering; (2) identify aims, objectives, and problems concerning the future direction of Michigan State University as suggested by a selected group of educational administrators responsible for policy decisions in this institution; and (3) show how some of these identified concerns and objectives were addressable to a Specific computer simulation program which is an implementation of the theoretical model des- cribing the university as a system. GEORGE MERRILL VAN DUSEN The theoretical model used in this study identifies the university as a total system composed of interacting sub-systems or components. Mathematical models have been constructed for selected representative components of the system and the interconnection pattern among components. Each component defines a specific operation or function of the university to the overall educational process and the associated units of production. It delineates how the university uses its resources—-personnel, space, and equipment--in the production of educated manpower and other services. The development of a simulation computer program (MSUSIMZ) which includes data for the College of Engineering was the tool used to conduct eXperiments. By using this program it was possible to vary selected parameters to reflect conditions and policies of Michigan State University as recognized by a se— lected group of educational administrators. Interviews were conducted with thirteen administrators at Michigan State University to offer input for the experiments accord- ing to aims, objectives, and problems regarding the future direction of this Institution. The interviews yielded a broad range of response from detailed and specific alternatives to generalized goals and obj ective s . GEORGE MERRILL VAN DUSEN As a result of the interviews the following conclusions were drawn. Administrators desire more descriptive information as an aid in planning and decision-making. The significance of enrollments and the importance of finances are recognized as critical elements in planning. Policy decisions are made in isolation without an awareness of other areas on campus. A need exists for evaluation of present programs and personnel before the development or expansion of innovations. Seven experiments were designed which were addressable to the conclusions drawn from the interviews. The parameters which were manipulated in the simulation program reflected changes in enrollments, finances and policies as suggested by the administra- tors. Experiments were conducted that reduced the number of new freshmen and increased the number of new sophomores, juniors, and seniors admitted to the University. One eXperiment was con- ducted to examine the effects of a change in graduation requirements for students in a specific major. The final experiment was a com- posite of changes plus simulated salary increases for faculty members in the College of Engineering. The following conclusions were drawn as a result of the experiments. A reasonable confidence in the calculations performed by the computer was developed by manual calculations of anticipated changes. The user of this simulation program can then be reasonably GEORGE MERRILL VAN DUSEN sure that the calculations are accurate and express reasonable relationships. The specific model used in this study can be used to simulate enrollment projections, calculate appropriate demands and costs, and change selected parameters. As changes are intro- duced it is possible to trace some interrelationships of the results of the changes. The interaction of the variables made it possible to observe that when policy changes are made in isolation the re- sults of these changes affect the total operations of the University. To the extent that an accurate data base exists, the simulation program provides a tool for administrators in the College of Engi- neering to conduct a number of eXperiments concerning the present and future direction of the College. The development of management information systems to aid administrators can only be effective if accompanied by an organizational structure to insure communications in the system. This communication linkage must be recognized as an important mechanism for decision-making in order for the administrator to make maximum use of the analytical tools in carrying out manage- ment functions. APPLICATIONS OF A COMPUTER SIMULATION MODEL TO LOGISTICAL DECISIONS IN A UNIVERSITY BY George Merrill Van Dusen A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Administration and Higher Education 1969 Chapter II III IV TABLE OF CONTENTS THE PROBLEM AND DESIGN OF THE STUDY. . Purpose of the study. . . ............. The need and importance of the study ....... Operational definitions . . . . . . ..... Design of the study ................. The interview guide .......... Data collection procedures ......... Analyzing the data . . . . ...... REVIEW OF RELATED LITERATURE . . . Literature on the systems approach Literature on the use of systems analysis in approaching educational problems ....... DESCRIPTION OF THE MODEL ........ . Structure of the model .......... Resources and production. . . . . Implementations of the model. . . ..... INPUT FOR THE MODEL . ............ General observations . . . ............ Enrollments............. ..... .. Finances.. .......... ........ Academic program ................ ii Page 1 2 2 ll 12 14 14 15 16 17 22 31 32 39 41 45 46 49 56 60 TABLE OF CONTENTS (continued) Chapter Page Social concerns. ..... . ........... . 63 Physicalfacilities................. 65 Analysis of the interview input as related tothemodel..... ..... 67 The model as an aid to examining alternatives ..... . . . . . .......... 70 V ANALYSIS OF THE EXPERIMENTS ........ 73 Experiment I ............... . . . . . 77 Experiment II. ..... . . . . . ......... 88 EXperiments III and IV . .............. 91 ExPerimentV............. ..... .. 98 Experiment VI ................... 101 Experiment VII ........... . ....... 106 Summary of the experiments ............ 111 VI SUMMARY, CONCLUSIONS, AND THEORETICAL CONSIDERATIONS. . ........... . . . 112 SELECTED BIBLIOGRAPHY ................. 119 iii LIST OF TABLES Table Page 1 Computer Usage for Advanced Analysis ..... 7 2 Interview Responses of 13 Administrators According to Selected Categories . . . . . . . 49 3 Simulated Enrollment Increases 1967-1971, College of Engineering. . . . . ..... . . . 79 4 Simulated University Enrollment Increases 1967-1971 (Not Engineering). . . . ..... . 82 5 Student Credit Hour Increases 1967-1971 . . . . 84 6 Increases in Faculty and Graduate Assistant Requirements for Engineering Departments 1967-1970. ........ . . ......... 85 7 Increases in Total Costs - College of Engineer- ing 1967-1970 0 ccccc o o ...... o o o 87 8 Net Effects on Enrollment for Engineering and the Remainder of the University Resulting from Two Parameter Changes. . . . . . . . . 89 9 Differences in the Production of Student Credit Hours for Engineering and the Remainder of the University as a Result of Two Parameter Changes. 0 I O O . C O O C O O O O O O C O O O O 90 10 Differences in Faculty Increases and Graduate Assistants for Engineering as a Result of Two Parameter Changes. . . . . . . ..... 90 11 Total Enrollment Change for the College of Engineering and the Remainder of the Univer- sity Resulting from a Reduction in the Number ofNewFreshmen................ 93 iv Table 12 13 14 15 16 l7 l8 19 20 21 LIST OF TABLES (continued) Page Comparative Cost Reductions for the College of Engineering According to Two Percent and Four Percent Decreases in Freshmen Enrollments 1967-1970. . . . . ........ 94 Comparative Cost Reductions for the Mechani- cal Engineering Department According to Two Percent and Four Percent Decreases in Freshmen Enrollments 1967-1970 . ...... 97 Changes in the Production of Student Credit Hours in the Electrical Engineering Depart- ment Resulting from an Increase of Thirteen Credit Hours Per Major 1967-1970 . . . . . . 99 Changes in the Number of Faculty and Teach- ing Graduate Assistants in the Electrical Engineering Department Resulting from an Increase of Thirteen Credit Hours Per Majorl967-l970................ 100 Changes in Teaching Costs for the Electrical Engineering Department Resulting from an Increase of Thirteen Credit Hours Per Major. 100 Enrollment Changes for the College of Engi- neering Resulting from Two Parameter Changes 1967-1971. C O O O O O O O O O O I O O 103 Enrollment Changes for the Remainder of the University Resulting from Two Parameter Changesl967-1971............... 104 Net Effects on Student Credit Hours, Teaching Requirements, and Total Costs for the College of Engineering Resulting from Two Parameter Changes..................... 106 Enrollment Changes for the College of Engi- neering Resulting from Several Parameter Changes 1967-1971. 0 o o o o c o o c o o o o o 107 Net Efforts of Several Parameter Changes on Student Credit Hours for Engineering . . . 108 V LIST OF TABLES (continued) Table Page 22 Effects of Several Parameter Changes on Faculty and Graduate Assistant Require- ments for Engineering. . . . . . . . . . . . . 109 23 Effects of Several Parameter Changes on Costs for Engineering. . . . . . . . . . . . . 110 vi LIST OF FIGURES Figure Page 1 Flowchart of a Systems Approach. . . . . . 19 2 Structure of the System Model . . . . . . . 34 3 Sectors of the Model . . . . . . ...... 4O LIST OF APPENDIC ES Appendex Page A Interview Guide. . . . . . . ..... . . . . 122 B MSUSIMZ User's Manual. . . . . . . . . . . 123 viii [[[rl CHAPTER I THE PROBLEM AND DESIGN OF THE STUDY Analytical tools and techniques such as systems analysis, modeling, and simulation, generally deve10ped by analytically trained people ”outside” the area of educational administration, have been suggested in recent years for assisting educational administrators in decisions concerning the present and future direc- tion of colleges and universities. At the present time, there is little evidence that educational administrators are becoming parti- cularly adept at or evidencing thrust toward possible adaptations of these analytical techniques to educational problems. One attempt to apply systems analysis to educational pro- blems has been the work of a research group in the College of Engineering at Michigan State University. This research effort describes a university as a system and exemplifies one approach to mathematically depict the operations of a university. If an approach of this type is to be understood by educational admini- strators for assessment and possible implementation, there is a need for an explanation and demonstration of its potential. A means was sought in this study of demonstrating how a specific model might be employed to describe the operations of a university so that educational administrators can become aware of the potential of a systems approach as an aid in rational decision-making. The study encompasses the following questions: (1) What is the systems model developed by the Systems Science Group at Michigan State University? (2) What are some of the Specific strengths and limi- tations of this model? (3) For what might it be useful? (4) How might the model be used ? Purpose of the Study It was the purpose of this study (1) to describe in non- technical language a systems model and an implementation of the model using data descriptive of Michigan State University as devel- oped by the Systems Science Group in the College of Engineering; (2) to identify aims, objectives, and problems concerning the future direction of Michigan State University as suggested by a selected group of educational administrators responsible for policy decisions in this institution; and (3) to show how some of these identified con- cerns and projects are addressable to a specific computer simulation program which is an implementation of the theoretical model des- cribing the university as a system. The Need and Importance of the Stud)! The need and importance of the study largely stems from three sources: (1) the expressed concern by educational leaders for decisions regarding educational planning to be based on factual data; (2) the expressed concern by educational leaders for admini- strators to become aware of analytical tools as aids in educational planning; (3) the current commentary which suggests the potential of systems analysis for approaching educational problems. The literature related to educational administration has long emphasized the need for careful planning in decision-making. Dodds, commenting about the university president's role in the decision- making process, presents a summary of the need for facts in decision—making: Any administrative action is based on a com- bination of established facts and conjecture. The circumstances leading up to a decision may be well established or known only in outline; it consequently may be relatively certain or almost totally obscure. Yet decisions must continually be made; the refusal to decide between alternatives constitutes a decision in itself . . . . This chronic predicament requires a degree of philosophical resignation on the part of the president, to be sure, but it also calls for efforts to enlarge the sc0pe of the known and reduce the sc0pe of the unknown. The more facts the better the hunches. Moore suggests there is a need to delineate between the administrative and leadership functions performed by the university president. He reasons that both functions cannot be conducted by the same man and one of the basic distinctions relates to planning. 1Harold W. Dodds, The Academic President--Educator or gretaker? (New York: McGraw-Hill Book Company, Inc., 1962), p. 177. The administrator, in dealing with the here and now of the institution, must be astute at effect- ing the best collage of data suggesting priority and direction. The descriptive data developed by the administrator provide the reference points for the leader as he attempts to predict trends and visualize what the future holds. The leader uses data as might an historiographer--his real world is the future. Francis A. Horn strongly emphasizes the importance of decisions involving the present and future direction of educational institutions being based on relevant and factual data when he states: . . . to continue to operate colleges and universities in the future as they have been operated in the past, to go on making decisions by such unscientific and ad hoc means as have prevailed, can only lead to the failure of higher education to meet the challenge and Opportunities ahead, if not, indeed to down right disaster. Educational decision makers face a difficult task in the formulation of policies to provide direction and leadership for colleges and universities. There is evidence in the literature which suggests that educational administrators should become familiar with tools and techniques which might aid in this endeavor. 4 . . Mauch was one of the early educational writers to urge educators to consider a systems analysis approach. He suggested Samual Moore, ”Leaders are Leavers, ” The Journal of icneral Education, XX, No. 4 (January, 1969), p. 293. Francis H. Horn, "A University President Looks at Insti- tutional Research, " The Role of Institutional Research in Planning (Madison, Wisc.: The University of Wisconsin Press, 1963), p. 4. 4(James Mauch, "A Systems Analysis Approach to Education, " Phi Delta Kailan XLIII, No. 5 (January, 1962-), pp. 153-161. that it offers an opportunity to consider alternatives in resource allocation and planning. Meals emphasizes the strengths, limi- tations and potential for systems analysis in examining educational planning and suggests a role for the administrator as follows: Systems analysis will not cure all the real and pre- sumed ills of education. It will not alone eliminate a single evil or replace one traditionally trained administrator. Moreover, such benefits as may be derived from a systems approach will occur gradually as educators adapt some new attitudes and adopt a few new tools. Bern presents a rationale for deveIOping ”educational engineers" and offers a two-step program to accomplish this task: 1. Consultation with and study of educational institutions such as M. I. T., military organizations such as the U.S. Naval Training Devices Center, research and development centers such as System Development Corporation, and education and train- ing research laboratories of industrial organizations such as the Hughes Aircraft Company. In effect this would be a survey and analysis of areas where considerable cross—fertilization of education and engineering has already taken place and where it is therefore likely that the seeds of educational engi- neering of the future are germinating. 2. The development and institution of courses and a curriculum leading to a professional degree in educational engineering. Judy and Levine suggest that educators need to have analytical tech- niques to work with: "In the administration of scarce resources, 5Donald W. Meals, "Heuristic Models for Systems Planning, " Iii Delta Kappan, XLVIII, No. 5 (January, 1967), 203. 