ABS TRACT A MODEL FOR PREDICTING THE UNIT COST OF DEVELOPING SELF-INSTRUCTIONAL MATERIALS IN HIGHER EDUCATION BIOLOGICAL AND PHYSICAL SCIENCES by Walter Grove Chappell Need Budgeting for the cost of developing self-in- structional materials for higher education biological and physical science courses is difficult, since instructional planners , administrators and funding agencies have had few guidelines to use in making accurate development cost pre- dictions . Purpose The purpose of this study was to design a predictive unit-cost model which could be used after be- havioral objectives have been initially stated and a slide-audio tape and film or an audio tape only media stra- tegy has been chosen by the instructional planner(s). All relevant costs for the initial development of the hard- ware and software systems prior to the first use of the materials to grant course credit to students should be able Walter Grove Chappell Application A.predictive, unit-cost model in.budget planning guide form was designed and validated in this study. It will predict development costs when applied in existing higher education institutions by biological or physical science instructional planners who anticipate de- signing a self-instructional system which utilizes a slide- audio tape and film or an audio-tape-only media strategy. The cost and non-cost categories included in the predic- tions are faculty and related content personnel costs, carrel equipment costs, software production and duplication costs, consultant costs, facilities modification costs, validation and revision costs, the amount of time required for faculty involvement in the materials development, and the number of months for the development process prior to implementation. Costs not accounted for in these predic- tions include developing the initial statement of objec- tives, the selection of an appropriate instructional strategy and the involvement of non-development faculty and staff to the extent that optimal use of the self-instruc- tional materials can occur once they are developed. How- ever, though not included in the model's predictions, these costs should be planned and budgeted for as necessary ex- penditures if the materials developed are to be effective and optimally utilized. Walter Grove Chappell to be predicted within plus or minus fifteen per cent of the actual development costs. Procedures A faculty interview instrument was designed and validated during a pilot study. After revision, the instrument was used to gather aggregate data for unit de- velopment cost calculations which represented a sample of nine courses in two higher education institutions. A.pre- dictive, unit-cost model was then designed and adapted for use as a budget planning guide by instructional planners beginning to project the unit costs of self-instructional materials development. The model's predictability in plann- ing guide form was validated on the unit development costs of self-instructional materials initially developed in a project at a third institution. The validity of the pre- dicted costs was determined by comparing these estimates with the results of a cost analysis of the same project. Findings The findings of this study indicate that a predictive cost model of higher education self—instruc- tional materials development in the biological and physical sciences can be designed with a predictive accuracy within plus or minus fifteen per cent of costs analysis results which are determined at the end of a development project. Walter Grove Chappell Once cost estimates have been made with the aid of the planning guide, it is suggested that project costs may be controlled by identifying cost centers and assigning appropriate administrative responsibility for each center. For effective cost control, however, this responsibility must be assigned to individuals at the point of functional cost incurrance. A MODEL FOR PREDICTING THE UNIT COST OF DEVELOPING SELF-INSTRUCTIONAL MATERIALS IN HIGHER EDUCATION BIOLOGICAL AND PHYSICAL SCIENCES by Walter Grove Chappell A THESIS Submitted to Michigan.$tate University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Secondary Education and Curriculum 1970 Liv") Copyright by WALTER GROVE CHAPPELL 1971 ACKNOWLEDGMENTS The assistance of the following individuals is sincerely appreciated: To Dr. Robert Davis, Dr. John Barson, Dr. Hilliard Jason, Dr. Gardner Jones, and Dr. Desmond Cook for their contributions to the formulation of the research design and systems parameters and constraints. To Dr. James Page for participating in the pilot study and graciously submitting to an interview the night before departing on sabbatical leave. To Dr. Raymond Johnston, Mrs. Kathy Mikan, Dr. Robert Echt, Miss Ruth Allen, Dr. Henry Foth, Dr. Howard Hagerman, and Dr. David Shull for supplying much of the sample data and for their enthusias- tic continued support. To committee members Dr. Erling Jorgensen as Doctoral and Thesis Committee Chairman for his encourage- ment and willingness to attempt a study of this magnitude, Dr. Norman Bell for support and in- sight, and Dr. Charles Blackman for stressing the human side of instructional systems analysis. To Mr. Wilfred Veenendaal and Mr. Martin Kies for supplying graphic production cost data and Mr. .Archie Watson and Mr. Somnath Chatterjee for supplying media equipment cost data. To Dr. Samuel Postlethwait for validating the Budget Planning Guide and providing insight into the history of self-instructional systems for the biological sciences, and to Mr. Dave Moses for critiquing the Planning Guide. ii Acknowledgments . . .continued .A very special note of thanks to my wife Jean for the encouragement, love, patience and time she gave to sustaining a positive mental attitude, and for her editorial and typing excellence. And finally to Dr. Ronald Richards, who had the foresight and courage to give me thinking room. iii TABLE OF CONTENTS CHAPTER I. INTRODUCTION Need for the Study Purpose General Design Overview II. REVIEW OF LITERATURE The USe of Unit-Cost Data Jurisdictional Cost Analyses in Higher Education Unit-Cost Studies in Higher Education Research to Date Cost and Cbst-Benefit Models of Education The Cost of Media Systems Summary of the Pertinent Litera- ture iv FREE 10 13 17 19 23 (HLAPTER TABLE OF CONTENTS - - continued RESEARCH DESIGN Present Systems Analysis Theory Hypothesis to be Tested Population Sample Definition of Terms Assumptions Limitations Instrumentation Data Gathering Data Analysis Model Construction Model Validation Summary ANALYSIS OF RESULTS Full, Relevant, Functional, Job Cost Analysis Results Unit Costs of the Self-Instructional Materials Developed Model I Validation Results Revision of Model I PAGE 25 27 28 28 29 34 38 39 4O 48 49 50 51 53 76 84 90 TABLE OF CONTENTS--continued CHAPTER Revised Model Results Summary V. SUMMARY AND CONCLUSIONS Study Design and Model Construction Results Limitations Conclusions Implications for Future Research BIBLIOGRAPHY APPENDICES A Aggregate Data Interview Instrument B Model I C A.Simulation of the Revised.Self- Instructional Materials Develop- ment Budget Planning Guide D Example Hardware and Software Unit Costs for Self-Instructional Ma- terials Development Cost Planning vi PAGE 93 101 103 104 106 109 109 123 131 140 143 194 TABLE 1.0 1.1 1.3 1.4 1.5 1.6 1.7 1.8 2.0 2.1 3.1 LIST OF TABLES Instructional Unit Development Data Carrel, Demonstration and Special Development Equipment Data Software Production Data Software Duplication Data Faculty, Secretarial and Content Assistant Data Instructional Development and Content Consultant Data Validation and Revision Data Facilities Modification Data Summary of Aggregate Development Data Unit Costs per Self-Instructional Unit Unit Costs per carrel Summary of Aggregate Development Data from the Validation Project that Costs per Self-Instructional Unit and Related Data from the Validation Project vii PAGE 56 58 63 65 68 70 71 72 74 77 80 86 88 TABLE 3.2 5.0 5.1 5.2 6.0 7.0 7.1 7.2 7.3 704 gsr OF TABLES Unit Costs per Carrel and Related Data from the Validation Project .A Comparison of the Unit Cost-Analy- sis Results of the Validation Project with Model I's Predicted Costs and the Revised Model's Predicted Costs General Development Descriptive Data Development Costs per Carrel Development Costs per Instructional Unit Faculty Salaries Content Assistant and Secretarial Wages Production Technician Wages Facilities Modification Costs Number of Students per Carrel Number of Hours per Student per Unit Instructional Strategy Mix Carrel, Demonstration, and Special Development Equipment 93 Carrel Equipment Only Carrel, Demonstration, and Special Development Equipment for Slide, Tape, and/or Film Strategy Mix viii PAGE 89 100 160 161 162 164 165 166 167 168 169 170 171 172 TABLE 7.5 7.9 7.10 7.11 7.12 7.13 8.0 8.1 8.2 8.3 8.4 1451‘ OF TABLES Carrel Equipment Only for Slide, Tape and/or Film.Strategy Mix Slide, Tape and/or Film Software Production Costs Slide, Tape and/or Film Software Duplication Costs Slide, Tape and/or Film Faculty Costs Slide, Tape and/or Film Content Assistant and Secretarial Costs Slide, Tape and/or Film Consultant Costs Slide, Tape and/or Film Validation and Revision Costs Hours of Faculty Time Spent During the Instructional Development Months for Initial Instructional De- velopment Carrel, Demonstration and.Special Development Equipment for Audio Tape Only Strategy Carrel Equipment Only for an Audio Tape Strategy Audio Tape Software Production Costs Audio Tape Software Duplication Costs Audio Tape Faculty Costs ix PAGE 173 174 175 176 178 179 180 181 182 183 184 185 186 187 TABLE 8.5 8.9 9.0 9.2 9.3 10.0 10.1 10.2 10.3 10.4 LIST OF TABLES Audio Tape Content Assistant and.Sec- retarial Costs Audio Tape Consultant Costs Audio Tape Validation and Revision Costs Hours of Faculty Time Spent During the Instructional Development Months for Initial Instructional De- velopment Camera Equipment Unit Cost Carrel Equipment Unit Cost Demonstration and Special Equipment Unit Cost Photographic Lighting Equipment Unit Cost Audio Tape Unit Cost 16mm Reduced to Super 8mm Film Unit Cost 2X2 Slide Unit Cost Photoprint Unit Cost Overhead Transparency Unit Cost PAGE 189 190 191 192 193 194 195 196 197 198 199 202 203 204 LJST OF TABLES-~continued TABLE 10.5 Model Unit Cost 10.6 Miscellaneous Supplies Unit Cost xi FIGURE 1.0 1.1 1.2 LIST OF FIGURES Major Steps in Revised Predictive Cost Model Revised Model's Sequential Steps Basic Prediction Equations xii PAGE 94 95 98 LIST OF APPENDICES APPENDIX PAGE A. Aggregate Data Interview Instrument 131 B Model I 140 C A Simulation of the Revised Self- 143 Instructional Materials Development Budget Planning Guide I) Example Hardware and Software Unit 194 Costs for Self-Instructional Materials Development Cost Planning xiii I . INTRODUCTION Need for the Study Use of self-instructional materials in the biological and physical sciences is increasing. With this increase, faculty, administrators, and funding agencies are finding it necessary but difficult to make accurate budget estimates of development costs. Experience with mediated self-instructional strategies has been limited, due partially to the fact that only recent technological advances have made them feasible. Therefore, comparative data have not had time to accumulate in sufficient quan- tities to be reported for cost predictive purposes . In addition, those data which might be present appear to be buried in departmental requisition orders, with few records kept of actual cumulative costs . Where data have been reported, they often deal with hardware systems exclusive of software development. Higher education's Combined lack of long-term experience with self-instruc- tional materials development and lack of reported cumu- lative data make cost comparisons and estimates diffi- cult for instructional planners who are writing develop- m o o ent progect budgets . A predictive, unit-cost model would 2 therefore be helpful as these planners attempt to project the cost of self-instructional materials development in various curricular areas. Pu ose The purpose of this study is to design a pre- dictive unit-cost model which can be used after behavioral objectives have been initially stated and a slide-audio tape-film or an audio-tape-only media strategy has been chosen by the instructional planner(s). All relevant costs from that point on until the self-instructional units are ready for student use should be able to be predicted within plus or minus fifteen per cent of cost analysis results which are determined at the end of a development project. These cost predictions should include faculty and related personnel costs, carrel equipment costs, software produc- tion and duplication costs, consultant costs, facilities modification costs and validation and revision costs. General Design_ A full, relevant, functional, job cost analysis of development was conducted on selected biological and physical science, self-instructional programs, which have been completed on the Michigan State University and lensing Community College campuses. On the basis of ag- gregate data collected during this cost-analysis phase, a 3 predictive unit-cost model was constructed and adapted so that instructional planners may use the model as a budgetary planning guide. Once adapted, the planning guide was vali— dated on the development costs of a biological science course taught with self-instructional materials at Purdue University. Overview Chapter I of this dissertation has contained a clarification of the need, purpose, and general design of the study. Chapter II contains a review of unit-cost analyses as they have been applied to education, a discussion of cost-benefit models and a review of various studies which attempt to clarify the cost of various mediated in- structional systems. Chapter III reviews current systems theory, outlines the research design in further depth and cfiscusses the construction and validation of the predictive model developed as a result of this research. Chapter IV is a review of the research results and model validation, ‘Mfile Chapter V contains the conclusions and educational implications of the research findings. Appended are the :hnerview instrument, a simulation of the predictive model, and example software and hardware costs for self-instruc- tional sys tems . II. PERTINENT LITERATURE The Use of Unit-Cost Data Private business organiza- tions were the first to use cost analysis on a program- by-program basis.1’2 By 1965 the federal government had instituted program planning budgeting systems (henceforth referred to as PPBS), as a means of analyzing costs for specific programs in its major departments.3 Recently, some educators have been applying a few of these systems analysis techniques such as program evaluation and review technique (henceforth referred to as PERT), linear pro- gramming, and utility/cost sensitivity analysis.4 As educators become more aware of the functional educational system parameters and thereby more clearly 1Francis Keppel, "Operations Analysis --- The Fkomise and the Pitfalls," Socio-Economic Planning Sciences, Vol. 2 (Pergamon Press, 1969), pp. 121-125. Roger L. Sisson, "Can We Model the Educational FTocess?," Socio-Economic Planning Sciences, Vol. 2 (Pergamon Press, 1969), pp. 109-119. 3Keppel, Loc. cit. _ 4C.K. Tanner, "Techniques and Application of Educa- t1Onal Systems Analysis: PERT, Linear Programming, and Utility/Cost Sensitivity Analysis," Audio-Visual Instruction, V01. 14 (March 1969), pp. 89-90. 4 5 identify the educational system within which these para- meters operate, it will undoubtedly become obvious that a systematic analysis of that system and its subsystems must be carried out in detail. These systematic analysis tech- niques will undoubtedly include functional unit—cost (costs related to specific instructional objectives and strategies) analysis as the basis for any eventual cost- benefit analysis. Witmer points to a number of basic reasons why unit-cost studies are essential to higher education de— cision making. "If the general ignorance of policy makers is not sufficient reason for continuing study of the unit cost of higher educa- tion, the many wastes and in- efficiencies in higher educa- tion are."5 Ihxaddition, Dr. Witmer suggests another reason "...for studying the unit costs of higher education is to accum- late information which can be used in making allocations to and within education. The fact is that crucial decisions on such 5David R. Witmer, Unit-Cost Studies (Madison, wiSconsin: Board of Regents of State Universities, 1967), ERIC #ED 013 492. matters as how much to spend on education, and on what pro— grams, are not made in any ra- tional manner. Although much of this irrationality can be blamed on excessive political influences and poor organization for planning and administration, some of that is due to inade- quate data."6 Various authors stress the importance of mak- ing decisions concerning resource allocations on the basis of program unit costs. With these data, work-load requirements and different levels of sup- port may be projected. Combined with desirable quality standards, these data can help develop an un- derstandable, objective budget procedure in which budget decisions are related to clearly defined goals. Knowledge concerning the effects and value of alternative investments is useless unless one also 6Witmer, Loc. cit. 7 knows the relative costs of a1ternatives.79 8’ 9’ 10’ 11’ 12 To further clarify how unit-cost data might be used, Dr. Witmer indicates that "Formulas are designed to form the basis for estimates on future budgetary requirements through the use of pre-determined program cost relationships coupled with estimates of future levels of pro- gram activity. Cost relationships are derived from unit costs which result from cost studies and analy- sis."13 7Witmer, Loc. cit. 8Clarence Scheps, "Systematic Financial Analysis and Budgetary Planning as Aids in the Attainment of College and University Purposes." In Smith, G. Kerry (ed.) Current Issues in Higher Education, 1961: Goals for Higher Education in a Decade of Decision. (Washington: Associa- tion for Higher Education, National Education Association, 1961), pp. 185-188. 9John Dale Russell and James I. Doi, "Analysis of Institutional Expenditures," College and University Busi- ness, 19 and 20 (September 1955 to August 1956), various pages. 1OJ. Harvey Cain, "How Unit Cost Accounting Can Serve the College Field," College and University Business, 32 (March 1962), pp. 63-65. 11M.M. Chambers, Financing Higher Education. (Wash- ington: The Center for Applied Research in Education, Inc., 1963), pp. 84-91. 12Robert W. Peden, "Is There an Educational Industry?" College and University Business, 21 (November 1956), pp 0 50-51 0 13Witmer, Loc. cit. 8 Jurisdictional Cost Analyses in Higher Education Though educators have been generally slow to adopt func- tional, unit-cost analysis and modern systems analysis techniques, such as PERT and PPBS, there is a long history of attempts on their part to develop uniform procedures to record the jurisdictional costs (costs within specific academic administrative units)cfi?education and, speci- fically, higher education. The first of these steps was taken by the Carnegie Foundation in a bulletin published in 1910. Since many of the institutions of higher education were trying to qualify for Carnegie Foundation grants, the Carnegie procedure had the effect of establishing a national accounting system for colleges and universities. In 1917, Christensen headed a committee which recommended that a standard classification of receipts and expenditures be established so that these records were more compatible with the common practice of municipal government accounting. By 1922 Arnett was emphasizing the desirability of separating current, endowment, and plant funds. This was Perhaps the first attempt to get at some of the overall Program categories in higher education. Lindsay and H01- land in 1930 argued that accounting builds an audit trail for fiscal procedures and legal review. As such, they 9 suggested using accounting data in the decision-making process on the basis of teaching loads, class size, floor space and various other comparisons. By 1935 a National Committee on Standard Reports had added loan funds and auxiliary funds to the 1922 proposals of Arnett. The American Council on Education, in 1938, suggested sepa- rating four categories of activity from other institutional operations. These categories consisted of auxilliary en- terprises, student financial aids, hospitals and contract research. In some ways, these recommendations were not far afield from those of Arnett. However, by 1955 Russell and Doi were arguing for the revision of the previously used definitions and for a greater degree of uniformity in data collection for higher education. They further suggested (flyiding the major categories previously recommended in an effort to get more concise data. Presently a national cmmmittee is attempting to revise the College and University Business.Administration.Forms #1 and #2 so that they take on more uniform and concise means of fiscal evaluation.14’ 15 14Witmer, Loc. cit. 15Russell and Doi, Loc. cit. 10 ggit-Cost Studies in Higher Education Though there are, as yet, no reliable inter-institutional means to evaluate unit costs, there is a fairly long history of studies dealing with jurisdictional instructional costs on an in- stitutional basis; these date as far back as 1894. Both Strayer and Elliott did unit-cost studies in 1905, and jurisdictional cost studies in various forms have been con- ducted since. Allen attempted a study in 1914 of the Uni- versity of Wisconsin state system on the basis of cost per full-time student. A study in the state of Washington in 1916 went a step further in attempting to analyze the cost of instruction in various disciplines on the basis of stu- dent clock-hour units. The concept of placing costs on the tmsis of full-time faculty and then cost per student credit tmmr has been in effect for most of the time since that 1914 s tudy .16 For example, the National Committee on Standard Reports for Institutions of Higher Education recommended inl935 that unit costs be computed on the basis of "costs Per full-time-student equivalent" and costs per student 16Witmer, Loc. cit. 11 credit hour.17 In 1938, McNeely reinforced the National Committee's recommendations by suggesting to the United States Office of Education the use of "student-credit- hour" as the basic unit of instruction for departments, schools and colleges.18 By 1960 the basic unit of analysis had not changed. This stability was exemplified by the Florida Cost Study Committee's use of "student semester hour of instruction" and "full-time equivalent student" as its basic units of analysis.19 In the same year, Walker was using "semester-credit-hour" to calculate turnover rates of direct instructional expenditures, permanent plant in- vestment, and indirect instructional expenditures.20 Minor 172National Committee on.Standard Reports for In- stitutions of Higher Education, Financial Reports for Col- leges and Universities. (Chicago: University of Chicago PTess, 1935), pp. 177-249. (Reprinted under the title 99mputation of Unit_Costs. Washington: American Council on Education, 1955) . 