Ask the Scholar

Document scope · 1 page
doc
Scholar
Ask about this object, its catalog metadata, its source description, or the page inventory. For page-specific OCR and visual context, open one of the page chats.

Scholar Source Context

Document identity
localId
351002347
label
"A Resource Handbook for Vocational Rehabilitation" [1974]
core
doc
dtoType
document
pageCount
1
Source metadata
Source extras
naId
351002347
levelOfDescription
fileUnit
recordType
description
ocrSource
nara-archive
Single page context
seq
1
pageIndex
0
type
document
mediaId
6be042126fff800d
ocrText
Originally Processed With FOIA(s): FOIA Number: S S FOIA MARKER This is not a textual record. This is used as an administrative marker by the George Bush Presidential Library Staff. Record Group/Collection: Donated Historical Materials Collection/Office of Origin: Frieden, Lex, Collection Series: Printed Materials Subseries: Manuals OA/ID Number: 52104 Folder ID Number: 52104-010 Folder Title: "A Resource Handbook for Vocational Rehabilitation" [1974] Stack: Row: Section: Shelf: Position: A RESOURCE HANDBOOK FOR VOCATIONAL REHABILITATION ICD Laboratory PROGRAM EVALUATION: A Resource Handbook For Vocational Rehabilitation Eleanor Carol Bennett Resource Analyst Research Utilization Laboratory ICD Rehabilitation and Research Center Marvin Weisinger Associate Project Supervisor Research Utilization Laboratory ICD Rehabilitation and Research Center ICD Rehabiliation and Research Center New York, N.Y. 10010 Copyright © 1974 ICD Rehabilitation and Research Center except for portions copyrighted by others, permission for the reprinting of which herein has been obtained from the respective copyright owners. Portions of this publication, except for those portions copyrighted by others, may be reprinted provided per- mission to do SO is obtained in writing from the ICD Rehabilitation and Research Center or from the Social and Rehabilitation Service, U.S. Department of Health, Educa- tion, and Welfare with whose support the project resulting in this publication was conducted. The development of this publication was supported, in part, by Research Grant No. SRS 22-P-55060/2 from the Division of Research and Demonstration Grants, Social and Rehabilitation Service, Department of Health, Education and Welfare, Washington, D.C. 20201. TABLE OF CONTENTS Page ACKNOWLEDGEMENTS V PREFACE vi I. PROGRAM ASSESSMENT 1 OBJECTIVITY 2 A PERSPECTIVE 5 A SUGGESTED MODEL 7 Classes of Measures 7 Program Level 9 A Word on Indices 11 Some Additional Considerations 11 SUMMARY 15 II. PROGRAM EVALUATION DESIGN 17 STATEMENT OF PROGRAM EVALUATION PROBLEM 18 Operationalization 19 SPECIFICATION OF METHODOLOGY 21 Approaches to Evaluation 21 Research Approach 22 Systems Approach 24 Reliability, Bias, Validity 27 Selection of the Sample 28 COLLECTION OF DATA 29 ANALYSIS OF DATA 32 Levels of Statistical Measures 32 Data Analysis Techniques 33 Research Approach 33 Systems Approach 40 INTERPRETATION AND CONCLUSIONS 43 III. THE NINTH PANACEA: PROGRAM EVALUATION 45 By Robert A. Walker IV. COMPENDIUM OF MEASURES 61 PROGRAM 63 Effort 63 Effectiveness 64 Efficiency 70 Quality 71 SERVICE DELIVERY 72 Counselor Performance 73 Evaluator, Job Trainer, Placement Officer 75 PROGRAM MANAGEMENT 76 Administrative Functions 77 Community Relations 78 Staffing 81 Inservice Training 82 Fiscal Administration 84 WEIGHTED CLOSURES 85 CASE FLOW (Monitoring) 86 . V. EXAMPLES OF PROGRAM EVALUATIONS 89 PLANS 90 RESEARCH APPROACH 95 SYSTEMS APPROACH 99 COST BENEFIT ANALYSIS 104 OTHER FIELD EXAMPLES 109 GLOSSARY 111 READING LIST 117 REFERENCE LIST 139 APPENDIX I. Other Classes of Measures Used in Program Evaluation 149 APPENDIX II. Standards 150 ACKNOWLEDGEMENTS Compilation of this handbook reflects the support and suggestions of many people. Special acknowledgment must go to the staff of the Research Utilization Laboratory whose aid and encouragement was instrumental in its completion- Dr. Isabel Robinault, Project Supervisor and Susan Ahrenhold, Information Specialist. Dr. Salvatore G. DiMichael, Project Director and Director of ICD Rehabilitation and Research Center provided valuable assistance to this effort by placing the Center's resources at our disposal. Helen Stonehill, Center Librarian, generously supplied us with much of the esoteric literature in the field of program evaluation. In addition, we must express our great appreciation for the fortitude and patience of our secretarial staff, Helen Carr and Christine Hicks, who typed numerous drafts of this document. Finally, we wish to thank the many program evaluation professionals whose substantial consultant effort is critical to enhancing the Handbook's value for the field. Any errors in the interpretations and compilations based on the consultant's input for the program evaluation examples and compedium of measures are ours alone. We particularly would like to