6H. A. Bern, "Wanted: Educational Engineers, " Phi Delta Kappan, XLVIII, No. 5 (January, 1967). p. 235. university officials deserve managerial tools as powerful and sophi- sticated as those available to managers in business and government. The necessity of reaping the maximum return from our educational investment is no less. "7 In 1966 Rourke and Brooks completed a study of computer usage in colleges and universities in the United States. They suggest that the applications of operations research and systems analysis to educational problems were in part reSponsible for the conclusion that there is a ”managerial revolution in higher education. "8 This same survey also provides some background on the use of computers in simulation and related activities at the time of the survey. Rourke and Brooks asked, "Are you now using or do you plan in the near future to use computers for the simulation of campus operations, heuristic problem solving, or other forms of advanced computer analysis ?” The response to this question is shown in Table 1. Because systems analysis has been developed primarily for use in business, government and the military to analyze resource management, there are some educators who dismiss these techniques 7Richard W. Judy and Jack B. Levine, A New Tool For EducationaLAdministrators (Toronto: University of Toronto Press, 1966), p. vii. 8Francis E. Rourke and Glenn E. Brooks, The Managerial Revolution in Higher Education (Baltimore, Md.: Johns Hopkins Press, 1966), p, vi. TABLE I COMPUTER USAGE FOR ADVANCED ANALYSIS9 Response Number Percent Not at present 53 37 Plans in progress 11 Yes 2 1 No response 77 54 TOTAL 143 100 as having no value for education. Millett eXpresses this point of view, ". . . ideas drawn from business and public administration have only a very limited applicability to colleges and universities. ”1 The differences in functions performed by education as compared to other institutions is the argument usually presented by critics such as Millet. Dill, in discussing Litchfield's global theory of admini- stration, also supports Millett's position: Despite Litchfield‘s arguments that much of a science of administration will be applicable to all kinds of or gani— zations, laboratory groups, business firms, and govern— ment administrative agencies differ in imgolrtant reSpects from schools, colleges, and universities. 9Ibid. 143. 0 1 John D. Millett, The Academic Community: An Essay on Organization (New York: McGraw-Hill Book Company, Inc., 1962), p. 4. 11William R. Dill, "Decision-Making, " Behavioral Science El Educational Administration, ,Sixty-third Yearbook of the National Society for the Study of Education, Part II (Chicago: University of Chicago Press, 1964), p. 205. An example of the counter argument is offered by Corson. He states, ”the assumption that the university is different and not subject to assistance from the considered experience of other institutions seems to be the crucial barrier to imaginative development of new and im- H12 13 . . proved means of governance. Caffrey and Mosmann indicate that most American colleges and universities lag behind government and industry in using systems techniques for two reasons: (1) A refusal to face the problem; and (2) a refusal to pay for its solution. The importance of this study is also suggested by the grow- ing interest for informing educators about analytical tools and tech- niques. A Symposium sponsored by the U. S. Office of Education was held in November, 1967 to bring together the analysts and the edu- cational administrators to discuss the implications and adaptations of systems analysis to educational problems. David S. Stoller identified the purpose of the meeting in his opening remarks: The symposium is set up to accommodate as wide a range as possible--from the educator interested to learn what is happening in operations analysis (and who may have little mathematical background) to the sophisticated model builder engaged in modeling the entire educational system of a country--and even he may find4there are developments he hadn't heard of before. 2 1 John J. Corson, Governance of College and Universities (New York: McGraw-Hill Book Company, Inc., 1960), p. 200. 13John Caffrey and Charles J. Mosmann, Computers On Campus (Washington, D. C.: American Council on Education, 1967), p. 38. 4 1 David S. Stoller, "Symposium Theme: Operations Analysis of Education, ” (Opening speech at the Simposium on 03erations Analysis of Education sponsored by the U.S. Department of Health, Education and Welfare, Washington, D. 0., November 19, 1967), p. 1. The discussions at the conference ranged through such topics as site locations of urban schools, bussing schedules, measuring stu- dent achievement, and modeling of universities. Chauncey in the foreword to Pfiffer's report states his assessment for the conference as follows: One of the most valuable outcomes of these sessions was the dispelling of some myths about the computer as a control instrument over individuals and over the educational process . . . . It must be said that a system does not of and by itself, produce better edu- cation. It should, however, if used seriously, present educators with the opportunity to face up more exactly to what they want to achieve, a program of how they hope to go about it, and the copgage to assess honestly the outcomes of their actions. The Ford Foundation has taken an active role in the finan- cial support for implementation of systems analysis projects at educational institutions. In April, 1968 grants totaling $2 million were received by Stanford University, The University of California at Berkeley, Princeton University, and The University of Toronto to support the development and testing of new management tech- niques in the solutions of problems of higher education.1 Generally, these programs are applications of systems analysis techniques to university problems. A potential aid for assisting the educational administrator in decision-making has grown out of the work of the Systems Science 15.l'ohn Pfeiffer, New Look at Education, with a Foreword by Henry Chauncey (New York: Odyssey Press, 1968), p. VIII. 16American Council on Education, Higher Education and National Affairs, XVIL No. 14 (April 19, 1968), p. 2. 10 Group in the College of Engineering at Michigan State University. In 1964 this groxp initiated a research project sponsored by the Economic Manpower Commission entitled "A Systems Approach to Higher Education. ” A progress report prepared in September, 1967, summarizes the original intent of the project as follows: . . . to determine, first, whether it was possible to develop a valid mathematical description, or systems model, of the university, and, second, whether usable and effective information processing programs based on this model could be implemented on a computer to answer important questions concerning allocation policies. The theoretical structure of such a model was developed in the early stages of the project and refined more recently. With the cooperation of the Office of Institutional Research considerable attention has already been devoted to the problemgpf providing an adequate data base from which to work. Contained in the same report is the challenge to educational admini- strators at Michigan State University which largely provides the impetus for the study undertaken: The project now stands on the threshold of practi- cal application of the system method in decision-making. The last steps towards the use of the model cannot be taken by system specialists and computer scientists alone. It has become increasingly important for the university administration to become acquainted with the objectives of the project, its potentials and limitations and to pigvide suggestions for the direction of future efforts. 17Herman E. Koenig, Martin G. Keeney, and Rita Zemach, Systems Analysis and Planning in University Administration (East Lansing, Mich.: Division of Engineering Research, Michigan State University, 1967), p. 4. 18Ibid . 11 The need for factual information as an aid for educational administrators has been suggested. Systems analysis and related techniques have been identified by some educators as offering pos- sible tools to aid in the decision-making process. The use and sharing of these analytical techniques is in the beginning stages. A specific project using systems analysis at Michigan State Univer- sity has been offered as a possible aid for administrators. Operational Definitions Components are the parts of a system. Mathematical models are constructed for each component of the system and the inter- connection pattern among components. Each component defines a specific operation or function of the university. The components of the university in this study are labeled sectors. A Model is a mathematical description of the system. Thus, the model in this study consists of a mathematical description of the identifiable interrelated sectors within the system. It is merely one conceptualization of how the components of the university are inte r r elated . A Parameter is avariable whose assigned value is changed to reflect different conditions of the system. Enrollment, Faculty Salary Scale, Cost of Supplies and Services are merely three examples of numerous parameters which can be manipulated in the simulation program used in this study. 12 Simulation is the process of manipulating the parameters of the model and noting the resulting condition of the system as described by the model. The MSUSIMZProgram is one implemen- tation of the model which allows the user to experiment with the system by assigning values to certain parameters. The process which enables the user to manipulate variables in the model is known as simulation. The mechanical means of accomplishing the manipulation is a computer program. A System is a collection of interacting identifiable parts. A university is viewed as a system in this study. Systems Analysis is a technique for mathematically identi- fying, representing, and studying the interrelationships of the parts which comprise the system. Thus, the tool employed by the Specialist in this study to examine the structure of the university. Design of the Study Prior to the design and development of the study it was necessary to describe, in nontechnical language, the systems model, and the implementation of the model, developed by the Systems Science Group in the College of Engineering at Michigan State Univer- sity. This was accomplished by reading, and discussions with participants in the research project and was essential in order to determine whether or not it was possible to gain the appropriate level of understanding necessary to undertake the study. Hare and Chorafas offered valuable sources in order to obtain a historical l3 perspective and understanding of systems analysis. 19 Kivat, Evans, 3231., Pfeiffer, and several journal articles served as basic sources for examining the application of systems analysis to engineering, physics, and socio-economic problems. The re- ports by Koenig, 3:31., served as the sources for understanding the Specific model used in this study. (The discussion of the model is presented in Chapter III. ) In order to identify aims, objectives, and problems con- cerning the future direction of Michigan State University, interviews were conducted with thirteen administrators who generally parti- cipate in long range planning as a normal part of their administrative responsibilities. Included in the interview group was one depart- ment chairman, one dean, and eleven administrators generally identified as members of the ”central administration. " Individuals from the "central administrative” group were selected from the Offices of the President, Provost, Secretary, and Registrar, plus representatives from the Graduate Office, Business Office, Insti- tutional Research, and Admissions and Scholarships. No teaching faculty members, students, or members of the Board of Trustees were included in the interview group even though their role in the future development of the University is recognized. It was reasoned 19These and all references which served as background information are identified in the bibliograph. 14 that the members of the interview group were more concerned with the management and operational aspects of the University as compared to the groups that were excluded. The Interview Guide A private, one-two hour, in-depth interview was the data gathering technique employed in the study. In deve10ping the inter- view guide, primary recognition was given to the central purpose of the investigation; to discover the long range plans for Michigan State University. A semi-standardized interview guide was con- structed to help answer this and related questions. The interview guide contained a minimum of structure because each individual included in the study represented a unique administrative unit in the University. Appendix A contains a copy of the interview guide. As a result of a pretest with three administrators, not in- cluded in the study, a number of minor revisions were made in the guide. This eXperience also suggested more effective approaches to be used by the interviewer. Data Collection Procedure 5 The interviews began June 18, and were completed Septem- ber 20, 1968. A tape recorder was used for two interviews, but was discontinued because its use tended to inhibit the flow of infor- mation. During the remaining eleven interviews careful notes were taken and an immediate reconstruction of the comments was written. 15 Analyzing the Data The data gathered in the interviews were classified accord- ing to categories identified by the thirteen administrators. Five categories were identified which included: (1) enrollments, (2) finances, (3) academic programs, (4) social concerns, (5) physical facilities. The information gathered from the inter- views was analyzed as follows: 1. The responses which fell within the framework of the model, but could not be directly applied to the simulation program, i. e., social concerns quality of programs. . The re3ponses which were found to be directly addressable to the computer simulation program. The responses which were found to be addressable to the com- puter simulation program, served as the basis for the design of seven experiments to demonstrate the possible use of the model in approaching educational problems. The experiments were carried out using a simulation pro- gram identified as MSUSIMZ and run on a Control Data 3600 Com- puter. The computer carried out the basic calculations according to the instructions of the program developed by the user. The analysis of the experiments was conducted to show the effects of the parameter changes within the simulation program. Graphs and tables were prepared to assist in the display of the changes and interactions which occurred. CHAPTER II REVIEW OF RELATED LITERATURE A review of related research was conducted to identify and examine studies where systems analysis has been used as a tool to approach educational problems. A limited number of investigators have used this technique for analysis of entire school systems and universities or recognized parts of these organizations. Of primary concern were those studies involved with colleges and universities, but selected studies pertaining to other school systems were also included. The review of literature also revealed a number of studies where the term "systems" was used to describe a particular research activity, but not in the same sense as in this study. Therefore, a general criterion was developed to determine the studies that were judged as relevant to this project. The section which follows is not intended to be a comprehensive review of the ”systems approach, " but a framework or strategy for identifying the types of educational studies included. in the review of literature. 16 17 Literature on the Systems Approach A systems approach is not new and there is evidence which suggests that a great deal of modern systems theory has been borrowed from the past. Blaschke reports, "We have returned to the use of the scientific approach to the method develOped in the days of the Greeks, refined the techniques of implementation where possible, and in twentieth-century style christened it, sometimes with too much glory, 'the systems approachfl'”l A similar conclusion has been reached by Hare; ”The system concept is as modern as ancient Egypt, where a crude form of today's system theory played a role in the construction operation of the pyramids. " Numerous strategies have evolved for systems analysis, but there is evidence which suggests there is no single approach. Pfeiffer offers the following: . . . there is no such thing as the systems approach, if that implies the existence of a formula or a special set of rules for handling problems. A wide range of procedures are available, and which turns out to be the most helpful depgnds on the nature of the problem under investigation. After listing seven constituents in the study of a system, Evans and others, suggest, "Naturally all studies need not conform to 1Charles L. Blaschke, "The DOD: Catalyst in Educational Technology, ” Phi Delta Kappan, XLVIII No. 5 (January, 1967), p. 211, 2Van Court Hare, Systems Analysis: A Diagnostic Approach (New York: Harcourt, Brace and World, 1967), p. 22, 3Pfeiffer, New Look at Education. 9. 12. 18 this organization . . . Thus, it sometimes is hard to say whether a given constituent of study plays the role of a system, a model, a method of solution, or a solution. Recognizing that there is no single systems approach and the type of problem determines what will be included in the study, it is nevertheless important to identify a generalized procedure. Figure 1 presents a flowchart which identifies a set of procedures generally involved in a systems approach. The flowchart emphasizes a num- ber of important principles, according to Pfeiffer: 1. Identification of the boundaries of a problem. (Define the problem. ) 2. Specification of the subfunctions and alternatives in relationship to the system. 3. The use of a model to clarify and to yield infor- mation. 4. Identification of the systems approach as a cyclical and continuing process. The application of the technique to educational problems is outlined by Blaschke as follows: 4George W. Evans, Graham F. Wallace, and Georgia L. Sutherland, Simulation UsingDigital Computers (Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1967), p. 3. 