Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Canadian scientists : their research department structure and research output in four types of organizations Pelton, Terrance Ronald 1970

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
831-UBC_1970_A2 P44.pdf [ 9.13MB ]
Metadata
JSON: 831-1.0102274.json
JSON-LD: 831-1.0102274-ld.json
RDF/XML (Pretty): 831-1.0102274-rdf.xml
RDF/JSON: 831-1.0102274-rdf.json
Turtle: 831-1.0102274-turtle.txt
N-Triples: 831-1.0102274-rdf-ntriples.txt
Original Record: 831-1.0102274-source.json
Full Text
831-1.0102274-fulltext.txt
Citation
831-1.0102274.ris

Full Text

CANADIAN SCIENTISTS: THEIR RESEARCH DEPARTMENT STRUCTURE AND RESEARCH OUTPUT IN FOUR TYPES OF ORGANIZATIONS by Terrance Ronald Pelton M.P.E., University of B r i t i s h Columbia, 1964 A Dissertation submitted i n P a r t i a l Fulfilment of The Requirements for the Degree of Doctor of Education i n the Center for the Study of Administration i n Education Faculty of Education We accept this d i s s e r t a t i o n as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July, 1970 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an advanced degree a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I a g r e e t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r ag ree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d by t h e Head o f my Depar tment o r by h i s r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Depar tment o f EDUCATION ADMINISTRATION The U n i v e r s i t y o f B r i t i s h Co lumbia Vancouver 8 , Canada D a t e J u ] , y j n , 1970 ABSTRACT Previous research has revealed a relationship between research department structure and s c i e n t i s t research output. Investigators have drawn on the findings of thi s research to make recommendations to research directors and administrators regarding the type of structure necessary to maintain high levels of s c i e n t i s t research output. Since the recommendations were made to research directors and administrators i n general, the implication i s that one type of research department struc-ture should be u t i l i z e d i n a l l types of organizations. This, i n turn, implies that the relat i o n s h i p between structure and output i s constant across organization types. Consideration of the goals and operating conditions i n di f f e r e n t types of organizations suggests that some organiza-tions would tend to place greater s t r u c t u r a l constraints on s c i e n t i s t s than others. In other words, differences i n goals and conditions of operation make i t almost impossible for research departments i n d i f f e r e n t types of organizations to be structured the same way. If t h i s i s the case, the hypo-thesis that follows inevitably i s that some types of organi-zations cannot have the one best structure and must suffer losses i n research output. An examination of some o r i g i n a l research, however, reveals that the responding s c i e n t i s t s were employed i n a i i i v a r i e t y of o r g a n i z a t i o n types. Moreover, the i n v e s t i g a t o r s made no attempt to examine the r e l a t i o n s h i p s between s t r u c t u r e and research output on an o r g a n i z a t i o n type-by-type b a s i s . There remains then, an e q u a l l y p l a u s i b l e hypothesis, t h a t s c i e n t i s t s i n d i f f e r e n t types of o r g a n i z a t i o n s accept the e x i s t i n g s t r u c t u r e and t h a t no b a s i c i n c o m p a t i b i l i t y e x i s t s . This i m p l i e s t h a t the r e l a t i o n s h i p between s t r u c t u r e and r e -search output i s not constant across o r g a n i z a t i o n types, but v a r i e s from type to type. The problem of t h i s study, t h e r e f o r e , was t o determine whether or not the r e l a t i o n s h i p s between research department s t r u c t u r e and research output was constant across o r g a n i z a t i o n types. The main hypothesis t e s t e d was: "The r e l a t i o n s h i p between research department s t r u c t u r e and research output v a r i e s across o r g a n i z a t i o n types." I m p l i c i t i n t h i s hypothesis were two p r i o r hypotheses. 1. There i s a r e l a t i o n s h i p between research department s t r u c t u r e and research output. 2. Research departments i n d i f f e r e n t types of parent o r g a n i z a t i o n s are s t r u c t u r e d d i f -f e r e n t l y . A l s o i m p l i c i t i n the main hypothesis was a type of summary hypothesis, which p r o p e r l y followed the main hypothesis. 3. R e l a t i o n s h i p s between research output and s t r u c t u r e found i n combined o r g a n i z a t i o n samples are d i f f e r e n t than r e l a t i o n s h i p s found i n separate o r g a n i z a t i o n samples. i v These general hypotheses were tested by examining i n -formation obtained from testing related s p e c i f i c hypotheses. The data necessary for the testing of the s p e c i f i c hypotheses was obtained from questionnaire responses provided by scien-t i s t s from four types of organizations—business, government, s o c i a l development, and university—who were mailed questionn-aires i n order to obtain measures of reported research output, and perceptions of research department structure. 523 scien-t i s t s or 45% of the sample, returned a completed questionnaire. Another 15% of the sample provided reasons for not responding. Examination of the data related to Hypothesis Number One indicated that i n a combined organization sample, levels of reported research output were: 1. p o s i t i v e l y associated with levels of perceived (a) influence to decide own work goals and objectives, (b) decentralized control of research a c t i v i t i e s , and 2. negatively associated with levels of perceived (a) supervisor influence to decide s c i e n t i s t work goals and objectives, and (b) centralized control of research a c t i v i t i e s . On the basis of thi s information, Hypothesis Number One was accepted. Examination of the data related to Hypothesis Number Two indicated that s c i e n t i s t s i n d i f f e r e n t types of organi-zations perceived d i f f e r e n t levels of: V 1. emphasis to be placed on p a r t i c u l a r c r i t e r i a used i n the sele c t i o n of research projects; 2. time expenditures i n basic and applied research; 3 . time pressure on t h e i r work; 4 . influence to decide work goals and objectives; 5. supervisor or department head influence i n deciding t h e i r work goals and objectives; 6. centralized and decentralized control of research a c t i v i t i e s ; 7. coordination of e f f o r t s for common objectives. On the basis of th i s information Hypothesis Number Two was accepted. Examination of the data related to Hypothesis Number Three indicated that high research output s c i e n t i s t s , i n d i f -ferent types of organizations perceived d i f f e r e n t levels of: 1. influence i n deciding t h e i r work goals and objectives (university s c i e n t i s t s only); 2. immediate supervisor or department head influence i n deciding t h e i r work goals and objectives; 3 . centralized and decentralized control of research a c t i -v i t i e s ; and 4 . coordination of e f f o r t s for common objectives. On the basis of th i s information, Hypothesis Number Three was accepted. Hypothesis Number Four was also accepted because examination of the data indicated that responses from high v i research output s c i e n t i s t s i n combined organization samples— as compared to responses from separate organization samples— d i f f e r e d i n the same ways as those l i s t e d above. In summary, this'study found relationships between research output and research department structure. Research departments i n d i f f e r e n t types of parent organizations appeared to be structured d i f f e r e n t l y . F i n a l l y , relationships between structure and research output varied across organization types, as well as between separate and combined organization samples. In conclusion, the present study indicated that there i s no 'best' type of research department structure for a l l organizations. TABLE OF CONTENTS Page Abstract i i Table of Contents v i i L i s t of Tables x L i s t of Appendices x i i Acknowledgements xiv CHAPTER I. INTRODUCTION TO THE PROBLEM 1 Questions 11 Statement of the Problem 11 Study Plan 13 Review of Literature ; 13 Research Methods and Procedures 13 Analysis of Data 14 I I . ORGANIZATION TYPES AND RESEARCH DEPARTMENT STRUCTURAL VARIABLES 16 Four Types of Organizations 16 The Business Organization 18 Government Organizations 22 Soci a l Development Organizations 29 The University. 33 Research Department Structural Variables. . . 39 Sp e c i f i c Hypotheses Related to Hypothesis Number One 39 Sp e c i f i c Hypotheses Related to Hypothesis Number Two 40 Sp e c i f i c Hypotheses Related to Hypothesis Number Three. 44 Sp e c i f i c Hypotheses Related to Hypothesis Number Four 45 v i i i CHAPTER Page I I I . RESEARCH METHODS AND PROCEDURES 47 I d e n t i f i c a t i o n of the Sample 47 The Questionnaire 52 Preparation of Data for Analysis 56 Analysis of Data: Testing for Significance 58 IV. RESULTS AND CONCLUSIONS 61 Section Number One: Relationships between Research Department Structure and Research Output 62 Sp e c i f i c Hypothesis Number 1A . . . . . . . 63 Sp e c i f i c Hypothesis Number IB 65 Sp e c i f i c Hypothesis Number IC . . . . . . . . 68 Sp e c i f i c Hypothesis Number ID 71 Hypothesis Number One: The Test 74 Section Number Two: Differences i n Structure Across Organization Types 75 Sp e c i f i c Hypothesis Number 2A 76 Sp e c i f i c Hypothesis Number 2B 86 Sp e c i f i c Hypothesis Number 2C 91 Sp e c i f i c Hypothesis Number 2D 93 Sp e c i f i c Hypothesis Number 2E 95 Sp e c i f i c Hypothesis Number 2F 97 Sp e c i f i c Hypothesis Number 2G 100 Hypothesis Number Two: The Test 10 2 Section Number Three: Relationships Between Research Output and Structure Across Organization Types 103 Sp e c i f i c Hypothesis Number 3A 10 4 S p e c i f i c Hypothesis Number 3B 107 Sp e c i f i c Hypothesis 3C 109 i x CHAPTER Page IV. (Continued) S p e c i f i c Hypothesis 3D 114 Hypothesis Number Three: The Test 116 Section Number Four: Differences i n Relationships between Combined and Separate Organization Samples 118 Spe c i f i c Hypothesis Number.4A. . 119 Spe c i f i c Hypothesis Number 4B 121 Spe c i f i c Hypothesis Number 4C. . 123 Spe c i f i c Hypothesis Number 4D. . . . . . . . 125 Hypothesis Number Four: The Test. 127 Summary 128 Limitations of the Study 133 Summary of the Results 137 Conclusions. . . . . ... . . . . . . . . . . 142 Implications of the Results and Conclusions for Research Directors . 143 Implications for Further Research 146 BIBLIOGRAPHY 14 8 APPENDICES . . . . . . . . . 158 LIST OF TABLES TABLE Page I. Levels of Research Output #1 versus Levels of S c i e n t i s t Influence i n Deciding Work Goals and Objectives 64 I I . Levels of Research Output #1 versus Levels of Supervisor Influence i n Deciding S c i e n t i s t Work Goals and Objectives 67 II I . Levels of Research Output #1 versus Levels of Centralization/Decentralization 70 IV. Levels of Research Output #2 versus Levels of Coordination 73 V. Organization Type versus Importance of Selec-ti n g Projects that Contribute to the Organization's F i n a n c i a l Weil-Being. 77 VI. Organization Type versus Importance of Selec-ti n g Projects that are Delegated by Some Higher Authority 79 VII. Organization Type versus Importance of Selec-t i n g Projects that Contribute to the Well-Being of Some So c i a l Group . 81 VIII. Organization Type versus Importance of Selec-t i n g Projects that are Interesting i n Themselves 83 IX. Organization Type versus Importance of Selec-t i n g Projects that Seek New Knowledge 85 X. Organization Type versus Levels of Time Expenditure on Applied Research 88 XI. Organization Type versus Levels of Time Expenditure on Basic Research. . 90 XII. Organization Type versus Levels of Time Pressure Worked Under . 92 x i TABLE Page XIII. Organization Type versus Levels of S c i e n t i s t Influence i n Deciding Work Goals and Objectives 9 4 XIV. Organization Type versus Levels of Supervisor Influence i n Deciding Work Goals and Objectives . . . 96 XV. Organization Type versus Levels of Centralized and Decentralized Control of Research A c t i v i t i e s . 98 XVI. Organization Type versus Levels of Emphasis on Coordination of E f f o r t 101 XVII. High Levels of Research Output versus Levels of S c i e n t i s t s Decision Making Influence 105 XVIII. High Levels of Research Output versus Levels of Supervisor Decision Making Influence 108 XIX. High Levels of Research Output versus Levels of Centralized and Decentralized Control,. . . . I l l XX. High Levels of Research Output versus Levels of Coordination. 115 XXI. High Levels of Research Output versus Levels of Decision Making Influence 120 XXII. High Levels of Research Output versus Levels of Supervisor Decision Making Influence 122 XXIII. High Levels of Research Output versus Levels of Centralized and Decentralized Control . . . .124 XXIV. High Levels of Research Output versus Levels of Coordination 126 x i i LIST OF APPENDICES APPENDIX Page (!) Random Sampling Procedure for Selection of University Sample 146 ( i i ) Four Types of Organizations: Their Sample Size and Questionnaire Return Data 14 8 ( i i i ) Reasons Given for Not Responding to the Questionnaire 149 (iv) The Questionnaire 150 (v) Questionnaire V a l i d i t y and R e l i a b i l i t y . . . . 155 (vi) Questionnaire Cover Letter 157 (vii) Questionnaire Follow-Up Letter Number One. . . 158 ( v i i i ) Questionnaire Follow-Up Letter Number Two. . .159 (ix) S c i e n t i s t Output: Rationale for Its Use. . . . 160 (x) Rationale for Using the Chi-Square S t a t i s t i c 161 (xi) The Contingency C o e f f i c i e n t : C 164 (xii) D e finitions of Research A c t i v i t i e s 166 ( x i i i ) Levels of Research Output #2 versus Levels of S c i e n t i s t Influence i n Deciding Work Goals and Objectives 16 8 (xiv) Levels of Research Output #2 versus Levels of Supervisor Influence i n Deciding S c i e n t i s t Work Goals and Objectives 169 (xv) Levels of Research Output #2 versus Levels of Centralization/Decentralization 170 (xvi) Levels of Research Output #1 versus Levels of Coordination 171 (xvii) Research Output versus Levels of S c i e n t i s t Decision-Making Influence 172 x i i i APPENDIX Page ( x v i i i ) Research Output versus Levels of Supervisor Decision Making influence 173 (xix) Research Output versus Levels of Centralized and Decentralized Control 174 (xx) Research Output #2 versus Levels of Coordination 175 x i v ACKNOWLEDGEMENTS To Dr. R. Jean H i l l s , , chairman of the doctoral committee, the writer wishes to express his sincere appreciation and gratitude for acting as chairman of the committee, and also for his c r i t i c a l evaluation as the study progressed. These served to sharpen the writer's thoughts and his writin g . The writer wishes to thank the other members of his committee, Dr. Walter Hartrick, Dr. Harold Dyck, Dr. Herman A. Wallin and Dr. G. N e i l Perry, for t h e i r valuable suggestions and for giving t h e i r time i n reading the study. Thanks are also due to Dr. Douglas McKie, for his assistance with the s t a t i s t i c a l procedures and suggestions for presenting the data i n thi s study. CHAPTER I INTRODUCTION TO THE PROBLEM Impli c i t i n the research work and writings of some investigators i s the assumption that there i s a r e l a t i o n s h i p between research department decision-making structure and s c i e n t i s t research output. These investigators indicate that s c i e n t i s t s require such things as autonomy, an absence of coordination, decentralized decision-making, p a r t i c i p a t i v e leadership, and so on, i n order to achieve high levels of research output.^ That i s , not only does there seem to be a rel a t i o n s h i p between structure and research output, but a p a r t i c u l a r kind of r e l a t i o n s h i p . In other words, previous research indicates that there i s a p a r t i c u l a r kind of re-search department structure related to high research output. Amitai E t z i o n i , Modern Organizations (Englewood C l i f f s , New Jersey: Prentice-Hall Inc., 1964), p. 76. ...Only i f immune from ordinary s o c i a l pressures and free to innovate, to experiment, to take r i s k s without the usual s o c i a l repercussions of f a i l u r e , can a professional carry out his work e f f e c t i v e l y . I t i s t h i s highly i n d i v i d u a l i z e d p r i n -c i p l e which i s diametrically opposed to the very essence of the organizational p r i n c i p a l of control and coordination by s u p e r i o r s — i . e . , the p r i n c i p l e of administrative authority. Also, P.R. Lawrence, Organizing for Innovation, an address before the Second Harvard Business School, Pittsburgh Regional Conference, A p r i l 4, 1964, pp. 3-4. There i s now beginning to be some research evidence to support what many i n t u i t i v e managers have known for a long time. . . There i s f a i r l y clear evidence that research and development units, i f they are to supply the r i g h t kind of working environment for innovators, need to follow t h e i r own type of organization. Research organizations seem to provide a most stimulating 2 The i d e n t i f i c a t i o n and study of the s t r u c t u r a l v a r i -ables associated with s c i e n t i s t research output has occupied Peiz and Andrews for many years. Their dedication to the search for these variables i s indicated i n the introduction to t h e i r book, S c i e n t i s t s i n Organizations. What constitutes a stimulating atmosphere for research and development? That was the guiding question for the six-year exploration described i n t h i s book. Each of the following chapters gives r e s u l t s on s p e c i f i c aspects of that question.^ This book i s based on the premise that R & D organizations provide more than f a c i l i t i e s for t h e i r members. They also provide an "en-vironment" which may either stimulate or i n h i b i t the s c i e n t i s t s ' performance.3 environment when they use a f a i r l y f l a t type of structure with f a i r l y wide spans of c o n t r o l , f a i r l y open channels of communi-cation. . . . The leadership s t y l e that seems to work best i s more p a r t i c i p a t i v e and e q u a l i t a r i a n . I t seems to work better when goals and targets are p r i m a r i l y set by the innovators, themselves, and when the goals are f a i r l y general and long range. Control i n such an organization i s carried out more by informal ground rules than by formal rules and prescribed procedures. Also, William H. Whyte, J r . , The Organization Man (Garden C i t y , New York: Doubleday and Co.Inc., 1957), p. 264. ... What we must concern ourselves with are the conditions  that are common to a l l s c i e n t i s t s , for the kind of environ-ment which stimulates the creative side of the average scien-t i s t i s the same environment i n which genius f l o u r i s h e s . 2 D.C. Pelz and F.M. Andrews, S c i e n t i s t s i n Organizations, Productive Climates for Research and Development (New York: John Wiley and Sons Inc.~^ 1966) , p. 1. 3 I b i d . 3 Some of the resu l t s and conclusions of the Pelz and Andrews study appeared to support the writings and research of other investigators. That i s , they found relationships between research department s t r u c t u r a l variables and scien-4 t i s t research output. Pelz and Andrews discussed t h e i r results and con-clusions at the end of each chapter i n t h e i r book, and made p r a c t i c a l suggestions to research supervisors and adminis-trators . These results are i n t r i g u i n g as s c i e n t i f i c data about research administration and i n addition, are also u s e f u l . They describe c h a r a c t e r i s t i c s of organizations, and organi-zations can be "changed" toward more favorable forms. Individual s c i e n t i s t s and engineers, for example, can a l t e r t h e i r working r e l a -tionships with colleagues; often they can steer themselves into (or out of) c e r t a i n types of commitments. Through such changes the i n -d i v i d u a l researcher can a l t e r his own environ-ment so as to gain more stimulation from i t . The d i r e c t o r or supervisor of R & D, using s i m i l a r techniques and his substantial power as an administrator, can produce changes i n his laboratory which increase the l i k e l i h o o d of achievement. Each chapter concludes with a  section on implications, usually i n the form of a dialogue, which attempts to translate the findings into p r a c t i c a l steps for the R & D manager. 5. Pelz and Andrews, op. c i t . , pp. 17-32. For example, Pelz and Andrews formed a p o s i t i v e association between levels pf reported research output and perceived lev e l s of influence to decide work goals and objectives. They also found that s c i e n t i s t s who reported high levels of research output, also perceived p a r t i c u l a r levels of research coordination ("moder-ate") , along with p a r t i c u l a r levels of influence to decide work goals and objectives ("high"). ~*Pelz and Andrews, op. c i t . , p. 2. 4 The samples of s c i e n t i s t s given questionnaires by Pelz and Andrews, for t h e i r study, however, came from several d i f f e r e n t types of organizations.^ In t h e i r book, they did not consider organization type as one of t h e i r units of analysis i n studying the relationships between leve l s of s c i e n t i s t research output and lev e l s of emphasis on various 7 s t r u c t u r a l variables. They stated that "We did not design g the research to test a systematic organization theory." Ibid. The data on which the study i s based come from 1311 s c i e n t i s t s and engineers located i n eleven d i f f e r e n t laboratories. Included were 641 professionals i n f i v e indus-t r i a l laboratories s p e c i a l i z i n g i n pharmaceuticals, glass and ceramics, e l e c t r o n i c s , and e l e c t r i c a l equipment. Also i n -cluded were 144 professors from seven departments of a mid-western unive r s i t y ( b i o l o g i c a l , physical, and s o c i a l sciences) and 526 s c i e n t i s t s and engineers from f i v e government labora-to r i e s (weapons guidance, animal diseases, commercial uses of a g r i c u l t u r a l products, and basic research i n several physical sciences). 7 Pelz and Andrews, op. c i t . , p. 4. From these three factors, f i v e "primary analysis groups" were defined as follows. These w i l l be used throughout the book. A. Ph.D.'s i n Development-Oriented Laboratories. Half of these were located i n industry, half i n government. B. Ph.D.'s i n Research-Oriented Laboratories. Two-thirds were i n the unive r s i t y ( a l l of bur academic s c i e n t i s t s were i n t h i s category) and one-third i n government. C. Non-Ph.D.'s i n Development-Oriented Labs not Dominated by  Ph.D.'s. About three-quarters (primarily engineers) were i n i n d u s t r i a l locations, one-quarter i n government. D. Non-Ph.D.'s i n Ph.D.-Dominated Laboratories (either re-search-or development-oriented). Half(of these "assistant" and "subordinate" s c i e n t i s t s ) were i n government, half i n industry. E. Nondoctoral S c i e n t i s t s i n Research-Oriented Labs not  Dominated by Ph.D.'s. A l l of these were i n government settings. 8 Pelz and Andrews, op. c i t . , p. 32. 5 Since p r a c t i c a l suggestions aand recommendations were made by Pelz and Andrews to research directors and adminis-trators i n general, the implication i s that one type of research department structure should be u t i l i z e d i n a l l types 9 of organizations. This assumption also suggests that scien-t i s t s are a unique group of people and require s p e c i a l or-ganizational treatment i f they are to be made productive. The assumption implied i n the work of Pelz and Andrews, and the writings of other investigators raises an i n t e r e s t i n g p r a c t i c a l question. If one were tryi n g to maximize research output i n a research department, would one tend to do the same things i n d i f f e r e n t types of organizations? Put another way, would the research supervisor i n the business organization tend to place the same levels of emphasis as the department head i n a un i v e r s i t y , on p a r t i c u l a r research department s t r u c t u r a l variables? Pelz and Andrews, op. c i t . , p. 2. This set of labora-to r i e s was de l i b e r a t e l y heterogeneous, but not a representative sample of research organizations. We went where we could gain entree. Some types of labs were missing—the basic research lab i n industry, for example, or the independent n o t - f o r - p r o f i t i n s t i t u t e . Although i t i s possible that findings would have been markedly d i f f e r e n t i n these types of labs, from what was found i n those that were studied, t h i s seems u n l i k e l y . Repor-ted here are conditions which enhanced performance for a wide variety of research personnel. Thus the idiosyncracies of p a r t i c u l a r locations became less important.* [*A11 data were co l l e c t e d i n American laboratories and i t i s not clear how applicable the findings would be i n other coun-t r i e s . Although i t i s possible that the motivational' and environmental conditions which enhance i n t e l l e c t u a l function-ing (including research and development) are si m i l a r every-where, the study which demonstrates t h i s has yet to be conducted.] 6 One would expect s c i e n t i s t s i n university research depart-ments, and i n business organization research departments, to be "organized" i n a d i f f e r e n t way. The business organization, with i t s " p r o f i t " goal, i s l i k e l y to be more " e f f i c i e n c y " conscious, and exercise more control over decisions about what the s c i e n t i s t does, than the u n i v e r s i t y . 1 ^ Any research carried out i n a business organization, that did not make some kind of contribu-tion to the organization's profit-making goal, would l i k e l y be considered " i n e f f i c i e n t . " To ensure that " i n e f f i c i e n c y " i s kept to a minimum, research supervisors would l i k e l y exert a high l e v e l of influence over s c i e n t i s t s ' decisions about work goals and objectives. The u n i v e r s i t i e s ' goal of "discovering new knowledge—for i t s own sake," however, would appear to require a lower l e v e l of research supervisor or department head influence i n deciding s c i e n t i s t work goals and objectives. 1"'" No matter what work goals or objectives s c i e n t i s t s selected for research, they would theo-r e t i c a l l y make some contribution to "new knowledge." Hence, because they would be making a contribution to the u n i v e r s i t i e s ' goal, there would be no need for department heads to exert more than a low l e v e l of influence over s c i e n t i s t s ' decisions about work goals and objectives. "^Peter M. Blau and R.W. Scott, Formal Organizations (San Francisco: Chandler Publishing Co., 1962), p. 49. ...Business concerns . . . p r i v a t e l y owned and operated for a p r o f i t . The dominant problem of the business concern i s operating e f f i c i e n c y — the achievement of maximum gain at minimum cost i n order to fur-ther s u r v i v a l and growth i n competition with other organizations. "'""''University of Waterloo, B r i e f to the Special Senate Committee on Science Policy, No. 47, May 28, 1969, p. 5972. 7 Consideration of the goals and operating conditions i n d i f f e r e n t types of organizations suggests that some organi-zations would have to place greater s t r u c t u r a l constraints on s c i e n t i s t s than others. In other words, differences i n goals and conditions of operation make i t almost impossible for re-search departments i n d i f f e r e n t types of organizations to be structured the same way. If t h i s i s the case, the hypothesis that follows inevitably i s that some types of organizations can-not have "the one best structure," and must suffer losses i n research output. A re-examination of Pelz and Andrews research, however, reveals that the responding s c i e n t i s t s were employed i n a variety of organization types. Moreover, the investigators made no attempt i n t h e i r text to examine the relationships between structure and 12 research output on an organization type-by-type basis. . . . much of the university research i s indeed of long-term i n -terest and must not be s a c r i f i c e d to immediate or short-term needs of society. This long term, often disinterested research or "quest for knowledge for i t s own sake". . . 