18John H. McNeely, University Unit Costs, U.S. Office of Education, Bulletin 1937, No. 21 (Washington: ELS. Government Printing Office, 1938), 35pp. 19Florida Cost Study Committee and the Office of the State Board of Control, A Manual for Analyzing Uni- versity Expenditures by Function. Revised 1960-61. (Tallahassee: State Board of Control, n.d.) 66 pp. 20Ernest W. Walker, "To Measure Operating Ef- :ficnency," College and University Business, 29 (August 1960) , pp. 24-29 . 12 alterations were made in 1962 by the Executive Secretary's Office of the University of Montana System of Higher Educa- tion in using "student credit hour registered for" as the basis for its study.21 It was not until later in that year that noticeable changes in Allen's 1914 recommendations appeared in the form of Tyndall and Barne's study in which they used "weekly teaching hours," "weekly student hours," and "semester hourly rate" for calculating workload and costs.22 Several authors have raised concern about the validity of unit-cost data and its use in appraising higher education spending. Rand, for one, cautions that precision and accuracy need to be used when classifying and distri- buting expenditures, and determining the number of cost units, while Hull raises concern with the misuse of unit- 3 cost data by administrators and faculty.2 He points out 21University of Montana System of Higher Education, ,Master P1an.Study: Status Report. (Helena, Montana: Office of the Executive Secretary, State Capitol, Room .139, October 8, 1962), pp. 51-53. 22D. Gordon Tyndall and Grant A. Barnes, "Unit Costs of Instruction in Higher Education," The Journal of Experimental Education, 31 (December 1962), pp. 114-118. 23Edson R. Rand, "If Unit Cost Calculations Are to be Valid,"_ College and University Business, 19 (August 1955), pp. 25-26. 13 that unit-cost data are quantitative and not qualitative and that present quantitative measures of faculty perfor— mance are not accurate. Further, the use of unit costs may give the false impression that costs are the most im- portant variable in the instructional setting.24 Pike also raised concern over the misuse of unit-cost data and proceeded to provide faculty with a means of analyzing standard costs and comparing them to actual costs.25 Research to Date Though a number of studies claim to deal with unit costs, it is quite obvious in reading their findings that their major parameters of analysis are jurisdictional rather than functional and are therefore far too general to be applicable to PERT, PPBS or other nmans of analyzing program costs, and eventually cost- benefits. Most of the studies cited deal with implemen- tation costs per department and rarely give much more than passing attention to comparative costs between in- stitutions or specific developmental costs for the 24L.E. Hull, "Pitfalls in the Use of Unit-Cost SStudies," Journal of Higher Education, 32 (October 1961), .PP- 371-376. 25Walter L. Pike, "What You Can Learn from Unit ‘305t3:" College and University Business, 37 (July 1964), pp. 39‘41- 14 instructional strategies used. This absence of develop- xental cost and functional data may be due in part to the fact that modern, media systems require some instruc- tional development, and the standard lecture-lab approach often operates on the principle that relatively little in- structional development is needed and therefore little is accounted for in the cost studies. For example, School Management Magazine has been attempting to gather cost data since 1962 on various types of instructional strategies and has developed a means by which they feel they can compare costs across K-12 dis- tricts. Their basic unit of analysis is expenditure per pupil. This unit of analysis is then used to determine a cost-of-education index. However, none of their studies apply to higher education, or include more than the most general reporting variables.26 A very extensive unit—cost study was done by Wit- mer for the University of Wisconsin system in which he attempted to use cost studies to identify information re- lated to policy formulation, to evaluate efficiency, to 26"How to USe the Cost of Education Index, 1968- 69," Sghool Management, Vol. 13 (January 1969), pp. 52- 54. 15 study alternatives and to prepare for the use of PPBS. His findings were reported in terms of contact periods, credits, major programs, curriculum, and students. The major analysis variables included faculty salaries, number of students per section, units per day, and the instruc- tional strategy mix. No mention was made, however, of at- tempts to analyze instructional development costs per course.27 In 1966 a study similar to the Wisconsin study was carried out for the State of Michigan. The basic unit of analysis for this study was cost per student credit hour kw department and student level. Only the eleven state universities were involved in this study and a vast number cfi'assumptions were used, making the data difficult to compare to other colleges within the state and in other parts of the country.28 This and other research findings suggest that attempts to compare costs across institutions may be quite difficult. For example, Witmer reports that an attempt to apply the 1935 National Committee's standards 27Witmer, Loc. cit. . . 28Linn Peltier, Institutional Research Office, Michigan.State University; interviewed January 1970. 16 to a study of nine universities, found great variation from institution to institution. He cites the 1966 study by Kilzer who found that among seventy-eight junior colleges which presumably had the same mission some had costs which ran as high as six times those in others.29 Attempts to compare costs to quality between in- stitutions have also met with ambiguous findings. Reeves and Russell, for example, mention the need to correct cost data to account for enrollment size variations before using costs as an indicator of quality in making inter-institu- tional comparisons.30 In analyzing the relationship be- tween cost and quality, Ikenberry notes that course pro- liferation is a detriment to quality programs and that class size is the crucial variable in reducing costs. As such, he contends that lower instructional costs per student need not lead to a proportional decrease in in- structional quality.31 29Witmer, Loc. cit. 3OIbid. 31Stanley Ikenberry, "Instructional Cost and Quality," College and University, 37 (Spring 1962), pp ° 242-250 0 17 Cost and Cost-Benefit Models of Education Model build- ing is a frequent outgrowth of systems analysis. With the analysis of educational systems have come recent attempts to develop cost models and cost-effectiveness models of instruction. Most of these are based on fairly large para- meters and do not deal with the specific functional concerns of individual courses. Likewise, Francis Keppel cautions that there is little sense in trying to build a model of broad scope with inconclusive «or incomplete data. Furthermore, once any model is developed, it needs to be adapted to the system being analyzed.32 Model building procedures have been used by gui- dance and counseling professionals for years. Predictive models used as a means of advising prospective employees of their probable area of employment success are good examples of the outgrowth of these procedures. As part of the model-building process, linear regression and step- wise, multiple regression equations have been used to determine the relevant predictive variables. However, as Cronbach and others point out, discriminate and step- Wise, discriminate analysis may be much more effective in determining the relative success of an employee in one 32Keppel, QB: cit., pp. 121-125. 18 job or another.33: 34: 35: 36 Discriminate and regression analysis techniques may have application in building predictive models of successful instructional development. However, major problems are encountered when attempting to use these pro- cedures in analyzing instructional systems in that the criteria for successful instruction are not clearly de- fined, and to carry out the longitudinal studies necessary to indicate whether benefit has actually been derived from the instruction developed would take considerable time and a large data base. One large-scale attempt was made to determine the cost-effectiveness of the jprojects funded during 1965-66 Inder Title I of the Elementary and Secondary Education Act. Zhlthis case the systems model for cost-effectiveness of the 33Henry Borow, Man in a World at Work, (Houghton, 34Wm. W. Cooley and Paul R. Lohnes, Multivariate Egocedures for the Behavioral Sciences (John Wiley and SODS, 1965), pp. 17-59. 35Lee J. Cronbach, Essentials of Psycholggigal ISEEEEE; 2nd Ed. (Harper and Brothers, 1960) pp. 247- 268. 325-359. 36N.R. Draper and H. Smith, Applied Regression Analysis (John Wiley and Sons, 1967)- 19 resource input of Title I funds was broken down into five sub-models: the schools affected, the community, the instructional process, the costs, and the cost effective- ness. These five major sub-models were used as input to a cost-effectiveness simulation model and the predicted outcome of the resource input was projected on the basis of data contained in the five sub-models. It is worth noting that most of the variables used in this attempt at cost-benefit analysis were operational, rather than capi- tal costs. As such, most of the cost-benefit outputs represented the implementation and operation of new pro- grams, rather than the development of materials to go into those programs.37 lye‘Cost of Media Systems There have been several recorded attempts to ascertain the unit cost of mediated instruction. One such attempt was the 1968 cost study of the Michigan State University Closed Circuit Television System in which the units of analysis were divided far below those of the previously mentioned Wisconsin and Michigan studies. Rather than using general departmental 37Clark C. Abt, AgCost-Effectiveness Model for the AE§1YSiS of Title I ESEA.Project Prqposals, Part I-VIII, ERIC #ED 014 018. 20 records as the main source of data, interviews were con- ducted, and records from various sources were analyzed to calculate the cost of producing, distributing and re- ceiving the ITV programming being produced at Michigan State. Very little of this study, however, was inclusive of developmental costs of the instructional units pro- duced.38 Another attempt to establish unit costs for media systems was made by the General Learning Corporation. In it data were compiled for numerous instructional strate- gies involving media components. Their data are represen- tative of a number of years, and express trends rather than specific unit costs. Their units of analysis were the average length of the instructional unit, the type of program, the quality of the production, and the physical format of the medium used. Facility costs were ignored unless the facilities were being built specifically to housethe instructional strategy utilized.39 This ; ! 38Gardner Jones, CCTV Cost Study, 1968-69, (East Lansing, Michigan: Michigan State University, Edu- cational Development Program) Project #1—073. 39General Learning Corporation, 92st Study of Edu- cational Media Systems and Their Equipment Components: jTechnical Report (ERIC #ED 024 286, 1968). 21 procedure was undoubtedly wise on their part, as facility costs vary greatly from region to region, depending upon the trade unions and the availability of materials.40 In categorizing their costs, the authors of the General Learning Corporation study separate costs into pro- duction, distribution and reception. They further classify costs as capital (those which are incurred during the ini- tial planning and development stages) and operational (those costs which are incurred on an annual basis to imple- ment the programs). Production costs were subdivided into duplication costs and general production costs.41 In analyzing the detailed, technical report, published to coincide with the General Learning Corpora- tion study, one finds that media cost estimates were based primarily on the equipment systems involved, rather than the planning and production. Although some account was made of instructional planning as simple line items in the cost itemization, the software production did not include faculty time and other key components in the 4O"Ten Deceptions in Building Cost Comparisons," Overview Management (July 1962). 44General Learning Corporation, Guidelines for Retermnichosts of Media Systems (ERIC #ED 024 273, 1968). 22 development process.42 After searching numerous sources for possible re- lated studies, the General Learning Corporation research and a cost analysis of instructional systems by Jones were the only references found which gave any indication of the relative unit costs of various instructional strategies in- volving media. These data do serve, perhaps, as first approximations for some of the major mediated instructional systems such as ITV, dial access, and 16mm film libraries, but they do not include any of the costs related to self- instructional materials strategies.43’ 44 The only closely related reference found was reported in School Management of costs for media systems which dealt strictly with the media equipment expenditures for K-12 districts based on . . . . 45 expenditure-per-pupil unit comparisons. 4zGeneral Learning Corporation, Loc. cit. 43Ibid. 4«Gardner Jones, A Procedural and Cost Analysis EStudy of Media in Instructional Systems Development, (East Lansing: United States Department of Health, Education and welfare, Office of Education Grant #OE- 3-16-030, September 1, 1965). 45"Cost of Audio-Visual Instruction, 1962-63, 1968-69," School Management, Vol. 12 (October 1968), pp . 67-72. 23 Summagy of Pertinent Literature To summarize, it appears that private business was the first major segment of our society to use cost analysis as a means of analyzing program development. The federal government and local and state branches of government were quick to follow with the advent of numerous publicly funded contracts. Only late in the 1960's did some educators attempt with much success to use sophisticated cost analysis procedures in their budget and program evaluation. Unit costs are the basis of most cost analysis procedures, and permit such techniques as PERT, PPBS and linear programming to be successfully utilized by the systems analysts if they contain functional program cost data. Cost/benefit analyses may follow and be used in conjunction with these various analysis procedures, but in the field of education the benefits section of the analysis are still quite speculative. Various attempts have been made to categorize jurisdictional costs in higher education since the Carnegie Foundation standardized its granting procedures back in 1910. These analyses are in reality not of functional units which lend themselves to instructional systems analysis. Furthermore, the unit costs of instructional development are generally unknown for any 24 given strategy. Instructional model building as an outgrowth of educational systems analysis is still in its infancy. Predictive models have been built and used by occupational counselors for a number of years and have been derived with the aid of a variety of sophisticated statistical equations, but predictive models of instructional costs are quite limited. One major attempt in this area was made, however, in relation to the Elementary and Secondary Education Act, Title I programs, but its results were only applicable to K-12 districts and do not take into consideration the con- straints of the higher education instructional system. First approximations of the functional costs of implemented mediated instructional systems have also been exclusive of developmental costs and have been geared primarily at the K-12 level. III. RESEARCH DESIGN Present Systems Analysis Theo£y_ Systems theory may be clarified by describing the flow of activities involved. The first step is to define the system under study. The next step involves the development of a theory to explain the system defined. From this point, research is necessary to clarify the constraints and parameters of the system. On the basis of the data derived from such research, a model may be designed in an attempt to predict the interaction of these parameters and constraints. Once the model has been put to the predictive test, an evaluation of its validity and reliability must be carried out. After model valida- tion, the procedure starts again with a re-definition of the system and appropriate modification of the theory.46 Cost-benefit analysis is a means of studying two basic aspects of a system. Before benefits can be as- sessed, it is necessary to analyze the system's costs. Benefits are much more difficult to determine, due to the 1en9th of time necessary to carry out the benefits analy- sis, as well as the subjective nature of the criteria 6Sisson, Loc. cit. 25 26 used to indicate what has been beneficial. For these reasons, cost benefit analyses become quite lengthy, and at present, their results are general in nature.47 Some authors have pointed out that before systems theory can function on an operational level, it is neces- sary to develop analysis techniques which can be used to clarify cost data. Once these techniques are refined, then PERT (program evaluation and review technique), PPBS (pro- gram planning budgeting systems), and other means of sys- tems analysis can be utilized with maximum effective- ness.48 For example, since PPBS links long-range planning udth fiscal planning, it is essential that adequate data be nude available before decisions can be made as to either fiscal or long-range goals.49 This need for concise data requires that when analyzing an educational setting, the institution's objectives and behavioral objectives for 47Alexander Mood and Richard Powers, Cost-Benefit Analysis of Education (1967) #ED 012 519. 48Desmond L. Cook, Interview re: PERT and Systems £23l2§i§_épplications to Instructional Develqpment (Columbia, Ohio: Ohio State University, October 1969). 49Charles J. Hitch, "What are the Programs in Planning, Programming, Budgeting," Socio-Economic Plannipg §2$EEEE§J Vol. 2 (Pergamon Press, 1969), pp. 465-472. F~ 27 specific instructional programs must be taken into account. Without this balance between institutional and behavioral objectives, the cost of attaining the latter will have little meaning.50 Hypothesis to be Tested The system under analysis has been defined as those functions relating to the develop- ment of self-instructional materials in the biological and physical sciences at the higher education level. As stated above, there is a body of theory relative to systems analysis, but few attempts have been made to analyze the developmental aspects of self-instructional systems. Therefore, the major hypothesis of this study was that the parameters and constraints of the system which encompass the development of biological and physical sci- ence self-instructional materials in higher education can be analyzed, and that once these parameters and constraints have been identified, unit-costs of development can be computed and a predictive unit-cost model for use in future cost estimates can be constructed and validated. 50Paul Harmon, "Curriculum Cost-Effectiveness Evaluation," Agdio-Visual Instruction, Vol. XV, No.1 (January 1970), pp. 24-26. 