5Pfeiffer, New Look at Education, pp. 21-32. l9 Define the problem 1) define objectives 2) measures of effectiveness 3) constraints, uncontrol- lable variables 4) controllable variables Define subfunctions Develop model Define alternatives for each subfunction Collect data 1 Synthesize subsystems Evaluate Select Fig. l -- Flowchart of a Systems Approach6 61bid., 32, 20 Its (systems approach) significance to education is that it forces the individual manager to define the problem precisely, note the alternatives available and their total costs, and choose the most efficient alternative according to performance criteria. To- day its merit lies in its conceptual approach; for the future, the need to refine implementing techniques depends on our ability to define our objectives clearly, delineate our problems accurately, and develop criteria for measuring how much success we can get for how many dollars. A general systems approach for examining educational pro- blems or other types of problems has been identified. The tools and techniques involved in implementing this systems approach is the most basic difference between research efforts related to this study and those that are not. The model developed by the research group and used in this study is a mathematical model. Chorafas states that mathematical models ”describe the equilibrium conditions among significant system variables. " He further suggests that "they can be either static or dynamic. The variables themselves can be deterministic or probabilistic. Their choice and the establishment of the logical structure are of capital importance. " Thus, as indicated in the flowchart, a model is meant to clarify, and to yield information. The research efforts included in this review describe studies of educational problems where the relationship of the components are expres s ed mathematically. 7 Blaschke, ”D. O. D.: Catalyst in Education, " p. 21 l. 8 . . . . Dimitris N. Chorafas, Systems and Simulation (New York: Academic Press, 1965), p. 21. 9Ibid . 21 Most of the studies included in the review of literature also involved computer simulation. "Computer simulation provides a means for studying systems. It can be applied to a wide variety of systems, both real and hypothetical, and it can be employed for many different purposes. " The following purposes are recog- nized by Chorafas: (1) for purposes of experimentation or evaluation; (2) as a means of learning about new systems in order to redesign or refine them; (3) as a tool in familiarizing personnel with a system or a situation which may, as yet, not exist in real life; (4) for the verification or demonstration of a new idea, system, or approach; and as a means for projecting into the future and thus pro- viding quantitative bases for planning and forecasting. (5) The research efforts listed in this review of literature tend to cut across these purposes because it is inherent in the design of the simulation model. Kiviat suggests the following reasons: Before a simulation model is designed, two important questions must be asked and answered: (1) What use will be made of the mode1?(what questions will be asked); and (2) What are the requirements of acCuracy and pre- cision? Answers to these questions determine the structure of a model, as they demand that certain assumptions be made, that certain boundaries be im- posed and reapected, that certain types of questions can and cannot be asked, that certain territories cannot be explpged, and that certain realities cannot be pre- dicted. 10 Evans, 2. _a_l_., Simulation Using Digital Computers, p. 13. l Chorafas, Systems and Simulation, p. 17. Modeling Con.- 12 P. J. Kiviat, Digital Computer Simulation: The Rand <:epts, Memorandum RM-5378-PR (Santa Monica, Calif.: Corporation, August, 1967), p. 14. 22 Criteria for the inclusion of investigations related to this study have been derived. Within a generalized systems approach, those studies judged to be relevant contain elements of mathematical modeling and, where appropriate, computer simu- lation is used to study educational problems. A discussion of these studies follows . Literature on the Use of Systems Analysis in Approaching Educational Problems The Systems Development Corporation has been active in the production of research related to instructional systems for 3 secondary schools. Pfeifferl credits Cogswell and associates with the first modeling effort (1963) using systems analysis and computer simulation to assess the effects of educational innovations. Cogswell summarizes the intent of the project, funded in part by the U.S. Office of Education, as follows: The purpose of the research is to find new solutions for implementing instructional media through analysis and simulation of school organization . . . . The SDC project, which studies the use of systems analysis and computer simulation in education, should yield techniques and provide design recommendations that are more carefully conceived, that involve more pervasive and integrated changes throughout the schools, and that employ instructional media more le‘ffectively than do current methods of school design. 13 . . Pfeiffer, New Look at Education, p. 119. 14 . John F. Cogswell, "Systems Technology in Education, " Man Machines Systems in Education, ed. by J. W. Loughary (New York and London: Harper 8: Row, 1966), p. 46. 23 Cogswell outlines the procedures in the project as follows: (1) survey and selection of high schools; (2) systems analysis of five high schools selected for study; (3) construction of a computer-simulation vehicle that will provide the capability of building detailed dynamic models of the schools and of hypothetical changes in the schools; and (4) simulation and study of tgxe five high schools with . . . l the Simulation vehicle. After the completion of steps (1) and (2), listed above, Cogswell developed a simulation model which described a high school in terms of instructional activities, student characteristics, and selected school activities. Yett made a major contribution to the project by expanding Cogswell's model to include the allocation of resources. The addition of the resource allocation processor to the previously developed simulation vehicle provided for the logical flow and the capacity for control of resources, persons, places, and things by analyzing the terminations, continuation, and activation of activities according to the logical demands of the simulation vehicle and the curregt exPression of the systems resource capabilities. The application of this technique was used to integrate real and simulated data for courses, counseling, the academic progress of individual students, and the exploration of a possible instructional management information system for the five schools that partici- pated in the project. 6 Frank A. Yett, Resource Allocation Processor For the Sihool Simulation Vehicle--Pilot Version (Santa Monica, Calif .: SYstems Deve10pment Corporation, 1964), p. 30. 24 The final report of this project contained the following recommendations: 1. The continued development of the computer-based system to assist students and counselors in planning; 2. The continued study of the use of information pro- ces sing for student instruction; and 3. The development of procedures for the management of change in schools. Reports of recent projects by the Systems Development Corporation indicate that present efforts are basically a continua- tion and extension of applications of systems analysis and computer simulation to individualized instruction at the elementary and secondary school levels, computerized instruction, and computer- ized counseling. Systems analysis and computer simulation have also been used by Clark and others to model urban education. The analytical model is an aggregation of sub-models which is intended to aid in decision—making regarding school location, enrollment, facilities, organization, programs, and costs. The model has the capability to handle the introduction of known data such as available money, staff allocation, and the present school plant. 17John F. Cogswell, New Solutions to Implementing Instruc- tional Media Through Analysis and Simulation of School Organization-- Final Report (Santa Monica, Calif.: Systems Development Corpora- tion, 1966), p. 51. 25 The aggregated model consists of the following sub-models: 18 1. An urban sub-model which combines pupil population, location, transportation, needs, and socio-economic characteristics of the community. 2. A school sub-model which describes the school program, site specifications, and Space and equipment provisions per pupil by instructional area. 3. A cost sub-model which helps to estimate the total per pupil expenditures for remodeling existing facilities compared to new site and construction proposals. The aggregated model then evaluates benefits and costs per pupil in relation to educational objectives. Attempts to implerrent the theoretical model have been limited. "The major constraint, " according to O'Brien, ”has been the limited data which are available for the estimation of parameters. ”19 This recognizes one of the basic necessities of implementing a theoretical model, namely, an accurate and available data base. Another general model is the effort by Reisman to develop a mathematical model to describe the flow of students in and out of a university system and to follow the progress of students through the system. The relationships of students to the rest of the system are expressed by differential equations. 18U.S., Department of Health, Education, and Welfare, Division of Operations Analysis for Educational Statistics, Urban Education Systems Analysis, by Stephen C. Clark, Richard J. O'Brien, and C. Marston Case, Technical Note No. 24 (Washington, D. C.: Government Printing Office, 1967) p. 10. 19U. S., Department of Health, Education, and Welfare, Division of Operations Analysis for Educational Statistics, Cost Model for Large Urban Schools, by Richard J. O'Brien, Technical Note No. 30 (Washington, D.C.: Government Printing Office, 1967), p. 13. 26 This model breaks the educational sector up into four segments: undergraduate programs, master's pro- grams, doctoral programs, and post-doctoral programs. It breaks the other sectors of society employing college- university trained people into segments according to the highest degree earned by those within the segment. Reisman does not attempt to offer any implementation of the model; instead, he recognizes the generality of his model as follows: Thus, we are left with a decision that most systems analysts of socio-economic systems sooner or later must face. The decision is two-fold. First, it is concerned with what portion of the universe one ought to subject to study--that is, where he should place his system boundaries. Second, it is concerned with the level of aggregation to be used both within the system's boundaries and within that portion of the remaining universe with which the system communicates. The investigations by Cogswell, Clark, and Reisman pro- vide evidence of efforts to model large scale educational problems; other researchers have examined more specific problems. Belin- . Z . . . . ski used systems analySis to describe the relationship between characteristics and cost elements in the procurement, installation, and operation of educational media and technology. He first id enti— fied the general system, detailed the physical and operational characteristics of the system, developed the mathematical models to describe (the relationship between characteristics and costs, and 2 0Arnold Reisman, ”A Population Flow Feedback Model, " Science, 153 (July 1, 1966), p. 89. 2 llbid., 91. 22 John Belinski, A Cost Study of Educational Media Systems (Washington, D.C.: General Learning Corporation, 1968). 27 built an aggregate model of the sub-models. The General Learn- ing Corporation has implemented the results of this effort to assist schools in determining cost alternatives in purchasing instructional equipment aids. Bowman has indicated that Yale University is in the process of developing a model of the operating growth budget which projects what “the University fiscal flows and structure might look like under varying conditions over an extended period, e. g., 20 years. " Bowman characterized the progress of this effort as follows: We have run many simulations, adjusted the program, refined the parameter estimates, and modified the questions we have been asking. . . . . Our work with the operating growth budget has already started to influence some of the decisions of the University including the yearly operating budget, the capital funds program, and the endowment in- vestment portfolio. 2' Another research effort on a specific problem was carried out by Yurkovich. He developed a computerized methodology for determining the physical facility requirements of a large university and implemented the model by using data collected for the University of Wisconsin. The results of the study indicate he was able to con- duct room utilization studies, project enrollments based on fixed or exPanded space needs, project staff needs, and project future 2 3Edward H. Bowman, "A Budget Model of a University, " S_ymposium of Operations Analysis of Education (Washington, D. C.: November 19-22, 1967), p. 7. Z 41bid. , 8. 28 physical facility requirements. ”Each element, including the space classifier, the perpetual Space inventory, the enrollment projector, and the staff projector, is an independent system. The integration of these elements allows for the projection of future space needs. ”25 Few investigators have used systems analysis to produce a global model of a system. The development of a descriptive model of this type forfeits detail, but enables the educator to observe the interrelationship of several subsystems. At the present stage of development, the work of Keller, Judy, and Koenig represents the major activity in this area at the college or university level. It is perhaps misleading to suggest that these efforts are the work of one man, for each activity represents an extensive project with sizeable financial and manpower efforts. Keller26 indicated in November, 1967 that during the years 1965-1967 cost simulation models, physical plant utilization models, capital outlay models, scholarship aid models, and models of the demand for higher education, had been developed at Berkeley. Since that time real data has been implemented to examine a number of real problems. Most of these studies have been internal cost 2 5John V. Yurkovich. A Methodology For Determiniii Future Physical Facilities Requirements (Madison, Wisc.: Univer- sity of Wisconsin Press, 1966), p. 154. 2 6John E. Keller, ”The Use of Models in University Decision- Making, “ Symposium on Operations Analysis of Education (Washington, D.C.: November 19-22, 1967), p. 7. 29 studies based on the financial problems facing higher education in California. In April, 1968 the University of California at Berkeley received a $500, 000 grant from the Ford Foundation to continue its efforts to apply systems analysis techniques to univer- sity problems. "The major use of this grant will be to develop planning models to deal with academic, economic, and physical . . 37 factors in relation to costs. ” Judy and Levine at the University of Toronto have made an effort to communicate their modeling efforts of the entire Univer- sity to educational administrators. A simulation model called C A M P U S has been built for the Faculty of Arts and Sciences to represent the implications of resource allocations as related to enrollments, resource demands, Space requirements, and budgetary calculations. Judy and Levine offer the following summary of the capability of the model: The model simulates university operations over a time period of any length. Loaded into the computer, the model accepts descriptions of the university's structure and statements of the levels of activities that the university is expected to perform. With these inputs, the model computes the resulting re- source requirements of staff, space, materials, and money. These requirements are displayed by several computer-prepared reports and graphs. The work at Toronto is no longer a pure research effort because it has been implemented as a normal part of the operations of the 2 7John Keller, A personal letter. 2 .. 8Judy, A New Tool, p. v11. 30 University through the Office of Institutional Research. At the pre- sent time the University is constructing a systems simulation model for the entire University. Part of the reason for the success in implementing the effort has been the ability of the Office of Insti- tutional Research to communicate with members of the faculty. According to Pfeiffer, ”Hansen (director of the Office) has made a special effort to speak in uncluttered English, a sound and strategi- cal policy judging by certain unfortunate eXperiences elsewhere. "29 A review of related literature located studies employing systems analysis to model global systems and specific components of schools and universities. The major research contributions in these areas were conducted by a few individuals. The results of the research efforts suggested that the use of systems analysis is in the early stage of development. Researchers who have intro- duced real data have produced some tangible results. The work at the University of Toronto was suggested as in the most advanced stage of development toward implementation. As stated previously, the model to be used in this study is a description of the university as a system by Koenig and others at Michigan State University. A detailed description of the model in non-technical language is presented in Chapter III. 2 9Pfeiffer, New Look at Education, p. 109. CHAPTER III DESCRIPTION OF THE MODEL The discussion in this section is limited to a description, in non-technical language, of a Specific modeling effort, which depicts a university as a system. It is important to emphasize that the model presented in this Chapter represents the thinking of one group as deve10ped by a series of logical steps according to the following considerations. 