12 F.M. Andrews and D.C. Pelz, Analysis Memos From the Study of S c i e n t i f i c Personnel, #7 and #8, Survey Research Center,' Ins t i t u t e for Social Research, The University of Michigan, Ann Arbor, Michigan, March 1961, p. 83. E a r l i e r , i n one of t h e i r analysis memos, however, Pelz and Andrews appeared to use organi-zation type as a unit of analysis. They stated that: In subse-quent analyses, we would not be j u s t i f i e d i n combining u n i v e r s i t y departments with those i n other locations, the differences i n departmental atmosphere are too great. . . Government s c i e n t i s t s are more d i v e r s i f i e d i n t h e i r departmental atmospheres than we f i n d among i n d u s t r i a l s c i e n t i s t s . . . .As one might expect, university s c i e n t i s t s perceive the greatest autonomy i n depart-mental operations; i n d u s t r i a l s c i e n t i s t s report the le a s t ; and government s c i e n t i s t s stand at an intermediate l e v e l . 8 There remains then an equally plausible hypothesis, that s c i e n t i s t s i n d i f f e r e n t types of organizations accept the e x i s t i n g structure and that no basic incompatibility • 4 . "13 e x i s t s . The s c i e n t i s t i n the business organization, for example, may perceive a lower l e v e l of influence to decide his work goals and objectives than the s c i e n t i s t i n the u n i v e r s i t y . The business organization s c i e n t i s t , however, may accept t h i s constraint as one of the norms of the business organization. He may even perceive t h i s constraint as necessary for maxi-mizing the research department's contribution to the organi-14 zation's goals and objectives. He may also have s i m i l a r Robert H. Knapp, Personality Committee Report i n 1956 Utah Conference, pp. 229-241. We know, from the analysis of s i t u a t i o n a l factors discussed here two or three times, the effects of d i f f e r e n t s o c i a l climates on c r e a t i v i t y and upon the type of man who w i l l be creative i n d i f f e r e n t climates. One man w i l l do well i n the i n d u s t r i a l s i t u a t i o n , another w i l l require to be i n an academic c a l l i n g . Also, B.G. Glaser, Organizational S c i e n t i s t s : Their Professional  Careers (Indianapolis; The Bobbs-Merrill Co. Inc., 1964), p.3. Glaser indicates that s c i e n t i s t s t y p i c a l l y work i n many d i f f e r -ent organizations. Perhaps t h i s i s a search for a compatible research department structure. ...Scientists standard career pattern i s one of moving about the u n i v e r s i t y departments, a f f i l i a t e d research groups, sometimes i n government, non-p r o f i t or i n d u s t r i a l research organizations, always seeking  better research conditions, and more i n t e r e s t i n g , more prestigious, and perhaps more p r o f i t a b l e p o s i t i o n s . 14 . . . . . . P.R. Lawrence and J.W. Lorsch, Organization and Environment (Boston: Harvard University Press, 1967), Abstract, (i.) What organizational c h a r a c t e r i s t i c s are required to deal e f f e c t i v e l y with d i f f e r e n t external market and technological conditions? Such a question i s q u i t e . d i f f e r e n t from the 9 personal interests and goals to those of the research depart-ment and be quite w i l l i n g to accept any "reasonable" con-15 s t r a m t s i n order to work on his research. To maximize the contribution of s c i e n t i s t s to organi-zational goals and objectives, therefore, research supervisors i n d i f f e r e n t types of organizations may exert d i f f e r e n t levels of emphasis on some research department s t r u c t u r a l v a r i a b l e s . These d i f f e r e n t levels of emphasis may be accepted by scien-t i s t s i n the d i f f e r e n t organizations as normal and have no adverse e f f e c t on s c i e n t i s t research output. This implies, therefore, that the r e l a t i o n s h i p between structure and re-search output i s not constant across organization types, but varies from type to type. central theme of most e a r l i e r organizational studies, which have tended to focus on the question of what i s the one best way to organize, i r r e s p e c t i v e of the external environmental conditions facing the business. ( i i ) . . . the authors have developed a contingency theory of organization which, rather than propounding one best way to organize under a l l conditions, focuses on the organizational c h a r a c t e r i s t i c s which lead to e f f e c t i v e performance given the s p e c i f i c demands of an organization's environment. 15 G. Steiner,"Concluding Remarks," The Creative  Organization (University of Chicago Press, 1965), p. 260. ~ '. '. I think you see that the c r u c i a l variable i s the degree to which the organization's own goals coincide with the goals and i n t r i n s i c i nterests of those members within i t who are supposed to create. . . . Now, at some points i n industry, you get a happy marriage between i n d i v i d u a l i n t e r e s t s and or-ganizational objectives. . . . The goals of the corporation are i d e n t i c a l with the interests of the people who w i l l provide the c r e a t i v i t y for i t . The advertising agency thrives to the extent that i t produces creative, outstanding advertising. Each i n d i v i d u a l i n that firm personally thrives to the extent that he creates outstanding advertising. There i s no clash; there i s no problem. The reason you get away with regimenta-t i o n i n t h i s case i s that people are already preregimented i n the same i n t e r e s t that you have. 10 Now i f one were to combine research departments from business organizations and u n i v e r s i t i e s into one sample, and examine the relat i o n s h i p between l e v e l of research output and another variable such as l e v e l of s c i e n t i s t influence to decide work goals and objectives, the relationships found i n the combined sample may d i f f e r from the relationships found separately i n each type of organization. That i s , although l e v e l of s c i e n t i s t influence to decide work goals and objectives may be p o s i t i v e l y associated with l e v e l of research output, the p a r t i c u l a r l e v e l of influence may vary from one type of organization to another, and be d i f f e r e n t again when the or-16 ganization types are combined. If thi s i s the case, the findings from combined organization samples would be d i f f i -c u l t f or research directors to u t i l i z e , because relationships Morris I. Stein, A Transactional Approach to C r e a t i v i t y , 1956 Utah Conference. If a company i s one i n which a high value i s placed on "basic" or "pure" research, and a man i s expected to f u l f i l l scien-t i f i c - p r o f e s s i o n a l r o l e s , then the c o n s t e l l a t i o n of factors that would predict c r e a t i v i t y i n t h i s environment would have greater weights placed on man's t h e o r e t i c a l values, capacity to deal i n abstractions and independence than would be the case i n a company where there i s greater emphasis on pro-f e s s i o n a l - s o c i a l r o l e s . Also, by considering both companies as a single unit, s i g n i f i c a n t differences "wash out", and we don't learn the true story. 11 s p e c i f i c to any one organization type would not be a v a i l a b l e . That i s , i f relationships between research department struc-t u r a l variables and research output vary across organization types, the relationships i n a combined organization sample would tend to be "washed out" or d i f f e r e n t than the r e l a t i o n -ships found i n separate organization samples. Questions The main question for consideration, therefore, i s whether or not the relationships between research department structure and research output vary across organization types. Related to t h i s question, are two p r i o r questions that also need to be answered. 1. Is there a r e l a t i o n s h i p between research depart-ment structure and research output? 2. Are research departments i n d i f f e r e n t types of parent organizations structured d i f f e r e n t l y ? Also suggested i n the main question i s a type of summary question, which properly follows the main question: 4. Are relationships between research output and structure found i n combined organization samples, d i f f e r e n t than relationships found i n separate organization samples? Statement of the Problem The problem of t h i s study i s to determine whether or not the r e l a t i o n s h i p between research department structure and research output are constant across organization types. 1 2 In order to study t h i s problem, the following hypothesis i s put forward for t e s t i n g . Main Hypothesis Relationships between research department structure and research output vary across organization types. Impl i c i t i n t h i s hypothesis are two p r i o r hypotheses. Pr i o r Hypothesis (Number One) There i s a relat i o n s h i p between research structure and research output. Pr i o r Hypothesis (Number Two) Research departments i n d i f f e r e n t types of parent or-ganizations are structured d i f f e r e n t l y . Also i m p l i c i t i n the main hypothesis i s a type of "summary hypothesis" which was alluded to i n the discussion. Since t h i s hypothesis properly follows the main hypothesis, i t i s c a l l e d Hypothesis Number Four. Summary Hypothesis (Number Four) Relationships between research output and structure found i n combined organization samples are d i f f e r e n t than relationships found i n separate organization samples. 13 STUDY PLAN To study the problem, the following plan w i l l be u t i l i z e d . Review of Literature The review of l i t e r a t u r e w i l l be concerned, f i r s t of a l l , with the i d e n t i f i c a t i o n or c l a s s i f i c a t i o n of several types of organizations. Literature relevant to each of these d i f f e r e n t types of organizations w i l l be studied to determine the levels of emphasis p a r t i c u l a r s t r u c t u r a l variables receive, i n order to maximize s c i e n t i s t research output. S p e c i f i c hypotheses relevant to the three general hypotheses w i l l be constructed from the information obtained. I t i s hoped that by testing each of the s p e c i f i c hypotheses enough information w i l l be obtained to test each of the general hypotheses. Research Methods and Procedures Questionnaire responses from s c i e n t i s t s i n research departments i n several types of organizations w i l l provide the data for the study. Research output w i l l be measured by the number of project completion reports; a r t i c l e s published i n journals; speeches; books and monographs each s c i e n t i s t s re-17 ports he has completed. Levels of emphasis on research department s t r u c t u r a l variables w i l l be measured by the 14 frequency of s c i e n t i s t s questionnaire responses regarding the l e v e l of emphasis they perceive p a r t i c u l a r s t r u c t u r a l v a r i -ables to receive i n t h e i r research departments. The question-naire respondents w i l l be selected from s o c i a l science depart-18 ments i n several types of organizations i n Canada. Analysis of Data Four separate analyses of data relevant to each of the four general hypotheses, w i l l be carried out. The f i r s t analysis w i l l be concerned about the r e l a t i o n -ships between levels of research output and leve l s of emphasis on research department s t r u c t u r a l v a r i a b l e s . Frequencies of s c i e n t i s t responses regarding levels of emphasis and level s of output perceived w i l l be examined i n order to determine whether or not relationships e x i s t . The second analysis w i l l examine the frequency of scien-t i s t responses regarding the l e v e l of emphasis they perceive to be placed on p a r t i c u l a r research department varia b l e s . Most studies that include measurements of research output u t i l i z e these items as measures of research output. For a more complete analysis of the v a l i d i t y and r e l i a b i l i t y of these items, see Appendix ix/. 18 Most studies of t h i s kind have been car r i e d out i n the United States, and involved s c i e n t i s t s from the physical, b i o l o g i c a l , chemical and medical sciences, p r i m a r i l y . The investigator was more interested i n the S o c i a l Sciences, and i n carrying out the study i n Canada. 15 The frequency of s c i e n t i s t responses at each l e v e l of emphasis w i l l be compared across organization types. The purpose of t h i s analysis i s to determine whether or not s c i e n t i s t s i n d i f f e r e n t types of organizations perceive d i f f e r e n t l e v e l s of emphasis to be placed on p a r t i c u l a r s t r u c t u r a l v a r i a b l e s . The t h i r d analysis w i l l examine the frequency with which s c i e n t i s t s who have high levels of research output, perceive p a r t i c u l a r levels of emphasis on s t r u c t u r a l v a r i a b l e s . The purpose of t h i s i n v e s t i g a t i o n s i s to determine whether or not high output s c i e n t i s t s i n d i f f e r e n t types of organizations perceive d i f f e r e n t levels of emphasis to be placed on struc-t u r a l v a r i a b l e s . The fourth analysis w i l l compare combined organization samples with separate organization samples, to determine whether or not relationships between high output and struc-ture vary from separate to combined samples. Data from the tables used i n the t h i r d analysis, w i l l be used for the fourth and f i n a l analysis. CHAPTER II ORGANIZATION TYPES AND RESEARCH DEPARTMENT STRUCTURAL VARIABLES Four Types of Organizations H i l l s , i n his text Toward a Science of Organization, i d e n t i f i e d four d i f f e r e n t types of organizations. He made d i s -t i n c t i o n s between types according to the s o c i e t a l function or goal around which they were organized. A section from his text provided some of the reasons why one might select (a) business organizations, (b) government organizations, (c) s o c i a l develop-ment organizations, and (d) u n i v e r s i t i e s , to represent four d i f f e r e n t types of organizations i n our society. Basis of C l a s s i f i c a t i o n The attainment of a goal by an organization i s i n the integrated case, the performance of a func-t i o n on behalf of the society of which i t i s a part. Hence, the f i r s t d i s t i n c t i o n that can be made among organizations i s i n terms of the type of s o c i a l function or goal, around which they are organized. Accordingly, we may d i s t i n g u i s h organizations with adaptive, goal-attainment, integrative, and pattern-maintenance goals, depending on the function performed for the society as a system. From t h i s point of view, the p r i n c i p a l types of organizations are: (1) organizations oriented to economic production. The business firm i s the most obvious example of organizations with economic primacy. (2) Organizations oriented to the attainment of c o l -l e c t i v e goals. Governmental organizations are the most prominent examples of t h i s type. (3) Organizations oriented to integration. These are organizations which on the s o c i e t a l l e v e l con-tribu t e to the adjustment of c o n f l i c t s and the d i r e c t i o n of motivation to the f u l f i l l m e n t of 17 i n s t i t u t i o n a l i z e d expectations. The courts, le g a l firms, p o l i t i c a l parties which mobilize support for government, i n t e r e s t groups, and hospitals are included here. (4) Organizations oriented to the expression and maintenance of c u l t u r a l patterns and the maintenance of the pattern of the units of the system. Churches and schools are the most clear cut examples.19 Br i e f s to the recent Special Senate Committee on Science Policy supported t h i s c l a s s i f i c a t i o n , i n that they also i d e n t i -f i e d (a) u n i v e r s i t i e s , (b) business and i n d u s t r i a l firms, and (c) government organizations as being the organization types 20 carrying out the bulk of research xn Canada. Although " s o c i a l development" agencies were not mentioned as frequently, they presented more b r i e f s and were referred to more often than any other organization type, and appeared to meet the c r i t e r i a i n H i l l s ' c l a s s i f i c a t i o n system for the "integrative" organization. Social development agencies, therefore, were the fourth type of organization to be considered i n t h i s review of l i t e r a t u r e . The goals of each of these organizations, t h e i r organiza-t i o n a l c h a r a c t e r i s t i c s , and the implications of these goals and organizational c h a r a c t e r i s t i c s for t h e i r research department organization and s c i e n t i s t research output, w i l l be discussed R. Jean H i l l s , Toward a Science of Organization (Eugene University of Oregon Press, 1968), p. 65. 20 Proceedings of the Special Senate Commxttee on Scxence Policy (Ottawa: Queen's Pr i n t e r , B r i e f No. 40) , p. 4943, B r i e f No. 50, p. 6282; Br i e f No. 48, pp. 6098, 6148, 6150; B r i e f No. 47, p. 5985, 6041 and 5947-48/ Br i e f No. 61, p. 7372, Br i e f No. 44, p. 5733. These b r i e f s contain examples of the references made about d i f f e r e n t organization types carrying out research. The l i s t i s by no means complete, but i s representative of what i s contained i n terms of references about research i n other b r i e 18 i n succeeding sections of t h i s chapter. The Business Organization The goal of the business organization, according to H i l l s , i s economic production—making a p r o f i t from the pro-21 duction of goods or services. Blau and Scott also stated that the goal of the business firm i s to make a p r o f i t and to do t h i s , management must be concerned about operating e f f i -ciency . Business concerns . . . p r i v a t e l y owned and operated for a p r o f i t . The dominant problem of the business concern i s operating e f f i c i e n c y — the achievement of maximum gain at minimum cost i n order to further s u r v i v a l and growth i n com-p e t i t i o n with other organizations.22 According to thi s statement, a l l a c t i v i t i e s c a r r i e d out within a business organization should make some contribution to the organization's profit-making goal. Any a c t i v i t i e s that do not make such a contribution are l i k e l y considered "unpro-f i t a b l e " and an " i n e f f i c i e n t " expenditure of the organization's resources. In a business organization, therefore, one would ex-pect a s c i e n t i s t ' s research a c t i v i t i e s to make a contribution 23 to the organization's profit-making goal. That i s , i n t h i s 21 H i l l s , l oc. c i t . 22 Peter M. Blau and Richard W. Scott,. Formal Organiza-tions (San Francisco: Chandler Pub. Co., 1962), p. 49. 23 See Appendix (xii) for d e f i n i t i o n s of research a c t i v i t i e s . 19 type of organization, the s c i e n t i s t would tend to s e l e c t pro-jects only i f they appeared to make a contribution to the organization's profit-making goal. Stein appeared to support t h i s thinking i n the following statement about s c i e n t i s t commitments: By accepting a pos i t i o n with a company, a researcher both i m p l i c i t l y and e x p l i c i t l y accepts the tasks of working on problems re- 24 lated to the products that the company produces. Quinn and Mueller gave further support to the idea that i n business firms, research a c t i v i t i e s should be directed to-ward the organization's profit-making goal. The second step i n tr a n s f e r r i n g technology e f f i c i e n t l y from research into operations i s to make sure that the research and development program i s s p e c i f i c a l l y designed to support the company's goals and f u l f i l l i t s technological needs: This "targeting" of research and develop-ment involves two e s s e n t i a l a c t i v i t i e s : (1) developing a company-wide long range plan into which R and D are properly integrated and (2) providing adequate commercial information to rank, and balance R and D programs to meet com-pany goals.25 Morris I. Stein, "Several Findings of a Transactional Approach to C r e a t i v i t y , " i n Gary A. Steiner, ed., The Creative  Organization (The University of Chicago Press, 1965), p. 157. James Brian Quinn and James A. Mueller, "Transferring Results to Operations," Harvard Business Review, January-February, 1963, p. 53. 20 The research s c i e n t i s t i n the business firm, therefore, would appear to u t i l i z e as his main c r i t e r i a i n the sel e c t i o n of his research projects, the c r i t e r i a that his research pro-jects make a contribution to the organization's profit-making goal. To make such a contribution, the s c i e n t i s t would l i k e l y r e s t r i c t his research a c t i v i t i e s to p a r t i c u l a r problems of an applied nature, that have commercial value. He would tend to be concerned about the organization's problem of "maximizing marketable outputs at minimum cost," and avoid " i n e f f i c i e n t " expenditures of his research e f f o r t s on research projects that did not make a contribution to the organization's f i n a n c i a l well-being. He would also tend to be concerned about completing his research work as fa s t as p o s s i b l e — s i n c e e f f i c i e n c y would require speed, as well as careful s e l e c t i o n of projects. To ensure that s c i e n t i s t s make a contribution to the organization's profit-making goal, research supervisors i n business organizations may exert considerable influence on 2 6 decisions about s c i e n t i s t work goals and objectives. These Carl E. Barnes, I n d u s t r i a l Research: Is i t Outmoded? i n Business Horizons (vol. 7, no. 1; Bloomington, Indiana: Graduate School of Business, Indiana University, Spring, 1964), p. 90. "What kind of new products should management be looking for? . . . A complete understanding of corporate objectives by the research s t a f f i s perhaps the most e s s e n t i a l factor i n improving research productivity. No research man wants to waste time on a project he knows w i l l not be accepted by man-agement... With t h i s understanding, the research department can i n s t i t u t e sound plans, r e c r u i t the r i g h t kind of personnel set up the proper screening t e s t s , and select appropriate ideas for further study. 21 research supervisors would also tend to be involved i n "co-ordinating" the research e f f o r t s of the s c i e n t i s t s , toward a 27 common o b j e c t i v e — a product, or service of commercial value. To be able to p a r t i c i p a t e e f f e c t i v e l y i n influencing s c i e n t i s t s ' decisions about work goals and objectives, and i n the coordination of s c i e n t i s t e f f o r t s , the research supervisor would l i k e l y have some delegated authority to make decisions about s c i e n t i s t work goals and objectives. There would l i k e l y be a "centralized" control of s c i e n t i s t research a c t i v i t i e s , therefore, i n research departments, i n t h i s type of organi-zation. Consequently, the s c i e n t i s t would not l i k e l y perceive himself to have high levels of influence to decide his own 2 8 work goals and objectives. Quinn, op. c i t . , p. 124. In s e l e c t i n g i n d i v i d u a l pro-jects and experiments, there i s no question that companies must ultimately r e l y on the judgement and imagination of i n d i v i d u a l researchers. But top-level s c i e n t i f i c and general managers must plan the framework within which the researcher makes his choice. They must: 1. e s t a b l i s h the broad objectives and strategies which the program should be designed to support, 2. see that researchers and research managers are selected on the basis of both t h e i r s c i e n t i f i c competence and how well t h e i r backgrounds and interests f i t the defined scope and interests of the company. 2 8 James C. March and Herbert A. Simon, 0rganizations (New York: Wiley, 1958), pp. 140-141. Once the objectives of the organization are formally established, the h i e r a r c h i c a l organization of r e s p o n s i b i l i t i e s serves as a framework for means-ends c h a i n s — s p e c i f y i n g for each o f f i c i a l the ends of his tasks and thus confining his duties to the s e l e c t i o n of the best means for achieving these ends. . .; also, Amitai E t z i o n i , Complex Organizations (New York: The Free Press of Glencoe Inc., 1961), pp. 80, 141. "There are several reasons why organizations that have economic goals function more 22 Government Organizations Government organizations, according to H i l l s ' c l a s s i -f i c a t i o n c r i t e r i a , are "oriented to the attainment of c o l l e c -29 t i v e goals." The attainment of c o l l e c t i v e goals would appear to involve the provision of various services, consid-ered necessary by the public-at-large, and too d i f f i c u l t to provide on an i n d i v i d u a l basis. A government organization would tend to be the type that i s concerned about providing many d i f f e r e n t services to the public. In Canada, each of the three levels of government (municipal, p r o v i n c i a l , and federal) have "departments" created to provide these d i f f e r e n t services for the public i t serves. For example, at the p r o v i n c i a l l e v e l we have a department of education whose function i s to see that educa-t i o n a l services are provided to the people of the province. How these services are provided, the nature of the services, who should receive them and so on, i s decided primarily by . e f f e c t i v e l y when they employ remuneration than when they em-ploy coercion or normative power as t h e i r predominant means of control. Production i s a r a t i o n a l a c t i v i t y , which requires systematic d i v i s i o n of labour, power, and communication, as well as a high l e v e l of coordination. I t therefore requires also a highly systematic and precise control of performance. U t i l i t a r i a n organizations emphasize v e r t i c a l instrumental communication as a condition of e f f e c t i v e production. Since th i s i s the most r a t i o n a l type of the three, and since coor-dination, planning and centralized decision making are emphasized here more than i n the other two types (normative and coercive). . ." H i l l s , l o c . c i t . 23 the p o l i t i c a l party i n power, through i t s representative i n the department concerned. That i s , the Minister of Education i s the representative of the p o l i t i c a l party " i n power" who supervises the services provided by the Department of Education. The public served by each l e v e l of government usually knows, to a ce r t a i n extent anyway, about the kinds of services that w i l l be provided by a p o l i t i c a l party, i f i t i s elected and "put i n power." Most p o l i t i c a l parties that "run" for ele c t i o n "advertise," p r i o r to e l e c t i o n , the goals and objec-tives of t h e i r party, and the services they intend to provide i f elected. By receiving votes and being elected to govern, the p o l i t i c a l party i s given authority, by the public concerned, to pursue the goals and provide the services that were adver-t i s e d . Through the legitimate use of i t s authority which defines the rights and obligations of the party i n power to do as i t has been authorized by the public that elected i t , the party makes decisions about the kinds of a c t i v i t i e s i t w i l l engage i n , i n order to implement i t s goals and provide the services i t advertised at e l e c t i o n time. I f "good d e c i -sions" are made, that i s , i f the party i s able to implement the goals and provide the services advertised, the party i s usually re-elected for another"term. I f "good decisions" are not made, the public may withdraw i t s authority to make de-ci s i o n s , by voting i n another party. In order to continue i n power, therefore, the p o l i t i c a l party w i l l usually extend 24 some e f f o r t toward implementing i t s stated goals and providing the services i t advertised. Hence, each department, supervised by a representative of the p o l i t i c a l party, w i l l be expected to implement a p a r t i c u l a r set of goals and provide some ser-vices advertised by the p o l i t i c a l party. In supervising the a c t i v i t i e s of a department, the party representative (or Minister, at the p r o v i n c i a l and fed-e r a l level) must make decisions about how best to u t i l i z e the resources at his disposal, i n order to maximize progress toward the goals, and provision of services, to which his department i s committed. In order to make good decisions, he may require information that he does not have on hand. Some of the information may be provided by c i v i l service o f f i -c i a l s from the various sub-departments within his department. These sub-department o f f i c i a l s may have the information on hand, or they may be required to d i r e c t members of t h e i r sub-department to carry out research, i n order to provide the information required by the supervising party representative. In any event, the supervising party representative of a government department may delegate, to those with less author-i t y , the r e s p o n s i b i l i t y of providing him with research i n f o r -mation that w i l l enable him and the government, to make good decisions. The following statements by government o f f i c i a l s appear to support t h i s kind of thinking. 