acknowledge the efforts of the following: Edward C. Baumheier, Ph.D. Leonard Miller, Ph.D. Janet M. Derr Division of Counselor Education Lalitha Raichur College of Education Social Work Research Institute University of Iowa University of Denver Iowa City, Iowa 52242 Denver, Colorado 80210 John Muthard, Ph.D. James Bitter, Ph.D. Regional Rehabilitation Research Institute Regional Rehabilitation Research Institute University of Florida University of Northern Colorado Gainesville, Florida 32601 Greeley, Colorado 80631 Robert Philbeck Gabriel Cifor Assistant Director for Planning & Evaluation Robert Struthers, Ph.D. Division of Vocational Rehabilitation Department of Education/ North Carolina State Department of Human Vocational Rehabilitation Services Resources Lansing, Michigan 48904 Raleigh, North Carolina 27602 Larry Dickerson, Ph.D. Sidney Stein Research & Training Center Program Evaluation University of Wisconsin-Stout Division of Vocational Rehabilitation Menomonie, Wisconsin 54751 Texas Rehabilitation Commission Austin, Texas 78705 William Hills, Ph.D. Barbara Chesney, Ph.D. Adam Zawada Regional Rehabilitation Research Institute Program Planning and Research The University of Oklahoma Division of Vocational Rehabilitation Norman, Oklahoma 73069 Tallahassee, Florida 32304 Adrian J. Marinelli Research Utilization Specialist New Jersey Rehabilitation Commission Trenton, New Jersey 08625 V PREFACE This handbook on program evaluation is a compilation of approaches, methods, and techniques culled from over 250 sources. It is an attempt to bring together a large array of information and organize it into a unified conceptual framework. The evaluator, whether novice or experienced, often faces a bewildering mountain of information when he attempts to examine the literature of program evaluation. There seems to be no one source where information on indices can be obtained. Indeed, from publication to publication, there seems to be little or no agreement even on definition of terms. Therefore, one of the major efforts of this manual was to review the diverse literature and opinion on program evaluation, to organize it, and reduce it to a useable form. However, it should be clearly understood that while the scope of the Handbook is comprehensive, we have not attempted to write the "magnum opus." Program Evaluation: A Resource Handbook is in no way a substitute for a text book, background training or experience. Some of the defined approaches and techniques require a good deal of technical sophistication to be properly applied, interpreted, and used in making program decisions for change. Some techniques and approaches are relatively simple to apply and this is indicated. Most important, we have tried to be cautionary where appropriate. The contents of this manual are directed to three audiences. One audience is comprised of the new evaluators. We assume they have been exposed to a course in statistics and a course in research methods, have read several program evaluations and are now wondering how to put it all together. The second audience consists of experienced evaluators who might be looking for some specific measures or techniques to solve an immediate problem. For the former, we hope this manual will serve as a major adjunct to his training and give him some important leads, cues and cautions on how to organize and proceed with a program evaluation. For the latter, the annotated list of indices in Chapter IV and the Bibliography and Reading List could be most helpful. The third audience consists of administrators. For them, we hope this handbook will serve as an aid to communication with program evaluation staff. Since program evaluation is in its relative infancy in some of the social services and since our review of the literature indicated little consensus among authors, we expect certain honest differences of opinion regarding the material contained in this handbook. We hope this document will be read, discussed and, perhaps, even argued over. New York Eleanor Carol Bennett June, 1974 Marvin Weisinger vi PROGRAM EVALUATION: A Resource Handbook For Vocational Rehabilitation PROGRAM ASSESSMENT Today, program assessment is a key concern of management-both govern- mental and private. Controversy surrounds it and published contradictions are rampant. Some of the reasons for this controversy and lack of understanding is the relatively recent use of program assessment in the management processes of human services where its purpose may seem antithetical to these services. In addition, the personnel involved in assessing programs and those utilizing the results of such assessments often have conflicting philosophies and differing goals. Consequently, the information-decision-making value of program assess- ment can get lost or may be interpreted as a spectre capable of damaging worthy programs, productive jobs and creditable