1. The purpose of the model. 2. The amount of detail incorporated into the model. 3. The assumptions required within the system. 4. The availability of necessary data. Within this general framework any number of models might be developed to describe the activities of a university, depending upon the Specifications of the model builders. Therefore, the model under consideration in this study should not be misinterpreted to be £12 model, but a single effort designed according to the purposes and objectives of a specific group. 31 32 The purpose of the model employed in this study is: . . . . to describe quantitatively the way in which univer- sity administrators collectively allocate resources in an effort to meet the demands placed upon them by a con- stantly changing student body, and to provide a tool for experimenting with alternative allocation policies in the face of these changes. Given this basic purpose, the model builders at Michigan State University progressed through a series of stages, outlined pre- viously, until the details of a Specific model were developed. The discussion which follows, of the structure and implementations of the model offers the necessary detail to describe the university as a system. Structure of the Model The total university is viewed as a system which contains a number of identifiable interrelated components called sectors. The Operation of the university is described in terms of the inter- relationships between and within these sectors as the university uses its resources in production. "The resource of the university are described, broadly, as personnel, Space, and equipment; its 'products' are regarded as educated manpower, research, and 2 public or technical services. " Students, faculty, office and 1H. E. Koenig 3311., A Systems Approach to Higher Edu- cation--Interim Report No. 3, Project C-396, National Science Foundation (East Lansing, Mich.: Division of Engineering Research, Michigan State University, 1966), p. 2. 2 Rita Zemach, A State-Space Model for Resource Allocation in Iiigher Education (East Lansing, Michigan: Division of Engi- neering Research, Michigan State University, 1967), p. 2. 33 maintenance staff are examples of personnel resources; classrooms, office Space, and residence halls are examples of space resources; computers, audio-visual aids, maintenance supplies, and motor pools exemplify equipment resources. The resources at work result in the production of educated students, internal and external ser- vices, and research conducted by members of the university. (Examples of internal services are data processing, faculty effort, counseling, and medical services; continuing education, consulting, and extension services are examples of external services.) The interactions between resources and production in the sectors and throughout the model are expressed by sets of equations. These relationships enable unit costs, units of effort, or other appro- priate units to be identified and associated with the sector. In this specific instance, the entire system is categorized into the following six components: 1. Administrative Control Sector .2. Personnel Sector 3. Physical Facilities Sector 4. Non-Academic Production Sector 5. Academic Production Sector 6. Student Sector All sectors contain identifiable interrelated parts and therefore qualify as sub-systems of the broader university system. The relationships of all sectors except the administrative control 34 it 51) s E 5 5 E 'a i administrative soetrel II \ [’1’ If“ :\ \\ ’ [’II'O' \ \ Personnel ,v hr—Wmmr roams») Sectors 44 :._—\_I Seetors I 'COI'U'b. \ k 7 V I) g: E I ‘ g 2 3 I, ’0” \‘ E .. 5 ' 3 E see-osodenis eseoenis 00M“ 3 i oroeesflse production serviees '- 5 sestor seoter . ._ A ”use! Sector \ developed stueers ()( 1" . . . ()J ”issuer ._..—. vesror "one of eeople ens servises ens essesletee mooted vetoes oer salt -->-- esnielstretlve eolloy oeetrels e IMerteses em renelelee seeie-eeeeeals presses lternieetel C) oeeeletlee groups see their mooted veleee (Interest stores) Figure 2. Structure of the System Model Source: H. E. Koenig, M. G. Keeney, and R. Zemach, Systems Analysis and Planning in University Administration (East Lansing, Michigan: Division of Engineering Research, Mini-vicar: State Universitv. 1967). Figure 2. p. 23. 35 sector, are expressed quantitatively. This means that the remain- ing five sectors are modeled independently and brought together through the descriptions of their interrelationships while the user functions as the administrative control as experiments are conducted. Figure 2 shows some elements of the interrelationships between the six sectors as they interact to form the university system. Student Sector The student sector forms the base of the schematic diagram, for it produces the internal demands of the system. This simply means that without students there would be no need for the other sectors which make up the system. The student sector takes in students from outside the system, uses the academic and non-academic production sectors and pro- duces developed manpower. Developed manpower includes all students who leave the university whether or not a degree has been completed. The production of developed manpower is a function of the entire system and results from the complex interrelationships of all the identified sectors. Within the student sector a description of the distribution of all students is developed according to major fields and class levels. The description identifies factors which influence student enrollments such as the number of students by class and level the previous year, the major choices of new students, the availability 36 of scholarships, graduate assistantships, housing, and other factors which may be identified as attracting students to the university. Student demand creates a pattern for courses in all fields and levels in addition to teaching and research associated with dissertation production. Demand from this sector is also evidence in the non- academic production services such as housing, counseling and medical services. Academic Production Sector The academic production sector consists of the relationship between the production of academic services and the faculty and graduate assistant effort, plus the environmental facilities required to produce such services. This sector therefore takes in resources from the personnel sector and physical facilities sector and produces academic services as a result of the demand fromtwo sources; 1. The student sector which creates the demand for the production of credit hours, academic advising, dissertation direction, and other related academic services. 2. The demand from outside the university system which comes from such sources as sponsored research, adult education, and other community needs. In order to meet the demand for academic services the effort of the faculty or graduate assistants is not enough; environmental facilities such as classrooms, instructional equipment, and library facilities are also needed. 37 Non-Academic Production Sector The non-academic production sector takes in staff and faculty from the personnel sector plus additional resources from the physi- cal facilities sector and produces non-academic services. The production of non-academic services such as registration, housing, health services, counseling, and placement result from the demand created by the student sector. The relationships be- tween the resources and the production within this sector are specified in terms of the efforts, facilities, and costs needed to meet the demands of the student sector. Personnel Sector Resources, in the form of faculty and staff manpower are received from outside the system, supplemented by student labor (including graduate assistants) from the student sector within the system. A third resource for the personnel sector consists of the environmental facilities,provided in turn by the physical facilities sector, and which are needed to support the activities of the per- sonnel sector itself. The resources of the personnel sector support the produc- tion sectors as well as the physical facilities sector by providing faculty, office and maintenance staff. The demand for the personnel sector is established by the academic and non-academic production sectors, administrative control sector, and the physical facilities sector. This sector therefore produces faculty and graduate 38 assistant effort for the academic production sector; administrative effort for the administrative control sector; maintenance and opera- tion effort for the physical facilities sector; and student non-academic services effort for the non-academic production sector. Phy sical Facilities Sector The resources of the physical facilities sector are received from outside the system in the form of space and equipment, or dollars, which the physical facilities sector converts to space and equipment, broadly categorized as environmental facilities. A second resource from within the system is received from the per- sonnel sector and is classified as construction, maintenance and Operational staff. The demands for this sector are established by the academic and non-academic production sectors, the personnel sector, and the administrative control sector. The physical facilities sector there- fore produces the space, equipment, and staff effort required by the interacting sectors listed previously. Administrative Control Sector The administrative control sector produces the policy de- cisions which allocate the resources to the various sectors. The dotted line arrows in Figure 2 show the flow to these sectors. The changes in administrative policy are reflected in the model by changes in all the other sectors. For example, an administrative 39 decision to alter the enrollment of the university results in changes in the demand within the student sector. Corresponding changes then result in all other sectors as this change in administrative policy moves throughout the system. No mathematical description of the units of production are calculated within the administrative control sector, but are changed within the sectors which are effected by policy alterations. This is perhaps a subtle distinction because administrative control is Specified in the model by the number of students admitted, the money available, the number of faculty, etc. , but as these controls are changed the computational adjustments are carried out in all the remaining sectors. R esources and Production Throughout the discussion of the components which make up the entire system, several examples from the university were used to explain the flows in and out of the various sectors. Figure 3 identifies additional examples of the resources and products as so- . ciated with the student sector, academic and non-academic production sectors, personnel sector, and physical facilities sector. It is the intent that the identification of resources and prouducts associated with each sector will offer additional assistance in visualizing the components and recognizing the interrelationships which must be developed in order to describe the entire system. Just how many, which ones, and the level of detail to be selected in any given imple- mentation, will be determined by the questions to be answered. Resources New students ——— Course credits, dissertation credits, academic advising, etc. -—-—-)———- V 40 (a) Financial aids (controls) > Registration, counselling, housing, etc. Faculty effort - \ Grad. assistant effort —— g : Library Facilities, study halls %‘ Classrooms, labs, etc. > Audio-visual aids, TV, computers, supplies, etc.->--——'— fil Products _....————>—-"Educated Manpower ‘ Student employees , Faculty and staff effort —->——-——a- Dorms, health center, food service (C) Student . Sector for teaching, etc. (b) P Course credit hours > Dissertation advising Academic : SpeCial academic pI'OJeCtS . 7 Sponsored research Production ‘ . . . , Extension, continuing Sector . education, etc. \ e I fiHousing, food \ r Medical and social service > Non-academici~———->-——— Registration, processing Misc. building space > Production ._——->——--Evaluation, placement, Equipment, supplies > *— Sector etc. Academic staff _ > fl Grid- asSistants _ > ~ (d) ‘— > wTeaching and research Office staff > \ - . ’ I Student serVices Maintenance staff a Personnel \ . . . r Maintenance and Operation Offices, desks, car Sectors . . . . . Pry—Administration pool, office equip- \ __ ment, etc. w ’ Building space for ( ) E ‘— Class and lab stations classes and labs % e ' Offices Storage, power plant, Physical ; Living space, etc. etc. .---- ‘r Facilities Audio-visual aids, etc. Dormitories > Sectors MLibrary facilities Equipment, supplies, etc.——— > “ L Figure 3. Sectors of the Model Source: H. E. Koenig, M. G. Keeney, and R. Zemach, Systems Analysis and Planning in University Administration (East Lansing, Michigan: Division of Engineering Research, Michigan State University, 1967), Figure l, p. 23. 41 Implementations of the Model Implementing the model requires the insertion of real data using the relationships which have been established by the theoretical model. The model builder implementing the theoretical model is faced with some real constraints. One of the basic constraints is related to the employment of a mathematical model. The model is limited to those components and variables of an educational system for which a quantitative measure or value can be established. The administrator in making policy decisions reacts to a large number of. factors; some of these factors can be expressed quantitatively and some cannot. For example, there are undoubtedly individuals on a university campus who would argue that a successful athletic program attracts students. It is not clear how anyone might identify the components and variables which would describe this effect quantitatively . There are also mechanical constraints in the implementation process re sulting from the necessity to have an accurate and avail- able data base which is addressable by a computer system. Pro- gress to this end has been reported by Rourke-Brooks3 and Caffrey- 4 . . . Mosmann, but they conclude that most universities have not yet fireloped extensive information systems, which are mandatory for detailed modeling using systems analysis. 3Rourke and Brooks, The Managerial Revolution in Higher Education. 4Caffrey and Mosmann, Computers on Campus. 42 The lack of a detailed information system is evidenced when an attempt is made to collect real data pertinent to Michigan State University. Central data files do not exist at Michigan State which identify the total production of faculty members and this is necessary to completely specify the academic production sector. Salary infor- mation for faculty members ia available and computer-addressable; however, consulting activities, research involvement, publications, and service to the community and University are not collected accurately and uniformly for each individual, and the data that is collected is not coded and aggregated for computer usage. To fur- ther complicate the collection process, this information, in a variety of forms, is scattered throughout the University. This is merely one example of a constraint in developing an accurate data base which forces the analyst to develop those sectors of the model in greater detail where the most information is available. Koenig, 3&3}. suggest an alternative for the model builder as follows: . . . many areas remain in which an adequate data base is not currently available, and probably will not be for the next few years. In these cases it is necessary, at least for the present, to use subjective estimates or to omit them from the model. 5 The constraints do not destroy the intent or implementation of the model, but limit the degree of detail that can be incorporated. 5H. E. Koenig, M. G. Keeney, and R. Zemach, A Systems Mpdel for Management, Planning and Resource Allocation in Insti- flions of Higher Education-—Fin-al Report Project C-SIB, National Science Foundation (East Lansing, Michigan: Division of Engineering Research, Michigan State University, 1968). P- 96- 43 Even with these limitations, the implementation Of the model using available data at Michigan State University has been developed to a point of sophistication where it is possible to demonstrate the way that changes in allocation policies are related to the changes in pro- duction as required resources. This development is significant, for if the theoretical model is to have any real value for the educational administrator as an aid in decision making, it must be translated into some mechanism which allows the administrator to manipulate policy changes. The process which allows experimentation in this manner is simulation. Developing a simulation program for this model involves the establishment of values for the base year of the variables or para- meters which are included in the model and assumed to accurately describe the system. These values represent the present condition or state (thus a state model) and operation of the university. It is possible to program a description of the university with these assigned values so as to reflect the Operation of the university over time. Through the efforts Of M. G. Keeney and associates, a simu- lation program (MSUSIMZ) has been written which employs a data base for the College of Engineering at Michigan State University. This is the tool used in conducting the experiments in this study. The MSUSIMZ User's Manual is included in Appendix B because it has been written so that a person with limited computer background can design and carry out experiments using this document. 