25 The basic purpose of the Research and Development Program, which consists of f i v e branches with the major research a c t i v i t i e s concentrated i n two of them, it. to improve the qu a l i t y of public and private decision-making on p o l i c y formulation and program administration i n the labour f i e l d through the provision of optimum research and i n t e l l i g e n c e functions.30 . . .[Research] programs are developed to meet the p o l i c y needs of the department and of the Federal Government, and are designed to increase understanding of i n d u s t r i a l r e l a t i o n s problems and issues, and to provide a basis for the i n -t e l l i g e n t formulation of policy.31 . . . The goal of D.B.S., as outlined i n the S t a t i s t i c s Act 50 years ago i s "generally to organize a scheme of coordinated s o c i a l and economic s t a t i s t i c s pertaining to the whole of Canada and to each of the provinces thereof . . . . The Bureau i s a major element of the information system upon which public and p r i -vate individuals and i n s t i t u t i o n s draw i n studying s o c i a l and economic conditions and problems and i n making l o g i c a l decisions.32 The system of placing decision-making authority, regar-ding the goals and objectives of an organization, i n the hands of one person at the top of a hierarchy, and having subordin-ates from d i f f e r e n t levels and departments below him provide 30 Department of Labour, B r i e f to the Special Senate Committee on Science Policy (Ottawa: Queens Pr i n t e r , No. 27, February 6, 19 6 9) , p. 3 817. 31 Canada Dept. of Labour, B r i e f to the Special Senate Committee on Science Policy (Ottawa: Queen's P r i n t e r , No. 27, February 6 , .1969) , p. 3849.. 32 Dominion Bureau of S t a t i s t i c s , B r i e f to the Special Senate Committee on Science Policy (Ottawa: Queen's Printer, No. 24, February 5, 1969), pp. 3481-3482, 3498. 26 him with decision-making information, i s t y p i c a l of the organization with a centralized control of i t s a c t i v i t i e s . Each person within the organization i s usually responsible for making some decisions that decide the a c t i v i t i e s of people below him i n the hierarchy. Each person, i n turn, i s usually responsible to someone above him for decisions that decide his a c t i v i t i e s . S c i e n t i s t s , i n t h e i r p a r t i c u l a r research departments, therefore, would l i k e l y have research projects delegated to them from someone i n a higher authority p o s i t i o n . That i s , they would l i k e l y be t o l d by someone of higher authority, within t h e i r research department, the research projects from 33 which to make a s e l e c t i o n . The c r i t e r i a used by the government research s c i e n t i s t , i n the s e l e c t i o n of research projects, therefore, would l i k e l y be whether or not the re-search project was "delegated by some higher authority." Harold Guetzkow, The Creative Person i n Organizations i n Gary A. Steiner, ed., The Creative Organization (The Uni-v e r s i t y of Chicago Press, 1965), pp. 37-38. The C e n t r a l i z a t i o n and Decentralization of Authority The authority system-provides premises for decision within the hierarchy of the organization. The commands of a superior are taken as "givens," thus decidedly l i m i t i n g the freedom for decision of the subordinates. When there i s c e n t r a l i z a t i o n within an organization, many decisions are taken at the top of the authority structure, so that d e c i s i o n a l l a t i t u d e i s narrower for those i n intermediate and lower positions, thereby r e s t r i c t i n g opportunities for non-conforming, creative de-c i s i o n s . 27 Close coordination of research a c t i v i t i e s would l i k e l y take place i n government research departments. Each scien-t i s t would l i k e l y be responsible to the research supervisor for providing part of the research information that was required for a research project or projects delegated to the research supervisor from someone of higher authority. In order to maximize the contribution from each s c i e n t i s t , the research supervisor would tend to avoid duplication or overlapping of s c i e n t i s t work goals and objectives. This would require the research supervisor to coordinate the s c i e n t i s t s ' e f f o r t s toward the common research o b j e c t i v e — providing information relevant to the delegated research . . 34 project. In such a department, therefore, the s c i e n t i s t would tend to perceive the research supervisor as having consider-able influence i n deciding his work goals and objectives. He would l i k e l y perceive himself, therefore, as having a low l e v e l of influence to decide his work goals and objec-tives . One would also expect s c i e n t i s t s i n government re-search departments to spend a considerable amount of time Edgar N. Schein, Organizational Psychology (Englewood C l i f f s , New Jersey: Prentice-Hall Inc., 1965), pp.7-8. Tied up with the concept of coordination and the r a t i o n a l achievement of mutually agreed upon goals i s the idea that such goals can best be achieved i f d i f f e r e n t 28 on research projects that provide s p e c i f i c information about p a r t i c u l a r problems, of an applied nature, of i n t e r e s t to the decision-makers. Research time spent on other projects would tend to be viewed as "wasteful" of the resources a v a i l -able to the department. The following statements tend to support t h i s view. Presumably most of the governmental i n t r a -mural research i s r e l a t i v e l y c l o s e l y mission-oriented, mainly applied and includes a small developmental component.3 5 Research projects for government departments and even Royal Commissions though limited i n scope and quantity are directed to immediate situations and are almost e n t i r e l y "mission-oriented. "36 As already indicated e a r l i e r i n thi s Brief,, our Department i s mission-oriented and the bulk of our research a c t i v i t i e s i s of an applied nature. As the need arises between 8% and 10% of the time of our s c i e n t i s t s may be devoted to fundamental research i n order to obtain the fundamental information required for the solu-t i o n of p r a c t i c a l problems.37 things i n a coordinated fashion. . . . the very idea of co-ordination implies that each unit submits to some kind of authority for the sake of achieving some common goal. . . . An organization i s the r a t i o n a l coordination of the a c t i v i t i e s of a number of people for the achievement of some common ex-p l i c i t purpose or goal, through d i v i s i o n of labour and func-t i o n , and through hierarchy of authority and r e s p o n s i b i l i t y . 35 . University of Guelph, Research Advisory Board, B r i e f submitted to the Special Senate Committee on Science Pol i c y (Ottawa: Queen's Pr i n t e r , No. 47, May 1969), p. 6041. 36 Soc i a l Science Research Council of Canada, B r i e f to the Special Senate Committee on Science Policy (Ottawa: Queen' Pri n t e r , No. 57, June 10, 1969), p. 7018. 37 Department of National Health and Welfare, Brief to the Special Senate Committee on Science Policy (Ottawa: Queen' P r i n t e r , No. 13, November 27-28, 1968), p. 13636. 29 Social Development Organizations Some of the c r i t e r i a used by H i l l s to describe the goals of his "integrative" organizations appeared to apply to s o c i a l development organizations. The " f u l f i l l m e n t of i n s t i -t u t i o n a l i z e d expectations" c r i t e r i a , i n p a r t i c u l a r , seemed to correspond to the " s o c i a l development" c r i t e r i a of the s o c i a l development organization. That i s , "developing" or improving the s o c i e t a l l i v i n g conditions of a p a r t i c u l a r s o c i a l group, appeared to be the same as " f u l f i l l i n g the i n s t i t u t i o n a l i z e d 3 8 expectations" of people i n some groups. Being "disadvan-taged" i n some way, and not capable of achieving on t h e i r own the standard of l i v i n g expected by the "average Canadian," these people are assisted by s o c i a l development organizations to move toward a " f u l l development of t h e i r p o t e n t i a l s . " In-cluded i n groups assisted by s o c i a l development organizations are people from s o c i a l l y disadvantaged areas such as the "poverty pockets" i n Canada; people from s p e c i a l i n s t i t u t i o n s for the various physical, mental and emotional disorders; people with inadequate education and vocational t r a i n i n g ; and 39 so on. H i l l s , l o c . c i t . Much of thi s discussion material was obtained from discussions with o f f i c i a l s from the Alberta Human Resources Research Council, Alberta Newstart Corp., and Nova Scotia Newstart Corp. as well as from correspondence with s o c i a l development research department d i r e c t o r s . 30 These s o c i a l development organizations which are pa r t l y financed by voluntary contributions from other members of society, p a r t l y by government and par t l y by the people served; provide services to a s s i s t the people concerned to cope with t h e i r problems, and raise t h e i r standard of l i v i n g . The services vary from organization to organization, but the i r goal of " s o c i a l development" appears to be the same. The c r i t e r i a that the research s c i e n t i s t i n a s o c i a l development organization would tend to use i n the sel e c t i o n of his research projects, therefore, would be whether or not the project would make a contribution to the well-being of some s o c i a l group. That i s , a p a r t i c u l a r s o c i a l group his organization was concerned about a s s i s t i n g . To make t h i s contribution, the research project would have to provide information of p r a c t i c a l value to the personnel providing the d i r e c t assistance to the s o c i a l l y disadvantaged group. People providing these services would tend to require s p e c i f i c information for the solution of p a r t i c u l a r problems of an applied nature. The s c i e n t i s t , therefore, would l i k e l y spend much of his time on problems of t h i s kind. An example of some s o c i a l development organizations are the Canada Newstart Corporations, which d i r e c t t h e i r e f f o r t s to solving the s o c i a l development problems of people i n p a r t i c u l a r l y disadvantaged areas, or "poverty-pockets," i n Canada. The goal of these newstart organizations i s to 31 "develop, through action research, methods and programs which w i l l prove t h e i r effectiveness i n helping to motivate and 40 prepare unemployed and under-employed adults for work." By "action research," Newstart o f f i c i a l s were r e f e r r i n g to research a c t i v i t i e s directed toward the i d e n t i f i c a t i o n and solution of p r a c t i c a l problems the s o c i a l group was confronted with. In order to make a contribution to the organization's s o c i a l development goal, some coordination of s c i e n t i s t re-search e f f o r t s would l i k e l y be required. That i s , given a common goal and objective, the s c i e n t i s t s ' e f f o r t s within a s o c i a l development research department would l i k e l y be co-ordinated, so that a maximum contribution to the groups' s o c i a l well-being was made. To e f f e c t i v e l y coordinate s c i e n t i s t research a c t i v i t i e s one would expect the research supervisor, therefore, to have some delegated authority, and exert considerable influence i n deciding s c i e n t i s t work goals and objectives. The s c i e n t i s t , i n turn, may perceive himself to have a low l e v e l of influence to decide his work goals and obje c t i v e s — a n d may even perceive his research department as having a "centralized" control of research a c t i v i t i e s . Department of Regional and Economic Expansion, Br i e f to the Special Senate Committee on Science Policy (Ottawa: Queen's Pr i n t e r , No. 4, A p r i l , 1969), p. 4943. In order to i d e n t i f y and solve the d i f f e r e n t s o c i a l development problems, however, new knowledge about the dynamics of various s o c i a l groups, t h e i r culture, values and possible reactions to s o c i a l development programs would have to be obtained. This new knowledge—probably obtained from basic research programs--would not be immediately applicable to p a r t i c u l a r problems, without further research of an applied nature. That i s , s o c i a l development research departments would probably require some basic research, to support t h e i r applied research program.^"'" Since basic research i s , by d e f i n i t i o n , "discovery of new knowledge, not immediately applicable to p a r t i c u l a r problems," i t could not be coordinated or controlled too much, without changing i t s nature. There would tend to be, therefore, somewhat less coordination and d i r e c t i o n of some s o c i a l development s c i e n t i s t s ' research a c t i v i t i e s . Some of the s c i e n t i s t s engaged i n some basic research would tend to perceive themselves to have considerably high levels of influence i n deciding t h e i r work goals and ob j e c t i v e s — a n d likewise would tend to perceive t h e i r research supervisors as not having high lev e l s of influence on decisions about t h e i r A Newstart research dir e c t o r t o l d the writer than some basic research was required to support t h e i r "action research" programs. 33 work goals and objectives. Feeling less supervisor control over t h e i r research a c t i v i t i e s , some s o c i a l development s c i e n t i s t s may perceive the research department to have a 42 "decentralized" control of s c i e n t i s t a c t i v i t i e s . There would tend to be, therefore, considerable v a r i -ation i n the organizational c h a r a c t e r i s t i c s and research a c t i v i t i e s of s c i e n t i s t s , i n s o c i a l development research departments. The University H i l l s i d e n t i f i e d schools and churches as "oriented 43 to the expression and maintenance of c u l t u r a l patterns." Presumably he was r e f e r r i n g to the goals these organizations have of both educating and transmitting the culture of the society to those i n attendance. The following statement by McDonald appears to support the idea that u n i v e r s i t i e s , as " r e l a t i v e s " of schools, also belong to organizations of t h i s type. . A decentralized decisional structure provides scope for innovative behavior through i t s emphasis upon the develop-ment of solutions appropriate to the d i f f e r e n t environments encountered i n the various extensions of the organization. When authority i s delegated, f u l l advantage can be taken of broad la t i t u d e by those able to develop creative responses to an ever-changing environment i n which t h e i r organization functions. . . Thus one speculates: a dispersed d i s t r i b u t i o n of authority within a firm provides more occasions for inno-vation, creative decision-making. . ., Guetzkow, lo c . c i t . H i l l s , l o c . c i t . 34 We see the u n i v e r s i t i e s having an important role to conduct basic research where there i s no application i n sight, research which has c u l t u r a l value and educational value to our society and without which our society would be destitute from a c u l t u r a l and educational standpoint.44 The u n i v e r s i t y , according to MacDonald and other authoritative sources, d i f f e r s from the other three types of organizations i n that the research a c t i v i t i e s c a r r i e d out by the s c i e n t i s t s are not of the "means-to-an-end" type. That i s , the univ e r s i t y appears to have no other goal (such as making a p r o f i t , a s s i s t i n g a s o c i a l group, or providing i n -formation f o r decision-makers) than teaching and research. Research i n such an organization would tend to be an end i n i t s e l f , because i n most cases i t would not be u t i l i z e d except for educational purposes. The following statements seem to give further support to these ideas. Ins t i t u t i o n s of higher education have two primary functions: teaching and research. These functions are concerned with "knowledge for i t s own sake," but they are also concerned with the s c i e n t i f i c , technological, economic, Dr. J.B. MacDonald, Executive Vice President, Committee of Presidents of U n i v e r s i t i e s of Ontario, B r i e f to the Special Senate Committee on Science Po l i c y (Ottawa: Queen's P r i n t e r , No. 43, May 21, 1969), p. 5670. 35 s o c i a l and c u l t u r a l development of a l l the communities which they serve, whether l o c a l , p r o v i n c i a l , regional, national or i n t e r -national. 45 . . . much of the univer s i t y research i s indeed of long-term i n t e r e s t and must not be s a c r i f i c e d to immediate or short-term needs of society. This long term, often d i s i n t e r -ested research or "quest for knowledge for i t s own sake". . .4° The main points i n the research role of the university are: a) much of the research should be fundamental, aimed at producing new knowledge on a broad front; b) considerable portion of the fundamental re-search can be concentrated i n selected areas, i . e . , broadly mission-oriented; c) applied research of a mission-oriented nature should be conducted i n the univer s i t y to the mutual advantage of the university and the nation; d) consulting a c t i v i t i e s on the part of the faculty should be provided. . . e) development research a c t i v i t i e s should not be car r i e d out i n any volume by the univer-s i t y . 47 Because his projects would be selected with l i t t l e condern for immediate app l i c a t i o n , one would expect the univers i t y research s c i e n t i s t to engage i n more "basic" Association of Un i v e r s i t i e s and Colleges of Canada (AUCC), Br i e f to the Special Senate Committee on Science Policy (Ottawa: Queen's P r i n t e r , No. 44, May 27, 1969),p.5733. 46 . University of Waterloo, B r i e f to the Special Senate Committee (Ottawa: Queen's Pr i n t e r , No. 47, May 28, 1969) , -p. 5972. University of Guelph Research Advisory Board, Brief to the Special Senate Committee on Science Po l i c y (Ottawa: Queen's P r i n t e r , No. 47, May 1969), pp. 6041-6042. 36 research a c t i v i t i e s than s c i e n t i s t s i n the other organization types. S c i e n t i s t s i n business firms, government organizations, and s o c i a l development agencies, because of the "means-to-an-end" a p p l i c a b i l i t y of t h e i r research, would tend to engage i n more applied research a c t i v i t i e s . ^ In a u n i v e r s i t y , a s c i e n t i s t would tend to have high levels of influence to decide his work goals and objectives. I t wouldn't r e a l l y matter what his research goals and objectives were, as long as he selected some research projects and engaged i n some research a c t i v i t i e s , he would, t h e o r e t i c a l l y be making a contribution to the university's goal of "discovering new knowledge." This freedom of the university s c i e n t i s t to decide his own "work goals and objectives, was considered by Steacie, to be very important. The chief reason why the unive r s i t y i s the "ideal place for. s c i e n t i f i c work (the s o c i a l sciences i n -cluded) i s that the work i s uncommitted. The univer s i t y man i s free to proceed i n any d i r e c t i o n which he sees f i t , and should not i n any way be influenced by p r a c t i c a l considerations. The uni-v e r s i t y i s , i n f a c t , v i r t u a l l y the only place where science can be pursued for i t s own sake.49 Amitai E t z i o n i , Modern Organizations- (Englewood C l i f f s , New Jersey: Prentice-Hall,Inc., 1964), p. 81. In organizations whose goal, i s no n p r o f e s s i o n a l (e.g. profit-, making), i t i s considered desirable for administrators to have the major (line) authority because they d i r e c t the major goal a c t i v i t y . Professionals deal only with means, with secondary a c t i v i t i e s . E.W. Steacie, A Speech at St. Georges School, Montreal, June 12, 1959, c i t e d i n J.B. Babbitt, ed., Science i n Canada (University of Toronto Press, 1965), p. 37. Berelson acknowledged that university research scien-t i s t s have considerable decision-making freedom regarding research. . . . the administrator i n an academic i n s t i t u t i o n has no r e a l authority and must defer the important decisions, about whom to h i r e , and what the c u r r i -culum w i l l be, and whom to promote to tenure, to the cumbersome machinery of faculty voting. Every-one knows that the dean cannot f i r e him or t e l l him exactly what to teach or influence his research or his decision to organize a McKinsey Seminar.50 In summary, the university s c i e n t i s t would tend to select projects that were "interesting i n themselves" and make some contribution to "new knowledge"—because he considered "the search for new knowledge to be i n t r i n s i c a l l y good." He would not l i k e l y be as concerned as his colleagues i n other types of organizations about the "discovery of s p e c i f i c knowledge, for the solution to p a r t i c u l a r problems (applied research)." He would tend to be more concerned about the "discovery of new knowledge, not immediately applicable to pa r t i c u l a r problems (basic research).". Free of most work goal constraints and the need for supervision or coordination, the university s c i e n t i s t would tend to perceive high levels of influence to decide his work goals and objectives. The preceding sections described the goals and objec-tives of s c i e n t i s t s i n research departments, i n four d i f f e r e n t Bernard Berelson, " C r e a t i v i t y and the Graduate School," i n Gary Steiner, ed., The Creative Organization, p. 228. 38 types of organizations. Also discussed were the s t r u c t u r a l constraints on s c i e n t i s t s brought about by the need for research supervisors to exert influence on s c i e n t i s t decisions and coordinate s c i e n t i s t e f f o r t s toward some common objectives Research evidence from the Pelz and Andrews study supported the discussion material regarding the d i f f e r e n c e s — i n scien-t i s t freedom to decide work goals and objectives, and coordin-ation of e f f o r t s toward some common objectives—one might expect to f i n d i n research departments i n d i f f e r e n t types of organizations. Their research evidence indicated that university research s c i e n t i s t s perceived more influence to decide work goals and objectives and less coordination of the i r research e f f o r t s than either i n d u s t r i a l or government research s c i e n t i s t s . I n d u s t r i a l research s c i e n t i s t s per-ceived the least amount of influence to decide work goals and objectives, and the most coordination of research a c t i v i t i e s , while government research s c i e n t i s t s stood at an intermediate l e v e l on both counts. Pelz and Andrews, op. c i t . , pp. 216-217. Pelz and Andrews did not use "organization type" as t h e i r unit of analysis. They used possession of Ph.D. etc. as indicated i n footnote number 7. The fac t that some of these units of analysis contained primarily government s c i e n t i s t s , u n i v e r s i t y s c i e n t i s t s , and i n d u s t r i a l s c i e n t i s t s , permitted these deduc-tions. Comparable results were also found i n t h e i r analysis memos (Toc. c i t , 1961). Research Department Structural Variables The preceding discussion and review of l i t e r a t u r e iden-t i f i e d some of the research department s t r u c t u r a l variables that might receive d i f f e r e n t levels of emphasis i n d i f f e r e n t types of organizations, i n order to enhance research output. As a r e s u l t of these d i f f e r e n t levels of emphasis, s c i e n t i s t s i n d i f f e r e n t types of organizations would l i k e l y perceive d i f f e r e n t : 1 . levels of influence to decide work goals and objectives; 2. levels of supervisor influence i n deciding t h e i r work goals and objectives; 3. levels of coordination of e f f o r t s for some common objec-tives ; 4. levels of centralized or decentralized control of research a c t i v i t i e s ; 5 . levels of importance i n u t i l i z i n g p a r t i c u l a r c r i t e r i a i n the selection of research projects; 6. levels of time expenditures on basic and applied research a c t i v i t i e s ; and 7. levels of time pressure under which they work. Spe c i f i c Hypotheses Related to Hypothesis Number One Hypothesis Number One states that there i s a rel a t i o n s h i p between research department structure and research output. The f i r s t four s t r u c t u r a l variables are u t i l i z e d i n the construc-txon of s p e c i f i c hypotheses, which then tested, are designed 40 to provide information with which to test Hypothesis Number One. S p e c i f i c Hypothesis Number 1A In a combined organization sample, levels of reported research output w i l l be p o s i t i v e l y associated with levels of perceived influence to decide work goals and objectives. S p e c i f i c Hypothesis Number IB In a combined organization sample, levels of reported research output w i l l be negatively associated with levels of perceived supervisor or department head influence i n deciding s c i e n t i s t work goals and objectives. S p e c i f i c Hypothesis Number IC In a combined organization sample, levels of reported research output w i l l be p o s i t i v e l y associated with levels of perceived decentralized control, and negatively associated with perceived le v e l s of centralized control of research department a c t i v i t i e s . S p e c i f i c Hypothesis Number ID In a combined organization sample, levels of reported research output w i l l be negatively associated with levels of coordination of s c i e n t i s t e f f o r t s f or common objectives. S p e c i f i c Hypotheses Related to Hypothesis Number Two Hypothesis Number Two states that research departments i n d i f f e r e n t types of parent organizations are structured d i f f e r e n t l y . The seven s t r u c t u r a l variables are u t i l i z e d i n 41 the construction of s p e c i f i c hypotheses, which when tested, are designed to provide information with which to tes t Hypothesis Number Two. Sp e c i f i c Hypothesis Number 2A Sc i e n t i s t s i n d i f f e r e n t types of organizations w i l l tend to perceive some project s e l e c t i o n c r i t e r i a to receive d i f f e r -ent levels of emphasis. S p e c i f i c a l l y : Hypothesis 2A^. More s c i e n t i s t s i n business organiza-tions, than i n the other types of organizations, w i l l respond that i t i s important to select research projects that c o n t r i -bute to the organization's f i n a n c i a l well-being. Hypothesis 2A?. More s c i e n t i s t s i n government organi-zations, than i n the other types of organizations, w i l l respond that i t i s important to sel e c t research projects that are delegated by some higher authority. Hypothesis 2A3. More s c i e n t i s t s i n s o c i a l development organizations, than i n the other types of organizations, w i l l respond that i t i s important to sel e c t research projects that contribute to the well-being of some s o c i a l group. Hypothesis 2 A 4 . More s c i e n t i s t s i n u n i v e r s i t i e s , than i n the other types of organizations, w i l l respond that i t i s important to sel e c t research projects that are in t e r e s t i n g i n themselves. 