28 Population The population to which the results of this analysis might be generalized includes those self- instructional material development projects which occur in the biological and physical sciences at the higher educa- tion level. Sample A selective sampling technique was used to select four biological science and two physical science courses which are taught primarily by self-instructional materials at Michigan State University, and one biological science and two physical science courses which are using self-instructional materials at Lansing Community College. The main selection criteria were: (1) the program chosen for study had to be in either the biological or physical science curriculum and (2) the program had to have a history of continued development and use by students for at least two years. Programs at two separate institutions were chosen to help reduce the influence of parameters and constraints which might be limited to a specific insti- tution. To provide a valid test of the unit-cost model's general predictability across institutions, a validation (n? the model was carried out on the development costs of 29 UKaIntroductory Botany course taught with self-instruc- tknml materials at a third institution, Purdue University. {Definition of Terms The following terms are used in the s tudy: 1. 4. Fully relevantLyfunctionali job cost analysis: an analysis of those direct and indirect func- tional costs which were considered by the de- velopment director and the researcher to be relevant to the initial self-instructional materials development project. Prqject unit costs: the total cost of the self-instructional materials developed, di- vided by the number of credit hour equivalents which the materials replaced in the previous instructional strategy. Functional costs: costs related to the achieve- ment of specific instructional objectives and strategies, including both development and im- plementation aspects. ,gyrisdictional costs: costs associated with the administration of various institutional sub- divis ions . 5. 30 Development upit ggsts: those separate total costs for equipment, software development, software duplication, faculty, consultants, content assistants, facilities modification or validation and revision, divided by the number of units developed and/or the number of carrels used. Self-instructional materials: those self-paced media used by the student to learn course con- cepts with little or no direct assistance from faculty or teaching assistants. Developmental production costs: those expendi- tures for the demonstration and special develop- ment equipment, software production, faculty, consultant and content assistants time, and facilities modification which were incurred in the process of producing the first copy of an instructional unit. _Qevelopmenta1 duplication costs: those costs incurred in duplicating the originally developed lmaterials and providing carrel equipment for the ‘use of those materials. 9. 10. 11. 12. 31 Ipstructionaleppit: any self-contained pack- age of materials used to teach a given concep- tual segment of the course. For example, a unit might contain any combination of films, tapes, slides, hand-outs, and demonstration materials used as integral parts to teach the concepts of a particular course segment. lpitial development: that development which takes place prior to the first student use of the materials being developed. Once students are using the materials developed for course credit, costs are not assigned as being part of the initial development. Self-instructional setting;_ the instructional facility or location where students use self-instructional materials with little or no formal contact with faculty or teaching assistants. Non-departmental funds: those financial re- sources which are derived from college, uni- versity, foundation or federal grants. These funds are not assumed to be a part of the nor- mal departmental operating budget. 13. 14. 15. 16. 32 Materials for student use: when the faculty developing the self-instructional units are satisfied that the materials are sufficiently valid and reliable to instruct students with academic credit in a self-instructional mode, then the materials are considered ready for student use. Student station: any location within the in- structional setting where a student can use the materials developed. These student sta- tions may be individual areas containing in- dividually assigned hardware and software, or locations at a demonstration area. Carrel: a student station containing an assigned number of hardware and software com- ponents which are used to transmit the concepts in a specific instructional unit. Demonstration area;_ student stations which are not individually assigned but are meant for all students in the course who are pro- gressing through a given unit of self-instruc- tion. Demonstration materials may be assigned to individual carrels, but those materials in 17. 18. 19. 33 the demonstration area are relatively few in number and are meant for all students to observe. Models and demonstration equipment: special equipment which is considered as part of the unit costs if it is specifically developed or purchased for the self-instructional unit to which it is assigned. Facility modification costs: any costs resulting from a change in the instructional setting which is necessitated by the development of the self- instructional materials. Such modifications may include wiring, lights, special carpet and the removal of existing equipment. However, the cost of installing carrels is included under developmental duplication costs rather than as facility modification costs. Production Technicians: any individual assigned the responsibility of producing or contributing to the production of a medium, such as cameramen, directors, editors, sound and light technicians, artists and other non-content specialists. 34 20. d d or ocessed software: any software which is produced and/or processed by members of the materials development team or by on-campus production service centers. 21. Ypiidatiop; an evaluation of the validity of instructional materials prior to the use of said materials in the instruction of students for academic credit. Agsgmptiog As may be apparent from the previous re- \dew of systems theory, this attempt to develop a predic- tivezmodel for the unit-costs of self-instructional ma- terials development was a first approximation. Many other Significant variables will obviously be found to operate as other parameters and constraints not included in this study are taken into consideration. Some of these parameters and constraints might in- <flude the psycho-sociological characteristics of the in- dnddua1(s) carrying out the development, and the academic mndronment in which that individual or group of indivi- dmfls is located. More specifically, motivational factors, I 0 I 0 ewards, tenure status, age, preVious instructional de- veloPment experience, the proportion of non-departmental 35 development funds and the quality of technical and consul- tant assistance might all influence the costs as well as the success of the development. However, these parameters and constraints were assumed to have an insignificant in- fluence on developmental unit costs . There may also be some question as to a model's reliability‘ based on a relatively small sample size. (Seven development projects comprise the sample in this study.) fbmevery conducting research prior to the initial construc- tion of’a predictive cost model requires that the original tmxameters and constraints be inclusive of a fairly genera- lized view of the system and, if possible, be representative cm'the costs, irrespective of benefits. The first design cm'the model may, therefore, be attempted, following the analysis of data from a limited, heterogeneous sample. Future research may then increase the sample size to test “Haassumptions of the model and to analyze the on-going Process and success of the self-instructional materials develoPfi‘d. Using common business management work-sampling techniques, for example, might provide more accurate fa— C‘flty and consultant time data as it relates to the development process. 36 51 Other assumptions which were made in relation to this study include the following: 1. Instructional development is a continual pro- cess involving the initial development phase, followed by student use, evaluation and further revision. For the purposes of this study, only the initial development phase is being analyzed. Developmental costs are divided into those costs incurred in the development, equipment procurement, and production of a master copy of a given medium. Duplication costs, a subset of the total development, are those costs incurred in reproducing the master and procuring additional equipment units to satisfy the instructional needs of the students en- rolled in a given course. The self-instructional material developed must replace, or add content to, the instructional strategy traditionally used to teach the course 51Charles H. Backstrom and Gerald D. Hursh, Sur- EQLBEEEEEQ.(Northwestern University Press, 1963), pp. 28-35. 37 concepts. The developmental costs for self-instructional materials in the biolOgical sciences are simi- lar to those costs of development in the phy- sical sciences in that both areas use a variety of equipment and expendable supplies in the instructional process, and their concepts are relatively easy to visualize. There is a significant difference in the amount of, and hence the cost of, instructional ma- terials used by those curricular areas which are highly visual, such as the biological and phy- sical sciences, and those areas of the curri- culum which are more conceptually abstract, such as the humanities. The experimental sample containing projects which developed self-instructional materials for five biological science courses and four physical science courses is sufficient, as an initial sample, to represent the costs in- curred in the development of self-instructional materials for these curricular areas. 38 Faculty responding to specific interview ques- tions give a reliable estimate of the time which they and their colleagues spent in the development of the materials under study. A fair appraisal of each unit's software pro- duction quantity and costs can be derived from an accounting of the total number of software items produced during the initial development and dividing this total by the number of instructional units initially developed. Commercial software production and processing costs may be determined once the materials pro- duced have been carefully described to appropriate vendors. Limitations The research design, and therefore the results and conclusions of this study, are subject to the following limitations: 1. The initial development of self-instructional ma- terials is the unit of analysis, including the re- finement of enabling objectives, the selection of appropriate media strategies to meet the objec- tives, but not the creation of a climate of acceptance among non-development faculty, or the 39 implementation and operation of a self—instruc tional setting. (It is recognized that these ac- tivities are not discreet processes and that over- lapping costs will necessarily exist.) 2. The experimental sample is drawn from biological and physical science curricular areas within higher education, and therefore the results do not at- tempt to generalize to broader spectra of curri- cular or academic levels. 3. Use of the predictive model for self-instructional materials development at the higher education le- vel for biological and physical science course de- velopment will project costs to the extent that the input data and step-wise decisions made in using the budget planning guide accurately reflect local conditions. Furthermore, the instructions in the planning guide must be used to adapt the accom- panying tables so that they will reflect the economic environment in which the model is to be used. IEEEEEESEEEEEQE. Interviewing the major faculty de- val°PerS in charge of each of the selected development PIOjects was considered the most accurate method of . 40 gathering most of the aggregate cost data. Therefore, an interview questionnaire was designed and a pilot study to validate the interview questions was conducted with the faculty who developed the self-instructional materials used in the media utilization section of the educational methods course for teacher preparation at Michigan State university. As a result of that pilot study, modifications were made in the interviewing instrument to help clarify the few ambiguities which arose from the original instru- ment, and to expand those areas where more data were needed to make a thorough analysis of all relevant costs. (Appendix A contains a copy of the interviewing instrument.) Qgta Gatheripg’ Once the interview items were tested, full, relevant, functional job cost data were gathered during interviews with the key faculty developers involved udth each of the following completed self-instructional nmterials development projects at Michigan State University: anatomy, soil science, nursing, physiology, Lyman Briggs C011ege (beginning biology), and biochemistry. Similar data were gathered on the three basic science courses twingtself-instructional materials at Lansing Community (kalege- (Full, relevant, functional job costing was 41 chosen rather than marginal or period process costing, due to the terminal aspects of the initial development pro- jects. If implementation or operational unit costs had been the units of analysis, then marginal or period pro- cess costing would have been more appropriate.)52 Additional supportive data were compiled from cata- logue price quotations, funding agency proposals and media production specialists. The aggregate cost data collected from the faculty and supportive sources were then divided into two separate groups consisting of developmental costs and duplication costs. Developmental costs were summarized under five major categories: (1) equipment, (2) soft- ware, (3) faculty and assistant, (4) consultant, and (5) revision costs. Duplication costs included those expen- ditures which resulted from the production of duplicate sets of the original software. (Since most previous \udt-cost studies of higher education courses have been relative to jurisdictional rather than functional costing, flxdr main input data have been faculty salaries. However, 52Jones, ALProcedural and Cost Analysis Study of Nbdia in Instructional Systems Development, Loc. cit. 42 for those experienced with the development of self-instruc- tional materials, it is obvious that faculty salaries -—- and equipment costs for that matter -—- are only a part of the parameters and constraints to be included in a cost analysis of the development process.) 53 Those parameters and constraints for which aggregate data from each project were gathered and later analyzed in determining the unit costs of development were the following: Developmental Production Costs Instructional Units: (1) The number of self-instructional units (2) (3) (4) (5) (6) 53 used with students after the initial de— velopment The date the initial development started The date the initial development was com- pleted for student use The average student time per week spent in the self-instructional setting The number of total credits offered for the course in which self-instructional materials were used The percentage of the total course credits taught by the self-instructional materials . Jones, A Procedural and Cost Analysis Study 2E_!Pd1a in Instructional Systems Development, Loc. cit. 43 developed Software Production Costs: (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) The average number of slides developed per instructional unit The number of silent films developed as opposed to the number of sound films de- veloped The number of films developed for reel- to-reel projection as opposed to the num- ber developed for cartridge projection The average running time per film The average cost per film The average number of photoprints developed per instructional unit The size of the photoprints produced The number of photoprints per unit in the student handouts, as opposed to the number per unit in centralized displays The average cost per photoprint The average number of audio tapes developed per instructional unit The number of tapes on cassettes as opposed to the number on reels The average audio or video tape playing time per unit The amount and type of other demonstration materials or equipment developed or pur- chased for the instructional units, but not included in the carrel setting The average cost per model built for in- structional use 44 (21) The average number of student handouts de- veloped for each instructional unit (22) The number of pages included in each hand- out (23) The type of graphic illustrations, if any, included in the student handout (24) The average cost per student handout (25) Whether the software was locally produced or purchased by type of software and quan- tity in each category (26) Whether the software locally produced were processed locally or whether they were pro- cessed commercially (27) The number of production technicians em- ployed during the initial development of the software (28) The salary these technicians were paid (29) The amount of time they spent on the initial development (30) The financial support received from non- departmental sources, as opposed to the support from departmental operating funds (31) The percentage of the total development costs received from non-departmental sources (32) The total cost of the software developed Eculty and Content Assistant Costs: (33) The number of faculty who worked on the development of the self-instructional materials (34) Their tenure level (35) (36) (37) (38) (39) (40) (41) (42) 45 The number of months per year of their appointment Their yearly salary The amount of faculty time assigned to the development of the instructional units The number of content assistants, such as graduate assistants and secretaries, who were used during the development of the instructional units Their salary level The amount of time they were assigned to spend on the initial development The amount of faculty time required to train the carrel room operators on the use of the equipment and distribution of materials The number of self-instructional units pre- viously developed by the major faculty developer Consultant Costs: (43) (44) (45) (46) (47) (48) The number of consultants used during the development of the instructional units Their area of expertise Their tenure level Their salary The amount of time spent by each consul- tant during the development The amount of travel cost incurred by consultants coming to the development site 46 (49) The amount of travel cost incurred by con- tent faculty traveling to exemplar projects Revision Costs: (50) The type of validation, if any, conducted on the materials prior to student use for course credit (51) The per cent of the software revised as a result of the validation (52) The method of production and processing used for materials revised as a result of the validation (53) The number of faculty and staff involved in the validation (54) The amount of time they were involved in the validation (55) The number of students who validated the materials, and the rate paid per student, if any (56) The estimated revision costs (57) The cost of modifying facilities as a result of the initial self-instructional materials developed Those parameters and constraints for which aggregate data from each project were gathered and later analyzed to determine the unit costs of producing duplicate sets of 1Ihe originally produced software and providing carrel equip- ment for the students enrolled were the following: 47 Developmental Duplication Cos ts : Software Duplication Costs: (1) The number of students per term the initial (2) (3) (4) (5) (6) (7) (8) self-instructional materials were developed to teach The number of students per term using the materials developed at the time of the interview The number of carrels the materials were originally planned to fill The number of carrels using the initially developed materials at the time of the interview The number of carrels which had been and were at the time of the interview being used for review units The number of hours per original instruc- tional unit for which the carrels were available for student use The frequency with which the units were rotated in the carrel facility and the type of rotation schedule The duplication cost for the software materials developed Eqpipment Costs: (9) (10) The type and model of instructional equip- ment and/or carrels used at each student station and in any adjacent demonstration area The amount of each type of equipment used in the student station, adjacent demon- stration area or as special development equipment 48 (11) The per item costs for each type of equip- ment used (12) The equipment purchased new, as opposed to the equipment drawn from existing sources (13) The total cost of the instructional equip- ment (14) The list-price discount given for the equipment purchased gpta Analysis As noted in the literature review, unit costs in higher education have traditionally been recorded as costs per student credit hour, cost per full-time equivalent student or a variation thereof. These units of analysis serve some utility in making jurisdictional com- parisons and decisions. However, the purpose of this study was to construct a predictive unit cost model of self- instructional materials development. As such, the data and unit costs had to reflect functional, rather than jurisdictional, relationships. Further, since develop- mental rather than implementation or operational parameters and constraints were necessary in the construction of this predictive model, greater emphasis had to be placed on costs per carrel, and particularly cost per instructional unit, than cost per student credit hour. If the traditional units of analysis had been chosen, there would be little 49 opportunity to compare and use developmental