reputations. Part of this difficulty lies in the differing backgrounds (and thus differing philosophies) of the evaluator and the practitioner-administrator. In many cases, the administrator who received his primary training as a practitioner in the helping services can be quite astute in systematically analyzing the complex problems of clients and recommending appropriate actions. Yet, when he is with systematically analyzing the complex problems of a program and recom- mending actions, he is somewhat at a loss. In addition, the training of a 1 practitioner emphasizes the solution of client problems with all due speed, without too much regard for economy. On the other hand, the evaluator's background stresses precision and control. He is often at a loss to understand why the boundaries for acceptable perform- ance for program activities are set SO wide. Communication between evaluator and administrator is often at cross pur- poses, even though they may be discussing the same things, they use different words. One purpose of this handbook is to provide an administrator with a conceptual framework which will aid him to effectively communicate with evaluation staff and to aid him in analyzing his program with the same acumen he analyzes client problems. It must be clearly understood that the final outcome of any evaluation process is change. Evaluation basically is a systematic and objective method of looking at where a program presently stands and estimating how a program can do better. Change is an integral part of this examination-change to serve the clients better and change in program economy. Only through an orderly process of documented program evaluation can justifiable alternatives for action be developed for the evolution of more efficacious and economical programs. OBJECTIVITY What sort of information should be used to document current program activity and to justify future change? The answer that comes quickly to mind is, "objective information". Two questions can then be asked, "How objective can an assessment be?" and "How does one objectively assess a program?" A program assessment can be conducted on one of two major levels of objectivity. The least objective, Level I, is the opinion of experts. That is, it is an assessment based upon the review of project documents or based on an onsite visit by a person recognized as "expert" in the field. The review may be superficial or in depth, but it relies on the previous experience of the reviewer for the ultimate statement of program worth. It is quite conceivable that two experts, reviewing the same program, could present two very different sets of findings. Therefore, choice of this level of objectivity should be avoided when- ever possible. The more objective approach, Level II, is the systematic collection and analysis of hard data. If this approach is followed, it should produce findings that are reproducible no matter who undertakes the assessment. Who administers the program assessment depends on the organization of the agency, the finances available for the assessment and the source of the request for assessment. Usually, there are three categories of people assessing programs: consultants or consulting firms, panels of knowledgeable people and depart- ments within the organization. 2 Exactly who administers the appraisal many times influences the impact of the assessment. For instance, an expert visiting the program on a site visit, or an outside researcher collecting "relevant" data to generate "meaningful" compar- isons can be open to attack on such dubious grounds as "lack of familiarity with local conditions" or the researcher's "Ivory Tower" approach. On the other hand, an "in-house" assessment may be accused of a "white wash" or bias. There is no simple solution to the selection of the evaluator. However, the use of a rigorous objective approach that is replicable can usually be defended against such attacks no matter who administers the assessment. The answer to the second question "How does one objectively assess a program?" is dependent on a number of factors: the type of information available, the precision required from the conclusions, and the funds available to undertake this assessment. There are three basic means of program assessment, that is, a program may be monitored, evaluated, or be the subject of a special study. Monitoring is ongoing feedback of information about professional perform- ance and program activities for the purpose of controlling those activities. This type of assessment serves as an indicator of disruption in program activities, but does not necessarily define the nature of the difficulty. No value judgement is offered concerning the "goodness" or "badness" of the program, although application of externally or internally derived standards may offer some kind of statement concerning professional activities. Examples of usual