44 As suggested previously, the policies for the allocation Of resources are outputs generated from the administrative control sector. The simulation program has been written to allow the user to experiment with changes in policies or variables, and it then yields information describing the effect on related components within the system. The simulation program using the engineering data base divides students into eight fields as follows: chemical, civil, mechanical, general, electrical and systems engineering, computer science, metallurgy, with all other fields within the University lumped into the remaining field. Students are also divided accord- ing to five levels as follows: freshmen, sophomores, juniors, seniors, and graduates. By using this program the experimenter is able to change selected parameters, project enrollments, and calculate appropriate demands and costs. Greater detail concerning the specific applica- tion of the program will be presented in Chapter V, in the discussion Of the experiments designed for purposes of this study. CHAPTER IV INPUT FOR THE MODEL The theoretical model was translated into a working simula- tion program to provide the tool for conducting experiments to demonstrate the way a university system behaves, through the mani- pulation of selected parameters. Rather than the arbitrary selection of the parameters to be changed, a method was devised to incorporate the collective judgements of thirteen university administrators. Be- cause the implementations Of the model incorporated data addressable to Michigan State University, all of the administrators selected were from that institution. The method used for obtaining information from the administrators was an interview technique. All thirteen individuals who agreed to be included in the inter- view group, generally participate in long—range planning as a normal part of their administrative responsibilities. The interview group consisted Of one department chairman, one dean, and eleven admini- strators representing the "central administration" of the University. Individuals from the "central administration" were selected from the offices of the President, Provost, Secretary and Registrar, plus 45 46 representatives from the Graduate Office, Business Office, Institu- tional Research, and Admissions and Scholarships. The interview guide (Appendix A) served as the means for soliciting responses from the administrators. NO attempt was made to explain or interpret the workings of the model to the interview group. The purpose was to Obtain input for the simulation program so that the experiments, designed to demonstrate the workings Of the model, approached a realistic condition. General Observations The interviews yielded a broad range Of responses from detailed and Specific alternatives to generalized goals and objectives for Michigan State University. The range Of response resulted from a combination of three factors that included the nature Of the individual administrator, the Open-ended structure of the interviews, and the administrative responsibilities associated with each area. The administrators in the interview group generally inter- preted long-range planning and specific aims and Objectives to be the seeking of solutions to present day problems. Thus, when asked, "What long range plans have you recently considered regarding the future direction Of the Univer sity?", a typical response identified a current problem associated with the entire University. Similarly, when the interview group was aked, ”What are the aims and Objectives of your area which have the highest priority ‘2’”, a typical response identified a problem associated with a specific administra- tive unit . 47 It can also be noted from the other two questions in the inter- view guide concerning alternatives, that the administrators tended to respond either by outlining a specific plan the individual had con- sidered or a series of questions which needed answers before alter- natives could be identified. An example of the Specific plan reSponse was a detailed outline Of a model academic budget presented by one member Of the interview group. An example of a response which suggested questions as alternatives was the following discussion of financial aids for undergraduate students Offered by one administra- tor: Nearly 43 percent of the freshman class had some kind Of financial aid last fall. Should we have an aid quota ? With the sliding-scale tuition plan, 65 percent of the freshmen applied for and received a reduction in fees. What should the economic composition Of our student body be ? A final general Observation was the eXpressed realization by the administrators of the existence of external and internal pressures which influence planning and decision-making. These factors were most frequently mentioned in association with the development and eXpansion of various educational programs, the allocation Of financial resources, and the control Of enrollments. One administrator stated, "Planning at universities may truly result from expedience and opportunism rather than a strong constructive realization of aims and Objectives. " Given these general Observations it is next appropriate to examine the responses of the administrators according to categories 48 which were established to offer a structure for reporting the find- ings. A category was broadly defined as a major grouping Of activities that appeared to fall naturally together. Five categories were established to divide the total response Of the interview group as follows: . Enrollments Finances l 2. 3. Academic programs 4. Social concerns 5. Physical facilities The enrollment category includes interview reSponses as so- ciated with the admission, retention, and characteristics of under- graduate and graduate students, plus the recordkeeping activities associated with students. Finances include the cost of education, the budgeting and recordkeeping activities associated with financing, and the allocation and justification Of financial resources. The academic program category includes responses concerning the pro- curement, retention, qualifications, and evaluation Of faculty; the development Of innovative educational activities and approaches; and research development for the total University. The social concern category includes those responses generally associated with the role of the University as related to the problems of society. Physical facilities includes the interview responses concerned with the allo- cation Of Space, justification for new construction, and the use of existing facilities. The analysis of each of these categories identi- fies the general and Specific responses associated with each area 49 and the potential appropriateness and application of this information to the theoretical model and the simulation program. The following table gives the number of administrators who provided information, interpreted to be within the five categories. TABLE 2 INTERVIEW RESPONSES OF 13 ADMINISTRATORS ACCORDING TO SELECTED CATEGORIES Category Number Enrollment. l 3 Finances 13 Academic Program 10 Social Concerns 5 Physical Facilities 4 Enrollments The enrollment category, consisting of interview responses associated with admission, retention, and characteristics of grad- uate and undergraduate students, plus recordkeeping activities for students, received the greatest attention from the thirteen admini- strators included in the study. It was perhaps not surprising for administrators concerned with long range planning at an institution which has experienced dramatic enrollment growth in recent years, to express a variety of concerns about the enrollment of students at all levels. The interviews clearly identified the need for greater 50 control and better projections of enrollments as aids in the manage- ment of enrollments. Control Of Enrollments The control of enrollments was specified as a desirable goal by nearly all administrators included in the study; however, the complexity Of exercising controls tO accomplish this end was also recognized. What policies can be deve10ped tO control enrollments ? How can we stabilize enrollments when we can control the number of new freshmen and transfer students, but departments admit graduate students, colleges readmit undergraduates, and sophomores, juniors, and seniors return at their own free choice ? The administrators responsible for the admission of new students were reluctant to identify a detailed plan for controlling enrollments. However, by "piecing together” a number Of comments, it appeared that the number Of new freshmen might be reduced from 2-4 percent in future years, accompanied by an increase in the num- ber of transfer students. Several administrators expressed interest in controlling enrollments by admitting students at the undergraduate and graduate levels where there is room because of existing staff, facilities, and financial support. One administrator stated, "Depart- ments must make a decision as to how big they are going to be; quotas to limit enrollment cannot be established by the central administration. " No Specific plan was Offered as to how departments might make this determination but a number of other factors were suggested which contribute to the complexity of controlling enrollments. 51 The interview group identified a number of internal (within the institution) and external factors associated with the control of enrollments. The internal factors generally included descriptive characteristics pertaining to students and policy decisions which influence enrollments. The external factors were exclusively policy alternatives over which the administrators felt they had little control. One Of the internal factors dealt with the transition Of stu- dents from one major to another. Some areas on the campus, parti- cularly in the sciences and engineering, were identified as big suppliers tO other majors on campus. HWe need better information concerning the mobility of students from one department to another, " suggested one administrator. Another administrator said, "to admit students where room is available at the undergraduate level, may lead to an overflow of students in other departments after they begin to change majors. " Internal policy decisions which effect enrollments were also discussed by some administrators. Those responsible for the con- struction on new residence halls expressed the need for and reliance of enrollment projections. These administrators also expressed concern regarding internal policy changes which influence their expected needs: Based on past experience we have made projections for the construction of residence balls to house new students. Virtually overnight a policy to liberalize requirements to live '11 these units was adopted. We are faced with the critical decision of whether or not to convert existing residence hall buildings for other purposes or keep the present number of housing 52 units. We don't have enough experience to deter- mine what might happen. The grading report, adopted for implementation in Fall, 1968, was suggested as an additional ramification for the control of enroll- ments. This report, primarily designed to change the system Of grading from a letter scale to a numerical scale, also contained a provision calling for the development of a four year academic pro- gress scale. The specific policy provided for the elimination of a Z. 00 grade point average requirement for undergraduate students at the time they reached junior standing and substituted a lesser require- ment. One administrator suggested that the academic progress scale might enable a greater number Of students to remain in school longer and therefore affect class size and teaching loads. The number Of services required by students as a result of this policy might also be eXpanded. The policy allowing a College Or Department to substitute other course requirements for the University College sequence closest to the student's major, Offered another example of an inter- nal policy which might affect enrollments. The flexibility for students in the College of Social Science, for example, to waive the University College Social Science series and replace these courses with credits inside or outside the College, might result in substantial changes in selected areas. External pressures were also identified by the interview group concerning the control Of enrollments. The control exercised by the 53 State Legislature in limiting the number of students from outside the State of Michigan to 20 percent of the entire student population, was suggested as a Specific example of an external pressure. A second example was the expressed concern for admitting more black students tO the University. A final external control was the concern about the effects of the draft policy on the enrollment of graduate students. As one administrator stated, "it is indeed difficult to deter- mine what effect the present policy Of not allowing deferments to graduate students will have on the total enrollment Of the graduate school. " Thus, the problem of controlling enrollments was recognized by the interview group as an important ingredient in planning. En- rollment controls were suggested as desirable, but the methods to accomplish this task are complicated by internal and external factors. Present admission policy calls for a reduction in numbers Of new freshmen and an increase in new transfer students, but it was suggested that for these enrollment controls to be effective, departmental quotas are needed. Student Characteristics A second element included in the enrollment category was the discussion of student characteristics. Those administrators pri- marily responsible for the admission Of students were most concerned about the quality and general composition of the student body. The responses received in this area were largely in the form of questions and reflected a need for greater descriptive information about 54 enrollments. Some Of the specific questions were as follows: 1. ”There has been a greater increase in the past two years of the number of females than of males being admitted at the freshmen level. What implications does this have? What does this mean about the image of M.S. U. ? What if we control this to provide a 60-40 ratio of males to females ?” 2. "What should the economic composition of our student body be ?" 3. "As we move ahead with better quality students, what does this do to marginal admission students ?" 4. ”What kind of student successfully completes a B.S. degree? Where does he go upon com- pletion of his degree ? " 5. '‘What kind of student body do we want ?” One administrator expressed concern over the increased number Of foreign students in certain departments on the campus. It was suggested that the number Of graduate students had not de- clined in those areas but the ratio of American to Foreign students had decreased compared to previous years. A study Of successful foreign students was offered as a method Of determining what foreign students might be admitted. The same individual recognized the role of the University has played in International Education, but was concerned about the prOSpects of the State Legislature examining the number of foreign graduate students. The identification of specific student characteristics was suggested by some administrators as an important element for long range planning. Generally, the major emphasis pertained to the admission of students as alternatives to the composition of the student body. 55 Student Records A final element in the enrollment category was the response associated with the recordkeeping activities for students. A central- ized recordkeeping system was judged to be desirable by one admini- strator, but the problem Of volume and the demand for services was recognized as follows: In 1962, there were 27, 000 current records for stu- dents while in 1967 there were 42., 000. The average number of current records per staff member has increased from 386 to 512 during the same period. The demand for services such as providing tran- scripts has reached a point where 196, 231 pages Of transcripts were produced in 1967. Additional demands for services such as the certification of an increasing number of teachers were also suggested as contri- buting tO this problem. The basic conern outlined by the Registrar was how to keep pace with the increased volume and whether or not certain services should be cut out. The hOpe for future planning rests in a sophisticated record- keeping system using an advanced computer system. One of the most promising alternatives suggested was a number Of teletypewriters connected to a centralized data system which would provide Offices throughout the campus with accurate information about students. Another administrator expressed interest in the desirability Of an automated information system to process changes of major and a general updating of students records as follows: "The amount of paper work that shuffles in and out of my office pertaining to the enrollment and status of students is rediculous. Certainly some system could be developed to cut down on this activity. ” 56 A final comment concerning the records of students was the recognized