project costs, since an increase or decrease in student course enrollment would affect only the duplication, implementation and opera- tion costs, not the development costs which are content and strategy related. Therefore, the aggregate data from the faculty interviews and supportive sources were grouped under the development and duplication cost parameter and constraint categories and sub-categories. Totals for each category by project were computed, and the mean for all projects by category calculated. Unit costs per carrel and unit costs per instructional unit were then computed from each project category total, and the mean unit costs per carrel and instructional unit recorded for all projects. Model Construction A predictive, unit-cost model was designed after the data were analyzed. A budget planning guide was then constructed, based on the model. High, low and average unit costs per category were grouped and cri- teria written for the user selection of the proper value for each category. Opportunities for the planning guide user to modify estimates due to inflation factors and local parameter or constraint variations from the aggregate data 50 were also made. The planning guide was designed to permit the user to answer a variety of questions about his local instructional and economic conditions, decide which value under each category is appropriate for his development pro- ject, modify that value to fit his economic environment and multiply the subtotals by the number of carrels and instruc- tional units to be used to determine the total development cost estimate. The user is also able to make an estimate of the amount of faculty time and the number of months needed to develop the materials. Model Validation A validation of the model (in planning guide form), and the cost selection criteria was conducted on the Introductory Botany course at Purdue University. The validation was conducted in conjunction with a cost analysis of the initial development costs of the same pro- ject for which the model was attempting to predict. The results of the cost analysis were then compared to 1me model predicted costs to determine the validity of the model. The two-stage validation used the same para- meters and constraints to analyze the development costs cfi'the Purdue project as were used in analyzing the Mi- chigan State and Lansing Community College projects . It 51 was decided that the model would be considered valid if it predicted the Purdue development costs within fifteen per cent of the costs resulting from the cost analysis of the same project. This value was chosen since: (1) correla- tions for predictive purposes are seldom considered signifi- cant unless they are greater than plus or minus .60, (2) the sample from which aggregate data were collected was relatively small, thus leaving the possibility of skewed results due to biased sampling, and (3) the nature of the input data in some cost categories was based on faculty estimates of time. Further, an override of ten per cent is often written into contracts to cover unexpected costs. It was therefore decided to accept a model with predictive accuracy of plus or minus fifteen per cent so that a first approximation could be derived by an instructional planner and leave the task of developing greater accuracy for re- search based on more exact input data. Summary A pilot study was conducted to standardize the interviewing instrument. Following this phase of the research, data were gathered from seven completed self- Ihstructional projects representing five biological science mnuses and four physical science courses. Interviews 52 (were conducted with the major faculty developers for each Iproject, and software production specialists. Additional ssupportive data were gathered from equipment specialists and hardware catalogues. The resulting aggregate, full, rele- 'vant, functional job cost data were analyzed and a predic- tive model of the self—instructional materials development costs was constructed and adopted as a budget planning guide. Validation of this model took place at Purdue University on the self-instructional materials used to teach the Introductory Botany course. IV. ANALYSIS OF RESULTS Full, Relevant, FunctionaiL Job Cost Analysis Results; Gardner Jones once described the frustration of gathering instructional cost data by comparing the process to attempts to put a rubber band around a cloud. The sense of frustra- tion behind his analogy must be shared by all those who attempt to analyze instructional systems and begin to place unit costs on the out-puts of the instructional process. Though the creative aspects of the instructional process balk at being standardized by unit-cost analysis, attempts to clarify the cost of various strategies for comparison purposes and decision making are nonetheless becoming more and more essential as attempts are made to use limited resources more effectively. Craig Johnson noted a ray of hope when he reported that "While no one argued that evalua- tion of education ought to be restricted to cost effect- tiveness, many programs found that cost data could be collected and analyzed meaningfully to evaluate instruc- tion projects ." 54 54F. Craig Johnson, An Evaluation of Educational De- velgment Progr_ams in Higher Education. USOE Grant No. OEG-O-8-070114-1856 (010), Project No. 7-E-ll4 (1968). 53 54 "Because of differing relationships of individual expense categories to the volume (of production in industry, or of enrollments in education) it is necessary,for planning purposes, to make studies of individual expense categories separately."55 Therefore, to arrive at an accurate appraisal of the development costs for self-instructional materials, separate expense categories were established and full, relevant, functional, job costs were analyzed in each cate- gory. Full costs imply both direct and indirect costing of the instructional development, and that labor and nmterials costs to carry out production and duplication :h1"service-oriented" audio-visual centers should be in- cfluded, even though the department for which the services were rendered was not charged. Faculty costs are also :hxduded in the full costing analysis, due to the fact that their involvement in the development process removes them from other productive activity. Relevancy indicates that only those direct and indirect costs which were Specifically relevant to the development process were in- cfluded. For example, indirect overhead expenses which would 55Jones, A Procedural and Cost Analysis Study of EEdia in Instructional Systems Development, 9p, cit., p. 28. 55 be assumed by the institution, regardless of the facility use, were not included in this analysis. Functional de- velopment costs, as opposed to jurisdictional costs, were also isolated in gathering and analyzing the data. Finally, a terminal point was sought for the analysis. Therefore, job or project costs for the initial develop- ment phase were used. Once the aggregate cost data were compiled, the parameters and constraints in each expense category were analyzed. As Table 1.0 illustrates, there were numerous variations from project to project in the number of units developed, the length of the development period, and the way in which the self-instructional materials were used. Projects 4 and 5, for example, used the materials they initially developed in addition to the existing ins truc- tional strategies, rather than as replacement for those strategies. Project 7 combined the faculty talents of a full department and developed three courses simultaneously, While the development on the rest of the projects studied Was carried out by primarily one faculty member on each Project. In addition to the way in which the initially developed materials were used, and the talents were used on the development team, variations in the length of the 56 one: mHmoOP pcwmupmcoo can umpoemumd .mmusoo nod mommum>m pomammu ow Adv mean an popw>ep .mmmusoo conundmm Amv ounce uom madeumpma HMGOfiposupmcwnmamm mo unmadoam>mp mco pm>Ho>cw Ahv cm>mm economic .muowmumcH undo Hum mm 00 wH m5 mm mm pmaoam>mp seamwpwcu mamfiumpms Anson» nosupmcfinmamm an undone mmuaoo mo R mpupmuo w.v mmuaoo cw newcmuo mo Honezz mHDOc o.N om. Om.H Oo.m afic3\pcmpzpm \musOc HmcowposuumcH mnecoa s.c pcmamoam>mp Hmwpwcfi mo mcucoz onusoo\OH Om OH VH mmu500\pmm0Hm> non mafia: dances nosupmafl mo nonasz mmuzou umm mmmum>< mu m pGOEQOHw>OQ Mpmn HHGD HmCOflpODHumCH pooh0um Hawmupmcou Ho nonmamumm O.H OHDMH 57 initial development period were also noted and seem to be due to the development competence demonstrated by the major faculty developer, and the availability of existing software to illustrate the concepts he wished to teach. As would likely be apparent to anyone familiar with the instructional media field, the cost of various hard- ware systems vary quite greatly, depending on a number of factors such as the sophistication and amount of the equip- ment used. The strategy of instruction also plays a sig- rnficant role, as indicated in Table 1.1. For example, ‘ste projects which used film in the strategy mix had the additional expense of a Technicolor Super 8mm silent or sound projector, or in some cases a Fairchild Mark IV pupjector. In most cases these hardware items were used Ihidemonstration areas rather than in carrels, and therefore their costs were distributed over a larger number of stu- dents. Since three equipment categories were established, rmtice in Table 1.1 that carrel equipment, demonstration (fimipment and special development equipment were first analyzed as an aggregate body, then carrel equipment was Separated. This procedure was partially due to the fact that in a few cases developers chose to produce their own 58 one vaccmsoucp com: um; muapmooua mficH use: an popw>wp mums manpo» ocwmupmcoo can umumEMuMQ .muomoumca .mmmooud mamaamcm dump .mmusoo Hod mommum>m pomflmmu Op Adv . wmefi—OU monummmm Amv mmucu How mumoo unmadoam>mp accumudmu mmoam> any cm>mm pomhoumm omomm mom.mmm Hovvw o m ovsvm omoom omosm omoom umoo unmadwsvm Hmeoa onHm ems.m 00m mamunmo owe cos com o unmadwaom cowpmupmcoamp 6cm vamaQOHm>op adsomdm mo pmoo . UCflPwfl oomsm pom mmm aoomm -xw cam: ooavm ooaom modem omoom ucmsaunum Hmuumo mo «moo OH on m OH OH 0H OH OH mmuooo \mdmuumo mo Monasz mmusoo mm c m o m N H pcamuvmcou Ham pomhoum Ho wounded umpmamumm demo pcmeawsvm HGGEQOHU>OQ Hafiommm Dam GOHQMHPmGOEmQ .Hmuumo H.H OHDGH 59 software rather than use campus (i.e. local) instructional media center facilities or commercial producers. In such cases, cameras for photomicroscopy, tape recorders for recording and duplicating tapes and an apparatus to freeze- dry tissue were examples of special development equip- ment which was purchased in addition to the carrel equip- nmnt per se. Further equipment cost variation was ex- emplified by Project 5 where materials were developed fcr use in an existing carrel facility and therefore there were no appreciable equipment costs to be recorded. (Average equipment costs were therefore computed on the tmsis of eight courses rather than nine, due to this variation.) Another factor influencing the cost of the car- rel equipment was the sophistication of the carrel it- self. For example, if a rear-projection reflex system was used to project the slides to the learner, as op- posed to a home-made screen on the side of the carrel, there was a considerable increase in expense. Like- wise, if carrels were made by campus craftsmen out of plywood and existing laboratory tables, the cost of the carrel itself was diminished considerably. Another 60 cost-reducing factor noted was that most of the projects studied were somewhat experimental in nature and there was a tendency on the part of carrel manufacturers to work quite closely with the project faculty. For the oppor- tunity to design and work with a self-instructional pro- gram using carrels, these manufacturers made project bids rather than per-item bids, and reduced the carrel costs. Dds procedure gave them an outlet for their product, experience in carrel design, and gave the faculty the oppor- tunity to reduce carrel costs for that particular project. Further variations in equipment costs were illus- tmated by one case where the tape recorders purchased did rmt have the proper out-put amplifiers for the headphones to be used. Specially adapted amplifiers had to be made by.a local sub-contractor to accomodate this mismatch. As equipment manufacturing procedures become more stan- dardized to meet the needs of self-instructional systems, variations in carrel and instructional equipment costs due to local conditions should be minimized. Software production costs were separated from dup- lication costs since the development and production of the original set of software for the self-instructional units 61 would necessarily be more expensive than the duplication thereof. As those experienced with instructional develop- ment are well aware, there are many variables which in- fluence the cost of software production. It was necessary, therefore, to arbitrarily place some constraints on the influence of a number of these parameters. For example, the per-unit costs of producing various types of software were derived from interviews with various software producers, both commercial and university-service oriented. Then the average number of slides, films, tapes, models and hand- outs per unit were determined from the faculty during the interview. The average number of a given medium was then multiplied by the appropriate unit cost and number of in- structional units to determine the total cost of producing a given type of software. To illustrate, if slides were to be shot without art work, they were given a standard unit cost. However, if art work was involved attempts were made to ascertain the length of time necessary for the artist to do the work and then ascribe a labor and materials charge to the particular work. (Examples of software unit costs used are appended in Appendix D.) The average number of slides with 62 artwork per instructional unit were then multiplied by the unit cost of producing slides with artwork. This value was then multiplied by the number of units, to estimate the total cost of producing slides with artwork. The same pro- cedure was used with slides without artwork, and the rest of the cost parameters. As one will notice in reviewing Table 1.2, three of the projects did not engage in film production at all. Of the four projects which did have some film involved in their instructional strategy, only Projects 3, 4 and 5 actually produced their own. Project developers on Project 3 shot their own 8mm footage and Projects 4 and 5 had a media center shoot 16mm footage which was then reduced to 8mm. Project 7 bought commercially produced 8mm footage and added their own sound to magnetic stripping. Photoprints were used sparingly by most projects, if at all, and models were usually purchased as part of the realia rather than locally developed and produced. It is important to note that Project 6 had an exceptionally high realia cost, due to the use of freeze-dried tissue samples and models which were purchased. Hand-out production costs were primarily included under secretarial labor, 63 .Honma Hmflumwmuomm Hops: pmpafiocw mpmoo coauospoud uncupcmmm mvmn GOHHOSUOHA mumeuom N.H OHDMH ommaw mmHVm Hmvmm Hemam 4004» came Hamm omsm mumoo cowuospoud mumBHMOm HMHOH ohm 00m no no mo ma mo mo mpmoo cowpospOud uncupcm: ovwm ow mmmm oo oma o 0 wow mpmoo coeposcoua madame mom co m ms 0mm me «we on mumoo acupuncOua menu owpq< been 00 o 0 com com am 0 mpmoo couposcouc pneumouosm mmoam some 0 mms ooov own 0 o mpmoo coupospoui saws omm m msamm mm w ammm mmam pamm mm m mvmm mumoo coauospoud mcuam wmusoo h o m w m N H pcwmuumcoo Hum Ho momu0>< pompoum umumamumm 64 unless there was some graphic art to be done in preparing the hand-outs. In most cases where illustrations were used, these were simple line sketches done by the faculty or secretaries, and included under their time rather than under a graphic artist's production time. Software duplication was a reflection of both the number of students and therefore the number of carrels and the number of instructional units produced. Obviously the materials for a strategy of instruction which included numerous slides and films would cost more to duplicate than a strategy which relied entirely, or at least substantially, on audio tape. However, as noted in Table 1.3, none of the projects studied duplicated films for individual carrel use, and therefore no duplication costs were ascribed for that particular item. Photoprints were also used primarily in demonstration areas and therefore limited if any duplica- tion was involved. The one project which did duplicate photoprints for individual carrel use was primarily con- cerned with specific laboratory techniques, and the student's ability to study a specific print at length and compare it to a number of slides being presented was felt essential. Realia were also used primarily in demonstration areas and therefore no duplication costs were incurred. Audio tape 65 oowmm Newmam mmmm momm Homvm spam soomm maaam mumoo coup nmowddop mumSumom HmuOH omom osom be mm mm sma mv omm memoo coupmoeadzp psouccmm om o o o o o o o mumoo coepmoaaasp auamom owsm ovOm on on new «ma one saw mumoo coapmowaasp mama o» o o o ooom o o o mumoc coapmoeadso enema 0» o o o o o o o mumoo coupmowaaac seam mus m mommm mmaw mamm mmomm one» mamam oamm mpmoo coupmoeadsp mceam mmuooo b o m e m. N H pcflmupmcoo Hod Ho imomum>< poo.oum nonmagnmm upon :owpmowdmza mumsumom m.H OHDmH 66 duplication costs were a factor of the number of units and the number of carrels involved. Hand-out duplication costs werea reflection of the stencils and paper, and that secretarial time which was used on the production of hand- outs per se. One of the expense categories in the cost of in- structional systems design and the eventual implementation of that design is the use of faculty and other personnel in the develoPment process. Jones, in his analysis of in- structional systems development, clarifies the need to analyze this resource by stating: "Some major university should under- take a continued, large-scale study in detail of faculty time usage, to determine really where this high- priced resource goes, and what tasks could be passed on to lower-cost personnel. To our way of thinking, this kind of data is the largest gap in the measurement of costs of university 'products.'"56 In attempting to analyze the cost of faculty in- volvement in instructional development, the process through which the faculty went in their development of materials 56Jones, A Procedural and Cost Analysis Study of Media in Instructional Systems Development, Op. Cit., p. 88. ‘ 67 was broken down, and they were asked to ascribe an estimate of the time which they spent on individual activities. Es- timates of secretarial time and graduate assistant time were also received in the same manner. Wage and salary data sheets were then used to determine the specific cost of the time which was used by these individuals. Where faculty members were involved in the development process, time estimates were gathered on each participant's amount of involvement, his tenure status, and the length of his yearly appointment. From these data calculations were then made as to each participant's hourly wage and con- sequently the cost of his involvement. Table 1.4 represents the cost values on these time estimates. As will be noted, faculty development time per course was separated from the secretarial and content assis- tant time. Heavy use of associate or full professors for consultation or development logically caused the cost of the faculty increment to increase if the amount of time necessary to carry out the development was the same as those projects which used primarily instructor-level faculty. Another factor in the total cost of the content-related personnel was obviously the length of the development pro- ject. 68 omosm msommm moon» oaovm msvom osomm madam «mus» mumou unaumamma acme Icoo can Hafiumumuomm .apdoomm HMoOH ommam NNom moon 0mm omow Good NONH Nmma mpmoo pcmwmfimmm pumpcoo can Hafiumemu loom How Havovcsm owHHw boom com 0 Damn Good 000 mama mmuooo \mawu paupmwmmm ucmucoo mo pmou monm puma mesa 0mm Coma coo mom cum mmu300\oawu HmHHMpmHomm Ho pmou ovmmm omaoam owmmm oommw mvmam oaaom ommom oomom mmuaoo \mawp pumamoam> Imp andsomw mo amoo mus OMOH 0N3. own ova 0mm new con ONmH own—.50 \dfiflu uGOEQOH0>mp aufisomm Mo muzom OmHQOU N. O m. d. m N H FCMGHPmGOU Hod no u m mum momuwe< pompoum u o e m memo accumflmm¢ pcmpcoo Dam Hmwumpmuomm .huHDOME V.H manna 69 Instructional development and content consultant personnel were separated from the major faculty developers and primary content assistants so that a clearer under- standing of the consultant role in the instructional de- velopment process could be determined. As will be noted in reviewing Table 1.5, the services of these consultants, both instructional development and content, were used sparingly by most developers. In a few cases extensive use of instructional development consultants was found as a result of faculty participating in a fairly extensive and detailed instructional development program without prior experience. It must be noted that instructional development consultants consisted of evaluation and learning theory specialists, as well as graphic and hardware systems specia- lists. In some cases faculty used a small portion of the wide variety of these talents; in other cases the faculty keyed on specific consultants and used substantially more of their services in developing the instructional ma- terials. These variations in the type and variety of con- sultants used are not reflected in the Table 1.5 results. The validation of the instructional materials developed, and the subsequent revision of those materials, 70 comm comm move 33m macaw Sam mom» 8mm mp moo acupdsmcoo HauOH pom m com Sm m8 0 mmm EA 0 m» moo Hanuflomcoo pcmpcoo omwm o w Adam who» macaw as m oamm momm mumoo pcmuasmcoo puma IAOHm>mp HmcowuoouumcH .mmusoo h o m v m N H ocwmuowcoo Hod no monum>< pothum umpoemudm damn peopaamcou pcmucou pom pddEQOHd>mQ HMGOflposupmcH m.H mHDMH 71 .pfipmuo mmusoo sumo ow mamaumpma one cow: mpcmpopm muomoc mamwumums HmGOAHUDMpmcw one popmoflam> adamopom omenspm muomnoud ammo Noam mavm 0» cm mom» om ow om mumoo coamw>mu can nonempfidm> Hauoa ONHm New 0 o on o o o mowoo :0fimw>mu mumzpmom ---m o m cm cm ommm cm on om mumoo cowpmcuam> ucmpapm cum apaaomm omusoo h o m e m N H ucwmupmcoo umd no momum>< pomhoum umpmamumm guano coflmw>om pun coeumpwam> o.H manna 72 was a process carried out by very few of the projects studied. Therefore in analyzing the data as presented in Table 1.6, it becomes evident that most of the materials were presented to students for credit prior to any re- visions and that the validation process itself was an evaluation of how well the students learned from the materials during the term in which they were first used. Facilities modification was necessary in only one of the seven projects studied, inasmuch as most of the modification, if any, was due to the installation of the carrel equipment within an existing space. Only one pro- ject, as reported in Table 1.7, reportedauurappreciable Table 1.7 Facilities Modification Dataa Parameter Project Average or per Cons traint 1 2 3 4 5 6 7 Course Facilities modi- fied O O 0 $1015 0 O 0 ---------- aIf instructional facilities were modified due to the materials developed, they were usually modi- fied to install the carrels. Carrels installation costs are included in the carrel equipment costs in Table 1.1 The one project reporting other modi- fications needed lights and wiring changed to accomodate the instructional equipment, and work study students were hired to clean old lab tables. 