monitoring methods are caseflow information and case review. In some instances, ongoing outcome measures are examined to indicate disruption of activities. Program Evaluation usually examines annual program information and can offer a value judgement on an entire program's worth. It can identify causes of problems or reasons for success. Usually this judgement is based on program outcomes, or the outcomes of program components. Special studies offer a judgement from the examination of one facet of a program. That facet may be universal, in that it pervades many portions of the program, or it may be highly circumscribed. The final impact of that assessment depends upon what aspect of the program has been examined. For instance, to examine "the effectiveness of counselor performance when serving the culturally disadvantaged" is a pervasive problem encompassing attitudes and program policy. On the other hand, to examine "the effectiveness of white counselor performance in dealing with aged, black alcoholics" is specific to the administration of one program and, as such, is so circumscribed as to be of doubtful utility. 3 Figure I illustrates some of the benefits and limitations of the combinations of level of objectivity and level of assessment. Since the methods, dynamics and value of monitoring are well understood as necessary mechanisms for admin- istrative control, our discussion of this approach is limited to a presentation of indices for monitoring in Chapter IV. It can be seen from the above discussion that objectivity is partially a matter of the degree of refinement of information from which judgements are to be made. The remainder of this monograph is addressed to that process which is based on the analysis of hard data and is used to evaluate programs. This technique, program evaluation, is just now emerging as a program-planning and decision-making tool. "Special Studies" may be considered as a limited or circumscribed case of the broader "Program Evaluation". Therefore, any of the following discussions which refer to Program Evaluation may be considered as perfectly applicable to the topic of Special Studies. FIGURE I BENEFITS AND LIMITATIONS OF PROGRAM ASSESSMENTS Level Level of Assessment of Objectivity Monitoring Special Studies Evaluation Benefits Benefits Benefits I. Expert's Inexpensive Inexpensive Inexpensive Opinion Not time consuming Not time consuming Not time consuming from May identify problems May be perceptive May be perceptive A. Site Visit B. Annual Limitations Limitations Limitations Report Not replicable Not replicable Not replicable May not identify problems May be imprecise May be imprecise May not identify cause Lack of adequate firm Lack of adequate data firm data May over-emphasize minor detail Benefits Benefits Benefits II. Analysis of Ongoing Precise Precise Hard Data Identifies problems Replicable Replicable Replicable Identifies cause Identifies cause Broad in scope Limitations Limitations Limitations Does not identify cause Expensive Expensive More expensive than May be time consuming Time consuming Level I Limited in scope 4 A PERSPECTIVE In reviewing the program evaluation literature, several problems confront the reader. A great deal of literature is directed to extolling program evaluation's virtues as an aid in management, and an equally large collection decries the damage it can and does do, if perspective and methodology are not sound. A smaller body of representative literature discusses technique and methodology. However, these methodology articles are written from what appears to be many conflicting points of view. Actually, close examination reveals only two or three points of view. The first point of view is that of administrators who have learned to analyze their programs by using several different models. Probably the most popular model speaks of input analysis, process evaluation and outcome evaluation. A second model refers to levels of program responsibility or control and mentions the institution (program), managerial functions, and the technical core (service delivery). (95) An administrator, when considering his program, may view it as a system to be controlled and regulated. He then speaks of specific parts such as inputs, process, outputs, and some sort of a "feedback loop". The analyses of these parts may be defined as follows: Input Analysis-the - study of the resources (men, money and materials) used by the program to accomplish its goal. Process Evaluation-the study of a combination of program operations and techniques used by a program to accomplish its goal. Criteria usually are addressed to questions of conformity to, and expectations of the respec- tive professions involved. These criteria are usually called Quality. Many times the statistics used to monitor programs are composed of these types of data. Outcome Evaluation-the study of the results of a program's