need for a common coding system which would unify the recordkeeping procedures for Institutional Research, the Registrar, and the Business Office. One administrator listed several problems connected with the lack of a single system to serve all areas. Appar— ently the problem has been the failure to designate some office to assume a leadership role in this area. Those administrators who eXpressed concerns about students‘ records were from Specific administrative units assigned to carry out those tasks. The problems of volume, complexity, and continuity were generally identified by the group. The solution to the problems appeared to be auniform centralized information system which would serve all areas of the campus. Finances There was considerable discussion by the interview group regarding the justification, allocation, and budgeting Of financial resources to support both academic and non-academic functions of the University. Most of the discussions centered around the aca- demic budget since the competition for finances was recognized as an important element in planning as a ”means to an end. ” This point of view was exemplified as follows: ”If we are to attract and retain faculty to conduct research and develop quality educational programs, money is an extremely important consideration to carry out this end. " 57 Academic Financ e s Concern was expressed by some administrators about the need to justify eXpenditures to "outsiders" on the basis of something other than enrollment. It was clear from the interviews that in the past Michigan State University has based its justification for apprOpriations from the State Legislature on an anticipated increase in enrollments. "If enrollments become stablized,_ " as was suggested earlier," then the University must seek ways of communicating the value Of quality educational programs at the undergraduate and graduate levels. " Justification for resources was also suggested as an impor- tant consideration for areas within the University. This general area was discussed by one administrator as related to the academic budget. ”What are we buying for our money in terms Of instruction, research, and service ?" Further, "What are the meaningful pro- grams in terms of resources and other considerations?” It was pointed out, historically, allocations have been examined over a period Of time as related to the number Of faculty, students, and graduate programs. This has been done primarily through an exami- nation Of load and section size, sometimes resulting in reduced expense through the comparison Of costs per student credit hour in relation to full-time equivalent faculty, but not necessarily without some cost in terms Of quality. It was revealed that there has been little attempt to justify departmental budgets. The services and supplies category in the departmental budget was Offered as a specific example where there 58 has been an absence Of planning. An alternative to justify depart- mental eXpenditures was offered to include the development of a formula for each department based on the number Of faculty, faculty efforts, secretarial costs, etc. This would involve the development of staff and instructional needs in relation to costs and would include the development of common criteria to cut across various college lines. The possibility of a model academic budget has been con- sidered to develOp averages across departmental lines to include a staff ratio, faculty travel ratio, equipment and supply ratio, which would be aggregated. The ultimate goal would be to produce a realistic and defensible academic budget. If differential treatment were necessary, then this would provide a means Of explaining why this might be the case. Some discussion was focused on the amount of research money that flows into the University. Concern was expressed by some over the prospects of a curtailment Of funding in certain areas by the Federal Government. Commitments have been made on the basis that a certain amount of research money would flow in from the outside, but if a serious cut-back occurs the University would have tO pick up a greater share Of the cost. Some questions were also raised related to research finding: 1. "What is the real commitment by the Univer- sity when we agree to participate in a match- ing grant?” 2. ”Overhead charges are accpeted, but where does the money flow and are these charges realistic ? " 59 3. ”What resources and facilities are being pro- vided to support consulting and other outside research efforts ?“ A final concern expressed by a few administrators was the increased money needed to attract and hold faculty and graduate stu— dents. Financial incentives were recognized as important tO the total development Of the University. Non-Academic Finances The responsibility for the financing and prioritizing Of re- sources for the non-academic area was recognized as clearly in the hands Of a few administrators at Michigan State University. Apparently most of the administrators included in this study assumed in their plan- ning that buildings, classrooms, residence halls, laboratories, equip- ment, and other auxiliary facilities and supportive programs would be available. Two administrators identified a number Of financial problems associated with non-academic financing, but no clear alternatives. Foremost in the minds of these people were the rising costs of labor. primarily for maintenance, service, and construction. An example of this concern was the following: ”A need is determined for a building and a dollar value is placed on it. Long before working draw- ings are prepared and the plans are finalized, the cost may increase nearly 20%. As the bids are finally let for the actual construction the building cost may have increased as high as an additional 20:70. it 60 A need was expressed to justify expenditures for non-academic areas to external sources. Externally, the importance of justifying needed construction projects to the State Legislature and outside agencies was offered as an important element for planning. A need was expressed for all types of information to support requests for construction. As one administrator noted, ”It is of paramount importance that we build a careful case for new construction because legislators are extremely sensitive to buildings that always visually remind them Of expenditures they have allocated. " Academic Program Academic programs were discussed by ten administrators and contained a great deal of overlap with the other categories. The elements included in this section were those pertaining to faculty, new educational approaches, and research development for the entire Univer sity . Faculty Several administrators suggested that the faculty is a very important ingredient in long range planning. There was a great deal of interest in attempting a variety of approaches to Obtain quality faculty. The basic suggestion was more money, but secretarial and research support were also mentioned as important factors not only to attract faculty, but retain them as well. One administrator asked, ”What are the implications of faculty mobility ? Where do faculty members go and do they leave for more money, or other reasons ?" 61 It was suggested that faculty needed to be evaluated critically, particularly in their early years. One member of the interview group noted that each position should be treated as ”sacred” and the indi- vidual in that spot be evaluated critically. Innovative Educational Activities and Programs The major interest concerning educational activities and pro- grams was the evaluation Of existing approaches and the development of new ones. Specifically, it was suggested that the new Residential Colleges needed to be evaluated to determine whether this movement should be expanded, disregarded, or continued. It was recognized that these Colleges were established as experimental but there has been little evidence tO date Of evaluation. Related to this concern was the question as to whether Michigan State should attempt to organize more diversified colleges such as an Antioch or an "Ivy League" selective college approach. There was considerable interest in the development of inno- vative teaching methods. It was suggested that little is being done with computer-assisted instruction and other innovative techniques which many public schools have attempted. Broadly, one admini- strator stated, "Are there alternatives to the lecture method of instruction?” The use of graduate assistants was applauded by one administrator and criticized by another. The general need for in- service training for the new instructor was emphasized as a possi- bility for upgrading teaching. 62 The need was suggested for flexible curriculums cutting across traditional departmental lines that would allow students the opportunity to be free to choose, with proper direction, the courses and competencies they desired to achieve. Experimentation with curricular Offerings, grades, independent study, etc., coupled with proper evaluation was suggested as an important alternative to present practice. One administrator inquired about the development of new programs and the elimination Of others. "Is M. S. U. a complete University? What fields are we not covering that have appreciable enrollment at other institutions ? Are these areas we are missing? Are there areas in the University we should get rid of based on enrollments, research, and other acti— vities ? " The development of new programs, the strengthening of existing programs, and the abandonment or reduction Of others were mentioned. It was generally agreed that one of the most important areas that should be strengthened was the graduate school. It was noted that Michigan State has reached a high point in the develop- ment Of undergraduate education, but a genuine effort to improve all phases of the graduate school was needed. One administrator noted the need to analyze the emphasis on agriculture with the hope that it would be brought into a realistic perspective with the rest of the Univer s ity . 63 Social Concerns Social concerns refer to requirements of the University to be sensitive to societal needs in planning for present and future direction. The greatest attention was directed toward minority groups, but concern with student dissatistfaction was also identified. The need for a variety of equal opportunity programs on the campus was identified as an important first step. The following definition of equal Opportunity programs was offered: Equal Opportunity programs are those directed at assuring non-discriminatory access of minority representatives to the student body, administrative staff, teaching faculty, and supporting staff Of this University. Equal Opportunity programs should be further concerned with the articulation of policies and the inauguration of projects that establish a body Of legalistic and quasi-legalistic statements against which equal Opportunity issues can be analyzed and judged, and which will preclude Michigan State University support Of discriminatory practices by organizations and individuals doing business with the University, its students and employees. A number Of objectives were identified by some administra- tors within the framework Of the above definition which included: an increase in the number of entering and graduating undergraduate and graduate minority students; an increase in the proportion of minority faculty members, administrators, and staff and supportive personnel; and the develOpment and enforcement of regulations against discrimi- natory practices in University involvements. The four recommendations for curricula, research, community action, and experimental functions, outlined by the committee Of six- teen at Michigan State University, were identified by one member of 64 the study as important areas for social concern. Some of the Objectives in these four areas were as follows: the coordinated development of socially relevant curricula; the need to conduct and coordinate urban related research; the development and coordination of degree programs in Afro-American Studies and urban affairs; the initiation and coordination of University participation in community action programs; and the dissemination of research findings and other information. A second element in the social concern category was the discussion Of student unrest. Three administrators indicated that future planning should include provision for the orderly involvement of students in the affairs of the University. One administrator pointed to the "Academic Freedom Report" as an important guide and foundation block. ”If procedures and policies need revision, the mechanism is provided for in that document. ” It was emphasized that students are concerned about the excellence of teaching, courses that are relevant to the problems of society, and programs where they can be actively involved in assisting others. The direction that student dissatisfaction will take in the future, was identified as an unknown quantity. One administrator summarized the problem as follows, ”I am not willing to turn over the reins of leadership to students, but I welcome the opportunity to listen and to establish mechanisms and procedures for their participation. " 65 Phy sical Facilitie s The finance category included a discussion of the financial consideration and justification for the construction of new facilities. In addition to the problem of financing. some administrators, res- ponsible for specific administrative areas, identified problems and alternatives associated with the use and allocation of existing facilities. Present policy identifies classroom Space as belonging to the University. Departmental Offices, laboratories, libraries, and Special Space needs belong to the University, but are assigned and designated to certain areas. The problems associated with classroom scheduling are largely due to the increase in the number Of large sections. As enrollments have eXpanded one of the most common ways to handle this growth has been by increasing class size. The need for large classrooms at popular times during any given term was identified as critical. During Fall term the demands for classrooms peak because enroll- ments and course Offerings are the highest. A 50 percent classroom utilization figure is judged to be nearly optimal. (This percentage is Obtained by assuming that a classroom will be used every day, during all periods between 8:00 a.m. and 5:00 p.m., five days per week.) Alternatives to reduce the use Of space included greater controls for forcing departments to spread out course Offerings throughout the day; a better use of classroom facilities at night; 66 and greater imagination for the scheduling of courses at times other than normal patterns. (For example, the question was raised, ”Why must nearly all three credit courses be Offered on Monday, Wednesday, and Friday ? ”) The increase in the number of graduate students and faculty has also presented critical space problems. Classrooms can be moved throughout the campus, but departments do not want their faculty to be split into small divisions at various locations. The office space problem is further complicated by differences that exist between departments and colleges. In some departments graduate students may have private offices, while in others professors may share a common facility. Alternatives for the use of Office Space were; the grouping Of graduate students together in large ”bull—pen" type areas, the elimination Of two Offices for faculty on dual appointments, and greater uniform policies for the assignment of Office Space which cut across departmental lines. Research grants calling for the development of specialized equipment and facilities Often cause space problems. The contract any be written, but Space considerations for the development of hard- ware may be a minor consideration. A similar problem was suggested for the development Of new programs. If new programs and activities are established and grow, too often space facilities are needed as a result of this expansion. It was suggested that space considerations are not generally involved in the original planning and become impor- tant at a later date. Departments and administrative areas are 67 reluctant to release Space once it has been assigned. The end result may be the development of new construction when consolidation may have provided wiser alternatives. Analysis of the Interview Input as Related to the Model The interview results generally revealed a range of problems associated with present University policies and practices. It was clear that several administrative units in the University have inde- pendently considered future planning, but not within a centralized framework or strategy for development. There was little evidence reported, for example, that much centralized direction has been given to the problem of control of enrollments, even though this concern was identified by all administrators as a critical problem. The administrators eXpressed a desire for more descriptive information as an aid in planning. The discussions of the character- istics and transitions of students, and the distribution of finances exemplify this generalized conclusion. The significance of enrollments and the importance of finances were recognized by all administrators as critical elements in planning. The administrators tended to discuss these as interrelated factors whether they had concerns about the quality of faculty, the develop- ment of innovative programs, the construction of physical facilities, or other concerns about the University. There was concern expressed by administrators that policy decisions have been made in isolation without an awareness Of other 68 areas on campus. The reduction of requirements for living in residence halls, the redistribution of student credit hours, and the change in academic standards were suggested as Specific examples Of isolated policy decisions. The need for evaluation of existing programs and personnel was suggested as an important need by administrators. The evalua- tion Of faculty, curricula, social needs, and educational programs and approaches were some of the identified concerns in this area. Three of the four generalized concerns identified by the administrators are judged to be directly addressable to the Specific model employed in this study. The need for more information can be approached by using the model as a descriptive tool; the inter- relationship Of enrollments and finances, and the problem of iso- lated decisions can be approached by using the model as an aid in examining the effects of alternative policies. The need for evaluation of existing programs and personnel would require greater refinement before these questions are directly addressable to this specific model. As emphasized in Chapter III, the model used in this study is limited to those aspects Of the educational system that can be measured. As a result, many problems related to broad goals and Objectives would need further clarification in order to be addressable to the model. The less tangible aspects of the educational system that influence decisions, such as social concerns and political influences, are not directly included in the model. This type of influence must be translated into some tangible measure. 