42 Hypothesis 2Ac;. More s c i e n t i s t s i n u n i v e r s i t i e s , than i n the other types of organizations, w i l l respond that i t i s important to select research projects that seek new knowledge, because the search for new knowledge i s i n t r i n s i c a l l y good. Hypothesis Number 2B Sc i e n t i s t s i n d i f f e r e n t types of organizations w i l l perceive d i f f e r e n t levels of emphasis on time expenditures i n basic and applied research a c t i v i t i e s . S p e c i f i c a l l y : Hypothesis Number 2Bi. More s c i e n t i s t s i n business, government and s o c i a l development organizations, than i n u n i v e r s i t i e s , w i l l perceive themselves to emphasize high ex-penditures of time i n applied research a c t i v i t i e s . Hypothesis Number 2B2« More s c i e n t i s t s i n u n i v e r s i t i e s , than i n the other types of organizations, w i l l perceive them-selves to emphasize high expenditures of time on basic r e- ' search a c t i v i t i e s . Hypothesis Number 2C More s c i e n t i s t s i n business organizations, than i n the other types of organizations, w i l l perceive themselves to work under high levels of time pressure. Hypothesis Number 2D More s c i e n t i s t s i n u n i v e r s i t i e s , than i n the other types of organizations, w i l l perceive high levels of influence to decide t h e i r work goals and objectives. 43 Hypothesis Number 2E More s c i e n t i s t s i n business, government and s o c i a l development organizations, than i n u n i v e r s i t i e s , w i l l perceive immediate supervisors or department heads to have high levels of influence to decide t h e i r work goals and objectives. Hypothesis Number 2F Sc i e n t i s t s i n each type of organization w i l l tend to perceive d i f f e r e n t levels of emphasis on centralized or de-centralized control of research department a c t i v i t i e s . S p e c i f i c a l l y : Hypothesis Number 2F-| . More s c i e n t i s t s i n u n i v e r s i t i e s , than i n the other types of organizations, w i l l perceive high levels of "decentralized" control of research department a c t i v i t i e s . Hypothesis Number 2F2. More s c i e n t i s t s i n business, government and s o c i a l development organizations, than i n uni-v e r s i t i e s , w i l l perceive high levels of "centralized" control of research department a c t i v i t i e s . Hypothesis Number 2G More s c i e n t i s t s i n business, government and s o c i a l development organizations, than i n u n i v e r s i t i e s , w i l l per-ceive high levels of coordination toward common objectives. 44 Sp e c i f i c Hypotheses Related to Hypothesis Number Three Hypothesis Number Three states that the rel a t i o n s h i p between research department structure and research output varies across organization types. The f i r s t four s t r u c t u r a l variables are u t i l i z e d i n the construction of s p e c i f i c hypo-thesis, which then tested, are designed to provide information with which to test Hypothesis Number Three. Hypothesis 3A There w i l l be a greater tendency i n u n i v e r s i t i e s , than i n the other types of organizations, for high levels of repor-ted research output to be associated with s c i e n t i s t perceptions of high levels of influence to decide t h e i r own work goals and objectives. Hypothesis 3B There w i l l be a greater tendency i n business, govern-ment, and s o c i a l development organizations, than i n univer-s i t i e s , for high levels of reported research output to be associated with high levels of perceived' supervisor or department head influence i n deciding s c i e n t i s t work goals and objectives. Hypothesis 3C Relationships between high levels of reported research output and perceived levels of emphasis on centralized and decentralized control of research department a c t i v i t i e s w i l l vary across organization types. 45 S p e c i f i c a l l y : Hypothesis 3C-^ . There w i l l be a greater tendency i n business, government and s o c i a l development organization^ than i n u n i v e r s i t i e s , for high levels of reported research output to be associated with perceptions of high levels of emphasis on centralized control of research department a c t i v i t i e s . Hypothesis 3C2. There w i l l be a greater tendency i n u n i v e r s i t i e s , than i n the other types of organizations, for high levels of reported research output to be associated with perceptions of high levels of emphasis on decentralized con-t r o l of research department a c t i v i t i e s . Hypothesis 3D There w i l l be a greater tendency i n business, govern-ment, and s o c i a l development organizations, than i n u n i v e r s i t i e s , for high lev e l s of reported research output to be associated with perceptions of high lev e l s of coordination of research e f f o r t s f or common objectives. S p e c i f i c Hypotheses Related to Hypothesis Number Four Hypothesis Number Four states that relationships be-tween research output and structure found i n combined organi-zation samples, are d i f f e r e n t than relationships found i n separate organization samples. The f i r s t four s t r u c t u r a l variables are again u t i l i z e d i n the construction of s p e c i f i c hypotheses, which when tested, are designed to provide i n f o r -46 matron with which to test Hypothesis Number Four. S p e c i f i c Hypothesis Number 4A Relationships between levels of reported research out-put and perceived le v e l s of influence to decide work goals and objectives found i n combined organization samples, w i l l d i f f e r s i g n i f i c a n t l y from the relationships found i n separate organization samples. S p e c i f i c Hypothesis Number 4B Relationships between leve l s of reported research out-put and perceived levels of supervisor or department head influence i n deciding work goals and objectives found i n combined organizational samples, w i l l d i f f e r s i g n i f i c a n t l y from the relationships found i n separate organization samples. Sp e c i f i c Hypothesis Number 4C Relationships between leve l s of reported research out-put and perceived levels of centralized and decentralized control of research a c t i v i t i e s found i n combined organization samples, w i l l d i f f e r s i g n i f i c a n t l y from the relationships found i n separate organization samples. S p e c i f i c Hypothesis Number 4 D Relationships between levels of reported research out-put and perceived le v e l s of coordination of research e f f o r t s for common objectives found i n combined organization samples, w i l l d i f f e r s i g n i f i c a n t l y from the relationships found i n separate organization samples. CHAPTER III RESEARCH METHODS AND PROCEDURES In order to study the various research department st r u c t u r a l variables and t h e i r association with s c i e n t i s t research output., as planned, the populations to be sampled had to be i d e n t i f i e d , a questionnaire had to be constructed and administered to the samples, and the data obtained had to be analyzed. The questionnaire was designed to obtain some information about the s c i e n t i s t s ' personal background, t h e i r research outputs, as well as perceptions about t h e i r research department decision-making structure. This ques-tionnaire was mailed to research s c i e n t i s t s i n business organizations, government organizations, s o c i a l development organizations, and u n i v e r s i t i e s . Responses to the ques-tionnaire items were coded and punched on computer cards. These "data" were then manipulated by a computer to provide frequency and percentage frequency tables of responses to the questionnaire items. Details of the methods and procedures u t i l i z e d w i l l be discussed i n succeeding sections of thi s chapter. T d e n t i f i c a t i o n of the Sample I t was decided that s c i e n t i s t s from research depart-ments i n u n i v e r s i t i e s , s o c i a l development organizations, government organizations and business organizations would be 48 included i n the samples to be studied. These s c i e n t i s t s would, according to the review of l i t e r a t u r e , represent s c i e n t i s t s from four d i f f e r e n t types of organizations. After an extensive search of the univer s i t y l i b r a r y and a f t e r enquiries directed to l i b r a r i a n s i n d i f f e r e n t depar-tments i n the l i b r a r y , i t was discovered that l i s t s of or-ganizations with research departments did not e x i s t . The most useful sources of information were the univer s i t y 52 calendars and the McGraw-Hill Directory for Canada. By corresponding with people i n the various technical and s c i e n t i f i c associations l i s t e d i n the d i r e c t o r y — a s well as corresponding with people i n departments i n some of the major u n i v e r s i t i e s across Canada and p r o v i n c i a l and federal government departments—leads were obtained regarding the id e n t i t y of some research departments. These leads provided a tentative l i s t of research departments for the sample. More information about these departments was required, so a short information sheet was sent to the research department directors and department heads of the research departments on the l i s t . The i n f o r -mation obtained from t h i s information sheet—and l a t e r used i n the c l a s s i f i c a t i o n of these research departments—was McGraw-Hill Publishing Co., Directory and Almanac  of Canada, 1969, pp. 338-365. 49 simply whether or not the research departments were part of (a) a u n i v e r s i t y , (b) a government organization, (c) a busi-ness organization, or (d) a s o c i a l development organization. Other information such as (a) date the research department started operations, (b) number of s c i e n t i s t s , (c) the main purpose and functions of the research unit, and (d) source of f i n a n c i a l support, was useful i n making decisions about whether or not a research unit had been i n operation long enough, whether or not the research unit was r e a l l y a depart-ment of a p a r t i c u l a r organization type, and how many question-naires would be needed. The information provided i n response to "Please des-cribe the main purposes and functions of your research unit" helped determine whether or not research departments were concerned, i n some way, with s o c i a l science research. Since i t r e a l l y did not matter whether the s c i e n t i s t s were s o c i a l s c i e n t i s t s or not, except as a way of r e s t r i c t i n g the sample size and providing a more homogeneous sample, no research department was withdrawn from the sample unless i t was 53 c l e a r l y not of the s o c i a l sciences. A response rate of eighty percent to a l l preliminary enquiries was obtained. A response rate of almost one hundred percent was obtained to the information sheet questions. About f i f t y percent of these research departments were v i s i t e d and the directors and department heads i n t e r -viewed. Two large research departments were withdrawn from 50 Directors and department heads of these research departments were promised the publication of a much-needed directory of s o c i a l science research departments, i n exchange for the information. In the correspondence with these directors and department heads, i t was pointed out that t h i s was the f i r s t stage of a study of Canadian research s c i e n t i s t s . Since the university calendars provided l i s t s of faculty members i n each of the s o c i a l science research departments and since the main objective of the correspon-dence was to c l a s s i f y the research departments and obtain research personnel l i s t s , i t was not considered necessary to correspond with university s o c i a l science department heads. The univ e r s i t y s o c i a l science research department l i s t was much larger than the research department l i s t s from the other organization types. A random sample from small, medium and large size departments was taken to reduce t h i s 54 sample sxze. The names of the research s c i e n t i s t s i n each of the other research departments on the sample l i s t s were obtained from the research department d i r e c t o r s . This was accom-the sample when i t became clear that they were i n d u s t r i a l research units, employing only non-social s c i e n t i s t s . 54 The random sample procedure xs described -in Appendix ( i ) . Also see Appendix ( i i ) for the sample break-down, according to organization type and research department s i z e . 51 plished by either v i s i t i n g them or by corresponding with them. These directors were promised a Directory of Canadian s o c i a l science research departments as well as a summary of the results of the study, i f they would provide a l i s t of t h e i r research personnel. They were shown the questionnaire during the interview, or during correspondence. In a l l cases, these research directors cooperated by providing a research scien-t i s t personnel l i s t . The v i s i t i n g and correspondence was carried out i n June, 1969. The questionnaires were mailed to the research scien-t i s t s as soon as the personnel l i s t s were obtained. The university sample was the only sample that did not receive questionnaires i n June, 1969. They received questionnaires i n mid-September i n order to avoid the p o s s i b i l i t y of questionnaires "going astray" during summer session, when personnel are on holiday, teaching i n other i n s t i t u t i o n s , or 55 involved i n research elsewhere. The questionnaire "package," sent to each Canadian research s c i e n t i s t , contained a self-addressed envelope, a self-addressed post-card, a cover l e t t e r and the question-naire. The cover l e t t e r introduced the objectives of the Several faculty members from U.B.C. strongly advised against sending questionnaires to u n i v e r s i t i e s during the summer. They stated that s t a f f are either on holiday, teaching, or catching up on t h e i r own research. 52 study to the research s c i e n t i s t s , and asked t h e i r cooperation i n completing and returning the questionnaire. They were promised that a copy of the summary of the resu l t s of the study would be forwarded to t h e i r research department for th e i r perusal. The s c i e n t i s t s were also informed that the i d e n t i t y of the research department, and t h e i r own personal i d e n t i t y , would not be revealed i n any written reports. S c i e n t i s t s were not asked to sign the questionnaire. They were asked, however, to sign t h e i r name and the name of t h e i r research department on the back of the self-addressed post-card. They were asked to send these post-cards i n the mail, separate from the questionnaires. This procedure kept t h e i r i d e n t i t y secret, provided a record of returns, and permitted follow-up procedures to be used. The number of post-cards returned corresponded almost exactly with the number of questionnaires received from each 5 6 research department. The Questionnaire A questionnaire designed by D.C. Pelz and F.M. Andrews — f o r the study of s c i e n t i s t s i n research and development 56 In some cases the handwriting was i l l e g i b l e and names could not be checked o f f , but c r e d i t for a return was given to the p a r t i c u l a r research department. Since the name of the research department and organization to which i t belonged was part of the questionnaire information, a check was made and the t a l l y of post-cards and questionnaires for each research department matched almost p e r f e c t l y . 5 3 laboratories i n the United States--was modified and:added 57 to for t h i s study. These authors had been studying the performance of s c i e n t i s t s i n organizations for almost twenty years, had written numerous papers, and f i n a l l y a book c a l l e d S c i e n t i s t s i n Organizations. Compared to other writers studying s c i e n t i s t s i n organizations, Pelz and Andrews had written more and been c i t e d more often than any other authors encountered i n the course of th i s study. Besides evidence that Pelz and Andrews were "pioneers" i n the f i e l d under study, the questionnaire they had con-structed was p a r t i c u l a r l y appropriate. I t had items which appeared to measure some of the variables t h i s study was concerned about. With minor a l t e r a t i o n s , therefore, some of the Pelz.and Andrews questions were included i n the prelimin-ary questionnaire for th i s study. The alte r a t i o n s were primarily concerned with: (a) a l t e r i n g the response system from percentage-type answers to f i v e point r a t i n g scales; and (b) re-wording questions s l i g h t l y , so that they were more appropriate for Canadian s o c i a l s c i e n t i s t s . I t was recognized that questionnaires, at best, mea-sure the "perceptions" of s c i e n t i s t s only, and that responses to questionnaires do not necessarily represent r e a l i t y . Other questionnaire items suggested by the th e o r e t i c a l d i s -cussion and review of l i t e r a t u r e , were also included with the Pelz-Andrews items and became part of the preliminary questionnaire. After the preliminary questionnaire was constructed, i t was f i e l d - t e s t e d by administering i t to ten s c i e n t i s t s . These s c i e n t i s t s were from s o c i a l science departments i n the University of B r i t i s h Columbia and from the Alberta Human Resources Research Council (HRRC). These people were asked to complete the questionnaire as i f they were respondents. They were also asked to comment, i n the margin of the ques-tionnaire, on items they considered to be i r r e l e v a n t , objec-tionable,, vague, poorly worded, and generally i n need of further consideration. The purpose of the f i e l d t e s t , i n this case, was to "polish up" the questionnaire items. Suggestions received from these s c i e n t i s t s were l i s t e d and reviewed with a s o c i o l o g i s t employed by the HRRC. This s o c i o l o g i s t was considered by s t a f f at HRRC to be an expert i n questionnaire design. When f i n a l revisions 5 8 were completed, the questionnaire was printed. Questionnaire administration. As mentioned e a r l i e r , on receiving the s c i e n t i s t personnel l i s t s , questionnaire 5 8 See Appendix (iv) for the questionnaire items. Also see Appendix (v) for a discussion of the v a l i d i t y and re-l i a b i l i t y of the Pelz-Andrews items. 55 packages were immediately mailed to the research s c i e n t i s t s . Returns were received quite s t e a d i l y for about a month before dropping o f f to one or two per day. When th i s occurred, a rather strong follow-up l e t t e r , with a l t r u i s t i c overtones, was mailed to each non-respondent. Returns picked up imme-di a t e l y for another two weeks. About ten percent of these s c i e n t i s t s responded, and some said they had "mislaid" t h e i r o r i g i n a l copies of the questionnaire and promised to return one, i f another copy was sent to them. This was done, of course. The second follow-up, aft e r two weeks, included a questionnaire "package" s i m i l a r to the f i r s t , but with a d i f f e r e n t cover l e t t e r . This l e t t e r explained the purpose of the study, mentioned that a number of questionnaires had 3 "done astray," and that t h i s package was provided should t h i s be the case. They were promised that a copy of the re s u l t s would be forwarded to t h e i r research department for t h e i r perusal and so on, as i n the f i r s t cover l e t t e r . This l e t t e r also asked them to respond on the post-card, i f . t h e r e was any reason why they would l i k e to be excluded from the sample. Returns increased again for about two weeks, then dropped o f f to a "dribble" for another month. About 45 percent of the t o t a l sample population (523 s c i e n t i s t s ) returned a question-See Appendices ( v i - v i i i ) for the l e t t e r s . 56 60 naire. Another 15 percent of the sample population gave a reason for not responding. Altogether, therefore, a res-ponse rate of 60 percent was obtained.^"*" Preparation of Data for Analysis The research output items i n the questionnaire required numerical coding before responses could be transferred to 6 2 computer data sheets. A sp e c i a l numbering system recom-mended by the authors of the computer program used i n thi s study, retained the numerical i d e n t i t y of each respondent, his research department and the organization type of which his research department belonged. After s c i e n t i s t s res-ponses to questionnaire items had been transferred to computer data sheets, the data from these sheets were keypunched on computer cards. These cards became the "data decks" for the four types of organizations. 60 See Appendix ( i i ) for the d i s t r i b u t i o n of question-naire returns . ^See Appendix ( i i i ) f o r a table of the reasons given for not responding to the questionnaire. 62 Missing data was coded as "0" i n a l l cases, as the computer program used for t h i s study could be sub-programmed to omit "0" responses i n s t a t i s t i c a l c a l c u l a t i o n s . ft Susan Boyer et a l . , U.B.C. MVTAB, The U.B.C. Comput-ing Center, Vancouver, B.C., 1966, pp. 8-9. Actual i d e n t i t y oif s c i e n t i s t s was not known—just the name of the research department i n which they- worked and the type of organization i n which they were c l a s s i f i e d . Questionnaires were number coded, therefore, to give each respondent a numerical i d e n t i t y containing t h i s information. Included i n the "control deck" for computer program-ming were provisions to have the response systems f o r each variable, or question under study, collapsed to "low," "medium," and "high" ratings, or scores. For example, a f i v e point rating scale became L M H (1,2) (3) (4,5) This was done i n order to reduce the number of " c e l l s " i n the tables of responses, and thus increase the c e l l frequency for l a t e r s t a t i s t i c a l work. A spe c i a l computer program was written for t h i s study i n order to combine the f i v e output measures into two measures, 64 and thus increase the frequencies for each measure of output. The number of speeches, project completion reports, and a r t i c l e s accepted by journals became "output number one." The number 65 of books or monograph completed, became "output number two." The .computer program provided by the U.B.C. Computing Center designed for questionnaire analysis, was included i n the "control deck" and submitted with each of the "data 6 6 decks" for computer "runs." Each control deck provided for 64 Courtesy of John Campbell, U.B.C. Computing Center. ^See Appendix (ix) for the rationale behind the selec-t i o n of these items to represent research output. See Appendix (v) for a discussion of the v a l i d i t y of these items. ^Susan Boyer, l o c . c i t . 58 the construction of univariate and b i v a r i a t e frequency and per-centage frequency tables of responses of s c i e n t i s t s to either one question (the univariate case) or two questions (the b i -variate case) at a time. Data from each computer print-out--regarding s c i e n t i s t responses to each question or p a i r of questions--were trans-ferred to comprehensive frequency and percentage frequency tables so that responses could be compared. Chi-square values and contingency c o e f f i c i e n t s were then computed for these 6 7 comprehensive tables on an e l e c t r i c c a l c u l a t o r . The f i n a l computer run involved combining the data decks of s c i e n t i s t responses from the four types of organi-zations into one data deck or sample. Data from the print-out, including the chi-square values for b i v a r i a t e tables were again transferred to a comprehensive frequency and percentage frequency table. Analysis of Data: Testing for Significance In keeping with the p r i n c i p l e of parsimony, which states that phenomena should be explained on the basis of the simplest explanation consistent with a l l the facts of the case, i t was assumed that any differences to be noted i n the way s c i e n t i s t s See Appendices (x) and (xi) for the formulas and rationale for using each of these s t a t i s t i c s . responded was due to chance. S p e c i f i c a l l y , n u l l hypotheses which deny the existence of any r e a l differences between the "expected" frequency of responses and those "observed"—until the factor of chance has been eliminated as the causative agent i n the discrepancy—were assumed for s t a t i s t i c a l t e s t -68 xng. According to the n u l l hypothesis, only when the d i f -ference' noted i s greater than might be accounted for on the basis of chance fluct u a t i o n s , can i t be assumed that the difference i s " s i g n i f i c a n t . " The n u l l hypothesis was /"rejected i f the p r o b a b i l i t y of obtaining such a difference on the basis of chance alone, was very small. The n u l l hypo-thesis was "accepted" of course, when the difference was within the range of differences adequately accounted for by , 69 chance. I t was decided that the n u l l hypothesis would be rejected i f the p r o b a b i l i t y of obtaining the difference i n responses observed was less than .05. That i s , i f the pro-b a b i l i t y of obtaining the observed chi-square value was .05 or l e s s , the difference i n s c i e n t i s t responses—from what G.J. Mouly, The Science of Educational Research (New York: The American Book Co., 1963), pp. 150-151. -69 ~ o y I b i d . 