activities. Usually, the criterion used is the degree to which programs have attained goals. Outcome evaluation is frequently obtained through follow-up studies. The second model relates to specific levels of program responsibility (responsibility model)-the institutional level, managerial level and technical level. This breakdown may be considered as a reduction to program level, that is: * *Program monitoring and evaluation are part of the feedback loop. 5 Institution-(a program) is supported by a larger social system which legitimatizes the program's goals. Stated another way, we can say society expects a vocational rehabilitation program to assist the disabled to become meaningfully employed. Technical-(service delivery) is a sub-organization or level of program. Its function is to provide the services to clients and SO accomplish the program's goal. Managerial-(program management) is another sub-organization or level of program. Its function is to mediate between services, and between services and clients. In addition, it procures resources and aids in assuring appro- priate outputs. A second point of view is that of the evaluator who usually addresses his perspective to methodology. This perspective can be divided into two camps- the researcher who speaks of comparisons and goal attainment, and the systems analyst-operations researcher who speaks of effectiveness and survival models. (34) For the purposes of practical illustration, the popular goal-attainment model may be considered as an inductive approach, while the systems models usually are deductive in nature. Research constructs hypothesis about the way programs work, and the systems approach builds models describing program structure, functions and goals. Below are our definition and distinctions of the two approaches: Research Approaches seek to evaluate programs through making certain comparisons between variables. The basic question is, has the program caused a change in a variable? For example, has the introduction of a PA/VR program caused more PA recipients to leave the welfare rolls? These comparisons may be: before VS. after introduction of a program, between programs operating in differing localities, between clients in a special program and clients who did not have the benefit of the special program, between client behavior before entering the program and after leaving the program, etc. Of course, certain steps must be taken to assure comparability of the data (See Chapter II). A special case of the research approach is the Goal Attainment Model. In this case, in order to determine program effectiveness, the broad objectives of the program are translated into measurable criteria and a comparison is made between these criteria and actual program performance, i.e., has the program attained its goal? Standard Setting or Comparison of Planned vs. Actual Performance are special cases of the Goal Attainment Model in which the goals are set by some formula or by professional judgement and 6 are not necessarily one-to-one reflections of total program objectives. (See Chapter III.) Systems Approaches-seek to evaluate programs by describing the organizational structure of a program, the stated functions of the program, and the process by which that organizational structure with those func- tions accomplishes the program's objective. That is, the analytic process tries to build a model of the program, uses the model to stimulate the program, and then offers more efficient and/or effective alternatives of action (optimization). An enormous amount of information is needed for this type of evaluation, together with a great deal of data processing effort. (See Chapter II.) A SUGGESTED MODEL "The purpose of the evaluation process is to provide policymakers with the basic data necessary for them to make decisions wisely. Impact evaluations of programs should provide five essential sets of information. First, they should provide all of the data necessary to determine if a particular program should be continued. Second, they should determine which of the alternative programs achieve greatest gains for a given cost. Third, evaluations should present informa- tion on the components of each program and the mixes of components which are most effective for a given expenditure SO that maximum operating efficiency can be achieved. Fourth, evaluation should provide the first three types of information for persons with different characteristics SO that a decision-maker may determine which individuals are best served by each program. Finally, in the course of evaluating existing programs, data should be gathered which will suggest new methods to attack problems. To date, no evaluation of programs has provided all this information." (11) In short, the primary goal of program evaluation is to ask "Which programs are worthwhile?" or "Which programs work best?" and then state "Why