73 modification costs. This cost resulted from re-wiring and lighting an old laboratory facility which was not adequate for the instructional equipment being imple- mented. The aggregate data in Table 1.8 reflect the totals involved with each major expense category. It is fair to say that each project was an individual entity unto its own, and showed variations from the "norm" even though the strategy of self-instructional presentation was basically the same in each case. Project-by-project comparisons of total costs are inappropriate at this point, however, due to the fact that these totals do not reflect how much each unit cost. The results of calculations with knowledge of how many units of materials were developed and how many carrels were equipped are reviewed in the next section of this chapter. In addition to total costs per expense category, Table 1.8 also includes a description of the amount of time taken by the faculty during the development process, as opposed to the time taken by other personnel. This differentiation was made so that a clearer understanding of the amount of faculty involvement could be ascertained. 74 _ moamm omosw msommm wommm oeowm msvom osomm «mus» memoo pampmfimmm Home -coo pun downwamu nomm .zpaoomm HMHOH oova NBVNH mMN mom Home but NOON mead mpmoo coaumOHH udop muMBHMOm HmpOH ommdm wmav HMVN HvNH coco ova HoN one momoo COAposp noun mumBuHOm Hmpoa omomm mamomm Hovvm o m owswm ommom omosm omoom mumoo ucmamwsvm HmHOH OH on m OH OH OH ca OH pom: mamuumo no uwnasz OH on N N OH 0 ma o pmaoam> -mp seamwpuca mafia: Mo nonaoz Hum I( , Ho monum>< pomh0um nonmamumm mama acmedoao>mo mummmumm< Mo aumaaam m.H mance 75 .08 5.0 acme -aoam>mp amends“ now mcpcoa amuoe .muc OMOH Omdfl Ofim 0mm mow 00m ONMH mafia puma udOHm>mU audsomm Ho mance HMuOH monomm sosomm mwvmam moommm vvosam Nansen womoam mumoo pamEQOHm>mp HmHOH mHOH F80 coupmuuuwpoa mmauuflwomw Hayes mom» mad mam memoo cowmw>mn pad coupmpwam> Haves 00mm comm mow» Hmoam macaw Howe momw momm mumoo acupflamcoo «cmpcoo pan acme IQOHm>mo Hmcowu noauemcw Hmpoa onusoo Hod monum>< m nomhoum pcwmupmcou Ho nonmamumm pmscwpcoo I m.H OHDMH 76 Likewise, a clarification of the length of time for the initial developmental process was analyzed separate from faculty time to note the average duration of the initial development process. As previously mentioned in reviewing the totals in Tables 1.0 - 1.8, it is very difficult to make comparisons between projects without knowing the number of units and carrels involved. The next section of this chapter will therefore attempt to analyze the data from a unit-cost standpoint so that meaningful comparisons can be made. gnit Cbsts of the Self-Instructional Materials Developed Once the aggregate, full, relevant, functional, job cost data were analyzed and grouped under appropriate para- meters and constraints in separate expense categories, calculations were made of the unit costs and related data per self-instructional unit and carrel. The results of these calculations are noted in Tables 2.0 and 2.1 and give an appropriate means of comparing the costs of the projects studied. For example, the range of in- structional units developed went from two to thirty. If one chose the aggregate data in Table 1.8 in an attempt to make a project-by-project comparison on the basis of 77 omom ovom omoam omsam «mam omoam mon ooom vac: \mpmoo awasomu ommm new sea sod one oma awe sma paca\mpmoo coup umowadsp mumzpmom cam» mma mama mmo mow Had am am uwc:\mpmoo soap nonpoud mumspmom moose oamam oammm mewuuao «we» ooaam some ems» pacs\mumoo acme mew umfiuvm unmEQOHm>mp upmwxm Hmwommm dam .GOHH pom: nmuumcoamp .Hmuumo .muc o.m v v s.a on. m m.a o.m was: \vCQUSu. m\mH30£ HMCOflHUSHHmflH ca om m m ea c we a cmdoam>mp seamauaca wwfifis H0 MODE—AZ OmHSOU B. O m d m N H PCfiMHflMflOU Add Ho monum>< pompoum umvmamumm memo ompmawm can pass HmcoaposupmcHuuHmw umm mpmoo was: o.N wanna 78 rH” .muc ooa sea CNN ONV mN ovH um and “was \mawp pcmEQOHm>0p Apfloomm mo musom ooemw oammm omsom ovomm comm» owmmw ommam owed» pacsxmpmoo acme IQOHm>mU wounded th ma 0 o ow o o o was: \mumoo cowww>mu cam cofipmpwfim> 05% NH «MN me NOH to ON Nm uwc:\mwmoo page nflsmcoo pcmucoo 6cm pumEQOHm> Imp Hmc0fiuosuvmcH mamdm Obs oooN OVON hvo OVNH man Now was: \mumoo unapmwmmm ucmpcoo gum .Hmwume umuomm .apdoomm ammm anew ommam mmmm movm ommm cam msam pacs\mpmoo paupmwmmm pcoucoo pun HmeMHmuomm mmusoo b o m w m N H ucfimuumcoo Hod Ho modumad pomhoum umpmamumm .UODCHPGOU I O . N WHDNH 79 .OE mm.H m. m.H m.v m. H vN. pwca\pcme ndoam>mp How» uflcfi mo wcpcoz amusoo h o m w m N ucwmuomcou umd Ho monum>< womhoum Hovmemumm emacapcoo . o.N manna 80 ovaw mmmw cmow ommw anew Haow «mow omow Houumo \mpmoo apHsomm vwaw ova sv ov omw ms mNH vHH Hmuum0\mumoo coHumoHHmzp mumspmom omHm mo and VNHm oov mm mH mu Hmuumo\mpwoo cowpospoud mumspmom mvmw mom own new oaw ado omm moo Houumo\mpmoo nomeo pcoEdHovm Houumu moow moww moow meouuoo vsvw moow vaw moow dogwooxmpmoo name new unwavo pcoEQOHo>mp upmem Hmwoomm pun .cowv pom: nmuomcoaop .Houumo OH NH 0 m mH mH m CH Houum0\mpcop=pm 0H #0 m OH OH OH OH OH mHmuumo mo Hocasz omuooo h o m w m N H pcwmupmcoo Hod . Ho oomuo>< poohOHL umpmamumm Mama pooMHmm pun .Hmuumo uom mpwoo HHGD H.N chma 81 .mnc Om mm mOH mm mm um Om NmH Hmuumo \mEHo pamEQOHm>mU AuHaumm mo mono: oaoaw oooaw ooomw omsw oommw vosdw omaaw smoaw Hmuumo\pmoo ucme IQOHo>mU momum>¢ mmw 0* How tumo\mpmoo :onH> -mu paw coupmoaam> mow do mOH NOH 0* mm GN Houum0\umoo acne uHamcoo pumpcoo 6cm pcmEQOHm> Imp HmcowuosupmcH moow mum OmHH How two how OHm mhh Houumo \vmoo pcmpmwmmm pumpcoo cum .Hmwu Impouomm .aoHuomm [I oamw sow mmmw mvw medw OQHm osw mmaw Houumo\mpmoo acupmwmmm pcmocoo cum Hmwumvouomm ”I _ 1V mmwooo i h o n v m N H pcwmupmcoo Hod no amouwe< Homemumm 1. [Huuuuuuuuuhm pomhoum pmscwucou I H.N OHDMH 82 .moa mm. HH. o. a. m. o. oH. a. Hmuumo\ocmaaon>mp HMHpHcH How mcpcoz mmuooo h o m o m N H chmemcoo Hod Ho mmmuo>< pompoum umpmamumm UmschGOU I H.N OHDMH 83 total costs, it would be impossible to say which project made the most efficient use of the investment made. Once placed on a per-unit basis, the data begin to show less variation. Where variation does exist, however, these differences can be analyzed with much more accuracy. For example, software duplication costs in Tables 2.0 or 2.1 show Project 4, which used a wide variety of media, including prints, audio tapes, hand-outs, and 2X2 slides, with a relatively expensive per-unit cost. Where several faculty were involved in the development of a single course, as in Projects 3, 5 and 6, the faculty costs were substan- tially higher than in the other projects studied. In addition, Project 6 had substantially higher unit costs for secretarial and content assistants, due to the in- creased use of secretarial assistance, and the relatively few number of units resulting from the development effort. It is worth following Project 6 results through to the total unit cost of development. It will be noted that the unit cost of producing those two units was approxi- mately twice the amount of any other development project. This increase is due almost entirely to the fact that the faculty chose to spend their own time producing software, plus there was an increased use of models and expensive, 84 sophisticated carrels. It will also be noted that Project 6 was next to the highest in terms of the amount of time it took to carry out the development process. Project 5, as Table 2.0 illustrates, took the long- est of any project to develop each of its two units. This time factor may be due to a lack of development experience on the part of the project director, plus the time con- sumption to coordinate schedules and meet with numerous instructional development consultants and other depart- mental faculty. This heavy use of departmental faculty time is reflected under faculty unit costs in Table 2.0 for Project 5 which had the highest per-unit costs in this category. Medel I Validation Results After a thorough analysis of the aggregate data and a review of the unit-cost per instructional unit and carrel, a predictive unit-cost model was designed, based upon the parameters and constraints analyzed. This model will hereafter be referred to as Model I. Once designed, Model I was adapted for use in the form of a budget planning guide. (The budget planning guide for Model I is appended in Appendix B.) High, low and average values were provided for selection from each expense cate- gory in the planning guide and criteria for their selection 85 were written. Once this model was constructed, it was validated by the key faculty developer of the self-instructional materials developed for use in the Introductory Botany course at Purdue University. The validator took thirty minutes to complete the decision steps in Model I, during which time the planning guide provided estimates as to the total development costs, the length of time for the develop- ment process and the amount of faculty time necessary to carry out the development. The development time period analyzed and for which Model I predicted costs was the initial development which took place during 1961-62. Prior to the model validation, data from the Validation Project were gathered with the same procedures which were used in collecting the aggregate data from the original seven sample projects. Also, the same parameters and constraints in each expense category were analyzed. Table 3.0 provides a summary of the aggregate development data from this Validation Project. In comparing these data with Table 1.8, it is apparent that far more carrels were used in this project, and yet the total development cost was substantially lower than any of the projects in the initial sample. Likewise, though the length of the 86 Table 3.0 Summary of Aggregate Development Data from the Validation Project Parameters or Data Constraints Number of units initially developed 14 Number of carrels used 22 Total equipment costs $3036 Total software production costs $143 Total software duplication costs $889 Total faculty, secretarial and content assistant costs $1611 Total instructional development and con- tent consultant costs $250 Total validation and revision costs $0 Total facilities modification cost $0 Total development costs $5929 Total hours of faculty development time 274 hrs. Total months for initial development 13 mo. 87 initial development was four months longer than any of the projects initially studied, the number of hours spent by the faculty during this development period was only eighteen hours short of being equal to the lowest of those initially studied projects. In comparing those data in Tables 3.1 and 3.2 with the data in Tables 2.0 and 2.1, it becomes apparent that the unit costs of the validation project were substantially less than any of the other projects studied. When these data were further analyzed, it became obvious that this difference was due to primarily three factors. First, the development was done during 1961-62 at a deflated price, as compared to the development expense during the 1967-68 period when the sample projects were being developed. Two, the strategy of instruction included primarily audio tape only, with the exception of botanical specimens given to the students in the carrels. And, three, the carrels were made by a craftsman in the university at a reduced labor and materials rate. Model I provided for the deflation influence and the self-made carrels, but no exclusive values were avail- able for a model user anticipating the use of an audio tape-only instructional strategy. This oversight contributed 88 Table 3.1 Unit Costs Per Self-Instructional Unit and Related Data from the Validation Project Parameters or Constraints Unit Costs .and Related Data Number of units initially developed 14 Instructional hours/student/unit 2.5 Carrel, demonstration, and special development equipment costs/unit $216 Software production costs/unit $10 Software deuplication costs/unit $64 Faculty costs/unit $113 Secretarial and content assistant costs/unit $2 Faculty, secretarial, and content assistant costs/unit $115 Instructional development and content consultant costs/unit $18 Validation and revision costs/unit $0 Average development costs/unit $424 ‘ Hours of faculty development time/ unit 19.5 hrs. Months of initial development/unit .93 mo. 89 Table 3.2 Unit Costs Per Carrel and Related Data from the Validation Project Parameters or Constraints Unit Cost and Related Data Number of carrels 22 Students/Carrel l8 Carrel, demonstration, and special development equipment costs/carrel $138 Carrel equipment costs/cartel $138 Software production costs/carrel $7 Software duplication costs/carrel $40 Faculty costs/carrel $72 Secretarial and content assistant costs/carrel $2 Faculty, secretarial, and content assistant costs/carrel $74 Instructional development and content consultant cost/carrel $11 Validation and revision costs/carrel $0 Average development cost/carrel $270 Hours of faculty development time/ carrel 12.4 hrs. Months for initial development/carrel .59 mo. 90 to an increased cost prediction, since less equipment and development time is necessary when audio tape is the main medium of instruction. Likewise, the assumption was made in designing Model I that the unit—cost/carrel would be the best means of anticipating expense fluctuations caused by the increase or decrease in student enrollment. AS was found after analyzing the validation results, this assumption was accurate to a point, in that it was a good indicator of equipment costs, but a poor indication of the software development costs per instructional unit. There- fore the predicted costs were substantially increased when the validator was forced to select per-carrel unit costs for software production and then multiply those unit costs by the number of carrels, as opposed to the number of in- structional units. As a result of these two basic flaws, the first validation of Model I predicted the total costs for the development process to be $9,995.00, as opposed to $5,929.00 as determined by the cost analysis. Revision of Model I The first validation results made it apparent that a number of revisions would be necessary to make Model I more predictive of the cost of development. A revised model resulted from these revisions. 91 Rather minor revisions included a more explicit introduction to the use of the model, and a clarification of the assumptions and limitations to which the model should be subjected. Of a more serious nature, a basic assumption in Model I was challenged. It stated that the costs per carrel should be the main unit cost for predic- tion. The Revised Model, however, provides for costs per carrel to be used in determining carrel equipment costs, but that costs per instructional unit should be used as the unit of analysis for the remainder of the parameters and constraints in each expense category. The exception to this procedure is the prediction of software duplication costs, which after further analysis were determined to be influenced by both the number of carrels and the number of units produced. Further modification provided for the Revised Model user to select either an instructional strategy which utilized slides, tapes, photoprints and film or a strategy which was comprised of primarily audio tape, with a few photoprints and specimens. A third strategy category was also provided for a user who anticipated no faculty sala- ries, graduate assistant or secretarial wages, consultant 92 fees, validation or revision costs, ppraculty time. As with Model I, a high, low and average per-unit cost was provided under each of the two major instructional strategies, those being the combined strategy mix of: (1) slide-tape and/or film, and (2) an audio tape strategy. The criteria for selecting the appropriate instructional strategy and the expense category values under each stra- tegy were modified to accomodate these revisions. In addition to changing (l) the basic unit costs from carrels to a combination of carrels and instructional units and (2) the ability to select from two basic instruc- tional strategies, an opportunity is provided for the user to use his own value for the number of carrels needed. This choice is made by the user based upon the unique con- straints of his present instructional setting. In the final analysis, the Revised Model provides three methods for the user to ascertain which carrel quantity he should select and use in calculating the equipment and software duplication costs: (1) a selection from values which re- present the previous experience of other developers, (2) a calculation of the number of carrels needed, based upon the average anticipated time for a student to use each instructional unit, the amount of time the carrel 93 facility will be open for student use, and the number of students needing to use the carrel facility, and (3) a prior determination of the number of carrels needed made from an appraisal of the local facility constraints. Figure 1.0 shows a schematic of the major steps in the Revised Model, and Figure 1.1 shows the sequential decisions in the Revised Model. A simulation of the re- vised planning guide is in Appendix C. Revised Model Results Following the revision of Model I, a validation of the Revised Model was conducted and the results of its predictions compared with the same aggre- gate and unit-cost data collected during the validation of Model I. The second validation used the same selections as chosen by the key faculty developer in the original validation at Purdue. The Revised Model predicted the total development costs to be $5,096, as compared to the cost analysis results of $5,929. This value compares favorably with the decision model specified in the re- search design, in that it predicted the "actual costs" within 14%. Table 3.3 illustrates a side-by-side comparison of the Model I validation results, the cost analysis results, and the Revised Model validation results. 