60 would be expected by chance—was said to be s i g n i f i c a n t at the .05 l e v e l . The .05 p r o b a b i l i t y l e v e l , therefore, was des-c r i p t i v e of the degree of confidence one could have that a 70 r e a l difference was observed. G.E. Ferguson, S t a t i s t i c a l Analysis i n Psychology  and Education (New York: McGraw-Hill Book Co. Inc., 1959), ppT 132-133. Ferguson stated that i t i s conventional to accept p r o b a b i l i t i e s of .05 or .01 as standards of s i g n i f i c a n c e . Also Mouly, op. c i t . , pp. 152-154. The level, ofa improbability necessary to lead to the r e j e c t i o n tof the n u l l hypothesis is""' obviously a matter of judgment, based on the nature of the pro-blem and the r i s k the investigator i s w i l l i n g to take. Two types of errors are involved here. Type 1 or Alpha errors -refer to the acceptance of the n u l l hypothesis when i t i s 'actually f a l s e . Type 2 or Beta errors, on the other hand, refer to the r e j e c t i o n of the n u l l hypothesis when i t i s actually true. . . Actually the only two ways i n which both types of errors can be reduced simultaneously—and not at the expense of one another—would be by taking larger samples and/or reducing the sampling v a r i a b i l i t y by selecting a more r e s t r i c t e d population. It i s , therefore, a matter of compromise. One type of error must be balanced against the other, and the l e v e l of acceptance and r e j e c t i o n of the n u l l hypothesis must be set at what might be considered the most opportune point, depending on the r e l a t i v e severity of the two types of error. Custom and t r a d i t i o n i n the f i e l d s of education and psychology favor balan-cing the two types of errors around the points at which there are either 5 chances out of a 100 or 1 chance out of a 100 of being i n error. . . . In the early stages of exploration, i t  might be advisable to set the l e v e l of r e j e c t i o n rather low so  that variables are not eliminated before they have had a chance  to prove themselves--that i s , so that they are not rejected pre-maturely. In the l a t e r stages of the investigation of a given problem, where precision i s e s s e n t i a l , the l e v e l of r e j e c t i o n should be set higher so that relationships that are not s i g n i f i -cant w i l l be excluded. CHAPTER IV RESULTS AND CONCLUSIONS The re s u l t s are i n the form of an examination of the contents of tables of s c i e n t i s t responses to questions r e l e -vant to each s p e c i f i c hypothesis. The conclusions are based on the evidence revealed i n the r e s u l t s , and take the form of (1) accepting the n u l l hypothesis, and re j e c t i n g the hypothesis, or (2) rej e c t i n g the n u l l hypothesis, and accepting the hypothesis. The chi-square values calculated for each table of responses assisted the investigator i n making deci -sions about whether or not to accept or r e j e c t the n u l l hypothesis. The r e s u l t s and conclusions chapter was divided into four'separate sections to correspond with each of the four general hypotheses. When examination of the s p e c i f i c hypo-theses i n each section was completed, the information gained was used to tes t the general hypothesis relevant to that section. Results and Conclusions Section Number One: Relationship's-5? between Research Department Structure and Research Output S p e c i f i c Hypothesis Number 1A In a combined organization sample, levels of reported research output w i l l be p o s i t i v e l y associated with lev e l s of perceived influence to decide work goals and objectives. Table I contains the frequency, horizontal percentage frequency and t o t a l percentage frequency tables of responses of s c i e n t i s t s , from four types of organizations, to two questions. The questions were: 1. How much influence do you feel'the following people have i n deciding your work goals and objectives? Rating scale Yourself 1 2 3 4 5 2. How many of the following have you completed? (a) project completion reports . . . . (b) speeches ''' (ti.) a r t i c l e s accepted by journals ' ' ' ' An examination of the table indicated that 92 percent (horizontal percentage) of the s c i e n t i s t s who responded that they had a "high" number of outputs, also responded that they f e l t they had "much" or "very much" influence i n deciding t h e i r work goals and objectives. Although 85 percent of those who had either a "low" or "medium" number of outputs responded that they f e l t they had "much" or "very much" i n f l u e n c e — o n l y 68 percent of the "non-productive" s c i e n t i s t s responded t h i s way. Examination of the table, therefore, indicated that most s c i e n t i s t s who had high levels of research output also per-ceived themselves to have a high l e v e l of influence i n deciding t h e i r work goals and objectives. 64 TABLE I Levels of Research Output #1 versus Levels of S c i e n t i s t Influence i n Deciding Work Goals and Objectives (Combined Samples) How much influence do you f e e l the following people have i n deciding your work goals and objectives? Yourself 1 2 3 4 5 Levels of Research Output #1 Level Low 1,2 of Influence Medium 3 High 4,5 N 0 f H 12 11 22 21 73 68 107 T 2 4 "l4 f 4 24 157 185 Low 1-4 H 2 13 85 T 1 5 30 f 4 16 111 131 Medium 5-8 H 3 12. 85 T 1 3 21 f 1 7 92 100 High 9+ H 1 7 92 T 0 1 18 21 69 433 523 X 2 = 29.3 (.001) C = .23 f = f r e q u e n c y ; H = h o r i z o n t a l p e r c e n t a g e f ; T = t o t a l p e r c e n t a g e f . Appendix ( x i i i ) contains the same elements as Table T, e'xcept that research output number one (number of project completion reports etc.) was exchanged for research output number two (number of books and monographs completed). The results were comparable to those i n Table I. The chi-square values computed for both tables were s i g n i f i c a n t beyond the .05 l e v e l . On the basis of t h i s information, the n u l l hypothesis that no r e l a t i o n s h i p existed between levels of reported re-search output and levels of perceived influence to decide work goals and objectives, was rejected. The S p e c i f i c Hypothesis Number 1A, therefore, was accepted,as plausible.• S p e c i f i c Hypothesis Number IB In a combined organization sample, levels of reported research output w i l l be negatively associated with levels of perceived supervisor or department head influence i n deciding s c i e n t i s t work goals and objectives. Table II contains the frequency, horizontal percentage frequency and t o t a l percentage frequency tables of responses of s c i e n t i s t s , from four types of organizations, to two ques-tions.. The questions were: 1. How much influence do you f e e l the following people have i n deciding your work goals and objectives? . . „ , J Rating Scale Immediate supervisor or department head 1 2 3 4 5 66 2. How many of the following have you completed? (a) project completion reports • • • • . (b) speeches • • • • . (c) a r t i c l e s accepted by journals ' ' ' ' An examination of the table indicated that 52 percent (horizontal percentage) of the s c i e n t i s t s who reported a "high" number of outputs also responded that supervisors and depart-ment heads had "low" levels of influence i n deciding t h e i r work goals and objectives. Only "23 percent of the s c i e n t i s t s who reported a "high" number of outputs also responded that t h e i r supervisors or department heads had "high" levels of influence i n deciding t h e i r work goals and objectives. These same kinds of relationships were apparent for s c i e n t i s t s re-porting a "medium" number of outputs. Only 34 percent of the s c i e n t i s t s who reported "0" output also responded that t h e i r supervisors or department heads had "low" levels of influence i n deciding t h e i r work goals and objectives. Forty percent of these "0" output s c i e n t i s t s responded that supervisors or department heads had "high" levels of influence i n deciding t h e i r work goals and objectives. These same kinds of relationships were apparent for s c i e n t i s t s reporting a "low" number of outputs. Examination of the table, therefore, indicated a ten-dency for s c i e n t i s t s who reported high levels of research out-put to also respond that supervisors or department heads had "low" levels of influence i n deciding t h e i r work goals and objectives. 67 TABLE II Levels of Research Output #1 versus Levels of Supervisor Influence i n Deciding S c i e n t i s t Work Goals and Objectives (Combined Samples) How much influence do you f e e l the following people have i n deciding your work goals and objectives? Immediate supervisor, or department head 1 2 3 4 5 Levels of Research Output #1 . Level Low 1,2 of Supervisor Medium 3 Influence High 4,5 • N 0 f H 36 34 29 27 42 39 107 T 7 6 8 f 69 45 71 185 Low 1-4 H 37 24 38 T 13 9 14 f 71 34 26 131 Medium 5-8 H 54 26 20 T 14 7 '5 f 52 25 23 100 High 9+ H 52 25 23 T 10 5 4 228 133 162 523 X 2 = 22.27 ( .01) f = frequency; H = horizontal percentage f; C = .2 T = t o t a l percentage f. 6 8 Appendix (xiv) contains the same elements as Table I I , except that research output number one (number of project completion reports etc.) was exchanged for research output number two (number of books and monographs completed). The results were comparable to those i n Table I I . The chi-square values computed for both tables were s i g n i f i c a n t well beyond the .05 l e v e l . On the basis of t h i s information, the n u l l hypothesis that no r e l a t i o n s h i p existed between levels of reported re-search output and levels of perceived supervisor or depart-ment head influence i n deciding s c i e n t i s t work goals and objectives, was rejected. The S p e c i f i c Hypothesis Number IB, therefore, was accepted1,as. p l a u s i b l e . S p e c i f i c Hypothesis Number IC In a combined organization sample, levels of reported research output w i l l be p o s i t i v e l y associated with levels of perceived decentralized control of research a c t i v i t i e s , and negatively associated with perceived levels of centralized control. Table III contains the frequency, horizontal percentage and t o t a l percentage frequency tables of responses of scien-t i s t s , from four types of organizations, to two questions. The questions were: 1. Which of the following best describes your research unit or department? Highly centralized control of a c t i v i t i e s 1 Centralized control 2 Neither centralized nor decentralized 3 Decentralized control 4 Highly decentralized control of a c t i v i t i e s 5 2. How many of the following have you completed? (a) project completion reports ' ' (b) speeches ' (c) a r t i c l e s accepted by journals ' An examination of the table indicated that 51 percent (horizontal percentage) of the s c i e n t i s t s who reported high levels of research output, also responded that they would describe t h e i r research department as either decentralized or highly decentralized. Only 21 percent of the s c i e n t i s t s who reported a high number of research outputs also responded that they would describe'their research department as either centralized or highly c e n t r a l i z e d . These same kinds of relationships were apparent for s c i e n t i s t s reporting a "medium" number of research outputs. Only 32 percent of the s c i e n t i s t s who reported "0" output also responded that they would describe t h e i r research department as either decentralized or highly decentralized. Thirty-seven percent of these "0" output s c i e n t i s t s responded that they would describe t h e i r research department as either centralized or highly c e n t r a l i z e d . These same kinds of relationships were apparent for the s c i e n t i s t s reporting a low number of outputs. 70 TABLE III Levels of Research Output #1 versus Levels of Centralization/Decentralization (Combined Samples) Which of the following best describes your research  u n i t or department? Highly centralized control of a c t i v i t i e s 1 Centralized control >'2 Neither centralized nor decentralized 3 Decentralized control 4 Highly decentralized control of a c t i v i t i e s 5 Level of Levels of Centralization/Decentralization Research Centralized Neither Decentralized Output #1 1,2 3 4,5 0 f 40 33 34 107 H 37 31 32 T 8 6 7 f 60 53 72 185 Low 1-4 H 32 29 39 T 11 10 14 f 28 37 66 131 Medium 5-8 H 21 28 50 T 5 7 13 f 21 28 51 100 High 9+ H 21 28 51 T 4 5 10 149 151 223 523 X = 15.51 (.02) f = frequency; H = horizontal percentage f; C = .17 T = t o t a l percentage f. 71 Examination of Table I I I , therefore, indicated a ten-dency for s c i e n t i s t s who report high levels of research output to also describe t h e i r research department as having a decen-t r a l i z e d control of research a c t i v i t i e s . Table III also i n d i -cated a tendency for s c i e n t i s t s who report low levels of research output to also describe t h e i r research departments as having a centralized control of research a c t i v i t i e s . Appendix (xv) contains the same elements as Table I I I , except that research output number one (number of project completion reports etc.) was exchanged for research output number two (number of books and monographs completed). The results were comparable to those i n Table I I I . The chi-square values computed for both tables were s i g n i f i c a n t beyond the .05 l e v e l . On the basis of t h i s information, the n u l l hypothesis that no rel a t i o n s h i p existed between levels of reported re-search output and perceived levels of centralized and decen-t r a l i z e d control of research a c t i v i t i e s , was rejected. The S p e c i f i c Hypothesis Number IC, therefore, was accepted as plaus i b l e . S p e c i f i c Hypothesis Number ID In a combined organization sample, levels of reported research output w i l l be negatively associated with lev e l s of coordination of s c i e n t i s t s e f f o r t s for common objectives. 72 Table IV contains the frequency, horizontal percentage frequency and t o t a l percentage frequency tables of responses of s c i e n t i s t s , from four types of organizations, to two ques-tion s . The questions were: 1. How much emphasis do members of your research unit or department place on the coordination of t h e i r e f f o r t s for some common objectives? Rating Scale 1 2 3 4 5 2. How many of the following have you completed? (a) Books (b) Monographs An examination of the table indicated that 69 percent (horizontal percentage) of the s c i e n t i s t s who responded that they had completed one or more outputs, also responded that they perceived a low or medium l e v e l of coordination. Only 31 percent of the s c i e n t i s t s who completed one or more outputs perceived high -levels of coordination. There was a tendency, therefore, for s c i e n t i s t s reporting one or more outputs to respond that they perceived a low l e v e l of coordination. The chi-square value for thi s table was s i g n i f i c a n t at the .05 l e v e l . Appendix (xvi) contains the same elements as Table TV, except that research output number two (number of books and monographs completed) was exchanged for research output number one (number of project completion reports e t c . ) . The res u l t s were comparable to those i n Table IV, but.the chi-square value for the table was not s i g n i f i c a n t at the .05 l e v e l . 73 TABLE IV Levels of Research Output #2 versus Levels of Coordin-ation (Combined Sample) How much emphasis do members of your research unit or department place on the coordination of t h e i r e f f o r t s for some common objectives? 1 2 3 4 5 Levels of Research Low and Medium High Output #2 Coordination Coordination N 1, 2, 3 4, 5 f 225 161 386 Low "0" H 58 42 T 43 31 f 94 43 137 High l+„ H 69 31 T 18 8 319 204 523 =• 4.52 (.05) = .029 f = frequency; H = horizontal percentage f; T = t o t a l percentage f. 74 On the basis of t h i s information, the n u l l hypothesis that no relat i o n s h i p existed between levels of reported re-search output and levels of perceived coordination of scien-t i s t e f f o r t s for common objectives, was accepted. The S p e c i f i c Hypothesis Number ID, therefore, was rejected. I t was noted, however, that the results were i n the predicted d i r e c t i o n . The analysis of s p e c i f i c hypotheses relevant to Hypo-thesis Number One i s completed. S u f f i c i e n t information appears to be available to tes t t h i s hypothesis. Hypothesis Number One; The Test There i s a relat i o n s h i p between research department structure and research output. The s p e c i f i c hypotheses re-lated to Hypothesis Number One (number 1A, IB and IC) were tested and accepted and S p e c i f i c Hypothesis Number ID res u l t s were i n the predicted d i r e c t i o n . The relationships tested i n these s p e c i f i c hypotheses were assumed to be i n d i c a t i v e of relationships between research output and research depart-ment structure. On the basis of th i s information, the n u l l hypothesis that no relat i o n s h i p existed between research department structure and research output, was rejected. Hypothesis Number One, therefore, was accepted as pla u s i b l e . Results and Conclusions Section Number Two: Differences i n Structure Across Organization Types 76 S p e c i f i c Hypothesis Number 2A S c i e n t i s t s i n d i f f e r e n t types of organizations w i l l tend to perceive some project selection c r i t e r i a to receive d i f f e r e n t l e v e l s of emphasis. In order to be more s p e c i f i c , s p e c i f i c hypotheses, relevant to the selection c r i t e r i a used by s c i e n t i s t s i n each type of organization were constructed. The f i r s t of these s p e c i f i c hypotheses was: S p e c i f i c Hypothesis 2A^. More s c i e n t i s t s i n business organizations than i n other types of organizations w i l l respond that i t i s important to select projects that contribute to the organization's f i n a n c i a l well-being. Table V contains the frequency and percentage frequency table of responses of s c i e n t i s t s from the four types of or-ganizations, to one question. The question was: How important are the following c r i t e r i a i n your selection of research projects? Projects that contribute to the organization's f i n a n c i a l well-being Rating Scale 1 2 3 4 5 An examination of the table indicated that 60 percent of the s c i e n t i s t s i n business organizations—as compared to 21 percent i n government, 14 percent i n s o c i a l development, and 5 percent i n university organizations—responded that i t was "important" or "very important" to select research projects that contribute to the organization's f i n a n c i a l well-being. Examination of the table, therefore, indicated that more 77 TABLE V Organization Type versus Importance of Selecting Projects That Contribute to the Organization's F i n a n c i a l Well-Being How important are the following c r i t e r i a i n your se l e c t i o n of research projects? Please c i r c l e the appropriate number. (a) Projects that contribute to the organization's f i n a n c i a l well-being Rating Scale 1 t o t a l l y unim-portant 2 unimportant 3 neither impor-tant nor unimportant 4 important 5 very important Importance Organization Unimportant Neither Important N Type "0" 1,2 3 4,5 f 7 1 11 28 40 Bus. T. 15 2 23 60 Gov. S.D. Univ. f 12 65 30 30 125 T 10 47 21 21 f 7 69 17 16 102 T 6 6 4 16 14 f 1 185 32 12 229 T .5 8.0 14 . /. . . 5 27 320 90 86 496 X 2 = 129.64 f = frequency; H = horizontal percentage f; C = .455 T = t o t a l percentage f. 78 s c i e n t i s t s i n business organizations than i n the other types of organizations responded t h i s way. The chi-square value (129.64), calculated for the table of responses was s i g n i f i c a n t well beyond the .05 l e v e l , The assumed n u l l hypothesis that no difference existed across or-ganization types, i n s c i e n t i s t responses to the degree of impor-tance involved i n selecting research "projects that contribute to the organization's f i n a n c i a l well-being," was rejected. The s p e c i f i c hypothesis 2Aj, therefore, was accepted as pla u s i b l e . S p e c i f i c Hypothesis 2A?. More s c i e n t i s t s i n government organizations than i n the other types of organizations w i l l respond that i t i s important to select research projects that are delegated by some higher authority. Table VI contains the frequency and percentage f r e -quency table of responses of s c i e n t i s t s , from the four types of organizations, to one question. The question was: How important are the following c r i t e r i a i n your selection of research projects? Rating Scale Projects that are delegated by some higher authority 1 2 3 4 5 An examination of the table indicated that 65 percent of the s c i e n t i s t s i n government organizations—as compared to 49 percent i n business, 23 percent i n s o c i a l development, and 3 percent i n university organizations—responded that i t was "important" or "very important" to select research "projects that are delegated by some higher authority." Examination of the table, therefore, indicated that more s c i e n t i s t s i n 79 TABLE VI Organization Type versus Importance of Selecting Projects that are Delegated by Some Higher Authority How important are the following c r i t e r i a Rating Scale i n your se l e c t i o n of research projects? , , . ,, .„ 2 * J 1 t o t a l l y unimpor-tant 2 unimportant 3 neither important nor unimportant Please c i r c l e the appropriate number. 4 important 5 very important (d) Projects that are delegated by some higher authority 1 2 3 4 5 Organi zation Type Unimportant . "0" 1,2 Importance Neither 3 Important 4,5 N Bus. f T 4 9 11 23 9 19 23 49 43 Gov. f T 6 4 17 12 27 19 87 65 131 f 3 56 26 27 106 S.D. T 3 51 23 23 Univ. f T 1 5 191 83 30 13 8 3 229 14 275 92 145 509 = 219.25 = .548 f = frequency; T = t o t a l percentage -f. r 80 government organizations than i n the other types of organi-zations, responded t h i s way. The chi-square value (219.5) calculated for the table of responses, was s i g n i f i c a n t well beyond the .01 l e v e l . The assumed n u l l hypothesis that no differences existed across organization types, i n s c i e n t i s t responses to the degree of importance involved i n selecting research "projects that are delegated by some higher authority," was rejected. The s p e c i f i c hypotheses 2K2, therefore, was accepted Las p l a u s i b l e . S p e c i f i c Hypothesis 2 A 3 . More s c i e n t i s t s i n s o c i a l development organizations, than i n the other, types of organi-zations w i l l respond that i t i s important to sel e c t research projects that contribute to the well-being of some s o c i a l group. Table VII contains the frequency and percentage f r e -quency table of responses of s c i e n t i s t s from the four types of organizations to one question. The question was: How important are the"following c r i t e r i a i n your selection of research projects? Rating Scale Projects that may contribute to the well-being of some s o c i a l group 1 2 3 4 5 An examination of the table indicated that 76 percent of the s c i e n t i s t s i n s o c i a l development organizations--as : compared to 4 3 percent i n business, 55 percent i n government, and 55 percent i n university organizations—responded that i t was "important" or "very important" to select research "pro-81 TABLE VII Organization Type versus Importance of Selecting Projects that Contribute to the Weil-Being of Some Soc i a l Group How important are the following c r i t e r i a i n your se l e c t i o n of research projects? Please c i r c l e the appropriate number. (c) Projects that may contribute to the well-being of some s o c i a l group Rating Scale 1 t o t a l l y unimportant 2 unimportant 3 neither important nor unimportant 4 important 5 very important Importance Organization Unimportant Neither Important N Type "0" 1,2 3 4,5 - f 8 9 10 20 39 Bus. T 17 19 21 43 f 10 27 25 75 127 Gov. T . 7 20 18 55 f 5 10 11 83 104 S.D. T 5 9 10 76 f 1 50 52 127 229 Univ. T 5 22 23 55 24 96 98 305 499 X 2 = 20.32 C = .17 f = frequency; T = t o t a l percentage f. 82 jects that contribute to the well-being of some s o c i a l group." Examination of the table, therefore, indicated that more s c i e n t i s t s i n s o c i a l development organizations than i n the other types of organizations responded th i s way. The chi-square value (20.32) calculated for the table of responses was s i g n i f i c a n t beyond the .05 l e v e l . The assumed n u l l hypothesis that no difference existed across organization types, i n s c i e n t i s t responses to the degree of importance i n -volved i n selecting research "projects that contribute to the well-being of some s o c i a l group," was rejected. The s p e c i f i c hypothesis 2A3, therefore, was acceptedvas p l a u s i b l e . S p e c i f i c Hypothesis 2A4. More s c i e n t i s t s i n univer-s i t i e s , . t h a n i n the other types of organizations, w i l l res-pond that i t i s important to select research projects that are i n t e r e s t i n g i n themselves. Table VIII contains the frequency"and percentage f r e -quency table of responses of s c i e n t i s t s from the four types of organizations to one question. The question was: How important are the following c r i t e r i a i n your se l e c t i o n of. research projects?. Rating Scale Projects that are i n t e r e s t i n g i n themselves 1 2 3 4 5 An examination of the table indicated that 81 percent of the s c i e n t i s t s i n u n i v e r s i t i e s — a s compared to 57 percent i n business, 38 percent i n government and 50 percent i n s o c i a l development organizations—responded that i t was "important" 83 TABLE VIII Organization Type versus Importance of Selecting Projects that are Interesting i n Themselves How important are the following c r i t e r i a i n Rating Scale your selection of research projects? Please c i r c l e the appropriate number (a) Projects that are i n t e r e s t i n g i n themselves 1 t o t a l l y unimpor-tant 2 unimportant 3 neither important nor unimportant 4 important 5 very important 1 2 3 4 5 Organization Type Unimportant "0" 1,2 Importance Neither 3 Important 4,5 N Bus. f T. 8 17 9 19 3 6 27 57 39 Gov. f T 7 6 43 31 34 25 38 130 f 7 22 26 •54 102 S .D. T 6 20 24 50 Univ. f T 0 22 10 21 9 187 81 230 22 96 84 321 501 X =81.3 C = .367 f = frequency; T = t o t a l percentage f, 84 or "very important" to select research "projects that are in t e r e s t i n g i n themselves." Examination of the table, there-fore, indicated that more s c i e n t i s t s i n u n i v e r s i t i e s than i n the other types of organizations, responded t h i s way. The chi-square value (81.3) calculated for the table o responses, was s i g n i f i c a n t well beyond the .01 l e v e l . The assumed n u l l hypothesis that no difference existed across organization types, i n s c i e n t i s t responses to the degree of importance involved i n selecting research "projects that are int e r e s t i n g i n themselves," was rejected. The s p e c i f i c hypo-thesis 2 A 4 , therefore was accepted',as p l a u s i b l e . v S p e c i f i c Hypothesis 2 A 5 . More s c i e n t i s t s i n u n i v e r s i -t i e s than i n the other types of organizations w i l l respond that i t i s important to sel e c t research projects that seek new knowledge—because the search for new knowledge i s i n t r i n s i c a l l y good. Table IX contains the frequency and percentage f r e -quency table of responses of s c i e n t i s t s from the four types of organizations to one question. The question was: How important are the following c r i t e r i a i n your se l e c t i o n of research projects? Rating Scale Projects that seek new knowledge— because the search for new knowledge i s i n t r i n s i c a l l y good. 1 2 3 4 5 An examination of the table indicated that 58 percent the s c i e n t i s t s i n u n i v e r s i t i e s — a s compared to 28 percent i n business, 24 percent i n government, and 42 percent i n s o c i a l 85 TABLE IX Organization Type versus Importance of Projects that Seek New Knowledge How important are the following c r i t e r i a i n your se l e c t i o n of research projects? Please c i r c l e the appropriate number (b) Projects that seek new knowledge— because the search for new knowledge i s i n t r i n s i c a l l y good Importance Organization Unimportant Neither Important N Type "0" 1,2 3 4,5 f 7 13 14 13 40 Bus. T 15 28 30 2 8 f 13 56 35 33 124 Gov. T 9 40 26 24 f 7 31 25 46 102 S.D. T 6 28 23 42 f 1 50 45 134 229 Univ. 22 20 58 T ...5. 28 150 129 226 495 . . . . Selecting Rating Scale 1 t o t a l l y unimpor-tant 2 unimportant 3 neither important nor unimportant 4 important 5 very important 1 2 3 4 5 = 29.29 = .22 f = frequency; T = t o t a l percentage f. 86 development organizations—responded that i t was "important" or "very important" to select research "projects that seek new knowledge. . ." Examination of the table, therefore, indicated that more s c i e n t i s t s i n u n i v e r s i t i e s than i n the other types of organizations, responded th i s way. The chi-square value (29.9) calculated for the table of responses was s i g n i f i c a n t beyond the .01 l e v e l . The assumed n u l l hypothesis that no differences existed across organization types, i n s c i e n t i s t responses to the degree of importance i n -volved i n selecting research "projects that seek new knowledge . . .," was rejected. The s p e c i f i c hypothesis 2 A 5 , therefore, was regarded as tenable. A l l of the s p e c i f i c hypotheses for S p e c i f i c Hypothesis Number 2A were accepted. On the basis of t h i s information, the assumed n u l l hypothesis that no difference existed across organization types, i n s c i e n t i s t responses to the l e v e l of importance of d i f f e r e n t c r i t e r i a i n the selection of research-projects, was rejected. S p e c i f i c Hypothesis Number 2A, there-fore, was accepted as p l a u s i b l e . S p e c i f i c Hypothesis Number 2 B S c i e n t i s t s i n d i f f e r e n t types of organizations w i l l perceive d i f f e r e n t levels of time expenditures i n basic and applied research a c t i v i t i e s . In order to be more s p e c i f i c , s p e c i f i c hypotheses, relevant to both basic research a c t i v i t i e s and to applied 87 research a c t i v i t i e s , were constructed. The f i r s t of these s p e c i f i c hypotheses relevant to applied research a c t i v i t i e s was: S p e c i f i c Hypothesis 2B-| . More s c i e n t i s t s i n business, government and s o c i a l development organizations than i n uni-v e r s i t i e s w i l l perceive themselves to emphasize high expendi-tures of time i n applied research a c t i v i t i e s . Table X contains the frequency and percentage frequency table of s c i e n t i s t responses to the question: How much of your working time do you spend i n the discovery of s p e c i f i c knowledge for the solution of p a r t i c u l a r problems (applied research)? An examination of the table indicated that 60 percent of the s c i e n t i s t s i n business, 48 percent i n government and 36 percent i n s o c i a l development organizations—as compared to 18 percent of the s c i e n t i s t s i n universities—responded that they spent "much" or "very much" of t h e i r time i n applied research. Examination of the table, therefore, indicated that more s c i e n t i s t s i n business, government, and s o c i a l development organizations than i n u n i v e r s i t i e s responded th i s way. The chi-square value (62.84) calculated for the t a b l e of responses was s i g n i f i c a n t well beyond the .05 l e v e l . The assumed n u l l hypothesis that no difference existed across organization types, i n s c i e n t i s t responses to the amount of time spent i n applied research a c t i v i t i e s , was rejected. The s p e c i f i c hypothesis 2Bj, therefore, was accepted.as p l a u s i b l e . 