they are worthwhile" or "Why they work best." These findings may produce information for program decision-making that may result in the implementation of a change. Suggested below is a model that approaches some of the issues for the systematic evaluation of programs. Classes of Measures Basically, there are five classes of measures or criteria that are used to estimate various aspects of program worthiness. The first class of measures enumerates the resources used by the program (effort). The second class ex- 7 amines the results of program service in terms of broad social change (impact). The third class attempts to estimate specific client outcomes or changes (effectiveness). The purpose of the fourth class is to determine the economics of program operation in comparison with program accomplishment (efficiency). The final class of measures, Quality, has come to mean different things to different people. One accepted global definition is that quality is a statement of program "goodness." A second definition used equates quality with effective- ness. That is, it uses process measures (see appendix I) to represent program effectiveness. The perspective taken in this handbook is that Quality estimates the conformance of program activity and professional activity to accepted standards of the day. It is one reading of ongoing program process. These five measures are derived from the administrator's system breakdown of input, process, and output, where effort represents input, quality is equivalent to process, impact and effectiveness represents output, and efficiency is an input- output ratio. More specifically: Effort is a statement of fact about the "amount and kinds of program activities used to reach program objectives." It refers to staff time, activity and commitment, and to the allocation and use of material resources (funds, space, equipment, etc.). In addition, it takes into account ancillary resources used (outside consultation, media, public relations, etc.). It answers the question "What did I do?" (94,97) Impact is a statement based on how the program was able to affect a broad social change together with the utility of this program technique. Usually, ost/benefit or cost/effectiveness is used for this type of evaluation. (11) Effectiveness is a statement of fact about how well the objectives of the program were met, or a statement of fact about how well the program components functioned to meet the program objectives or the program components' objectives. Effectiveness measures answer the question "Has Vocational Rehabilitation program accomplished its objective?" or "Have clients, who are rehabilitated by Vocational Rehabilitation, gotten jobs?" In other words, "How has the program changed clients behavior or performance?" (27,41,94,97) *Impact is the same as effectiveness, but on a much broader scale; that is, it can be thought of as a long-range socially pervasive change, whereas, effectiveness is usually thought of as simply attaining goals or a circumscribed change. In many cases, impact criteria are very difficult to establish and data very costly to collect. Therefore, this monograph only considers effectiveness. 8 Efficiency is a rate that is produced when effort (inputs, staff time, activities, salaries, physical facilities and purchased service) expended by a program is compared to the program's effectiveness (output). It answers the question "Can the same program results (output) be achieved by either reducing the effort expended or by choosing other less costly alternatives of action?" (26,49,97) Quality is a judgemental statement made about professional competence, acceptability of services to the client and the setting of minimum perform- ance standards by legislative and regulatory bodies. It answers the question "How well did you do what you did?" or "How well did you use the knowledge you had?" (2,29,97) Program Level Having established and defined proxy measures for the system elements (input, process, outcome) the "responsibility model" (program, program management and service delivery), was selected as a suitable outline for analysis. The goals of each organizational element are defined below. Program-goals are to meet society's expectation of a Vocational Rehabili- tation organization. Program Management-goals are to secure and organize inputs for the system (funds, staff, facilities, referrals) and then assure appropriate out- puts (closures). Service Delivery-(technical component)-goals are to perform the process of rehabilitating the disabled. Figure II visually represents the conceptual outline for a program evaluation model that will include input analysis, process analysis, and outcome analysis (goal attainment) or measures of effort, effectiveness, efficiency and quality broken down by program level. 9 FIGURE II PROPOSED MODEL FOR PROGRAM EVALUATION IN VOCATIONAL REHABILITATION MEASURE