69 An adequate and accurate data base enables the administra- tor to obtain information about the present operations of the Univer- sity. Because each sector in the model has been modeled as an independent system and these sub-systems interconnected to form the total system, it is possible for the administrator to examine a particular segment Of the system. For example, it is possible to examine the number of new students, the transition of students, the cost of faculty by rank, the faculty effort for teaching, the cost of equipment and/ or supplies and services, or the number of credit hours produced, etc. , all of these yield descriptive information which the administrator may desire to have about the present Operation of the university. A great deal of this information is available in the present model and the capability to enlarge the data base is present. Expanding the data base Of the model is obviously contingent upon the availability of information and the capability to categorize it in quantitative units. The transition of students from one area to another was identified as a Specific concern by one administrator in the inter- views and this information is directly available in the present model. Depending upon the specifications of the administrator, it would be possible to expand the data base to include additional student characteristics which were suggested in the interviews. For example, the identification of students by sex, race, economic backgrounds, and geographical locations are not included in the present model, but could be included. 70 Thus, the concern by the interview group for more descrip- tive information can be approached by the use of the model as a descriptive tool. This is the simplest use Of the model because the segments of the model are examined in isolation and not as they interact with the rest of the system. The Model as an Aid to Examining Alternatives Several alternatives to present University practices and policies were suggested in the interviews. While it is not the intent to examine how each alternative might be answered by the model, two examples have been selected tO illustrate the use of the model in this manner. Again, it must be emphasized that the alternatives must be reduced to measurable quantities. One administrator suggested that the graduate school should be expanded. TO consider this alternative a number Of questions need answers: 1. What areas will the new students enter ? 2. Will this increase require additional faculty? 3. Will the students receive graduate assistantships ? 4. What kinds Of demands will they place on the entire university for courses, equipment, space, and services ? 5. If these students are graduate assistants will they provide a source of manpower for teaching undergraduates ? 6. Will any of these students be working on research projects ? If so, can research projects help to carry the financial load ? 71 7. Where will these students live ? Will they create increased demands for married housing or graduate dormitories ? These questions could be carried further, but they do illustrate the interrelationship between enrollment, costs, faculty effort, and physical facilities, which is one intent of the model in this study. To the extent that the data base and the interrelationships expressed in the simulation are realistic, the administrator can eXperiment with alternatives to existing conditions. A second example is the concern about policies being developed in isolation. Specifically, the question of the redistribution of course hours because students are no longer required to take the University College sequence in social science, can be examined. Again, a number of questions surround the policy change: 1. What will be the reduction in the faculty effort for teaching in the Department Of Social Science as a result of the policy change? 2. Will the faculty members be available to do other activities in the department or shifted tO new areas ? 3. What is the associated cost of this shift? 4. Will new faculty positions be available to the areas receiving the students ? This example suggests that policy decisions affect the Operation Of the entire university. The model can be used as a tool to examine policy alternatives before they are enacted, by using simulation. The process of simulation requires the development of a series of computer programs to establish a data base and carry out the computations which represent interactions Of the sectors Of the model. 72 A simulation program has been deve10ped using the College of Engineering as an example of how policy alternatives affect the Operation Of the system. Experiments have been designed to demon- strate how the system behaves when subjected to changes in present policy. The experiments and the resultant interactions are reported in Chapter V. CHAPTER V ANALYSIS OF THE EXPERIMENTS The analysis of the interview data plus the intent to com- municate the use of the model, led to the design Of seven experi- ments to demonstrate the use of a simulation model as a potential aid for educational decision-makers. Due to the concern by administrators in the interview group regarding the management and control of enrollments, four experi- ments were designed to simulate the magnitude Of changes and the sensitivity of the system to the admission of new students. Two specific enrollment changes were identified by administrators as possible alternatives for the future direction Of Michigan State University. A trend was identified to increase the number of trans- fer students admitted at the upper levels to the University. The expansion of the number Of community colleges in the State Of Michigan was suggested as the basic reason for this projected increase. Therefore, experiments I and II were designed to reflect varying percentage increases in the number of new sophomores, juniors, and seniors admitted to Michigan State University. Trans- fer students from other institutions are not Specifically identified 73 74 in the simulation model used in this study; the new student category in the model includes transfer students and students readmitted to the University. Because transfer students comprise the largest number in the new student category, it was expanded. Administrators also expressed considerable interest in limiting the number of new freshmen admitted to the system. The rationale for this judgement was generally that quality should be emphasized at the first year level and one of the means to insure quality was to establish selective admission policies. Thus, experiments I, II, III, and IV reflect the response of the system to increases in the number of sophomores, juniors, and seniors, and decreases in the number of freshmen admitted as new students. The concern by the administrators about the effects of changes in graduation requirements for students in one area on supporting departments, provided the background for experiment V. Admini- strators in the interview group indicated that policies Of this type were Often made in isolation. The design of experiment V was an attempt to Show the effects on the system due to a change in the curriculum for students in a representative department. The remaining experiments were included to demonstrate the working Of the simulation model and to emphasize the notion that decisions are rarely made independent Of one another. Experi- ment VI combined the effects of the policy changes incorporated in 75 experiments I and III. The intent of this eXperiment was to demon- strate the effects of two policy changes operating at the same time. It was primarily related to financial considerations because of the reactions by the administrators regarding the importance of the allocation of resources. Four simulated policy changes were made simultaneously in eXperiment VII. The purpose of this experiment was to illustrate a more realistic situation where several changes are Operating at one time. The complexity of the interaction of the changes result from differing time constants and unclear relationships between effects that augment or cancel one another. Calculating the net change on the entire system is so complex that a computer is required if the interactions are to be evaluated in a reasonable amount of time for a variety of conditions. In particular, an illustration of the effect Of a decrease in freshmen enrollments, an increase in the number Of new students at the upper levels, an increase in Salaries and projected costs, and curriculum changes in one department, in combination, was exhibited for the College of Engineering. The College of Engineering is the basic administrative unit used to demonstrate the workings Of the simulation model. However, in selected instances, it is possible to demonstrate the relationship Of the College to the rest of the University. The extent to which this type of detail can be examined is directly dependent upon the information detail incorporated in the simulation program. Only 76 enrollment information is included for the entire University in the simulation program used in this study, while additional detail from other sectors is incorporated for the College of Engineering. This is because student records are available and computer addressable for the entire University, but cost information and faculty records are generally not accessible in this form and so were calculated for only one College. The presentation Of the experiments in this section represent a gradual movement from the simple to the complex through the manipulation of selected parameters in the simulation program. The suggested policy changes by the administrators become the parameter changes in the administrative sector when applied to the simulation model. Experiments I, II, III, IV, and V present a single parameter change where the results of the manipulation are clearly visible and can be traced directly to the change. The discussion of experiment I is presented in greater detail than the rest of the experiments because of the necessity to clarify terminology and Show the step by step pro- cedures carried out by the simulation program. Experiment VI re- flects two parameter changes where the alterations are visible, but the identification Of the changes result from the interactions of two variables instead of one. Experiment VII demonstrates the result- ing interactions of four parameter changes and exemplifies the complexity of a University system. Before examining the experiments, it is important to intro- duce two concepts which underlie experimentation with the simulation 77 model. The consistency of the data in the simulation program us ed in this study has been tested and carried out to a point in time where the calculations performed in the program do not alter the numbers or the results.1 This status is identified as the steady state condition and serves as the base year of the simulation model. The base year, labeled 1967, represents conditions prior to the intro- duction of any parameter changes. It is not practical to diSplay all the base year information supplied by the simulation program because of the detail incorporated in the model. Instead, only the base year quantities are presented which are pertinent to the experiments con- ducted in this study. It is important in these experiments to emphasize that the actual numbers are not to be interpreted too literally because the available records were not necessarily coordinated but trends and orders of magnitude do approximate realistic conditions. Experiment I The interviews with administrators at Michigan State Univer- sity revealed an interest in expanding the number Of transfer students from community colleges in the State by 25-33 percent. As explained earlier, it is not possible to single out transfer students 1Documented on October 20, 1968 by the creation of Data File 6. Data File 5 in the simulation program was projected 10 years to a point where the data base did not change. 78 in this particular program, but the number Of new students admitted to the University can be changed. Therefore, eXperiment 1 con- siders the following questions: What are the implications of a 25 percent increase in the number of new sophomores, juniors, and seniors admitted to Michigan State University? How many additional student credit hours will be produced ? What are the associated costs, enrollments, faculty additions, and student credit hours for the College of Engineering as a result of this increase? The answers to these questions can be diSplayed by a com- parison between quantities which result from present administrative policy and the quantities which result from a simulated policy change which alters the number Of new students. New students are defined as those students new to the University for a given year and includes freshmen, transfer, graduate, and all readmitted students. Table 3 presents the changes in enrollments as a result Of a simulated increase of 25 percent in the number of new SOphomore, junior, and senior students admitted to the College Of Engineering. The number of new students and the total number of students by class levels are identified. It is significant to note that the enroll- ment figures at the freshman level do not change because the new administrative policy affects only three class levels. The base year (1967) shows 129 new students presently being admitted at the sophomore, junior, and senior levels. A change in that policy is introduced which allows an additional 32. 79 TABLE 3 SIMULATED ENROLLMENT INCREASES 1967-1971 COLLEGE OF ENGINEERING EXPERIMENT I Freshmen Sophomores Juniors Seniors Grad 8 Total 1967 New Students 672 35 80 14 106 907 Total Students 687 382 310 343 287 2009 1968 New Students 672 43 101 17 106 939 Total Students 687 390 331 346 287 2041 1969 New Students 672 43 101 17 106 939 Total Students 687 392- 337 361 287 7-064 1970 New Students 672 43 101 17 106 939 Total Students 687 392 337 371 288 2075 1971 New Students 672 43 101 17 106 939 Total Students 687 392 337 373 2-91 2080 new students to be admitted to the system. The changes can be Observed by increases in the new students, at the three middle levels, and the total number Of students for the year 1968. The enrollment data reported for 1968 reflect the new policy. The movement of the students through the system for subse- quent years is also evidenced in Table 3, but the effects cannot be isolated. For example, the increase in the number of juniors in 80 1968 (331—310 = 21) partially accounts for the increase in the number of seniors in 1969; however, the increase may also partially account for the increase in juniors in 1969 because some may be retained at the same level. The increase in graduate students which first appears in 1970 and again in 1971, demonstrates two factors which contribute to the growth Of graduate enrollments. First, the undergraduate enroll- ment increases create demands for graduate assistants who are used in instruction; second, an increase in undergraduate enrollments produces students who continue in the College Of Engineering upon completion of the B.S degree. The effects of the new policy are virtually stabilized by 1971, for only the total enrollment at the senior and graduate levels increased slightly over 1970. It can be noted that the aggregated enrollment effects would produce a total of (2080-2009 = 71) additional students three years following the initiation Of the new policy. It is also possible to Observe similar effects for the rest Of the University by an examination of the data reported in Table 4. As a result Of the policy change introduced in 1968, a total Of (14, 107- 13, 403 = 704) additional students are initially admitted to the Univer- sity. The comparable figure for the College of Engineering is 32, The enrollment increases resulting from the policy change do not stabilize as quickly as reported for the College Of Engineering. For example, 62 more seniors appear in 1971 than were present in 1970. In spite of this limitation it is possible to state that the system has 81 returned to a relatively stable condition. A comparison between the total enrollment in 1967 and 1971 shows the net effects of the change in enrollment policy. The total enrollment in 1971 is 1, 539 students higher than the figure reported for 1967. The effect of an increase in undergraduate enrollment over graduate enrollment is evidenced earlier in Table 4 than in Table 3. The need for 10 graduate assistants to teach the additional undergraduate students accounts for the increase in graduate enrollments for 1969. The additional increases in graduate enrollments can be Observed in the remaining two years. The enrollment increases produce greater demands for services and instruction which results in greater costs. Examples of the effects of enrollment increases in instruction are the increases in the number of faculty and student credit hours. Student credit hours are determined by multiplying the num- ber Of students enrolled in a particular course by the credit weight assigned to each course. Thus, a 3 credit course with an enroll- ment of 30 students yields a total of 90 student credit hours. Table 5 shows the increase in the number of student credit hours, by levels, resulting from the simulated enrollment increase. The five levels associated with student credit hours are different than the enroll- ment levels for students. The following shows the grouping of courses used to determine the level of student credit hours: 82 TABLE 4 SIMULATED UNIVERSITY ENROLLMENT INCREASES 1967-1971 (NOT ENGINEERING) EXPERIMENT I Freshmen Sophomores Juniors Seniors Grads Totals 1967 New Students 7, 438 783 1, 432 599 3,151 13, 403 Total Students 7, 812 6, 615 6, 549 7, 757 7, 564 36, 297 1968 New Students 7, 438 979 l, 790 749 3,151 14,107 Total Students 7, 812 6, 811 6, 907 7, 907 7, 564 37, 001 1969 New Students 7, 438 979 l, 790 749 3,151 14,107 Total Students 7, 812 6, 825 7, 066 8, 222 7, 574 37, 499 1970 New Students 7, 438 979 l, 790 749 3,151 14,107 Total Students 7, 812 6, 828 7, 087 8, 416 7, 602 37, 745 1971 New Students 7, 438 979 l, 790 749 3,151 14,107 Total Students 7, 812 6, 828 7, 090 8, 478 7, 628 37, 836 Courses numbered: Level 100 - 199 Freshmen 200 - 299 Sophomore 300 - 399 Junior 400 - 499 Senior 800 - 999 Graduate Students at all enrollment levels register for a variety of courses with designated course levels. It is not uncommon for some graduate programs to require collateral work in 400 (senior) level courses. 