88 TABLE X Organization Type versus Levels of Time Expenditure on Applied Research How much of your working time do you spend on ^ a t i n g Scale the following a c t i v i t i e s i n your present re- ~ n o ^ e search unit or department? \ n o t v e r ^ m u c h 3 some Please c i r c l e the appropriate number. (a) Discovery of s p e c i f i c knowledge forrthe 4 much 5 very much solution of p a r t i c u l a r problems (applied research) 1 2 3 4 5 Levels of Time Expenditure Organization Low Medium High Type 0 1,2 3 4,5 N f 1 9 9 28 46 B U S * . T 2 19 19 60 f 6 32 33 66 131 G O V ' T 4 23 24 48 f 3 29 38 39 106 S * D * T 3 27 34 36 f 0 115 72 43 230 U n i V - T 0 50 31 18 10 185 152 176 513 62.84 .33 f = frequency; T = t o t a l percentage f. 89 S p e c i f i c Hypothesis 2B2. More s c i e n t i s t s i n u n i v e r s i -t i e s than i n the other types of organizations w i l l perceive themselves to emphasize high expenditures of time i n basic research a c t i v i t i e s . Table XI contains the frequency and percentage frequency table of s c i e n t i s t responses to the question: How much of your working time do you spend i n the discovery of new knowledge, not immediately applicable to p a r t i c u l a r problems (basic research)? An examination of the table indicated that 39 percent of the s c i e n t i s t s i n u n i v e r s i t i e s — a s compared to 21 percent i n s o c i a l development, 7 percent i n government and 2 percent i n business organizations—responded that they spent either "much" or "very much" of t h e i r time i n basic research. Examin-ation of the table, therefore, indicated that more s c i e n t i s t s i n u n i v e r s i t i e s than i n the other types of organizations, responded t h i s way. The chi-square value (85.2) calculated for the table of responses was s i g n i f i c a n t well beyond the .05 l e v e l . The assumed n u l l hypothesis that no differences existed across organization types i n s c i e n t i s t responses to how much of t h e i r time they spent i n basic research a c t i v i t i e s , was rejected. The s p e c i f i c hypothesis 2B2, therefore, was accepted'as p l a u s i b l e . Both s p e c i f i c hypotheses for S p e c i f i c Hypothesis Number 2B were accepted. On the basis of t h i s information, the assumed n u l l hypothesis that no differences existed across organization 90 TABLE XI Organization Type versus Levels of Time Expenditure on Basic Research How much of your working time do you spend Rating Scale on the following a c t i v i t i e s , i n your 1 none present research unit or department? 2 not very much Please c i r c l e the appropriate number ^ much 55very much (b) Discovery of new knowledge, not immediately applicable to p a r t i c u l a r problems (basic research). 1 2 3 4 5 • • - Levels of Time Expenditure Organization Low Medium High N Type. 0 1,2 3 4,5 f 2 35 9 1 45 B U S * T 4 74 19 2 f 9 94 25 9 128 G O V \ . T 7 68 18 7 f 5 .51 29 24 104 S \ D V . " . . T 5 47 27 21 f 1 73 66 90 229 U n i V * T .5 32 29 39 17 253 129 124 506 X 2 = 85.2 f = frequency; T = t o t a l percentage f. types i n s c i e n t i s t responses, to the levels of time expenditure i n basic and applied research a c t i v i t i e s , was rejected. S p e c i f i c Hypothesis Number 2B, therefore, was accepted. S p e c i f i c Hypothesis Number 2C More s c i e n t i s t s i n business organizations than i n the other types of organizations w i l l perceive themselves to work under high levels of time pressure. Table XII contains the frequency and percentage f r e -quency table of responses to the question: How much time pressure do you work under ( i . e . , r e s u l t s needed i n a hurry, deadlines to be met, etc.)? An examination of the table indicated that 91 percent of the s c i e n t i s t s i n business organizations—as compared to 39 percent.in government, 37 percent i n s o c i a l development, and 32 percent i n university organizations—responded that they worked under either "much" or "very much" time pressure. Examination of the table, therefore, indicated that more s c i e n t i s t s i n business organizations than i n the other types of organizations responded t h i s way. The chi-square value (62.87) calculated for the table of responses was s i g n i f i c a n t well beyond the .05 l e v e l . The assumed n u l l hypothesis that no differences existed across organization types i n s c i e n t i s t responses to the l e v e l s of time pressure worked under, was rejected. S p e c i f i c Hypothesis Number 2C, therefore, was accepted;as p l a u s i b l e . 9 2 TABLE XII Organization Type versus Levels of Time Pressure Worked Under Rating Scale 1 none 2 not very much Please c i r c l e the appropriate number. How much time pressure do you work under (I.e., results needed i n a hurry, dead-lin e s to be met, etc.)? 3 some 4 much 5 very much Organization Type Levels of Time Pressure Low Medium High 0 1 , 2 3 4 , 5 N Bus f T 4 3 9 1 4 7 f 2 3 1 5 1 5 3 1 3 5 G o V \ T 1 2 3 3 7 3 9 f 1 2 7 4 0 4 1 1 0 8 S \ D * T 1 2 5 3 7 3 7 f . l 8 0 7 5 7 4 2 2 9 U n i V * T 5 3 5 3 3 3 2 4 1 3 9 1 6 9 2 1 1 5 1 9 = 6 7 . 8 7 = . 3 2 8 f = frequency; T = t o t a l percentage f. 93 S p e c i f i c Hypothesis Number 2D More s c i e n t i s t s i n u n i v e r s i t i e s than i n the other types of organizations, w i l l perceive high lev e l s of influence to decide t h e i r work goals and objectives. Table XIII contains the frequency and percentage frequency tables of responses to one question. The question was: How much influence do you f e e l the following people have i n deciding your work goals and objectives? Rating Scale Yourself 1. 2 3 4 5 An examination of the table indicated that 93 percent of the s c i e n t i s t s i n u n i v e r s i t i e s — a s compared to 81 percent i n s o c i a l development, 69 percent i n government, and 79 per-cent i n business organizations—responded that they f e l t they had either "much" or "very much" "influence to decide t h e i r work goals and objectives. Examination of the table, there-fore, indicated that more s c i e n t i s t s i n u n i v e r s i t i e s than i n the other types of organizations, responded t h i s way. The chi-square value (40.59) calculated for the table of responses was s i g n i f i c a n t well beyond the .05 l e v e l . The assumed n u l l hypothesis that no differences existed across organization types i n s c i e n t i s t responses to the amount of influence they perceive themselves to have i n deciding work goals and objectives, was rejected. S p e c i f i c Hypothesis Number 2D, therefore, was accepted as p l a u s i b l e . 94 TABLE XIII Organization Type versus Levels of S c i e n t i s t Influence i n Deciding Work Goals and Objectives How itiuch influence do you f e e l the following R a t i n g Scale people have i n deciding your work goals and 1 none objectives? 2 not very much 3 some 4 much 5 very much (c) Yourself 1 2 3 4 5 Levels of Influence Organization Low Medium High N Type 0 1,2 3 4,5 f. 0 " 0 10 37 47 B U S * T 0 0 21 79 f 2 8 33 9.4 135 G O V ; T 1 6 24 69 . f 2 3 15 89 107 S \ D \ T 2 3 14 81 f 3 5 11 213 227 U n i v V T 1 2 5 ' 93 7 16 69 433 513 X 2 = 40.59 C = .268 f = frequency; T = t o t a l frequency f. S p e c i f i c Hypothesis Number 2E More s c i e n t i s t s i n business, government, and s o c i a l development organizations than i n u n i v e r s i t i e s w i l l perceive immediate supervisors or department heads to have high lev e l s of influence to decide t h e i r work goals and objectives. Table XIV contains the frequency and perdentage f r e -quency table of responses of s c i e n t i s t s from the four types of organizations to one question. The question was: How much influence do you f e e l the following people have i n deciding your work goals and objectives? Rating Scale Immediate supervisor or department head 1 2 3 4 5 An examination of the table indicated 47 percent of the s c i e n t i s t s i n business, 60 percent of the s c i e n t i s t s i n government, and. 33 percent of the s c i e n t i s t s i n s o c i a l develop-ment organizations—as compared to 10 percent of the scien-t i s t s i n universities—responded that they f e l t t h e i r imme-diate supervisor or department head had "much" or "very much" influence i n deciding t h e i r work goals and objectives. Examination of the table, therefore, indicated that more s c i e n t i s t s i n business, government, and s o c i a l development organizations than i n u n i v e r s i t i e s , responded t h i s way. The chi-square value (179.92) calculated for the table of responses was s i g n i f i c a n t well beyond the .05 l e v e l . The assumed n u l l hypothesis that no differences existed across 96 TABLE XIV Organization Type versus Levels of Supervisor Influence i n Deciding Work Goals and Objectives How much influence do you f e e l the following Rating Scale people have i n deciding your work goals *1 none and objectives? 2 not very much 3 some 4 much (d) Immediate supervisor or department 5 very much head 1 2 3 4 5 Levels of Influence Organization Low Medium High N Type 0 1,2 3 4,5 f 3 7 15 22 44 Bus. T 6 15 32 47 f 9 9 37 82 128 Gov. T 7 7 27 60 f 8 30 35 36 101 S.D. T 7 28 32 33-f 3 159 46 22 227 Univ. T 1 69 20 10. , 23 205 133 162 500 X = 179.92 f = frequency; T = t o t a l frequency f. C = .513 organization types, i n s c i e n t i s t responses about perceived levels of influence immediate supervisors and department, heads have i n deciding s c i e n t i s t work goals and objectives, was rejected. S p e c i f i c Hypothesis Number 2E, therefore, was accepted as pla u s i b l e . Hypothesis Number 2F Sc i e n t i s t s i n each type of organization w i l l tend to perceive d i f f e r e n t levels of emphasis on centralized or decen-t r a l i z e d control of research department a c t i v i t i e s . S p e c i f i c a l l y : Hypothesis Number 2F-[ . More s c i e n t i s t s i n universities', than i n the other types of organizations, w i l l perceive a decentralized control of research department a c t i v i t i e s . Table XV contains the frequency and percentage frequency table of responses of s c i e n t i s t s from the four types of organi-zations, t o one question. The question was: Which of the following best describes your research unit or department? Highly centralized control of a c t i v i t i e s 1 Centralized control 2 Neither centralized nor decentralized 3 Decentralized control 4 Highly decentralized control of a c t i v i t i e s 5 An examination of the table of responses indicated that 40 percent of the s c i e n t i s t s i n business,. 23. percent, i n govern-ment, and 29 percent i n s o c i a l development organizations—as compared to 61 percent of the s c i e n t i s t s i n u n i v e r s i t i e s — responded that they would describe t h e i r research unit or 98 TABLE XV Organization Type versus Levels of Centralized and Decentralized Control of Research A c t i v i t i e s Which of the following best describes your research unit or department? Please c i r c l e the appropriate number Highly centralized control of a c t i v i t i e s 1 Centralized control 2 Neither centralized nor decentralized 3 Decentralized control 4 Highly decentralized control of a c t i v i t i e s 5 Organization Centralized Neither Decentralized N Type 0 1,2 3 4,5 f 0 16 12 19 47 B U S * T 0 34 26 40 Gov. f T 1 5 61 45 43 31 32 23 136 S.D. f 1 40 36 32 108 T 1 37 33 29 Univ. f T 1 5 29 13 60 26 140 61 229 3 146 151 213 520 = 72.47 = .348 f = frequency; T = t o t a l percentage f. 99 department as having "decentralized" or "highly decentralized" control of research a c t i v i t i e s . Examination of the table, therefore, indicated that more s c i e n t i s t s i n u n i v e r s i t i e s than i n the other types of organizations, responded t h i s way. The chi-square value (72.47) calculated for the table of responses was s i g n i f i c a n t well beyond the .05 l e v e l . The assumed n u l l hypothesis that no differences existed across organization types, i n s c i e n t i s t s responses to the amount of "decentralized" control of research department a c t i v i t i e s , was rejected. S p e c i f i c Hypothesis Number 2Fi, therefore, was accepted^,as p l a u s i b l e . S p e c i f i c Hypothesis Number 2F2. More s c i e n t i s t s i n business, government and s o c i a l development organizations, than i n u n i v e r s i t i e s , w i l l perceive a centralized control of research department a c t i v i t i e s . A further examination of Table XV indicated that 34 percent of the s c i e n t i s t s in. business, 45 percent i n govern-ment, and 37 percent i n s o c i a l development organizations—as compared to 13 percent of the s c i e n t i s t s i n u n i v e r s i t i e s — responded that they would describe t h e i r research unit.or. department.as.having centralized or highly centralized control of research a c t i v i t i e s . Examination of the table, therefore, indicated..that more s c i e n t i s t s i n business, government, and s o c i a l development organizations than i n u n i v e r s i t i e s respon-ded t h i s way. 100 The chi-square value (72.47) calculated for the table of responses was s i g n i f i c a n t well beyond the .05 l e v e l . The assumed n u l l hypothesis that no differences existed across organization types, i n s c i e n t i s t s responses to the level s of centralized control of re s e a r c h a a c t i v i t i e s , was rejected. S p e c i f i c Hypothesis Number 2F2, therefore, was accepted. S p e c i f i c Hypotheses Number 2Fi and 2F2 were accepted. On the basis of thi s information, the assumed n u l l hypothesis that no differences existed across organization types, i n s c i e n t i s t responses to the levels of centralized and decentra-l i z e d control of research a c t i v i t i e s , was rejected. S p e c i f i c Hypothesis Number 2F, therefore, was accepted as pl a u s i b l e . S p e c i f i c Hypothesis Number 2G More s c i e n t i s t s i n business, government and s o c i a l development organizations than i n u n i v e r s i t i e s w i l l perceive high levels of coordination toward common objectives. Table XVI contains the frequency and percentage f r e -quency table of s c i e n t i s t responses to the question: How much emphasis do members of your research unit or department place on the coordination of t h e i r e f f o r t s for some common objectives? An examination of the table indicated that 70 percent of the s c i e n t i s t s i n business, 61 percent i n government, and 56 percent i n s o c i a l development organizations—as compared to 11 percent of the s c i e n t i s t s i n universities—responded that they perceived members of t h e i r research department to 101 TABLE XVI Organization Type versus Levels of Emphasis on Coordin-ation of E f f o r t Rating Scale 1 none 2 not very much Please c i r c l e the appropriate number ^ much How much emphasis do members of your 5 very much research unit or department place on the coordination of t h e i r e f f o r t s f or some common objectives? 1 2 3 4 5 Organization Type LOW . 0 Levels 1,2 of Emphasis Medium High 3 4,5 N Bus. f T 0 0 4 8 10 21 33 70 47 Gov. f T 4 3 22 16 28 20 83 61 133 S.D. f T 4 4 17 16 27 24 61 56 105 Univ. f T 0 0 144 62 60 26 26 11 230 8 187 125 203 515 2 X =16 5.5 f = frequency; T = t o t a l percentage f. C = .49 102 put "much" or "very much" emphasis on coordinating t h e i r e f f o r t s for some common objectives. Examination of the table, therefore, indicated that more s c i e n t i s t s i n business, government, and s o c i a l development organizations, than i n u n i v e r s i t i e s , res-ponded th i s way. The chi-square value (165.5) calculated for the table of responses was s i g n i f i c a n t well beyond the .05 l e v e l . The assumed n u l l hypothesis that no differences existed across organization types i n levels of coordination of e f f o r t s toward common objectives, was rejected. S p e c i f i c Hypothesis Number 2G, therefore, was accepted as pl a u s i b l e . The" analysis of data related to Hypothesis Number Two i s completed. S u f f i c i e n t information i s avail a b l e , therefore, with which to tes t t h i s hypothesis. Hypothesis Number Two: The Test Research departments i n d i f f e r e n t types of parent or-ganizations are structured d i f f e r e n t l y . The s p e c i f i c hypotheses related to Hypothesis Number Two, i n a l l cases, were accepted. On the basis'of t h i s i n f o r -mation, the assumed n u l l hypothesis that no differences existed across organization types i n research department structures, was rejected. Hypothesis Number Two, therefore, was accepted as p l a u s i b l e . Results and Conclusions Section Number Three: Relationships Between Research Output and Structure Across Organization Types 104 Sp e c i f i c Hypothesis Number 3A There w i l l be a greater tendency i n u n i v e r s i t i e s , than i n other types of organizations, for high levels of reported research output to be associated with s c i e n t i s t perceptions of high levels of influence to decide t h e i r own work goals and objectives. Table XVII contains the frequency and percentage f r e -quency tables of responses of high research output s c i e n t i s t s , 71 from the four types of organizations, to one question. The question was: How much influence do you f e e l the following people have i n deciding your work goals and objectives? „ .. „ ,^ J Rating Scale Yourself 1 2 3 4 5 An examination of the table indicated that 92 percent of the high output s c i e n t i s t s i n u n i v e r s i t i e s — a s compared to 86 percent i n s o c i a l development, 82 percent i n government, and 75 percent i n business organizations—responded that they perceived a high l e v e l of influence i n deciding t h e i r work goals and objectives. Examination of Table XVII, therefore, indicated that more s c i e n t i s t s from u n i v e r s i t i e s than from the other types of organizations, responded t h i s way. 71 High research output respondents were those s c i e n t i s t s who reported that they had completed 5 or more of output #1 within a f i v e year period. Approximately half of the sample had completed 5 or more outputs. TABLE XVII 105 High Levels of Research Output versus Levels of S c i e n t i s t s Decision Making Influence How much influence do you f e e l the following people have i n deciding your work goals and objectives? Yourself 1 2 3 4 5 Organization Type High Levels of Low and Medium Influence 1,2,3 Research Output High levels of Influence (5+) 4,5. N Bus. f T • 5 25 15 75 20 Gov. f T 6 18 28 82 3.4 S.D. f T 7 14 .43 86 50 Univ. f T 10 8 117 92 127 28 203 231 X = 6.4 (.10) High Research Output = 5 or more of: project c _ completion reports; a r t i c l e s accepted by journals; and speeches'delivered (completed within the l a s t 5 years, i n the s c i e n t i s t ' s present research department). Chi-Square Values: Separate Sample Compared to One Another Un. vs. Bus. X 2 = 5.53 (.02 Un. vs. Gov. X 2 = 2.86 (.10) Un. vs. S.D. X 2 = 1.55 Bus.vs.S.D. X2 = 2.36 (.20) Bus. vs. Gov. x 2 = .419 S.D. vs.Gov. X^ = .2 f = frequency; T = t o t a l percentage f. 106 Appendix (xvii) contains the same elements as Table XVII, except that output number one (project completion reports etc.) i s exchanged for output number two (number of books and monographs completed). The re s u l t s were comparable to those i n Table XVII. The chi-square values for the two tables of responses were s i g n i f i c a n t at only the .10 l e v e l , not the .05 l e v e l . Examination of the chi-square values l i s t e d below Table XVII and Appendix (xvii) indicated that the unive r s i t y and business organization s c i e n t i s t responses d i f f e r e d s i g n i -f i c a n t l y from one another. The values l i s t e d below Appendix (xvii) also indicated that university and s o c i a l development s c i e n t i s t responses d i f f e r e d s i g n i f i c a n t l y from one another. The other organization samples, however, did not appear to d i f f e r s i g n i f i c a n t l y from each other. On the basis of t h i s information, the n u l l hypothesis that no differences existed across organization types, i n the rela t i o n s h i p between leve l s of reported research output and perceived lev e l s of influence to decide work-goals and objectives, was accepted. S p e c i f i c Hypothesis Number 3A, therefore, was rejected. I t was noted, however, that the results did not give complete support to r e j e c t i o n of the h y p o t h e s i s — e s p e c i a l l y as regards the unive r s i t y sample.-107 S p e c i f i c Hypothesis Number 3B There w i l l be a greater tendency i n business, govern-ment, and s o c i a l development organizations, than i n univer-s i t i e s , for high levels of reported research output to be associated with high levels of perceived supervisor or department head influence i n deciding work goals and objectives, Table XVIII contains the frequency and percentage f r e -quency tables of responses of high research output s c i e n t i s t s , from the four types of organizations, to one question. The question was: How much influence do you f e e l the following people have i n deciding your work goals and objectives? _. , . „ , J Rating Scale Immediate supervisor or department head 1 2 3 4 5 An examination of the table indicated that 30 percent of the high output s c i e n t i s t s i n business organizations, 56 percent i n government, and 22 percent i n s o c i a l develop-ment organizations—as compared to 10 percent i n u n i v e r s i t i e s — responded that they perceived immediate supervisors or depart-ment heads to have high lev e l s of influence i n deciding t h e i r work goals and objectives. Examination of Table XIV, there-fore, indicated that fewer s c i e n t i s t s from u n i v e r s i t i e s than from the other types of organizations, responded t h i s way. Appendix ( x v i i i ) contains the same elements as Table XVIII except that output number one (number of project com-108 TABLE XVIII High Levels of Research Output versus Levels of Supervisor Decision Making Influence How .much influence do you f e e l the following people have i n deciding your work goals and objectives Immediate Supervisor or department head 1 2 3 4 5 Organization Type High Levels of Low and Medium Supervisor Inf. Research Output (5+) High Supervisor 1,2,3 Influence-4,5 N Bus. f T 14 70 6 ;.\ 30 20 Gov. f T 15 44 , 19 56 . .34 S.D. f T 39 78 11. 22 . 50 Univ. f T 114 90 13 10 127 182 49 231 X = 34.54 (.001) High Research Output = 5 or more of: c _ project completion reports; a r t i c l e s _ * accepted by journals; and speeches delivered Chi-square Values: Separate Samples vs. Separate Samples • Un. vs. Bus. x 2 = 5.99 (.02) Un.vs. Gov. x 2 = 35.08(.001) Un. vs.S.D. x 2 = 4.23 (.05) Gov. vs. S.D.x2 = 10.11(.01) Gov. vs. Bus.x 2 = 3.39 (.10) S.D. vs.Bus. x 2 = .49 f = frequency; T = t o t a l percentage f. 109 p l e t i o n reports etc.) i s exchanged for output number two (number of books and monographs completed). The r e s u l t s were comparable to those i n Table XVIII. The chi-square values for the two tables of responses were s i g n i f i c a n t well beyond the .05 l e v e l . Examination of the chi-square values l i s t e d below Table XVIII and Appendix ( x v i i i ) also indicated that the university s c i e n t i s t responses d i f f e r e d s i g n i f i c a n t l y from the responses of s c i e n t i s t s i n most of the other organization samples. On the basis of t h i s information, the n u l l hypothesis that no differences existed across organization types, i n the rel a t i o n s h i p between high levels of reported research output and perceived levels of immediate supervisor or depart-ment head influence to decide s c i e n t i s t work goals and objectives, was rejected. S p e c i f i c Hypothesis Number 3B, therefore, was accepted as p l a u s i b l e . S p e c i f i c Hypothesis 3C Relationship between high levels of reported research output and perceived lev e l s of emphasis on centralized and decentralized control of research department a c t i v i t i e s , w i l l vary across organization types. S p e c i f i c a l l y : 110 S p e c i f i c Hypothesis 3C-^ . There w i l l be a greater tendency i n business, government and s o c i a l development organizations, than i n u n i v e r s i t i e s , for high lev e l s of reported research output to be associated with perceived leve l s of emphasis on centralized control of research depart-ment a c t i v i t i e s . Table XIX contains the frequency and percentage f r e -quency tables of responses of high research output s c i e n t i s t s , from the four types of organizations, to one question. The question was: Which of the following best describes your research unit or department? Highly centralized control of a c t i v i t i e s 1 Centralized control 2 Neither centralized nor decentralized 3 Decentralized control 4 Highly decentralized control of a c t i v i t i e s 5 An examination of the table indicated that 30 percent of the high research output s c i e n t i s t s i n business organi-zations, 38 percent i n government, and 30 percent i n s o c i a l development organizations—as compared to 12 percent i n universities—responded that they perceived a centralized control of a c t i v i t i e s i n t h e i r research departments. Examin-ation of Table XV, therefore, indicated that fewer s c i e n t i s t s from u n i v e r s i t i e s than from the other types of organizations, responded th i s way. I l l TABLE XIX High Levels of Research Output versus Levels of Centralized and Decentralized Control Which of the following best describes your research unit or department? Highly centralized control of a c t i v i t i e s Centralized control Neither centralized nor decentralized Decentralized control Highly decentralized control of a c t i v i t i e s 1 2 3 4 5 Organization Type -Centralized 1,2 Neither 3 Decentrali zed 4,5 N Bus . f T 6 30 3. 15 11 55 20 Gov. f T 13 38 11 32 10 30 34 S.D. f T 15 30 20 40 15 30 50 f 15 31 81 127 U n i v ' T 12 24 . 64 49 65 117 231 2 X =28.68(.001) High Research Output = 5 or more of: _ 232 Project completion, reports; a r t i c l e s ~~ accepted by journals; and speeches delivered. Chi-square Values: Separate Samples vs. Separate Samples Un.vs.Bus. x 2 = 4.87(.10) Un.vs.Gov. x 2 = .17 .02 (...00.1) Un.vs.S.D. x 2 = 17.57(.001) S.D.vs.Gov. x 2 = .73 S.D.vs.Bus. x 2 = 5.12C.10) Gov.vs.Bus. x 2 = 3 . 8 2 (...20) f = frequency; T = t o t a l percentage f. 112 Appendix (xix) contains the same elements as Table XIX except that output number one (number of project com-pl e t i o n reports etc.) i s exchanged for output number two (number of books and monographs completed). The r e s u l t s were comparable to those i n Table XIX. The chi-square values for the two tables of responses were s i g n i f i c a n t well beyond the .05 l e v e l . Examination of the chi-square values l i s t e d below Table XIX and Appendix (xix) also indicated that s c i e n t i s t responses i n most of the government, s o c i a l development and business organization samples d i f f e r e d s i g n i f i c a n t l y from the responses of the university s c i e n t i s t sample. On the basis of t h i s information, the n u l l hypothesis that no differences existed across organization types, i n the association between high levels of reported research output and perceived l e v e l of centralized control of research a c t i -v i t i e s , was rejected. S p e c i f i c Hypothesis Number 3Ci, there-fore, was accepted as p l a u s i b l e . S p e c i f i c Hypothesis 3C2 There w i l l be a greater tendency i n u n i v e r s i t i e s than i n the other types of organizations, for high levels of re-ported research output to be associated with perceived lev e l s of emphasis on decentralized control of research department a c t i v i t i e s . 