94 Homo: pmoo m>HpoHpoum pomH>om CH mdmom mono: O.H musmHm mmmooum - acme mopmawpmm mumou mumou momoo IQOHmSmQ ‘4 mafia. fucmEQOHm>mQ pumEQOHm>mQ OHHmoHHm-fi ‘l. mo comcwq apHsomm Hooch mumSumom mumsowom mpHcD COHuomHmw mHouumo r‘ulleCOHuosuumcH ‘IIII mpmoo amwpmupw ‘IIII mo Illa mo umcEDZ acmaQanm mcoHuosuumcH Honeuz 95 REVISED MCDEL'S SEQUENTIAL STEPS nter lanning Guide Write # of student ‘ l r—‘ l I Write predetermined Calculate Select # # of carrels # of of carrels carrels 1 Determine # of review carrels Select media strategy Determine slide + Determine audio audio tape + film tape only 1 Select equipment I strategy I 96 Determine carrel Determine carrel and equipment other equipment only Select software duplication unit costs f ‘1 Write # Select of units ~* revision costs Calculate duplicatio Total production costs unit costs Select Calculate total faculty production costs costs Select Total hardware assistant and costs software costs Select Adjust for consultant inflation costs 7 ‘ l Select faculty time per unit Calculate total faculty time Select length of development/ unit Calculate total length of development EXi with cost and time estimates Figure 1.1 Revised Model's Sequential Steps ma. .mb 98 Figure 1.2 represents the basic equations used to predict development costs, faculty time and the months necessary for development. l. (g of Students) (Ayg; Time/Student/Unit) = . # of Carrels # of Hours Carrel Room Open/Unit needed for Regular In- struction 2. (# of Carrels needed for Regular Instruction) + (# Review Carrels needed) = Total number of Carrels needed 3. (Equipment costs/carrel) (Total # of Carrels needed) = Total Equipment Costs 4. (Software Duplication costs/Instructional Unit/Carrel) + (# of Instructional Units to develop)+(# of Carrels needed for Regular Instruction) = Total Software Dupli- cation Costs 5. (Software Production costs/Instructional Unit) + (Faculty Salary costs/Instructional Unit) + (Content Assistant and Secretarial Wage costs/Instructional Unit) + (Consultant costs/Instructional Unit) + (Validation and Revision costs/Instructional Unit) = Development Costs/Instructional Unit 6. (Development costs/Instructional Unit? + (# of In- structional Units being developed) + (Total Software Duplication Costs) + (Total Equipment Costs) = Total Development Costs 7. (Total Development COStS)+(% Inflation) = Cost In- crease due to Inflation 8. (Cost Increase due to Inflation) + (Total Develop- ment COSts) = Adjusted Estimate of Total Development Costs 99 Figure 1.2...continued 9. (Faculty Time/Instructional Unit) + (# Instructional Units) = Total Estimated hours of Faculty Time 10. (# Months for Development/Instructional Unit) + (# Instructional Units) = Total Estimated # of Months for Development Figure 1.2 Basic Prediction Equations 100 TABLE 3.3 A.COMPARISON OF THE UNIT COST-ANALYSIS RESULTS OF THE VALIDATION PROJECT WITH THE PREDICTED COSTS FROM MODEL I AND THE PREDICTED COSTS FROM THE REVISED MODEL lopment Parameter Model I Cost Ilevised or Predic- Analy- Model Cons traint tion sis Pr edic- LE: Results 'tions Number of students 400 400 400 Number of carrels 27 22 22 Number of instructional 14 l4 14 units Total development costs $9,995 $5,929 $5123a Hours of faculty time 1350 hr 274 hr 294 hr spent on initial develop- ment Months for initial deve- 24.3 mo 13 mo 13.3 mo aThis value is within 14% mined by the unit cost-analysis of the validation project. of the value as deter- 101 Summary Full, relevant, functional, job cost data were collected from seven self—instructional materials develop- ment projects. The parameters and constraints of these aggregate data were then grouped and analyzed under eight expense categories and four non-expense categories. Fol- lowing an analysis of the aggregate data, unit costs per carrel and per instructional unit were calculated for each of the seven projects. Model I was then designed, based upon the sample project unit costs. The Model provided the user with the opportunity to select high, low and average unit costs per carrel for each of the expense cate- gories. Criteria for the proper selection of these unit- cost values were also provided. Once designed, Model I was validated in a two-step process which included (1) a trial run of the Model by the faculty developer of the initially developed self-instructional materials used to teach the Introductory Botany Course at Purdue University, and (2) a unit cost—analysis of the initial development period for this Validation project which was conducted, and the results of the cost analysis compared with the pre- dicted costs suggested by the Model. An analysis of the first validation results showed that a number of oversights 102 and fallacious assumptions had been made in the design of Model I. Therefore a Revised Model was constructed, which took into account a more accurate view of the system for which predicted development costs were being made. The selection criteria were appropriately modified, branching possibilities which permitted greater selection flexibility were added and additional instructional strategy options were included. Once Model I was revised, a second vali- dation was conducted, and the predicted costs again com- pared to the results of the validation project unit cost- analysis. The Revised Model predicted costs within 14% of the validation project's cost analysis results. CHAPTER V SUMMARY AND CONCLUSIONS Study Design and Model Construction The purpose of this study was to design a model for predicting the unit costs of self-instructional materials development in higher education biological and physical science curri- cula. Prior to collecting aggregate sample data, a pilot study was conducted to test an interview instrument. Once modified, this instrument was used to interview the key faculty developers from each of seven self-instructional development projects, six self-instructional materials development projects at Michigan State University and one self-instructional materials development project at Lansing Community College. Materials for nine courses were de- veloped as a result of these seven projects. In addition to the full, relevant, functional, job cost and related data collected during the faculty interviews, supportive data were gathered from production specialists, equipment catalogues and other participants in the initial project development. 103 104 Once the aggregate data were gathered, various parameters and constraints were grouped and analyzed under eight expense categories and four non-expense cate- gories. Cost per carrel and cost per instructional unit developed were then calculated from the aggregate data for each parameter and constraint. Unit cost comparisons were then made between projects to note their similarity and differences. Once the analysis was complete, a predictive model (Model I), was designed and modified for use as a budget planning guide for instructional planners. This model took into consideration the aggregate, supportive and unit-cost data from the seven projects. Criteria were written for selection between high, low or average unit costs for each development parameter. Results After Model I was designed and adapted, its predictive validity was tested on the self-instructional materials initially developed to teach the Introductory Botany course at Purdue University. A unit-cost analysis of the validation project was simultaneously conducted so the validity of the costs predicted by the model could be determined. Since the assumption was made in designing "Iv ‘ 'l! b E H 105 Model I that costs per carrel would be the best source of predictive values for projects with wide variations in stu- dent enrollment and since provisions were made for only a slide, tape and film self-instructional strategy, the Model I predicted costs of the validation project's initial deve- lopment were approximately $4,000 more than those costs re- sulting from the unit cost-analysis. This value was not accurate enough and it was apparent that revisions were essential for the model to be valid. Therefore, revisions were made in Model I to make it more responsive to individual variations within projects. Cost per carrel values were used as predictors of carrel costs and as factors in the software duplication costs in- stead of for all expense categories. Costs per instruc- tional unit were used to predict the cost of producing and duplicating software, faculty time, content and secretarial assistant time, consultant time, validation and revision, and the amount of faculty involvement, plus the number of months for the development process. In addition to the above, the Revised Model per- mitted the developer to select between two basic strate- gies: (1) slide, audio tape and film or (2) audio tape only. A third strategy category was also provided for on“ IA“? 'll' 1.» -. .. mi u‘ 1:; Ev rm - Us. «U 106 those planners who would be using no faculty time, consul- tant time, content or secretarial assistant time or doing any validation or revision. Due to the flexibility provided by the design modi- fications, the Revised Model predicted development costs during a second validation within 14% of those development costs determined by the unit cost-analysis conducted at the time of the first validation. Since acceptable pre- dictive accuracy had been pre-determined to be within a range of ‘515% of the results of a cost analysis of a given project, the Revised Model was concluded to be a valid pre- dictor of self-instructional materials development costs. Limitations A number of limitations must be considered in using the Revised Model budget planning guide to predict development costs. One of the most important of these limi- tations is the approximate nature of the aggregate data upon which the model was designed. Jones has clarified the reason for the approximate nature of cost-analysis data in stating that: "Cumulative costs are embodied in the pro- duct and passed along with the product as it progresses through departments toward completion. The end result is an agglomera- tion of some carefully detailed experienced costs and some allocations and re-allocations, to reach what we call 'final product cost.' 107 No one knows better than the accountant that this product costs is an approxi- mation, even in the best of cost account- ing systems. The rocess of cost alloca- tion makes it so." 7 With this background, it should be clear that basing a predictive model upon approximations makes that model's predictions approximations by definition. Second, the model was developed from aggregate data which reflect costs for only higher education biological and physical science development projects. Therefore, generalization to other curricula and other levels of aca- demia is not possible. Third, the Revised Model also assumes that where film is used the footage will be reduced from 16mm to 8mm. Fourth, due to the fact that the Revised Model is designed for inexperienced developers who are starting the development of self-instructional materials, the input data can only be relatively general in nature. The vagueness of this input data contributes to the approximate nature of the predicted costs. In addition to the generalities surrounding the initial decisions, it is important to note that standardized costs for a creative process such as 57Jones, A Procedural and Cost Analysis Study of Media in InstructionalpSystems Develgpment,‘gp, cit., p. 21. 108 instructional development are difficult at best, and there- fore allowance for project variation from the model esti- mate should be made as the development progresses. A fifth limitation exists in that only two instruc- tional media strategies were accounted for in collecting aggregate data and developing the model and planning guide. Therefore planners who anticipate using strategies which rely heavily on programmed, printed material or video trans- mission are excluded from using this budget planning guide in making cost estimates until such time as unit costs are available for these strategies as well as the ones studied. Sixth, the aggregate data and the basic assumptions in the model are reflective of the development, rather than the implementation or operation of self-instructional systems. Therefore, predicted cost estimates are for only develop- ment activities, rather than implementation or operation cost parameters. Another limiting factor of a rather minor nature is the fact that the aggregate data and the assumptions in the model are based upon the use of reel-to—reel audio tape recorders. As advances in cassette recorder tech- nology occur, modifications of various standard unit costs within the model will likely be necessary. 109 The model user should be aware of the above-men- tioned limitations, and if major diversions from the assump- tions in the design of the model are planned, these modifi- cations should be taken into consideration in developing a project budget. Conclusions It is concluded that a cost model with a predictive validity of within 14% of cost analysis results of the same initial development period can be designed for specific mediated strategies of self-instruction in the biological and physical sciences at the higher education level. It is also concluded that the predictions resulting from the use of this model reflect present concepts of self- instructional systems theory, and that as technological and procedural changes modify the system upon which that theory is based, the model will need revision to be valid. Implications for Future Research There are numerous questions resulting from this initial attempt to analyze and model the self-instructional materials development system. A summary of future research implicated by this study includes the following: 1. The extent to which development costs may be controlled by assigning administrative llO responsibility to cost centers revealed by the use of predictive cost models. Whether unit costs go down as faculty develop- ment experience increases with each additional instructional unit. Whether multiple course use of common self- instructional settings will reduce develop- ment and operation costs. The variations in development costs which result from academic departments establishing decentralized graphic production centers. Whether development costs are reduced by pub- lishers assuming the responsibility of pro- ducing software components for insertion into self-instructional development projects. Whether breakthroughs in cassette recorder and other technology will reduce the equip- ment costs for non-print strategies of self- instruction. At what stages of instructional development should consultants be involved to optimize the learning effectiveness of the materials developed while keeping the development costs 111 at a minimum. 8. How development costs vary in new as compared to existing higher education institutions. 9. The extent to which released time for faculty developers results in reduced development costs and more effective self-instructional materials. 10. The extent to which an increased sample size improves the predictability of the model de- signed in this study. 11. The extent to which the model is adaptable to other media strategies other than slide- audio tape-film or audio tape only. 12. The extent of implementation and operation costs of self-instructional systems as com- pared to other strategies of instruction. Inasmuch as the model designed as a result of this study provides a cost control function, in that it sug- gests cost centers to which administrative responsibility at the point of cost incurrence can be assigned, it would be interesting to note the way in which prior knowledge of functional costs are able to control the 112 overall costs of instructional development. Research com- paring (1) development projects which have been budgeted without the use of the predictive model with (2) those pro- jects which have made use of the functional cost parameters revealed by the model expense categories to assign cost control centers throughout the development phases should be of great benefit to instructional systems designers and have application in other aspects of academic institutional management. Another area needing further research is the compara- tive difference between the initial time to get started and produce one instructional unit, as opposed to the amount of time necessary to develop two or more instructional units. As will be noted from Table 2.0, Projects 5 and 6 had relatively high unit costs in nearly all phases of the development. This increase may be due to the high initial costs of developing any instructional unit, rather than an efficient use of the development time. Previous experience with self-instructional materials development may also decrease the amount of time, and therefore the costs of future development projects. Therefore, since the study 113 reported dealt with initial development costs and with faculty who had generally no previous development experience, it would be important to note the influence on costs re- sulting from previous experience on (1) the later develop- ment of instructional units within the same project and (2) additional future projects. Another area for unit cost-analysis research would be to clarify any reduction in development costs which re- sult from the use of learning resources centers in which various self-instructional units from different courses can be taught in the same instructional facility. A mul- tiple-use facility, as illustrated by Project 5 in Table 2.0, would obviously reduce equipment and facility costs, but if the cost of coordinating such a facility would eventually outweigh the savings in development costs, it would be a questionable practice. Software development, production and duplication is, as is noted in Tables 2.0 and 2.1, a relatively expensive process, particularly as one gets into a large carrel system with numerous instructional units. Simple duplica- ting processes for audio tape or slides, which can be done by technicians in the academic departments, may provide an 114 eventual savings in total development costs. It will be important for research to note, however, the relative in- crease in labor costs which result in shifting these roles to the departments, rather than maintaining these functions in a media service center or a commercial outlet. As was noted with Project 6, in Table 2.0, the average costs of development per instructional unit were nearly twice that of any other project, due imipart to the fact that,thefl project faculty did most of their own software production, rather than hiring it out to media center or commercial technicians. Therefore a close cost analysis of decentrali- zation expenses for production and duplication services should be carried out before this shift in function is re- commended as a cost savings to developers. Other research related to the software production and duplication cost aspects would attempt to indicate the cost benefits derived from mass production of various com- ponents for self-instructional systems by publishers or by academic institutions. These components could be available for purchase by developers in various colleges and uni- versities and should therefore reduce local development costs by distributing these costs over a much larger number of individual projects. Some publishers and universities 115 seem to be producing software components for incorporation in local development projects. As more individuals become aware of, and satisfied with, the results of using a self- instructional strategy in the biological and physical sci- ences, significant moves in this direction will likely take place. However, the costs and marketability of published software will likely need to be identified by research before publishers will get into the market in sufficient depth to provide the type of software which has been ade- quately validated for the concept areas being developed. Along the line of commercial manufacturer-developer relations, the use of cassette recorders as opposed to reel- to-reel recorders may reduce the costs of equipment consi- derably if the reliability of the cassette and the recorder can be improved. At the present time, reports from the field indicate that some cheaper cassette recorders are not re- liable, and that much student and faculty time is lost in trying to repair or modify the system to accomodate for their inefficiency. If technological breakthroughs in the cassette field are able to be made within the next few years, research may show that these improvements will modify the equipment cost parameters of the development system. 116 To those experienced with instructional develop— ment, it is obvious in looking over the aggregate data in Tables 1.5 and 1.8 that evaluation, learning theory and production consultants were used in varying degrees. As the parameters and constraints of instructional development become clearer through future research, it would be impor- tant to know when the input of a professional instructional development consultant would be most beneficial in reducing the costs of the development while maintaining or improving the quality and effectiveness of the developed materials. Indications from Projects 4 and 5 seem to suggest that in- appropriate use of this expensive resource will increase the costs of development considerably without much appreci- able increase in quality. Techniques need to be developed to train faculty in the basic aspects of instructional development so that consultants are in fact consulting, rather than carrying out most of the development for the content faculty. Research which would identify the key stages in the development process in which the input of instructional development consultants would be most cost effective would provide a direct service to instructional systems designers. 