83 The increase in the number of student credit hours at the freshmen level reported in Table 5, offers an appropriate example. Even though the policy change to increase enrollment did not involve fresh- men students, the student credit hours required at the freshmen level increases. The admission Of new students at some levels, then, may create needs in the total system for the production of greater numbers of student credit hours at all levels. Three years following the initial policy change (1971), the net effect for the College of Engineering is that an additional (13, 049- 12, 321 = 728) student credit hours would be needed. It is important to emphasize that the majority of the demand would come from within the College, but increased demand from areas outside the College could also contribute. The net effect for the remainder of the University during this same period would be the production Of 22, 226 additional student credit hours. In addition to student credit hours, the effects of the enroll- ment increase on instruction can be identified with the need for additional faculty. The simulation program has the capability to calculate faculty and graduate assistant needs for all departments in the College of Engineering from student credit hour demands. The changes in faculty and graduate assistant needs from 1967-1970 are shown in Table 6. The increase column refers to the differences between faculty and graduate assistant needs in 1967 and those that would be required in 1970. 84 TABLE 5 STUDENT CREDIT HOUR INCREASES 1967—1971 EXPERIMENT I Fresh- Sopho- Juniors Seniors Grads Totals men mores 1967 Engineering 2, 236 2, 060 3, 318 2, 835 l, 872 12, 321 Not Engineering 138, 916 142, 960 84, 732 51, 589 51, 589 497, 394 1968 Engineering 2, 277 2, 142. 3, 380 Z, 874 l, 872 12, 545 Not Engineering 140, 514 147, 278 87, 563 80, 999 51, 624 507, 978 1969 Engineering 2, 299 Z, 181 3, 541 Z, 995 1, 878 12., 894 Not Engineering 141, 254 148, 972 89, 879 83, 237 51, 755 515, 097 1970 Engineering 2, 308 2,192 3, 564 3, 057 1, 890 13, 011 Not Engineering 141, 541 149, 531 90, 915 84, 472 51, 968 518, 427 1971 Engineering 2, 311 2,197 3, 571 3, 073 l, 897 13, 049 Not Engineering 141, 638 149, 706 91, 233 84, 888 52,155 519, 620 85 TABLE 6 INCREASES IN FACULTY AND GRADUATE ASSISTANT REQUIREMENTS FOR ENGINEERING DEPARTMENTS 1967-1970 EXPERIMENT I b 1967 and Materials Science. The figures are expressed in terms of full-time-equivalent faculty. 1968 1969 1970 Increase Chem. Engr. Faculty 7.2 7.2 7.3 7.3 .1 Grad. Asst. 2.5 2.5 2.6 2.6 .1 Civil Engr. Faculty 8.9 9.0 9.1 9.2 .3 Grad. Asst. 9.0 9.3 9.5 9.7 .7 Mech. Engr. Faculty 14.6 14.9 15.2 15.4 .8 Grad. Asst. 13.4 13.8 14.3 14.4 1.0 Gen. Engr. Faculty 5.8 5.9 6.0 6.0 ,2 Grad. Asst. 1.9 2.0 2.0 2.0 .1 Elec. Engr. Faculty 19. 3 19.6 20.0 20.2 .9 Grad. Asst. 24.9 25.8 26.6 27.0 2.1 M.M.M.a Faculty 14.5 14.8 14.9 15.0 .5 Grad. Asst. 11.4 11.9 12.2 12.3 .9 Computer Sci. Faculty 10.9 11.1 11.3 11.3 .4 Grad. Asst. 5.4 5.5 5.6 5.6 .2 3LM. M. M. refers to the Department of Metallurgy, Mechanics, 86 The collective faculty and graduate assistant needs of the College of Engineering provide a means of comparison with other data expressed in experiment I. The aggregated need for the College would be an additional 3. 2 full-time-equivalent faculty members and 5.1 graduate assistants. It is questionable, however, particularly for faculty members, that the total College needs could be met in this aggregated manner. It is more realistic to observe that mechanical and electrical engineering by 1970 would be approaching the point where an additional faculty member would be needed. A similar Observation can be made with reference to graduate assistants, but a part-time graduate assistant is a more realistic consideration than a part-time faculty member. An examination Of the effects of the simulated enrollment increase on costs is the next step in analysis. No cost data is avail- able for the entire University, but detailed cost information is avail- able for all departments in the College of Engineering. No changes in co st parameters have been introduced in this experiment, there- fore’, the changes in costs are a result Of the interactions among the enrollments resulting from the policy change. It is possible to examine the changes in total costs for the College Of Engineering from 1967- 1970, as presented in Table 7. Numerous calculations are performed in the computer program to reach the total distributions reported in this table. Total Undergraduate Costs includes the cost Of under- graduate instruction by faculty and graduate assistants and the 87 TABLE 7 INCREASES IN TOTAL COSTS - COLLEGE OF ENGINEERING 1967-1970 (IN DOLLARS) EXPERIMENT I 1967 1968 1969 1970 Increases Total Undergraduate 841, 263 863, 519 882, 572 890, 328 $49, 065 Costs Total Graduate 306, 408 306, 965 308,197 309, 738 3, 330 Costs Total Thesis 8: Research 355,658 355,648 355,699 355, 921 263 Costs Total Costs, Other ActivitieSI 159,112 161, 611 163, 722 164, 763 5, 651 Total Costs $1, 662, 441 $1, 687, 743 $1, 710,190 $1, 720, 750 $58, 309 equipment and supplies required to carry out that instruction. Total Graduate Costs includes the cost Of instruction by faculty and the equipment and supplies required to carry out that instruction. Total Thesis and Research Costs includes the faculty and graduate research assistants, equipment and supplies required to conduct research and direct dissertations. Total Costs, Other Activities includes secre- tarial services, special equipment and supportive services which cannot be directly attributed to any of the other categories. 88 The greatest increase in costs during the three year period is logically in Total Undergraduate Costs. The Total Costs of $58, 309 represent the net effects Of the enrollment policy change after three years. Thus, Experiment I shows the net effects Of introducing a single parameter change. The 25 percent increase in the number of new students produced enrollment changes for the College of Engineer- ing and the remainder Of the University. The new students create demands for the production Of additional student credit hours, and faculty and graduate assistants to carry out that instruction. Finally, the effects Of the increases are translated into the total dollar costs, illustrated by using the College of Engineering as an example. Experiment 11 Experiment 11 is designed to show a different magnitude Of the enrollment policy change introduced in experiment I. The only difference between experiment I and II is that the simulated percentage increase in the number of new sophomore, junior, and senior students admitted to the University is 33 percent instead Of 25 percent. It is possible to develop user confidence in the model in that calculations can be manually carried out from the results Of experi- ment 1. Table 8 shows the enrollment comparisons for the differences between the two experiments. Thus, -2% = l. 333 which is the ratio between the 33 and 25 percent enrollment changes. If a 25 percent increase causes a change of (939-907 = 32) additional new students, 88 The greatest increase in costs during the three year period is logically in Total Undergraduate Costs. The Total Costs of $58, 309 represent the net effects of the enrollment policy change after three years. Thus, Experiment I shows the net effects of introducing a single parameter change. The 25 percent increase in the number Of new students produced enrollment changes for the College Of Engineer- ing and the remainder Of the University. The new students create demands for the production Of additional student credit hours, and faculty and graduate assistants to carry out that instruction. Finally, the effects Of the increases are translated into the total dollar costs, illustrated by using the College of Engineering as an example. Experiment 11 EXperiment II is designed to show a different magnitude Of the enrollment policy change introduced in experiment I. The only difference between experiment I and II is that the simulated percentage increase in the number Of new sophomore, junior, and senior students admitted to the University is 33 percent instead of 25 percent. It is possible to develop user confidence in the model in that calculations can be manually carried out from the results of experi- ment 1. Table 8 shows the enrollment comparisons for the differences between the two experiments. Thus, :—: = l. 333 which is the ratio between the 33 and 25 percent enrollment changes. If a 25 percent increase causes a change Of (939-907 = 32) additional new students, 89 TABLE 8 NET EFFECTS ON ENROLLMENT FOR ENGINEERING AND THE REMAINDER OF THE UNIVERSITY RESULTING FROM TWO PARAMETER CHANGES EXPERIMENT II 1968 1969 Difference 1967 25 33 2,5 33 25 33 percent percent percent percent percent percent New Students’ 907 939 950 939 950 +32 +43 Engr. TOtal Students“ 2 009 2,041 2,052 2,080 2,103 +71 +94 Engr. ’ New Students" 13,403 14,107 14,341 14,107 14,341 +704 +938 Non Engr. TOtaIStudents' 36,297 37,001 37,135 37,836 38,356 +1.539 +2,059 Non Engr. then the expected number of new students according to the 33 percent increase should be (1. 33x32 = 43). is (950-907 = 43). Tables 9 and 10. Table 9 reports an increase Of (13, 049-12, 321 : 728) addi- As recorded in Table 8, the increase Similar calculations can be made with reference to tional student credit hours as a result Of the 25 percent enrollment change. cent change would be (1.333x728 = 970). Table 9 for the 33 percent change is (13, 294-12, 321 = 973). The expected number Of student credit hours for a 33 per- The increase reported in Table 10 presents the differences between the 25 and 33 percent enrollment increase for the College Of Engineering as they apply to teaching 90 TABLE 9 DIFFERENCES IN THE PRODUCTION OF STUDENT CREDIT HOURS FOR ENGINEERING AND THE REMAINDER OF THE UNIVERSITY AS A RESULT OF TWO PARAMETER CHANGES EXPERIMENT II 1968 1971 Difference 1 67 9 25 33 25 33 25 33 percent percent percent percent percent percent Engr. 12, 321 12, 545 12, 737 13, 049 13, 294 728 973 Non Engr. 497, 394 507,178 511, 506 519, 620 527, 032 22, 226 29, 638 TABLE 10 DIFFERENCES IN FACULTY INCREASES AND GRADUATE ASSISTANTS FOR ENGINEERING AS A RESULT OF TWO PARAMETER CHANGESa 6 1968 1970 Difference 7 19 25 33 25 33 25 33 percent percent percent ercent percent percent Engr. Faculty 81.2 82,5 84.7 84.4 85.5 3.2 4.3 Grad. Asst. 68.5 70.8 74. 3 73.6 75.5 5.1 7.0 aExPressed in full-time-equivalent faculty 91 requirements. The 25 percent enrollment change shows the difference to be (84.4-81. 2 = 3. 2) additional engineering faculty members. The expected number of engineering faculty members for a 33 percent enrollment increase would be (1. 333x 3. 2 : 4. 3). The difference noted in Table 10 is (85. 5-81.2 = 4. 3). It would now be possible to manually calculate the change in any of the tables for any percentage change simply by the ratio '27; x (change due to 25%). This is due to the linearity of the relation- ships used in the model. This means a model is probably valid over only a limited range because actual relationships are non-linear. If other (non-linear) relationships were incorporated in the model, this simple linear relationship would not hold. The illustrations Show that the model does give expected answers which are easy to verify for this simple change. Later, more complex changes are made which are not easy to verify manually. It points to one advantage Of simulation, that is, it allows a large amount Of calculations to be carried out more rapidly and accurately than can be done manually. Experiments III and IV Experiments III and IV are designed to simulate reductions in the number of freshmen admitted to the University. The members of the interview group expressed interest in reducing the number of students admitted at this level in an effort to increase the quality 92 of students attracted to Michigan State University. It is not possible to Simulate possible quality changes in the nature Of the first year population, but it is possible to reduce the number Of new students admitted at this level. In particular, experiments III and IV are concerned with the changes in dollar costs resulting from two and four percent reduc- tions in the number Of new freshmen admitted to Michigan State University. Cost variations in the College of Engineering due to the enrollment change are studied. EXperiment 111 provides for a two percent reduction in the number Of new freshmen admitted to the system while in experiment IV a four percent reduction is simu- lated. The results of the two experiments are presented together for comparison of the two alternatives. It is important to emphasize that no other parameter changes such as faculty salary or load adjustment are introduced. The cost differentials can only be attributed to the reduction in the number Of new freshmen. The primary purpose Of this experiment is to examine cost information; however, it is important, for background information, to Show the effects of the policy change on enrollments. Instead Of a detailed presentation Of enrollment changes by year and level, only the total enrollment change is presented in Table 11. The net change in enrollments by 1971 would result in 29 fewer engineering students and 420 fewer non-engineering students in the system due to the two percent reduction would be double these figures, or 58 fewer engineers and 839 fewer students in the rest of the University. 93 TABLE 11 TOTAL ENROLLMENT CHANGE FOR THE COLLEGE OF ENGINEERING AND THE REMAINDER OF THE UNIVERSITY RESULTING FROM A REDUCTION IN THE NUMBER OF NEW FRESHMEN 1968 1971 Change 1967 2 4 2 4 3 4 percent percent percentl percent percent percent t d - To a1 Stu ents 2, 009 l, 994 1’ 980 1, 980 1, 951 —29 -58 Engr. T t d - 0 all St“ ents 36,297 36,149 36, 001 35.877 35.458 -420 -339 Non Engr. Table 12 presents the computation of the freshmen enrollment reductions into cost information for the College Of Engineering. The differences in costs are caused by the two percent and four percent reductions in freshmen enrollment for the period 1967-1970. The slight reduction in costs during 1968 shows the limited number Of freshmen who take courses in the College of Engineering. The largest producer Of freshmen credit hours for engineering stu- dents is the computer science area and nearly all decreases for the College occur there. A larger reduction in costs begins to appear in 1970 because engineering students have completed the first full year of engineering courses. The full effects of the enrollment change on costs would not appear until 1971. The significance here is not so much the actual numbers, but the trends and the cost areas that would be reduced by the enrollment change. 94 an... .22 MS .63 So .83 .H was .36 .H a: .63 .2 mom .03 .2 ms. .03 4 8m .38 .H at .36 .5 330 38. Anon—gnu... m3 .7 So. So .me $6.me 5. .me moo .02 mi .02 314.2 Nam .403 3&9 mumoO Hon—OH. R300 :1 an- :6 .mmm $6.3... o3 .mmm 88.3... $6.3m $8.33 38 .mmm sonnets a ammonfi H.308 R300 93 NE..- N5 .mom 6: .2: $0 .nom mum .03 n3 .8... mom .8... n3. .nom 386.20 :38. mumoO NE .24 8m .3” as .on NS .mmm a? .mmm movnmw 4:. .omw Sn 6% SN .3 a «28623665 Enos unmouom ”=80qu usoonom usoonom ucoouem «coupon acoouom annoys e k u. k v N H N 53 monotonic tz 2.2 $3 $3 Ammadofl 7H: EAL-hem: mHZMZAAOMZH ZEEwMMh ZH mMmHHHfi