113 Examination of Table XIX also indicated that 64 percent of the high output s c i e n t i s t s i n u n i v e r s i t i e s — a s compared to 30 percent i n s o c i a l development, 30 percent i n government, and 55 percent i n business organizations—responded that they perceived a decentralized control of a c t i v i t i e s i n t h e i r research departments. Further examination of Table XIX, therefore, indicated that more s c i e n t i s t s from u n i v e r s i t i e s than from the other types of organizations responded th i s way. Further examination of Appendix (xix) indicated com-parable results to those i n Table XIX. The chi-square values for the two tables of responses were s i g n i f i c a n t well beyond the .05 l e v e l . On the basis of t h i s information, the n u l l hypothesis that no difference existed across organization types, i n the association between high levels of reported research output and perceived levels of decentralized control of research a c t i v i t i e s , was rejected. S p e c i f i c Hypothesis Number 3C2/ therefore, was accepted as p l a u s i b l e . S p e c i f i c Hypotheses number 3CX and 3C2 were accepted. On the basis of t h i s information, the n u l l hypothesis that no differences existed across organization types, i n the re-lationships between high levels of reported research output and perceived levels of centralized and decentralized control of research a c t i v i t i e s , was rejected. S p e c i f i c Hypothesis Number 3C, therefore, was accepted as,plausible. 114 Hypothesis 3D There w i l l be a greater tendency i n business, government, and s o c i a l development organizations, than i n u n i v e r s i t i e s , for high levels bf reported research output to be associated with perceptions of high levels of coordination of e f f o r t s for common objectives. Table XX contains the frequency and percentage frequency tables of responses of high research output s c i e n t i s t s from the four types of organizations, to one question. The question i s : How much emphasis do members of your research unit or department place on the coordination of t h e i r e f f o r t s for some common objectives? Rating Scale 1 2 3 4 5 An examination of the table indicated that 70 percent of the high output s c i e n t i s t s i n business organizations, 74 percent i n government, and 56 percent i n s o c i a l development organizations—asccompared to 14 percent i n u n i v e r s i t i e s -responded that they perceived high levels of coordination of research e f f o r t s for common objectives. Examination of Table XX, therefore, indicated that fewer s c i e n t i s t s from univer-s i t i e s than from the other types of organizations, responded thi s way. Appendix (xx) contains the same elements as Table XX except that research output number one (number of project 115 TABLE XX High Levels of Research Output vs. Levels of Coordination How much emphasis do members of.your research unit or department place on the coordination of t h e i r e f f o r t s for some common objectives? 1 2 3 4 5 Organi zation --— 'Type High Levels of Research Output. (5+) Low and Medium High Levels of Coordination 1,2,3 Coordination 4,5 N Bus . f T 6 30 14 70 20 Gov. f T 9 26 2,5. 7-4 - -34-S.D. f T 22 44 28 56 50 Univ. f T 109 86 18 14 127 146 85 231 X 2 = 65.08(. 001) High Research Output = 5 or more of: Project completion reports; a r t i c l e s accepted by journals; and speeches delivered. Chi-square Values; Separate Samples vs. Separate Samples. Un.vs. Gov. x 2 = •-48- . -27 (..00.1) Un.vs.S.D. x 2 = 32.63 (.001) Un.vs.Bus. x 2 = 31.62 (.001) Gov.vs.S.D. x? = 2.67 (.20) Gov.vs.Bus. x 2 = -07 S.D.vs.Bus. x = .1.16 f = frequency; T = t o t a l percentage f. 116 completion reports etc.) i s exchanged for research output number two (number of books and monographs completed). The results were comparable to those i n Table XX. The chi-square values for the two tables were s i g n i f i c a n t well beyond the .05 l e v e l . Examination of the chi-square values l i s t e d below Table XX and Appendix (xx) also indicated that s c i e n t i s t res-ponses i n each of the government, s o c i a l development and business organization samples d i f f e r e d s i g n i f i c a n t l y from the responses of the university s c i e n t i s t sample. On the basis of t h i s information, the n u l l hypothesis that no differences existed across organization types, i n the association between high levels of reported research output and perceived l e v e l of coordination of research e f f o r t s for common objectives, was rejected. S p e c i f i c Hypothesis Number 3D, therefore, was accepted:as pl a u s i b l e . The analyses of the data d i r e c t l y related to Hypothesis Number Three i s completed. S u f f i c i e n t information i s a v a i l -able, therefore, with which to test this hypothesis. Hypothesis Number Three: The Test The r e l a t i o n s h i p between research department structure and research output varies across organization types. The s p e c i f i c hypotheses related to Hypothesis Number Three (3B, 3C and 3D) were accepted, and r e s u l t s for S p e c i f i c Hypothesis Number 3A were i n the predicted d i r e c t i o n . On the 117 basis of t h i s information, the n u l l hypothesis that no d i f f -erence existed across organization types, i n the r e l a t i o n s h i p between research department structure and research output, was rejected. Hypothesis Number Three, therefore, was accepted,as p l a u s i b l e . Results and Conclusions Section Number Four: Differences i n Relationships Between Combined and Separate Organization Samples 119 S p e c i f i c Hypothesis Number 4A Relationships between levels of reported research out-put and perceived lev e l s of influence to decide work goals and objectives found i n combined organization samples, w i l l d i f f e r s i g n i f i c a n t l y from relationships found i n separate or-ganization samples. Table XXI contains the frequency, horizontal percentage frequency and t o t a l percentage frequency tables of responses of high research output s c i e n t i s t s , from four types of organi-zations, to one question. The question was: How much influence do you f e e l the following people have i n deciding your work goals and objectives? _ . . „ , J Rating Scale Yourself 1 2 3 4 5 Examination of the chi-square values l i s t e d below Table XXI and Appendix (xvii) indicated that university s c i e n t i s t responses d i f f e r e d s i g n i f i c a n t l y from responses of s c i e n t i s t s i n combined organization samples. The responses of s c i e n t i s t s i n other separate organization samples, however, did not d i f f e r s i g n i f i c a n t l y from the responses of s c i e n t i s t s i n combined organization samples. On the basis of t h i s information, the assumed n u l l hypothesis that no differences existed between s c i e n t i s t res-ponses i n combined and separate organization samples, i n the relationships between levels of reported research output and perceived levels of influence to decide work goals and objec-120 TABLE XXI High Levels of Research Output versus Levels of Decision-Making Influence How much influence do you f e e l the following people have inOdeciding your work goals and objectives? Yourself 1 2 3 4 5 Organization Type High Levels of Research Output Low and Medium High Levels Influence 1,2,3 Influence of 4,5 N Bus. f T 5 25 15 75 20 Gov. f T 6 18 28 82 34 S.D. f T 7 14 43 86 50 Univ. f T 10 8 117 92 127 28 203 231 Chi-square Values: Separate vs. Combined Samples. 2 University vs. Other three X = 4.77 (.05) 2 Business vs. Other Three x = 3 . 4 (.10) 2 Government vs. Other Three x = 1 . 1 4 2 Soc i a l Development vs.Other Three x = .'--21 f = frequency; T = t o t a l percentage f. 121 t i v e s , was accepted. S p e c i f i c Hypothesis Number 4A, therefore, was rejected. I t was noted, however, that the frequency d i s -t r i b u t i o n of university s c i e n t i s t responses did not support rej e c t i o n of the hypothesis. S p e c i f i c Hypothesis Number 4B Relationships between levels of reported research output and perceived levels of supervisor or department head influence i n deciding work goals and objectives found i n combined organi-zation samples, w i l l d i f f e r s i g n i f i c a n t l y from relationships found i n separate organization samples. Table XXII contains the frequency, horizontal percentage frequency and t o t a l percentage frequency tables of responses of high research output s c i e n t i s t s , from four types of organi-zations, to one question. The question was: How much influence do you f e e l the following people have i n deciding your work goals and objectives? _ .. „ . . ., J . Rating Scale Immediate supervisor or department head 1 2 3 4 5 Examination of the chi-square values l i s t e d below Table XXII and Appendix ( x v i i i ) indicated that most separate organi-zation sample responses d i f f e r e d s i g n i f i c a n t l y from combined organization sample responses. On the basis of t h i s information, the assumed n u l l hypo-thesis that no differences existed between s c i e n t i s t responses i n combined and separate organization samples, i n the r e l a -122 TABLE XXII High Levels of Research Output versus Levels of Supervisor Decision-Making Influence How much influence do you f e e l the following people have i n deciding your work goals and objectives? Immediate supervisor or department head 1 2 3 4 5 Organization Type High Levels of Low and Medium Supervisor I n f l Research Output (5+) High Supervisor . 1,2,-3. Influence 4,5 N Bus. f T 14 70 6 30 20 Gov. f T 15 44 19 56 34 S.D. f T 39 78 11 22 50 Univ. f T 114 90 13 10 127 182 49 2.3.1. Chi-square Values: Separate Samples vs. Combined Samples University vs. Three Others Combined So c i a l Development vs. Three Others Comb. Government vs. Three Others Combined Business vs. Three Others Combined f = frequency; T = t o t a l percentage f. 2 X = X2 X2 , X2 . 20.33 (.001) .02 28.67(.001) 1.01 123 tionships between levels of reported research output and per-ceived levels of supervisor or department head influence to decide work goals and objectives, was rejected. S p e c i f i c Hypothesis Number 4B, therefore, was accepted.as p l a u s i b l e . S p e c i f i c Hypothesis Number 4C Relationships between levels of reported research out-put and perceived levels of centralized and decentralized con-t r o l of research department a c t i v i t i e s found i n combined organization samples, w i l l d i f f e r s i g n i f i c a n t l y from r e l a t i o n -ships found i n separate organization samples. Table XXIII contains the frequency, horizontal percen-tage frequency and t o t a l percentage frequency tables of res-ponses of high research output s c i e n t i s t s , from four types of organizations, to one question. The question was: Which of the following best describes your research unit or department? Highly centralized control of a c t i v i t i e s 1 Centralized control 2 Neither centralized nor decentralized 3 Decentralized control 4 Highly decentralized control of a c t i v i t i e s 5 Examination of the chi-square values l i s t e d below Table XXIII and Appendix (xix) indicated that most separate organi-zation sample responses d i f f e r e d s i g n i f i c a n t l y from combined organization sample responses. On the basis of t h i s information, the assumed n u l l hypothesis that no differences existed between s c i e n t i s t 124 TABLE XXIII High Levels of Research Output versus Levels of Centralized and Decentralized Control Which of the following best describes your research unit or department? Highly centralized control of a c t i v i t i e s Centralized control Neither centralized nor decentralized Decentralized control Highly decentralized control of a c t i v i t i e s 1 2 3 4 5 Organi zation Type High Levels of Centralized 1,2 Research Neither 3 Output Decentrali zed 4,5 N Bus. f T 6 30 3 15 11 55 20 Gov. f T 13 38 11 32 10 30 34 f 15 20 15 • SO S .D. T 30 40 30 Uni v. f T 15 12 31 24 81 6.4 127 49 65 117 231 Chi-square Values: Separate Samples: vs.' Combined Samples University vs. Three Others Combined x 2 = .22 .74 (.001) 2 - • • S o c i a l Development vs. Three Others Comb, x - 10.88(.01) 2 Government vs. Three Others Combined X = 9.24(.01) Business vs. Three Others Combined x = 2.22 f = frequency; T = t o t a l percentage f. 125 responses i n combined and separate organization samples, i n relationships between levels of reported research output and perceived levels of centralized and decentralized control of research a c t i v i t i e s , was rejected. S p e c i f i c Hypothesis Number 4C, therefore, was accepted ,as p l a u s i b l e . • Sp e c i f i c Hypothesis Number 4D Relationships between levels of reported research output and perceived le v e l s of coordination of research e f f o r t s for common objectives found i n combined organization samples, w i l l d i f f e r s i g n i f i c a n t l y from relationships found i n separate organization samples. Table XXIV contains the frequency, horizontal percen-tage frequency and t o t a l percentage frequency tables of responses of s c i e n t i s t s , from four types of organizations, to one question. The question was: How much emphasis do members of your research unit or department place on the coordination of t h e i r e f f o r t s f or some common objectives? Rating Scale 1 2 3 4 5 Examination of the chi-square values l i s t e d below Table XXIV and Appendix (xx) indicated that a l l separate organization sample responses d i f f e r e d s i g n i f i c a n t l y from the combined organization sample responses. On the basis of thi s information, the assumed n u l l hypothesis that no differences existed between s c i e n t i s t res-TABLE XXIV 126 High Levels of Research Output versus Levels of Coordination How much emphasis do members of your research unit or department place on the coordination of t h e i r e f f o r t s for some common objectives? 1 2 3 4 5 Organi zation Type High Levels of Low and Medium Coordination 1,2, Research Output(5+) High Levels of 3 Coordination 4,5 N Bus. f T 6 30 14 70 20 Gov. f T 9 26 25 74 34 S.D. f T 22 44 28 56 50 Univ. f T 109 86 18 14 127 146 85 231 Chi-square Values: Separate vs. Combined Samples University vs. Three Others Combined x 2 = 62.07(.001) Soci a l Development vs. Three Others Combined x2= 10.11 (.01) 2 Government vs. Three Others Combined x =23.13(.001) Business vs. Three Others Combined x = 10.37 (.01) f .= frequency; T = t o t a l percentage f. 127 ponses i n combined and separate organization samples, i n the relationships between levels of reported research output and perceived levels of coordination of research e f f o r t s f or common objectives, was rejected. S p e c i f i c Hypothesis Number 4D, therefore, was accepted1.as p l a u s i b l e . The analysis of s p e c i f i c hypotheses relevant to Hypothesis Number Four i s completed. S u f f i c i e n t information i s a vailable, therefore, with which to tes t t h i s hypothesis. Hypothesis Number Four: The Test Relationships between research output and structure found i n combined organization samples w i l l be d i f f e r e n t than relationships found i n separate organization samples. The s p e c i f i c hypotheses related to Hypothesis Number Four (Specific Hypotheses 4B, 4C, and 4D) were tested and accepted. S p e c i f i c Hypothesis Number 4A was rejected, but results appeared to be i n the predicted d i r e c t i o n . On the basis of thi s information, the assumed n u l l hypothesis that no differences existed between combined and separate organi-zation samples, i n the relationships between research output and structure, was rejected. Hypothesis Number Four, there-fore, was accepted,as p l a u s i b l e . ' 128 SUMMARY Previous research reveals a relat i o n s h i p between re-search department structure and s c i e n t i s t research output. Investigators have drawn on the findings of t h i s research to make recommendations to research directors and administrators regarding the type of structure necessary to maximize scien-t i s t research output. Since the recommendations were made to research directors and administrators i n general, the implication i s that one type of research department structure should be u t i l i z e d i n a l l types of organizations. Consideration of the goals and operating conditions i n d i f f e r e n t types of organizations suggests that some organi-zations would tend to place greater s t r u c t u r a l constraints on s c i e n t i s t s than others. In other words, differences i n goals and conditions of operation make i t almost impossible for research departments i n d i f f e r e n t types of organizations to be structured the same way. If thi s i s the case, the hypo-thesis that follows i n e v i t a b l y i s that some types of organi-zations cannot have the one best structure, and must suffer losses i n research output. An examination of the o r i g i n a l research, however, reveals that the responding s c i e n t i s t s were employed i n a variety of organization types. Moreover, the investigators made no attempt to examine the relationships between structure and research output on an organization type-by-type basis. 129 There remained then an equally plausible hypothesis that s c i e n t i s t s i n d i f f e r e n t types of organizations accept the e x i s t i n g structure and that no basic incompatibility e x i s t s . This implies that the relationship between structure and research output i s not constant across organization types, but varies from type to type. The problem of t h i s study, therefore, was to determine whether or not the r e l a t i o n s h i p between research department structure and research output was constant across organization types. The main hypothesis tested was: The relationship between research department structure and research output varies across organization types. Implicit i n t h i s hypothesis were two p r i o r hypotheses. 1. There i s a r e l a t i o n s h i p between research department structure and research output. 2. Research departments i n d i f f e r e n t types of parent organizations are structured d i f f e r e n t l y . Also i m p l i c i t i n the main hypothesis was a type of summary hypothesis, which properly followed the main hypothesis: 4. Relationships between research output and structure found i n combined organization samples are d i f f e r e n t than relationships found i n separate organization samples. These general hypotheses were tested by examining information obtained from te s t i n g related s p e c i f i c hypotheses. The data necessary for the t e s t i n g of the s p e c i f i c hypotheses was obtained from s c i e n t i s t questionnaire responses. 130 S c i e n t i s t s from four types of organizations—business, government, s o c i a l development, and university—were mailed questionnaires i n order to obtain measures of reported re-search output, and perceptions of research department struc-ture. 523 s c i e n t i s t s , or 45% of the sample, returned a completed questionnaire. Another 15% of the sample provided reasons for not responding. Examination of the data relevant to Hypothesis Number One indicated that, i n a combined organization sample: 1. leve l s of reported research output was p o s i t i v e l y associated with levels of perceived influence to decide work goals and objectives; 2. levels of reported research output was negatively associated with levels of perceived supervisor or department head influence i n deciding s c i e n t i s t work goals and objec-tives ; 3. levels of reported research output was p o s i t i v e l y associated with levels of perceived decentralized control of research a c t i v i t i e s , and negatively associated with per-ceived levels of centralized control. On the basis of t h i s information, Hypothesis Number One was accepted. Examination of the data related to Hypothesis Number Two indicated that s c i e n t i s t s i n d i f f e r e n t types, of organi-zations : 131 1. perceived d i f f e r e n t levels of emphasis to be placed on p a r t i c u l a r c r i t e r i a used i n the selection of research projects. S p e c i f i c a l l y , the project selection c r i t e r i a were: (a) projects that contribute to the organization's f i n a n c i a l well-being; (b) projects that are delegated by some higher authority; (c) projects that may contribute to the well-being of some s o c i a l group; (d) projects that are in t e r e s t i n g i n themselves; and (e) projects that seek new knowledge—because the search for new knowledge i s i n t r i n s i c a l l y good. 2. perceived d i f f e r e n t levels of time expenditures i n basic and applied research a c t i v i t i e s ; 3. perceived d i f f e r e n t l e v e l s of time pressure on th e i r work; 4. perceived d i f f e r e n t levels of influence to decide work goals and objectives; 5. perceived immediate supervisors and department heads to have d i f f e r e n t levels of influence i n deciding t h e i r work goals and objectives; 6. perceived d i f f e r e n t levels of centralized and de-centralized control of research a c t i v i t i e s ; and o 7. perceived d i f f e r e n t levels of coordination of ef f o r t s f or common objectives. 132 On the basis of thi s information, Hypothesis Number Two was accepted. Examination of the data related to Hypothesis Number Three indicated that high research output s c i e n t i s t s , i n d i f -ferent types of organizations: 1. perceived d i f f e r e n t levels of influence i n deciding t h e i r work goals and objectives (university s c i e n t i s t s only); 2. perceived immediate supervisors or department heads to have d i f f e r e n t levels of influence i n deciding t h e i r work goals and objectives; 3. perceived d i f f e r e n t levels of centralized and de-centralized control of research a c t i v i t i e s ; and 4. perceived d i f f e r e n t levels of coordination of ef f o r t s for common objectives. On the basis of thi s information, Hypothesis Number Three was accepted. Examination of the data relevant to Hypothesis Number Four indicated that high research output s c i e n t i s t s i n combined organization samples—as compared to separate organization samples: 1. perceived d i f f e r e n t levels of influence i n deciding t h e i r work goals and objectives (university s c i e n t i s t s only); 2. perceived immediate supervisors or department heads to have d i f f e r e n t levels of influence i n deciding t h e i r work goals and objectives; 133 3. perceived d i f f e r e n t levels of centralized and de-centralized control of research a c t i v i t i e s ; and 4. perceived d i f f e r e n t levels of coordination of e f f o r t s for common objectives. On the basis of t h i s information, Hypothesis Number Four was accepted. In summary, thi s study found relationships between re-search output and research department structure. Research departments i n d i f f e r e n t types of parent organizations appeared to be structured d i f f e r e n t l y . F i n a l l y , relationships between structure and research output varied across organization types, as well as between separate and combined organization samples. Limitations of the Study There were some limi t a t i o n s to the study which should be discussed before the conclusions and implications a r i s i n g out of the study are considered. 1. Questionnaire R e l i a b i l i t y . The questionnaire used i n t h i s study, and the r e l i a b i l i t y and v a l i d i t y of the measurer-ments obtained may tend to l i m i t the strength of the conclusions one can.reach. The questions asked were necessarily general so that they would apply to s c i e n t i s t s i n .four types of organi-zations. I t may"have been possible, therefore, for s c i e n t i s t s in. d i f f e r e n t types of organizations, because of organizational differences, to place d i f f e r e n t interpretations on these ques-tions. If t h i s was the case, the questionnaire could appear to have v a l i d i t y and r e l i a b i l i t y a n d . s t i l l not present the 134 true picture for comparisons across organization types. The questionnaire used was the same as one designed by Pelz and Andrews, except for minor modifications and additions. Pelz and Andrews had checked t h e i r questionnaire for r e l i a -b i l i t y by. r.e.administering. i t to 52 s c i e n t i s t s after two months. Mean response scores were correlated .97. That i s , the mean score of the 52 s c i e n t i s t s ' on each questionnaire item found i n the f i r s t administration of the questionnaire was compared to the group mean score on the second administration. The corre-l a t i o n between the two sets of mean scores was .97. When i n -div i d u a l correlations were computed,'83 per cent were .5 or 72 better. A test to determine the r e l i a b i l i t y of the questionnaire items used i n the present study", however, was not carried out. That i s , since there, were only minor modifications and addi-tions made to the Pelz and Andrews questionnaire items, i t was assumed that the instrument would.have a sim i l a r l e v e l of r e l i a b i l i t y . This was possibly a risky assumption to make, as there were not only modifications and additions to the questionnaire items, but the questionnaire was administered to s c i e n t i s t s i n another country-~Canada instead of the United States. 2. : V a l i d i t y of. Research Output Measurements. The measurement of research output, has t y p i c a l l y been a contror v e r s i a l ' task. The controversy has revolved around the r e l a t i v e merits of u t i l i z i n g a quantitative measure of research output, 72 See Appendix (v). 135 such as the number of research papers that a s c i e n t i s t reports he has completed—and a q u a l i t a t i v e measure of research output, such as the quality of a research paper as determined by judges. Both methods were examined by Pelz and Andrews, but they con-cluded that quantitative measures were highly correlated with 73 q u a l i t a t i v e measures and much easier to obtain. The.impli-cations of t h e i r results were that quantitative measures of research output were v a l i d measures of research q u a l i t y . Pelz and Andrews did a" further v a l i d i t y check on t h e i r research output measures. They examined laboratory records to determine whether or not s c i e n t i s t s reported t h e i r research output accurately. High correlations between reported research output and lab records indicated that s c i e n t i s t s reported 74 research output with high and consistent accuracy. In u t i l i z i n g quantitative measures of research output for the present study without checking the v a l i d i t y of these measures, i t was again assumed that such measures would also apply for Canadian s c i e n t i s t s . Although the NRC did support the use of such measures for research output i n a b r i e f to the Special Senate Committee on Science Policy, i t i s an 75 assumption that should be examined i n l a t e r research. 3... . Questionnaire Return Rate. Forty-five per cent of the sample, returned a completed questionnaire. F i f t e e n per Pelz and Andrews, op. c i t . , pp. 261-270. See Appendix (v). See Appendix (v). 136 cent of the sample gave reasons for not completing the ques-tionnaires, and these reasons indicated that they should not 76 have been included i n the sample i n the f i r s t place. The remaining forty per cent of the sample did not respond i n any way, so there was no way of t e l l i n g what biases might have been introduced into the data as a r e s u l t of the absence of t h e i r responses. 4. Control of Background Variables. Although there were s i g n i f i c a n t and p o s i t i v e relationships between research output and such variables as age, education, previous research experience, and p o s i t i o n , these variables were not controlled 77 i n this study. Control of these variables i n t h i s study would have reduced the table frequencies to a point where the analysis of data was almost impossible. Other variables such as the incentive systems used to motivate s c i e n t i s t s , and t h e i r risk-taking behaviour, were not examined i n relationship with research output. Since *. these and other variables, besides structure, may be related to research output, more studies and more sophisticated measurement and s t a t i s t i c a l procedures w i l l be required to determine the relationships and i n t e r - r e l a t i o n s h i p s of these and other variables with research output. 