117 As this study dealt with existing institutions only, future research is needed which involves new junior colleges and four-year institutions which are developing self-instruc- tional programs. Though some phases of the development, particularly in the area of the software development, should remain the same in either an existing or new institution, it is suspected that there will be an increase in facility modification, and equipment costs incurred by new institu- tions which are not present in existing systems develop- ment. Therefore the similarities and differences between these two instructional settings should be researched, and modifications in the predictive model be made to accommodate both types of instructional settings. Released time for faculty has been recognized as a worthwhile investment in.some institutions, but with the exception of Projects 3 and 7 this study revealed re- latively little assigned time for faculty to develop self- instructional materials. The results of these two pro- jects seem to indicate, however, that more effective ma- terials may be developed at a reduced cost as a result 0f the faculty's freedom to concentrate their time for development. Table 2.0, for example, shows that Project 3 and 7, both of which provided some released time for faculty developers, were able to produce a substantially 118 well-balanced instructional system with arI average cost per instructional unit. Inasmuch as most faculty seem to be unable to develop instructional systems as effectively if they are asked or take on the responsibility of develop- ing materials in addition to their regular duties, research would be of great benefit to administrators if findings showed what an investment in released time should produce in terms of a more effective and efficient development process. If, in fact, reduced costs for the materials de- veloped will make up the difference in cost due to faculty released time from other duties, this information should supply greater incentive for administrators to release faculty for development. Industry has been profiting from investments in research and development for years. In- vestments by educational institutions in instructional re- search and development may also be found to be cost effec- tive over time. In addition to the above suggestions for future research, it is important to note that this study delibera- tely chose to limit the sample size and to design a pre- dictive cost model from the sample aggregate data so that the interaction of the various self-instructional systems 119 parameters and constraints could be analyzed upon further testing of the model. This procedure has several advantages as reported by Sisson: "It will tend to consolidate the few miscellaneous concepts we have learned about the educational process. It will focus attention on the kind of data which must be gathered; the data required to obtain parameters for the model and to validate them. More important, perhaps, it will in- dicate which kinds of data are not important. Finally, the model will fix our attention on the key para- meters which might have universa- lity."5 New research should now increase the sample size and use random sampling techniques to reduce the sampling bias inherent in the study reported. Research which keys on the parameters identified and used in the Revised Model should modify and improve the model's predictability over a more generalized population. Once the Revised Model is refined, several of the limitations to its use should be eliminated by parallel stu- dies conducted on other instructional media strategies than (1) slide, audio tape and film or (2) audio tape only. 58.Sisson, Loc. cit. 120 The unit costs of developing printed, programmed instruc- tional materials, video-taped and film-chain materials, computer-aided instruction materials and dial-access materials for self-instructional strategies are also needed. Analysis techniques, having been tested and refined by this and other research, should permit future analyses of other instructional media strategies to build new models or adapt this one to fit the parameters involved. Other areas for future research are implied by the results of this study. However, one last suggestion should be given special note. As previously mentioned, this study dealt with only development costs and there- fore the eventual cost of implementing and operating a self-instructional system needs to be known so that in- formed administrative decisions can be made. The Jones study of media in instructional systems development in higher education as well as the General Learning Corpora- tion study should permit comparisons between self-instruc- tional operation costs and other instructional strate- gies. In conclusion, it would appear from the results of this study that the cost of developing self-instruc- tional materials represents a substantial investment 121 in money and time. For example, the cost of producing a unit of instruction in the projects studied averaged $2795. There were many parameters and constraints affecting this value, but thirty-eight per cent of the investment went for carrel, demonstration and special development equipment, with faculty costs close behind with thirty-three per cent. The rule-of-thumb used by some media specialists which calls for every dollar of hardware costs to be matched by a dollar of software production and duplication costs did not seem to hold in practice, for only seventeen per cent of the development costs went to these latter two expense categories. Secretarial and content assistant costs com- prised nine per cent, with consultant, validation and revision costs representing relatively small investments of three and one per cent respectively. Time investment averages of 109 hours of faculty time per unit and seven months per project represent committment which developers and administrators need to take seriously so that costs are controlled, and an instructional "profit" results from the continued use of the materials developed. This "profit" can be achieved if quality and cost control pro- cedures are implemented during the development and operation process to make certain that the materials produced are indeed 122 valid and continue as an effective strategy of instruction. BI BLI OGRAPHY 10. BIBLIOGRAPHY Abt, Clark C. A Qgst-Effectiveness Mod§i_for the Analysisyof Title I ESEA Project Prgposgls, Eggt I-yIII. ERIC #ED 014 018. Backstrom, Charles H., and Hursh, Gerald D. Survey Reseazgn, Northwestern University Press, 1963, pp. 28-35. Borow, Henry. Mag ip A World at ngk. Houghton, Mifflin Co., 1964. Cain, J. Harvey. "How Unit Cost Accounting Can Serve the College Field." College and Uniyergity Busi- es , 32 (March 1962), 63-65. Chambers, M.M. Einagcing Higher Education. Washing- ton: The Center for Applied Research in Education, Inc., 1963. Cooley, Wm. W., and Lohnes, Paul R. Mpltivariate Pgo- ggdgges for the Behavioral Sciences. John Wiley and Sons, 1965. "Cost of Audio-Visual Instruction, 1962-63, 1968-69." Sghogl Manggemegt, Vol. 12 (October 1968), pp. 67-72. Cronbach, Lee J. Essentials of Psychological Testing. 2nd edition, Harper and Brothers, 1960. Draper, N.R., and Smith, H. Applied Regression Analy- sis. John Wiley and Sons, 1967. Florida Cost Study Committee and the Office of the State Board of Control. A Manual for Apalyzigg University Expendituresypy Function. Revised 1960-61. Talla- hassee: State Board of Control, n.d. 123 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 124 General Learning Corporation. Guidelines for Deter- mining Costs of Media Systems. ERIC #ED 024 273, 1968. General Learning Corporation. Cost Study of Educa- tional Media Systems and Thei£,Equipment Components: Technical Report. ERIC #ED 024 286, 1968. Harmon, Paul. "Curriculum Cost-Effectiveness Evalua- tion." Audio-Visual Instruction, Vol. XV (January 1970, No. 1. Hitch, Charles J. "What are the Programs in Plann- ing Programming, Budgeting." Socio-Economic Planning_Sciences, Pergamon Press, 1969, Vol. 2. "How to Use the Cost of Education Index, 1968-69." School Management, Vol. 13 (January 1969, pp. 52-54. Hull, L.D. "Pitfalls in the Use of Unit-Cost Studies." Journal of Higher Education, 32 (October 1961), pp. 371-376. Ikenberry, Stanley. "Instructional Cost and Quality." College and University, 37 (Spring 1962), pp. 242- 250. Jones, Gardner. A Procedural and Cost Analysis Study_ g£_Media in Instructional Systems Development. East Lansing, Michigan: United States Department of Health, Education and Welfare, Office of Educa- tion.Grant #OE-3-l6-O3O (September 1, 1965). Jones, Gardner. CCTV Cost Study, 1968-69. East Lan- sing, Michigan: Educational Development Program, Michigan State University, Project #I-O73. Keppel, Francis. "Operations Analysis --- The Promise and the Pitfalls." Socio-Economic Plannipg Sciences, Vol. 2, Pergamon Press (1969), pp. 121- 125. 21. 22. 23. 24. 25. 26. 27. 28. 29. 125 McNeely, John H. UniversityAUnit Costs. Washington, D.C.: U.S. Government Printing Office, U.S. Office of Education Bulletin 1937, No. 21 (1938). Mood, Alexander, and Powers, Richard. Cost-Benefit Analysis of Education. #ED 012 519 (1967). National Committee on Standard Reports for Institutions of Higher Education. Eigancial Reports for Colleges gnd Uniyersities. Chicago: University of Chicago Press (1935), pp. 177-249. (Reprinted under the title Cgmputation of Unit Costs. Washington: American Council on Education (1955). Peden, Robert W. "Is There an Educational Industry?" Cpllege and University Business, 21 (November 1956), pp. 50-51. Pike, Walter L. "What You Can Learn from Unit Costs." lelege and University Business, 37 (July 1964), pp. 39-410 Rand, Edson R. "If Unit Cost Calculations Are to be Valid." College and University Business, 19 (August 1955), pp. 25-26. Russell, John Dale and Doi, James I. "Analysis of Institutional Expenditures." College and Univer- sity Business, 19 and 20 (September 1955 to August 1956). Scheps, Clarence. "Systematic Financial Analysis and Budgetary Planning as Aids in the Attainment of College and University Purposes." In Smith, G. Kerry (ed.) Current Issues in Higper Education, 1961: Goals for Higher Education in a Decade of Decision. Washington: Association for Higher Education, National Education Association (1961). Sisson, Roger L. "Can We Model the Educational Pro- cess?" Socio-Economic Planning_Sciences. Per- gamon Press, Vol. 2 (1969). 126 "Techniques and Application of Educa- tional Systems Analysis: PERT, Linear Programming, and Utility/Cost Sensitivity Analysis." Audio- yisual Instruction, Vol. 14 (March 1969), pp. 89- 30. Tanner, C.K. 90. 31. "Ten Deceptions in Building Cost Comparisons." Over- view Management (July 1962). 32. Tyndall, D. Gordon and Barnes, Grant A. "Unit Costs of Instruction in Higher Education." The Journal of Experimental Education, 31 (December 1962), pp. 114-118. 33. University of Montana System of Higher Education. Helena, Master Plan Study: Status Report. Montana: Office of the Executive Secretary, State Capitol, Room 139 (October 8, 1962). 34. Walker, Ernest W. "To Measure Operating Efficiency." lelege and University Business, 29 (August 1960), pp. 24-29. 35. Witmer, David R. Unit Cost Studies. Madison, Wis- consin: Board of Regents of State Universities, ERIC #ED 013 492 (1967). GENERAL REFERENCES GENERAL REFERENCES Alkin, Marvin C. Towards an Evaluation Model ---#A _Systems Approach. ERIC #ED 014 150 (1967). Blalock, Hubert M. Social Statistics. McGraw-Hill Co., Inc. (1960). Ciancone, E.S. "New Technique for Instructional Analysis." Industrial Arts and Vocational Edu- cation, Vol. 57 (April 1968), pp. 35—39. Cook, Desmond L. "An Introduction to PERT." Colum- bus, Ohio: Monograph, Educational Program Manage- ment Center, O.S.U} (February 1964). Cook, Desmond L. "Better Project Planning and Control through the Use of System Analysis and Management Techniques." Columbus, Ohio: Monograph, Educa- tional Program Management Center, O.S.U} (November 1967). Cook, Desmond L. "The Impact of Systems Analysis on Education." Columbus, Ohio: Monograph, Educational Program Management Center, O.S.U. (April 1968). Cook, Desmond L. "The Nature of Project Management." Columbus, Ohio: Monograph, Educational Program Management Center, O.S.U. (June 1968). Cook, Desmond L. "Program Evaluation and Review Tech- nique." Washington, D.C.: U.S. Department of Health, Education and Welfare, Cooperative Re- search Monograph No. 17 (1966). Cook, Desmond L. "Some Economic Considerations in Edu- cational Project Planning." Columbus, Ohio: Monograph, Educational Program Management Center, O.S.U; (July 1968). 127 10. 11. 12. 13. 14. 15. 16. 128 Cox, D.R. Plannipg of Experiments. John Wiley and Sons, Inc. (1958). Durstine, Richard M. "In Quest of Useful Models for Educational Systems." Socio-Economic Planning Sciences, Vol. 2, Pergamon Press (1969). Froomkin, Joseph. "Cost/Effectiveness and Cost/ Benefit Analysis of Educational Programs." Socio-Economic Planning Sciences, Vol. 2, Pergamon Press (1969). Hays, William L. Statistics. Holt, Rinehart and Win- ston. (1963). Johnson, Craig. An Evaluation of Educational Develgp- ment Programs in Higher Education. USOE Grant No. OEG-O-8-070ll4—l856(OlO), Project No. 7-E- 114 (1968). Lyden, Fremont J., and Miller, Ernest G. Planning Programming Budgeting: A.Systems Approach to Management. Markham Publishing Co. (1969). Miller, Robert W. Schedule, Cost,_and Profit Control with PERT: A.Comprehensive Guide for Program Management. McGraw-Hill Book Co. (1963). INTERVIEWS 10. 11. 12. INTERVIEWS Allen, Ruth. Biochemistry SLATE Development. Michigan State University, 1970. Cole, Richard. Commercial Film Prodpction Costipg, Lansing, Michigan, 1970 Cook, Desmond L. PERT and Systems Analysis to Instruc- tional Development. Ohio State University, 1969. Echt, Robert. ,Anatomy SLATE Development. Michigan State University, 1970. Fiel, Nick. Physiolggy,Pre-Laboratory,SLAIES. Michigan State University, 1969. Foth, Henry. Soil Science SLATE Development. Michigan State University, 1970. Hagerman, Howard H. Biology SLATE Development. Michigan State University, 1970. Johnston, Raymond. Physiology SLATE Development. Michigan State University, 1970. Jones, Gardner. Instructional Cost Accountipg, Michi- gan State University, 1970. Kies, Martin. Commercial Software Production Costipg. Lansing, Michigan, 1970. Mikan, Kathy. Nursipg SLATE Developmept, Michigan State University, 1969. Page, James. AHV Methods SLATE Development. Michigan State University, 1970. 129 13. 14. 15. 16. 17. 130 Peltier, Linn. lnstitutional Research Techniques. Michigan State University, 1970. Postlethwait, Sam. Botany SLATES. Purdue University, 1969 and 1970. Shull, David. General Science SLATE Development. Lansing Community College, 1970. Veenenddal, Wilferd. Graphic Production Costs. Michi- gan State University Instructional Media Center, 1970. Watson, Archie. Hardware System Costs. Michigan State University Instructional Media Center, 1970. APPENDIX A Aggregate Data Interview Instrument SELFHINSTRUCTIONAL MATERIALS DEVELOPMENT UNIT COST-ANALYSIS INTERVIEW FORM I. DEVELOPMENTAL COSTS: A. Instructional Units: (Describe an Instructional Unit as those materials which are used and sequen- ced as a package to teach one or more concepts.) QUESTIONS RESPONSES 1. How many self-instruction- units al units were used with students after the initial development? What date was the initial started development started? What date was the develop- completed ment completed for student use? 2. What was the average student hours time per week spent in the self-instructional setting? 3. How many credits were offered credits for the course in which self- instructional materials were used? 4. What per cent of the total per cent course credits were being taught by the self-instruc- tional materials developed? 131 B. 132 _§guipment Costs: 5. ta 3 r What type and model of instructional equipment and/or carrels were used at each student sta- tion and in any adjacent demonstration areas? (Type, Model, Car., Dem.) How many of each type of equipment were used in the student station or adjacent demonstra- tion area? (#) What were the per/item costs for each type of equipment used? (S/I) What equipment was purchased (P) new, and what equipment was drawn from existing sources (E)? 8 8 carrel headset F O.H. 16mm screen 10. GRAND TOTAL How much did all the instruc- dollars tional equipment cost? What per cent discount, if discount any, was given on the list price of the equipment pur- chased? 11. 12. 13. 14. 15. 133 What was the average number (#) of films de- veloped per instructional unit? Were these films silent (Si) or sound (80)? Were these films reel-to-reel (R) or cart- ridge (C)? What was the average running time (X) per film? What was the average cost ($) per film? What was the average number (#) of photoprints developed per instructional unit? What size (52) were these prints? How many per unit were in the student hand-outs (H) and how many per unit were in centralized displays (D)? What was the average cost ($) per print? What were the average number (#) of audio tapes developed per instructional unit? Were these tapes on cassettes (C) or reels (R)? What was the average playing time (X) per unit? What was the average cost ($) per tape? How many (#) and what (type) models and other demonstration materials were developed for these instructional units? What was the average cost ($) per model? What was the average number (#) of student hand-outs developed for each instructional unit? How many pages (pg) were included in each hand-out? 134 Were graphic illustrations (ill) included in the student hand-outs? What was the average cost ($) per hand-out? 16. Which of the software were locally produced and which were purchased? 17. Which of the software locally produced were processed locally and which were processed commercially? (See "Software Response Form" next page.) 18. How many (#) produc- # Salary Time tion technicians were employed during the initial development of the software? What salary were they paid? How much time did they spend on the initial development? 19. Hovvmuch financial support dollars was from non-departmental sources such as federal, foundation, or university grants? 20. What per cent of the total per cent development costs were from non-departmental sources? 21. How much did the software dollars cost? 135 HHH n mpoopcmm u maumom x m u a m am a ax we axmpcaui ZmOm mmzommmm mm<3hm0m % D\mODHHm 136 D. Faculty and Content Assistants: 22. 23. 24. 25. How many faculty worked on the development of these self-in— structional materials? What was their tenure level? How many months per year was their appointment? How much faculty time was assigned to the de- velopment of these instructional units? How much faculty time was actually spent on the development of these units? How many content assistants such # as graduate assis- A. SEC. LEVEL TIME tants and secre- taries were used during the develop- ment of these in- structional units? What was their salary level? How much time were they assigned to spend on the development? How much faculty time hours was required to train the carrel room operators as to the use of the equipment and distribution procedures? 137 26. How many self-instruc- units tional units had you helped develop prior to these units? E. Consultants: 27. How many consul- tants were used during the develop- ment of these in- structional units? # Tenure Time Trav What type of exper- tise did they poss- ess, i.e., media, evaluation and/or learning theory? What was their ten- ure level by type? How much time did each spend dur- ing the develop- ment? 28. What travel costs were incurred by consul- tants coming to your campus or for you going to their location? F. Revision Costs: 29. Was a validation con- Yes No ducted on the ma- terials developed? 30. 31. 32. 138 What per cent (%) of the soft- ware materials were revised? Was the same met- hod of production and processing used for materi- als revised as was used during the initial pro- duction? If not, what pro- cedures were changed? How many faculty and other staff were involved in the validation of the materials developed? How much time were they in- volved in the #Fac. Time 10 12 tud. validation? What tenure level of faculty or staff were in- volved? How many students were the materials tried on? What do you estimate the revision costs to be? What facilities were modi- facilities fied as a result of the materials initially de- veloped? How much did these modi- fications cost? dollars II. 139 DUPLICATION COSTS: 33. 35. 36. 37. 38. How many students per term were the ma- Original # Present # terials developed ’17 to teach? I How many students per term are presently using the materials developed? How many carrels were the materi- Ori inall Present #Review als originally planned to fill? How many carrels are presently being used? How many of these carrels were (W) and are presently (P) used for review units? How many hours per original hours instructional unit were these carrels available to the students? How often were the units rotated in the Day Week Request Other ; carrel facility? L__ I .L_ By the day, week, or on request? What was the duplication cost for dollars these software materials? APPENDIX B Model I SELF-INSTRUCTIONAL MATERIALS DEVELOPMENT BUDGET PLANNING GUIDE FOR A FIRST APPROXIMATION ITEM VARIABLE AMOUNT I. The number of students who will be students using the self—instructional ma- terials developed equals II. Using the criteria in Figure 1.0, students select the number of students per per carrel carrel from Table 4.1.3 III. Divide the number of students under carrels Item I by the number of students per carrel under Item II to deter- mine the number of carrels needed. IV. Using the criteria in Figure 1.1 carrel, select an appropriate cost esti- demonstration mate from Table 4.3.1 and special equipment costs V. Using the criteria in Figure 1.3, software select an appropriate cost es- production timate from Table 4.3.3 VI. USing the criteria in Figure 1.4, software select an appropriate cost esti- duplication mate from Table 4.3.4 VII. Using the criteria in Figure 1.5, faculty select an appropriate cost esti- salaries mate from Table 4.3.5 140 141 SELF-INSTRUCTIONAL MATERIALS DEVELOPMENT BUDGET PLANNING GUIDE FOR A.FIRST APPROXIMATION (continued) ITEM VARIABLE AMOUNT VIII. Using the criteria in Figure 1.6, graduate select an appropriate cost estimate assistant and from Table 4.3.6 secretarial salaries IX. Using the criteria in Figure 1.7, consultant select an appropriate cost esti- fees mate from Table 4.3.7 X. Using the criteria in Figure 1.8, valida- select an appropriate cost esti- tion and re- mate from Table 4.3.8 vision costs XI. Add the unit cost/carrel from total Items IV through X cost per carrel XII. Since the baseline for the data in % inflation Tables 4.1, 4.2 and 4.3 is 1967, subtract 1967 from the year which your development will take place and multiply the difference by the average inflation factor of 7.7% per year. XIII. Multiply the total cost per carrel amount under Item XI by the inflation increased percentage determined under Item XII. XIV. Add the amount of increase under adjusted Item XIII to the total cost per cost per carrel carrel determined under Item XI. XV. Multiply the total cost per car- total de- rel by the number of carrels velopment costs under Item III. XVI. USing the criteria in Figure 1.9, faculty select an appropriate time esti- mate for faculty involvement from Table 4.3.9 time per car- rel 142 SELF-INSTRUCTIONAL MATERIALS DEVELOPMENT BUDGET PLANNING GUIDE FOR A FIRST APPROXIMATION (continued) ITEM VARIABLE AMOUNT XVII. Multiply the number of hours of total hours faculty time per carrel under Item of faculty time XVI by the number of carrels under Item III. XVIII. Using the criteria in Figure 1.10, months for select an appropriate time esti- development per mate for the number of months of carrel development/carrel from Table 4.3.10 XIX. Multiply the number of months per total # of carrel for development under Item XVIII by the number of carrels under Item III. months for de- velopment APPENDIX C A Simulation of the Revised Self-Instructional Materials Development Budget Planning Guide A.SIMULATION OF THE REVISED SELF-INSTRUCTIONAL MATERIALS DEVELOPMENT BUDGET PLANNING GUIDE Dr. Edgar P. Pennyworth is an Assistant Professor in charge of developing ten self-instructional units during the 1971-72 academic year for two hundred basic biology students at XYZ University. He has had no experience with self-instructional materials development but feels that those biological concepts which lend themselves to self- instruction are best taught with a balanced audio and visual media strategy. He anticipates some content and secretarial assistance and moderate consultative support from the campus media center, and his faculty colleagues. He also plans to validate the materials with the help of some student volunteers. Appropriate revisions are then anticipated. The following is a simulation of Dr. Pennyworth's use of the "Self-Instructional Materials Development Budget Planning Guide." 143 HOW TO USE THE SELF-INSTRUCTIONAL MATERIALS BUDGET PLANNING GUIDE Introduction The following budget planning guide is based on a predictive cost model for self-instructional materials development in higher education for the biological and physical sciences. In using this planning guide, you will be asked to make a variety of decisions based upon input data which reflect your local conditions. The guide will take you step—by-step through the various cost para- meters and will refer you at each decision point to a num- bered criteria Table which will help you make the approp- riate selection of specific values from the accompanying data tables. Once a value is selected, you should write it down and then proceed to the next step on the planning guide for instructions as to how to proceed. (The follow- ing sequence is used: (1) read the first item in the plann— ing guide, (2) then read the appropriate criteria Table before making a decision, (3) decide which value to select from the appropriate data Table, (4) write down the value in the guide and (5) proceed to the next step in the plann- ing guide.) 144 Assumptions It must be emphasized that instructional development is a relatively creative process and to ascribe standard costs to specific items is difficult at best, since individual programs and local economic conditions will vary from institution to institution. Therefore the usershould realize that the cost estimates derived from the use of this planning guide should be considered approximations, rather than exact values. Further, it is assumed that your instructional planning is in the initial phases, meaning that instructional objectives and strategies are still quite general. Specific quantities of tape, slides, films or hand-outs are yet to be determined, and as the instructional planner, you are likely at the stage where you are attempting to determine a cost estimate for a spe- cific number of units and students before proceeding with the development. A third assumption used in this budget planning guide is that you are developing materials in an existing institution which is changing an instructional strategy for a specific series of biological or physical science concepts. Therefore, this guide does not include cost estimates of building and equipping a new biological or 145 physical science laboratory facility. If the planning guide is to be used by new institutions, the cost of adding additional equipment and laboratory supplies should be added to the cost estimate determined by the use of this planning guide. It is also important to realize that the model upon which this planning guide is based was designed from higher education biological and physical science pro- gram data, and therefore the use of this planning guide is limited at the present time to those projects in these academic curricular areas. Finally, only two basic media strategies are accounted for in this planning guide. The first strategy consists of a mix of slide, audio tapes, hand-outs and film production. Films, however, are optional, depending upon the concepts to be taught. The second media strategy in- volves the use of audio tapes, with only a few photoprints. Microscope slides and other materials to be included in the carrel may be a part of this second strategy, but no other hardware or visual software is involved. It should be noted that there are other strategies of self-instruc- tion available to the biological and physical sciences such as programmed texts, computer-aided instruction, closed-circuit video recordings, and motion picture films. 146 If you are interested in a strategy of instruction other than the slide-tape-film combination, or tape-only stra- tegy, then this particular planning guide will not suit your purpose. If you feel that the concepts which you are developing materials for will be effectively taught by either of the two strategies included in this planning guide, you are on your way. 147 148 SELF-INSTRUCTIONAL MATERIALS DEVELOPMENT BUDGET PLANNING GUIDE ITEM FPARAMETER OR CONSTRAINT VALUE (1) How many students will be using 200 students the self-instructional materials developed? (2) If you have already determined Carrels the number of carrels which you needed (3) (4) will need for the materials you intend to develop, enter that value to the right, and proceed to Item 9. If you are undecided about (1) the approximate length of time each student will need to spend per instructional unit, but dg know (2) the length of time which the carrel room will be open for student use, then enter this value and use the criteria in Table 7.1 to select the proper value from Table 4.0.3. Enter this value also and proceed to Item 7. If, however, you are undecided as to the number of carrels you will need but d2 know both (1) the approximate length of time each student will need to spend per instructional unit, apd (2) the length of time which the carrel room will be open for student use, then enter these values and proceed to Item 7. 2-6 student hours/unit 60 hours car- rel room open/unit student hours/unit hours car- rel room open/ unit lTEM ( 149 PARAMTER OR CONSTRAINT VALUE (5) (6) (7) (3) (9) If you have neither predetermined the number of carrels needed, nor know (1) the approximate length of time each student will need per instructional unit apd (2) the length of time which the carrel room will be open for student use, then use the cri- teria in Table 7.0 to select the number of students per carrel from Table 4.0.2. Enter the value and proceed to Item 6. Divide the number of students under Item 1 by the number of students per Carrel under Item 5 to determine the number of carrels needed. Enter the value and proceed to Item 9. Multiply (l) the approximate length of time each student will need to spend per instructional unit py (2) the number of stu- dents under Item 1. Enter the value and proceed to Item 8. Divide (1) the number of "stu- dent hours needed in carrels" py_(2) the length of time which the carrel room will be open for student use as recorded in Item 3 or 4. Enter the value and proceed to Item 9. Will any separate carrels be assigned for review purposes in addition to the carrels being used to present current in- structional units? If so, enter the value and proceed to Item 10. If not, proceed to Item 11. students per carrel carrels needed 520 student hours needed in carrels 9 carrels needed 1 review carrels ITEM 150 PARAMETER OR CONSTRAINT VALUE (10) (11) (12) Add_(l) the number of carrels needed which you determined in Item 2, 6, o;_8 plus (2) the number of review carrels de- termined in Item 9. Enter the total number of carrels needed to the right and proceed to Item ll, Using the criteria in Table 7.2, select an instructional strategy mix which fits the concepts and behavioral objectives which you wish to achieve with self- instructional materials. Check your decision to the right, and proceed to Item 12. Using the criteria in Table 7.3, select the equipment mix for which you wish to budget in your planning. Check your decision to the right. (a) If you Chose slide, tape and film under Item 11 apd_carrel, demonstration and special de- velopment equipment under Item 12, then proceed to Item 13. (b) If you chose slide, tape and film under Item 11, but carrel equipment only, then proceed to Item 14. (c) If you chose audio tape only under Item 11 apd chose carrel, demonstration, and spe- cial development equipment under Item 12, then proceed to Item 15. 10 total num- ber of carrels needed 1 slide + tape + film (optional) audio tape only \/ carrel , de- monstration and special develop- ment equipment carrel equip- ment only 151 ITEM PARAMETER OR CONSTRAINT VALUE (d) If, however, you chose audio tape only under Item 11 apd carrel equipment only under Item 12, then proceed to Item 16. (13) Using the criteria in Table 7.4, 605 unit cost select an appropriate cost es- of carrel, demons- timate from Table 4.1.1. Enter tration and special the value and proceed to Item equipment l1, (14) Using the criteria in Table unit cost of 7.5, select the appropriate carrel equipment unit cost estimate from Table only 4.1.2. Enter the value and proceed to Item 17. (15) Using the criteria in Table unit cost of 8.0, select an appropriate carrel, demonstra- cost estimate from Table 4.1.1. tion and special Enter the value and proceed to equipment Item 17. (16) Using the criteria in Table unit cost 8.1, select the appropriate of carrel equipment unit cost estimate from Table only 4.1.2. Enter the value and proceed to Item l7. (l7) Multiply (l) the total number .total equip- of carrels needed under Item 2L6J 8 or 10 by (2) the value you selected under Item 13, 14, 15, p£_l6. Enter the value to the right. (a) If you selected the slide, tape and/or film strategy mix under Item 11, then proceed to Item 18. ment costs 152 ITEM PARAMETER OR CONSTRAINT VALUE (b) If, however, you selected the audio tape only strategy under Item 11, then proceed to Item 37. (18) ' Using the criteria in Table 246 software ( 7.6, select an appropriate cost production unit estimate from Table 4.2.1. costs Enter the value and proceed to Item 19. (19) Using the criteria in Table $18-30unit soft- 7.7, select an appropriate cost ware duplication estimate from Table 4.2.2. costs/instructional Enter the value and proceed to unit/carrel Item 20. (20) How many self-instructional 10 instruc- units will be developed during tional units the initial development period? Enter the value and proceed to Item 21. (21) Multiply (l) the value in Item $183 software l2_by (2) the number of units duplication in Item 20. Enter the value cost/carrel ‘ and proceed to Item 22. (22) Multiply (1) the value in Item .total soft- gl_by (2) the number of carrels ware duplication needed for regular instruction costs in Item 2, 6, pp 8. Enter the value and proceed to Item 23. (23) Using the criteria in Table $920 faculty ZLQJ select an appropriate cost estimate from Table 4.2.3. Enter the value and proceed to Item 24. salary unit costs 153 ITEM PARAMETER OR CONSTRAINT VALUE (24) Using the criteria in Table $90 content 2&2, select an appropriate cost assistant and sec- estimate from Table 4.2.4. retarial wage unit Enter the value and proceed to costs Item 25. (25) Using the criteria in Table $79 consultant 7.10, select an-appropriate cost unit costs estimate from Table 4.2.5. Enter the value and proceed to Item 26. (26) USing the criteria in Table $27 validation 7.11, select an appropriate and revision unit cost estimate from Table costs 4.2.6. Enter the value and proceed to Item 27. (27) Add the unit cost/instructional $1362 total soft- unit from Items 18, 23, 24, ware production 25, apd_26. Enter the total costs per instruc- and proceed to Item 28. 'oo.l unit (28) Multiply (1) the total software otal soft- production costs per instruc- ware production tional unit in Item 27 by (2) costs the number of instructional units in Item 20. Enter this value and proceed to Item 29. (29) Add.(l) the total equipment $21I3I7total de- costs in Item 17, (2) the total software duplication costs in Item 22 apd_(3) the total soft- ware production costs in Item gp, Enter the total and pro- ceed to Item 30. velopment costs 154 ITEM PARAMETER OR CONSTRAINT VALUE (30) Since the baseline for the __;Z;L% infla- data in Tables 4.0, 4.1 and tion 4.2 is 1967, subtract 1967 from the year which your de- velopment will take place and multiply (l) the difference by (2) an average inflation factor of 7.7% per year. En- ter the per cent and proceed to Item 31. (31) Multiply (l) the total develop- $1641 cost increase (33) ment costs in Item 29 by (2) the inflation percentage de- termined in Item 30. Enter the cost increase and proceed to Item 32. Add the cost increase in Item BI—to the total development costs determined in Item 29. Enter the value and proceed to Item 33. Using the criteria in Table 7.12, select an appropriate time estimate for faculty in- volvement from Table 4.2.7. Enter the value and proceed to Item 34. Multiply (l) the number of hours of faculty time per in- structional unit in Item 33 by (2) the number of instruc- tional units in Item 20. Enter this value and proceed to Item 35. due to inflation (:§22,95;:Ldjusted es- timate of total development costs 109 faculty time per instructional unit ($1090 )total es- timated hours of faculty time ITEM 155 PARAMETER OR CONSTRAINT VALUE (35) Using the criteria in Table 7.13, select an appropriate time estimate for the number of months of development/instructional unit from Table 4.2.8. Enter this value and proceed to ltem 36. Multiply (l) the number of months per instructional unit develop- ment in Item 35 by (2) the num- ber of instructional units in Item 20. Enter this value. Recheck your calculations, assump- tions and decisions, then read the following: Items 32, 34, and 36 provide an estimate of (1) the total de- velopment costs, (2) the number of hours of faculty time and (3) the total number of months for development. In using these totals, realize that these values are only estimates. As you get into the development, you may want to adjust your budget up or down to meet Changes in the behavioral objectives, instruc- tional strategies or economic conditions. Items 34 and 36 should help in determining the faculty time necessary to com- plete the development and in determining the total number of months you should plan in advance of using the materials you de- velop. 1 .33 months for development per instructional unit otal es- timated # of months for de- velopment 156 AUDIO TAPE ONLY (continued from Item 17) ITEM PARAMETER OR CONSTRAINT VALUE (37) Using the criteria in Table software 8.2, select an appropriate production unit cost estimate from Table costs 4.2.1. Enter the value and proceed to Item 38. (38) Using the criteria in Table 8.3, unit soft- select an appropriate cost ware duplication estimate from Table 4.2.2. costs/instructional Enter the value and proceed to unit/carrel Item 39. (39) How many self-instructional instruc- units will be developed during tional units the initial development period? Enter the value and proceed to Item 40. (40) Multiply (1) the value in Item software §§_by (2) the number of units duplication in Item 39. Enter the value cost/carrel and proceed to Item 41. (41) Multiply (1) the value in Item .kotal soft- gQ_by (2) the number of carrels ware duplication needed for regular instruction costs in Item 2, 6 g£_8. Enter the value and proceed to Item 42. (42) Using the criteria in Table 8.4, faculty select an appropriate cost salary unit estimate from Table 4.2.3. costs Enter the value and proceed to Item 43. (43) Using the criteria in Table 8.5, content select an appropriate cost es- assistant and timate from Table 4.2.4. Enter secretarial the value and proceed to Item salaries .42- ITEM 157 PARAMETER OR CONSTRAINT VALUE (44) (45) (46) (47) (48) (49) Using the criteria in Table 8.6, select an appropriate cost es- timate from Table 4.2.5. Enter the value and proceed to item 45. Using the criteria in Table 8.7, select an appropriate cost es- timate from Table 4.2.6. Enter the value and proceed to Item 46. Add the unit cost/instructional 63?} from Items 37, 42, 43, 44 app 45. Enter the total and proceed to Item 47. Multiply (l) the total software production costs per instruc- tional unit in Item 46 by (2) the number of instructional units in Item 39. Enter this value and proceed to Item 48. Add_(l) the total equipment costs in Item 17, (2) the total software duplication costs in Item 41, and (3) the total software production costs in Item 47. Enter this total and proceed to Item 49. Since the baseline for the data in Tables 4.0, 4.1 and 4.2 is 1967, subtract 1967 from the year which your development will take place and multiply (l) the difference by (2) an average inflation factor of 7.7% per year. Enter the per cent and proceed to Item 50. consultant unit costs validation and revision unit costs total soft- ware production costs per instruc- .00-1 unit ware production costs total de- velopment costs % inflation 158 ITEM PARAMETER OR CONSTRAINT VALUE (SO) Multiply (l) the total develop- cost in- ment costs in Item 48 by (2) crease due to in- the inflation percentage de- flation (52) (54) termined in Item 49. Enter the cost increase and proceed to Item_519 Add the cost increase in Item 50_to the total development costs determined in Item 48. Enter the value and proceed to Item_52. Using the criteria in Table 8.8, select an appropriate time estimate for faculty in- volvement from Table 4.2.7. Enter the value and proceed to Item 53. Multiply (1) the number of hours of faculty time per in- structional unit in Item 52 by (2) the number of instruc- tional units in Item 39. Enter this value and proceed to ltem 54, Using the criteria in Table 8.9, select an appropriate time estimate for the number of months of development/instruc- tional unit from Table 4.2.8. Enter this value and proceed to ltem 55. Multiply (l) the number of months per instructional unit for development in Item 54 by (2) the number of instruc- tional units in Item 39. Enter this value. Recheck your cal- culations, decisions, assumptions Cases estimate of total development costs faculty time per instructional unit timated hours of faculty time months for development per instructional unit timated # of months for development I ITEM 159 PARAMETER OR CONSTRAINT VALUE and then read the following: Items 51, 53, and 55 provide an estimate of (l) the total development costs, (2) the number of hours of faculty time and (3) the total number of months for development. In using these totals, realize that these values are only estimates. As you get into the development, you may want to adjust your budget up or down to meet changes in the behavioral objectives, instruc- tional strategies or economic conditions. Items 53 and 55 should help in determining the faculty time necessary to com- plete the development and in determining the total number of months you should plan in advance of using the materials you develop. 160 TABLE 4.0 GENERAL DEVELOPMENT DESCRIPTIVE DATA ITEM HIGH LOW AVERAGE (1) Months for initial 9 months 3 months 6.7 months development per course (2) Students/carrel 15 stdts/ 5 stdts/ lO stdts/ carrel carrel carrel (3) Student hours in cartel/unit 4 stdt. .56 stdt. 2.6 stdt. hours hours hours TABLE 4.1 DEVELOPMENT COSTS PER CARREL 161 Slide + Tape + Film Audio Tape ITEM DESCRIPTION (optional) Only High Low Ave High Low Avg. 1 Carrel, demons- $883 $441 $605 $420 $270 $300 tration and special develop- ment equipment costs 2 Carrel equip- $780 $386 $543 $337 $195 $238 ment costs alone 162 mumoo conH>mu 0w 00.nmw 00.4Hw 00.0vw 00.nmw 00.4Hw 00.0vw paw soaemowaa> o momoo -0- 00.ma 00.HH 00.0w 00.05 00.mH 00.mmm unmeasmsoo m mpmoo HMHHMpmu Iomm can macaw -0- 00.Hm 00.m 00.m0 00.4mm 00.00 00.0mmH -mammm esopsoo v -0- 00.mmm 00.nsa 00.0om 00.0mm 00.va 00.0osa memos seasons 0 mo moo Hmuum0\pwcs Mom :OHpmo -- 0m.s m0.v 0o.v 0m.0e m0.v 0o.ms -wHeso mumsemom m mpmoo coHp -- 00.naw 00.4Hw 00.mmw 00.o¢mw 00.Hmw 00.mHme -osooee oemzemom H o>¢ 33 sou: .92 33 coal 28.850me 28H mzoz >320 AHmsoaeeov mmmc m.v mqm -- .5 mm. .03 mm. .03 m.m ca mm.H .oa em. .03 m.w -00 How mascoz 0 -0- muss um .mus Hm .mus ms mus 00H .mus om .mes owe wane seasons s 0>¢ 38 5H3 0>< 33 ZOHE 20H: HHmUwMQ 2m“: mzozm NAZO AHmcodeov mmmc N.v mqm