5. Questionnaire Development. Except for the addition of a few questions, the questionnaire used i n t h i s study was See Appendix ( i i i ) . 77 A separate analysis of the data revealed these r e l a -tionships . 137 made up of the same questions used i n the Pelz and Andrews questionnaire. Their questions appeared to be appropriate for this study and were used with only minor modifications. It i s quite possible, however, that other questionnaire items might have been more appropriate for t h i s study. Pelz and Andrews were not concerned about differences across organi-zational types i n the relationships between structure and output. More appropriate questionnaire items might have been discovered i f more time had been spent interviewing s c i e n t i s t s from each type of organization. The use of a questionnaire instead of other data gath-ering techniques can be questioned. An interview guide and s c i e n t i s t observation would l i k e l y have provided more v a l i d data than the questionnaire. The questionnaire, however, provided responses from s c i e n t i s t s that would not otherwise have been reached—given the resource l i m i t a t i o n s of the study. Summary of the Results The evidence from the, present study tends to support previous research evidence i n that the data revealed a r e l a -tionship between research department structure and research o u t p u t — f o r a combined organization sample. That is> both previous research evidence and the findings i n the present study indicated that when s c i e n t i s t s from several d i f f e r e n t types of organizations were combined into one sample, scien-t i s t s who reported high levels of research output also perceived themselves to work under lower levels of decision-138 making s t r u c t u r a l constraints than low output s c i e n t i s t s . In the present study, for example, there was a tendency for scien-t i s t s who perceived themselves to have high levels of influence to decide t h e i r work goals and objectives to have higher levels of research output than s c i e n t i s t s who perceived them-7 8 selves to have low levels of influence. As one might expect from the above r e s u l t s , there was also a tendency for s c i e n t i s t s who perceived research super-visors to have high levels of influence to decide t h e i r work goals and objectives to have, lower levels of research output than s c i e n t i s t s who perceived supervisors to have low levels 79 of influence. Again m the present study, there was a ten-dency for s c i e n t i s t s who perceived themselves to work i n a research department with a decentralized control of research a c t i v i t i e s to have higher levels of research output than s c i e n t i s t s who perceived themselves to work under centralized . , 80 control. Previous investigators drew on findings such as these from combined organization samples to make recommendations to research directors i n general regarding the type of research department structure to u t i l i z e i n order to obtain high levels of research output. . In other words, previous investigators suggested that a l l s c i e n t i s t s , regardless of the organization type i n which they worked, must have such things as autonomy 7 8See Table I 7 9See T a b l e l l l 8 0See Table III 139 and decentralized decision-making i n t h e i r research departments i n order for them to at t a i n high levels of research output. Put another way, previous investigators indicated that there was one p a r t i c u l a r type of research department s t r u c t u r e — i n which there was a low l e v e l of decision-making s t r u c t u r a l c o n s t r a i n t s — t h a t was appropriate for s c i e n t i s t s i n research departments i n a l l types of organizations. Further analysis of the data i n the present study, however, indicated that the relationships between structure and research output found i n combined organization samples d i f f e r e d from the relationships found i n separate organization samples. For example, a greater proportion of high output university s cientists—when compared to a combined sample of high output business, government and s o c i a l development organization s c i e n t i s t s — p e r c e i v e d themselves to have high g levels of influence to decide t h e i r work goals and objectives. A smaller proportion of high output univ e r s i t y students—when See Table XXI., also Table XXII. A much smaller pro-portion of high output university scientists—when compared to a combined sample of high output business, government and s o c i a l development s c i e n t i s t s — p e r c e i v e d research supervisors to have high levels of influence to decide t h e i r work goals and objectives. Similar results were found when perceptions of centralized control of research a c t i v i t i e s and co-ordination of research a c t i v i t i e s were examined. 140 compared to a combined sample of high output business, govern-ment and s o c i a l development s c i e n t i s t s — p e r c e i v e d ce n t r a l i z e d control of research a c t i v i t i e s (and vice versa regarding per-8 2 ceptions of decentralized control of research a c t i v i t i e s ) . Data i n the present study also suggested that research departments i n d i f f e r e n t types of organizations have d i f f e r e n t decision-making structures. For example, a higher proportion of university scientists—when compared to s c i e n t i s t s i n the other three types of organizations—perceived themselves to have high levels of influence to decide t h e i r own work goals and objectives. In the. three other types of organizations, more s o c i a l development organization s c i e n t i s t s than business, and more business than government s c i e n t i s t s perceived high 8 3 levels of influence to decide work goals and objectives. See Table XXIII, also Table XXIV. Likewise, a smaller proportion of high output university scientists--when compared to a combined sample of high output s c i e n t i s t s from the other three types of organizations—perceived high levels of co-ordination of research e f f o r t s for some common objectives (and vice versa regarding perceptions of low levels of co-ordination). 8 3 See Table XIIT, also Table XIV. As one might expect from the above r e s u l t s , a lower proportion of university s cientists—when compared to s c i e n t i s t s in.the other three types of organizations—perceived research supervisors or department heads to have high levels of influence to decide t h e i r work goals and objectives. In the three other types of organizations, fewer s o c i a l development organization s c i e n t i s t s than business, and fewer business than government s c i e n t i s t s perceived research supervisors or department heads to have . high levels of influence to decide t h e i r work goals and objec-tives . 141 As regards perceptions of decentralized and centralized control of research a c t i v i t i e s , more university s c i e n t i s t s and fewer government s c i e n t i s t s perceived decentralized con-t r o l , while the perceptions of s c i e n t i s t s i n the other two types of organizations f e l l between. The data were reversed g for perceptions of centralized control of research a c t i v i t i e s ' . The data from the present study also indicated that relationships between structure and research output varied across organization types. That i s , s c i e n t i s t s i n d i f f e r e n t types of organizations attained high levels of research output under d i f f e r e n t decision-making s t r u c t u r a l conditions. For example, a high number of high output government s c i e n t i s t s and a low number of high output university s c i e n t i s t s per-ceived research supervisors to have high levels of influence to decide t h e i r work goals and objectives. That i s , only a few high output s c i e n t i s t s i n the university research departments perceived supervisors to have high levels of influence to decide t h e i r work goals and objectives, whereas the opposite was apparent i n government research departments. The percep-tions of s c i e n t i s t s i n research departments i n the other two types of organizations f e l l between these two extreme posi-85 tions. 84 See Table XV, also Table .XVI. S t i l l considering ex^ amples of s t r u c t u r a l differences across organization types, the present study also provided data to suggest differences i n co-ordination. Fewer university s c i e n t i s t s and more business s c i e n t i s t s perceived high levels of co-ordination of research e f f o r t s .for common objectives. The perceptions of the scien-t i s t s i n the other two types of organizations f e l l between. The data were reversed for perceptions of low co-ordination. . . 142 These same relationships were also apparent with regard to the centralized control of research a c t i v i t i e s . Only a few of the high output univ e r s i t y s c i e n t i s t s perceived a cen-t r a l i z e d control of research a c t i v i t i e s , whereas a high pro-portion of the high output s c i e n t i s t s from government organi-zations perceived a centralized control of research department a c t i v i t i e s (vice versa regarding decentralized c o n t r o l ) . The perceptions of s c i e n t i s t s i n the other two.types of organi-86 zations f e l l between these two extreme positions. Conclusions Analysis of the data i n the present study, beyond that carried out i n previous research, did not appear to support previous research findings. That i s , the present study con-tradicted evidence that there i s one p a r t i c u l a r type of research department structure-—in which there are low decision-making s t r u c t u r a l c o n s t r a i n t s — t h a t . i s associated with high s c i e n t i s t research output i n a l l types of organizations. The present study indicated that high output s c i e n t i s t s i n re-search "departments i n d i f f e r e n t types of organizations had perceptions of working under d i f f e r e n t levels of decision-8 5 See Table XVIII, also Table XX. Similar r e s u l t s were apparent with regard to the l e v e l of co-ordination of research e f f o r t s of high output s c i e n t i s t s i n d i f f e r e n t types of organi-zations.. Only a few of the high output univ e r s i t y s c i e n t i s t s perceived high levels of co-ordination, whereas a high propor-t i o n of the high output s c i e n t i s t s from the other three types of organizations'perceived high levels of co-ordination. 8 6See Table XXI. making s t r u c t u r a l constraints. For example, s c i e n t i s t s i n one type of organization appeared to achieve high levels of research output while working under low levels of s t r u c t u r a l constraints, whereas s c i e n t i s t s i n another type of organiza-ti o n achieved high levels of research output while working under high levels of s t r u c t u r a l constraints. In conclusion, the present study indicated that there i s no 'best' type of research department structure for a l l organizations. Implications of the Results and Conclusions for  Research Directors There appear to be some obvious implications for re-search directors that are based on the results and conclusions of t h i s study. The f i r s t one i s that a research dire c t o r should be wary of recommendations to u t i l i z e a p a r t i c u l a r type of research department decision-making structure. This would appear to be a necessary precaution, e s p e c i a l l y i f the recommendations are based on data: (a) obtained from s c i e n t i s t s i n a d i f f e r e n t type of organization from the one i n which he i s dir e c t o r ; or (b) obtained from s c i e n t i s t s from several types of organizations combined into one sample. In other words, a research director should probably give more weight to recommendations based on data obtained from scien-t i s t s i n organizations which are of the same type as the one 144 i n which he i s di r e c t o r . That i s , although one cannot i n f e r causal relationships, the data from the present study strongly suggest that research directors i n d i f f e r e n t types of organi-zations may fin d that d i f f e r e n t decision-making s t r u c t u r a l constraints are associated with high s c i e n t i s t research out-put. For example, research directors i n univer s i t y and s o c i a l development organizations are l i k e l y to f i n d that when they provide s c i e n t i s t s with autonomy, decentralized decision-making and a low l e v e l of' co-ordination of research e f f o r t s , the productivity of t h e i r s c i e n t i s t s w i l l tend to be high. Conversely, research directors i n business and government or-ganizations are l i k e l y to fi n d that when they influence t h e i r s c i e n t i s t s ' decision-making, have centralized control over research a c t i v i t i e s and c a r e f u l l y co-ordinate s c i e n t i s t e f f o r t s toward some common organizational objective, t h e i r s c i e n t i s t s w i l l tend to have high levels of research output. Research d i r e c t o r s , to date, have been reluctant to deviate too far from the s t r u c t u r a l model of the university research department—probably because of previous research evidence and opinions of vociferous advocates of research s c i e n t i s t autonomy. The results of the present study indicate that s c i e n t i s t s can have high levels of research output while working under d i f f e r e n t levels of decision-making s t r u c t u r a l constraints. Research directors may "find, therefore, that a structure that i s best for t h e i r organization i s d i f f e r e n t 145 from what i s best for another type of organization. If re-search directors f i n d that t h i s i s the case, they may become more innovative regarding the type of research department structures they use, or more w i l l i n g to a l t e r e x i s t i n g struc-tures . The data i n the present study indicated that there were marked s t r u c t u r a l differences i n research departments i n d i f f e r e n t types of organizations. In spite of these d i f -ferences, most s c i e n t i s t s s t i l l perceived themselves to have r e l a t i v e l y high levels of autonomy to decide t h e i r own work goals and objectives. For example, a high proportion of business and government s c i e n t i s t s had perceptions of working i n research departments with (a) centralized control of research a c t i v i t i e s ; (b) high levels of co-ordination of e f f o r t for common objectives; as well as (c) research supervisors with high levels of influence to decide t h e i r work goals and objec-t i v e s . In spite of the high proportion of these perceptions, a r e l a t i v e l y . h i g h proportion of business and government s c i e n t i s t s s t i l l had perceptions of autonomy to decide t h e i r own 8 7 work goals and objectives. 8 7 A s i g n i f i c a n t l y lower proportion -of business and government s c i e n t i s t s — t h a n the university and s o c i a l develop-ment s c i e n t i s t s — h a d perceptions of autonomy to decide t h e i r own work goals and objectives. 146 I t would appear from t h i s evidence t h a t s c i e n t i s t s perceived no marked decision-making s t r u c t u r a l i n t e r f e r e n c e i n t h e i r d e c i s i o n s about work goals and o b j e c t i v e s . This suggests t h a t s c i e n t i s t s are e i t h e r extremely v e r s a t i l e and able to work under almost any s t r u c t u r a l c o n s t r a i n t s without l o s s of perceived autonomy, or th a t s c i e n t i s t s seek out and are r e c r u i t e d by research d i r e c t o r s who d i r e c t research departments wi t h structures, and work goals and o b j e c t i v e s compatible to those p r e f e r r e d by the s c i e n t i s t . The l a t t e r p r o p o s i t i o n seems more reasonable s i n c e people tend to vary i n t h e i r preferences and opinions about the best way to do t h i n g s . I f t h i s i s the case,. research d i r e c t o r s may f i n d t h a t a p a r t i c u l a r type of s t r u c t u r e i s more appropriate not only because of the needs of the p a r t i c u l a r o r g a n i z a t i o n , but be-cause some s c i e n t i s t s s e l f - s e l e c t a p a r t i c u l a r o r g a n i z a t i o n on the b a s i s t h a t the working c o n d i t i o n s are compatible to t h e i r p r e f e r r e d c o n d i t i o n s of work. I m p l i c a t i o n s f o r Further Research The qu e s t i o n n a i r e items used i n t h i s study were n e c e s s a r i l y g e n e r a l : so'that'they would apply to s c i e n t i s t s i n d i f f e r e n t types of parent o r g a n i z a t i o n s . Refinement of the items could probably be achieved i f research departments i n only one o r g a n i z a t i o n type were st u d i e d i n depth. In depth studi e s may a l s o r e v e a l other measures of s t r u c t u r e and 147 research output peculiar to each organization type. To a s s i s t i n the development of more s p e c i f i c studies, an organization c l a s s i f i c a t i o n procedure may be h e l p f u l . The results and conclusions of t h i s study provided some support for the procedure of c l a s s i f y i n g organizations and developing a t h e o r e t i c a l framework from which to approach the study of departments or other units of analysis, within organizations. That i s a c l a s s i f i c a t i o n procedure could also be applied to a p a r t i c u l a r organization type, to a p a r t i c u l a r organization, to a p a r t i c u l a r research department, or to a p a r t i c u l a r scien-t i s t , as a way of developing a t h e o r e t i c a l framework from which to launch a s t u d y . ^ H i l l s , op. c i t . , p. 120. "There i s considerable number of r e l a t i v e l y s p e c i f i c hypotheses that might be advanced concerning differences i n values, norms, goals, evaluative standards and standards of successful performance character-i z i n g each of the "differentiated. subsystems within organiza-tions, but to l i s t them here would be highly r e p e t i t i o u s . Perhaps i t w i l l be s u f f i c i e n t to point out that, according to the Parsoniah. model, we should f i n d i t possible to d i f f e r -entiate not just an administrative and a technical orientation, but four distinguishable orientations. That i s to say, we should f i n d i t possible to d i s t i n g u i s h , on any given l e v e l of  analysis, four d i f f e r e n t ways of seeing things, arid the corres-ponding categories of expectations, and communication." B I B L I O G R A P H Y 148 BOOKS Alexander, F. "Observations on Organizational Factors Affec-ti n g C r e a t i v i t y . " The Creative Organization, ed. G.A. Steiner, University of Chicago Press, 1965. Babbitt, J.D. (ed.). Science i n Canada. University of Toronto Press, 1965. Berelson, B. " C r e a t i v i t y and the Graduate School." The Creative Organization, ed. G.A. Steiner, University of Chicago Press, 1965. Blau, Peter M. and Richard W. Scott. Formal Organizations. San Francisco: Chandler Publishing Co., 1962". Bloom, B.S. "Report on C r e a t i v i t y Research." Research Con-ference on the I d e n t i f i c a t i o n of Creative S c i e n t i f i c  Talent, ed. C.W. Taylor, University of Utah Press, 1956. Chorness, M.H. "An Interim Report on C r e a t i v i t y Research." Research Conference on the I d e n t i f i c a t i o n of Creative  S c i e n t i f i c Talent, ed. C.W. Taylor, University of Utah Press, 1956. E t z i o n i , Amitai. A Comparative Analysis of Complex Organiza-tions . New York: The Free Press of" Glencoe Ltd., 1961. E t z i o n i , Amitai. Modern Organizations. New Jersey: Prentice-H a l l Inc., 1964. Ferguson, G.E. S t a t i s t i c a l Analysis i n Psychology and Education. New York: McGraw-Hill Book Co. Inc., 1959. Glaser, B.G. Organizational S c i e n t i s t s : Their Professional  Careers" Indianapolis: The Bobbs-Merrill Co. Inc., 1964. Gustzkow, H. "The Creative Person i n Organizations. The Creative Organization, ed. G.A. Steiner, University of Chicago Press, 1965. Harmon, L.R. "Social and Technical Determiners of C r e a t i v i t y . " Research Conference on the I d e n t i f i c a t i o n of Creative  S c i e n t i f i c Talent, ed. C.W. Taylor, University of Utah Press, 1956. 149 H i l l s , R. Jean. Toward a Science of Organization. Eugene: University of Oregon Press, 196 8. Knapp, R.H. "Demographic, C u l t u r a l and Personality Attributes : of S c i e n t i s t s . " Research Conference on the I d e n t i f i -cation of Creative S c i e n t i f i c Talent, ed. CW. Taylor, University of Utah Press, 1956. Kerlinger, F.N. Foundations of Behavioral Research. New York: Holt, Rinehart and Winston Inc., 1967. Lawrence, P.R. and Lorsch, J.W. Organizations and Environment: Managing D i f f e r e n t i a t i o n and Integration. Boston: Graduate School of Business Admi n i s t r a t ion," "l'9 6'7'V' Learned, E.P. and Sproat, A.J. Organization Theory and P o l i c y . I l l i n o i s : Richard D. Irwin Inc. , 1966\ ' ~\ L i k e r t , Rensis. The Human Organization. New York: McGraw-H i l l Book Co., 1967. McClelland, D.C. "The Calculated Risk: An"Aspect of Scien-t i s t Performance." Research Conference on the Identi-f i c a t i o n of Creative S c i e n t i f i c Talent, ed. CW. Taylor, University of Utah Press, 1956." ' McGraw-Hill Publishing Co. Directory and Almanac of Canada. 1969 . McPherson, J.H. "A Proposal for Establishing Ultimate C r i t e r i a for Measuring Creative Output." Research Conference on the I d e n t i f i c a t i o n of Creative S c i e n t i f i c Talent, ed. CW. Taylor, University of Utah Press, 1956. March, J.G. and Simon, H.A. Organizations. New York: John Wiley and Sons, 1958. Mouly, G.J. The Science of Educational Research. New York: American Book Co., 1963. Pelz, D.C. "Relationships between Measures of S c i e n t i s t Performance and Other Variables." Research Conference  on the I d e n t i f i c a t i o n of Creative S c i e n t i f i c Talent, ed. CW. Taylor, University of Utah Press, 1956. Pelz, D.C. and Andrews, F.M. S c i e n t i s t s i n Organizations. New York: John Wiley and Sons, 1966". 150 P e t r u l l o , L. and B.M. Bass (eds.). Leadership and Inter-personal Behavior. New York: Holt, Rinehart and Winston Inc., 1961. Rubenstein, A.H. and Haberstroh, C.L. Some Theories of Organisation. I l l i n o i s : R.D. Irwin and Dorsey -Press, 1966. Schein, E.H. Organizational Psychology. New Jersey: Prentice-Hall Inc., 1965. ~w"**'*"*""'>;""; Siegel, Sidney.. Nonparametric S t a t i s t i c s . New York: McGraw-H i l l Book' Co. Inc., 1956. Smith, A l f r e d G. Communication and Status: The Dynamics of a Research Center. Eugene: University of Oregon Press, 1966. Stein, Morris I. "A Transactional Approach to C r e a t i v i t y . " Research Conference on the I d e n t i f i c a t i o n of Creative  S c i e n t i f i c Talent, ed. C.W. Taylor, UniversTty of Utah Press, 1956. Steiner, G.A.(ed.). The Creative Organization. University of Chicago Press , 1963 . Tannenbaum, A.S. S o c i a l Psychology of the Work Organization. Belmont, C a l i f . : Wadsworth Publishing Co. Inc., and Tavistock Publications Ltd., 1967. Taylor, C.W. C r e a t i v i t y : Progress and P o t e n t i a l . New York:'5 McGraw-Hill Book Co., 1964. Taylor, C.W. (ed.). Research Conference on the I d e n t i f i c a t i o n  of Creative S c i e n t i f i c Talent. The University of Utah Press, 1956. Torrance, E.P. Guiding Creative Talent. New Jersey: Prentice-H a l l Inc., 1962. Whyte, W.H. J r . The Organization Man. New York: Doubleday Book Co. Inc., 1957. Woodward, Joan. Management and Technology. London: Her Majesty's Stationary O f f i c e , 1958. 151 ARTICLES IN PERIODICALS Alderfer, C P . "An Organizational Syndrome." Administrative  Science Quarterly, December 1967, pp. 440-460. Barnes, Carl E. " i n d u s t r i a l Research: Is i t Outmoded?" Business Horizons, v o l . 7, no. 1 (Spring, 1964), p. 90. Baumgartel, Howard. "Leadership, Motivations and Attitudes i n Research Laboratories." Journal of So c i a l Issues, v o l . 12, no. 2 (1956), pp. 24-31. Baumgartel, Howard. "Leadership Style as a Variable i n Research Administration." Administrative Science  Quarterly, December, 1957, pp. 34-60. Blankenship, L.V. and R.E. Miles. "Organizational Structure and Managerial Decision Behavior." Administrative  Science Quarterly, June 1968, p. 107-120. Bush, C P . and L.H. Hattery. "Teamwork and C r e a t i v i t y i n Research." Administrative.Science Quarterly, v o l . 1 1956 , pp. 361-372. " "" "' ' Cooper, A.C. "R and D i s More E f f i c i e n t i n Small Companies." Harvard Business Review, May-June, 1964, pp. 75-83. Crane, Diana. "Sc i e n t i s t s at Major and Minor U n i v e r s i t i e s : A Study of Productivity and Recognition." American  So c i o l o g i c a l Review, v o l . 30 (1965), pp. 699-714. Dohrenwend, Barbara and Bruce P. "Sources'of Refusals i n Surveys." Public Opinion Quarterly, Spring 1968, pp. 81-83. Eaton, J.W. "Social Processes of Professional Teamwork." American S o c i o l o g i c a l Review, v o l . 16 (1951), pp. 707-713. E t z i o n i , Amitai. "Authority Structure and Organizational Effectiveness." Administrative Science Quarterly, v o l . 4 (June 1959), pp. 43-67. E t z i o n i , Amitai. "Two Approaches to Organizational Analysis: A Critique and a Suggestion." Administratiye Science  Quarterly, v o l . 5 (1960/61), p. 262. 152 Evan, W.M. "Superior-Subordinate C o n f l i c t i n Research Organizations." Admini s t r a t i v e Science Quarterly, June 1965, pp. 52-64. " " Feldvebel, A.M. "Organizational Climate, S o c i a l Class and Educational Output." Administrator's Notebook, no. 8 (April 1964) , pp. 22-36. " '""""*" Friedlander, Frank. "Positive and Negative Motivations Toward Work." Administrative Science Quarterly, v o l . 9 (1964), pp. 194-207. ~ " . ~ v Glaser, B.G. "Attraction, Autonomy and Reciprocity i n the Scientist-Supervisor Relationship." Admi n i s t r ative  Science Quarterly, v o l . 8, 1963, pp. 379-398. Hagstrom, W.O. "Traditional and Modern Forms of S c i e n t i f i c Teamwork." Administrat1ve Science Quarterly, v o l . 9, December 196 4, p. 2 41. ~ H a l l , Richard H. and others. "An Examination of the E-lau-Scott and E t z i o n i Typologies." Administrative Science  Quarterly, June 1967, pp. 119-139. H a l l , R.H. and Thompson, J.D. "What Do You Mean, Business i s a Profession?" Business Horizons, v o l . 7, no. 1 (Spring 1964), p. 39. Hodge, M.H. J r . "Rate Your Company's Research Productivity." Harvard Business Review, November-December, 1963, pp. 109-122. Indik, B.P. "The Relationship Between Organization Size and Supervisor Ratio." Administrative Science Quarterly, v o l . 9, 196 4, pp. 301-312. Kaczka, E.E. and R.V. Kirk. "Managerial Climate, Work Groups and Organizational Performance." Admi n i s t r at ive  Science Quarterly, September 1967, pp. 253-272. Kaplan, Norman. "The Role of the Research Administrator." Admini s t r a t i v e Science Quarterly, v o l . 4 (June 1959), pp. 20-42. La Porte, Todd R. "Conditions of Strain and Accommodation i n I n d u s t r i a l Research Organizations." Administrative  Science Quarterly, v o l . 10 (June 1965), p. 21. 153 Levinson, Harry. "Reciprocation: The Relationship Between Man and Organizations." Administrative Science  Quarterly, v o l . 9 (1964), pp. 370-390. " Meltzer, Leo. " S c i e n t i f i c Productivity i n Organizational Settings." Journal of S o c i a l Issues, no. 2 (1956), pp. 32-40. Myers, M. Scott. "Who are Your Motivated Workers?" Harvard  Business Review, v o l . 42 (January-February 1964), pp. 75-78. Pelz, D.C. "Some Soc i a l Factors Related to Performance i n a Research Organization." Admi n i s t r a t i v e Science  Quarterly, v o l . 1 (1956), pp. 310-325. Prince, Richard. "Individual Values and Administrative Effectiveness." Administrator's Notebook, v o l . 6, (December 1957), pp. 2-10. Pugh, D.E. And Others. "Dimensions of Organization Structure." Administrative Science Quarterly, June 196 8, pp. 65-105. Quinn, J.E. and Cavanaugh, H.M. "Fundamental Research Can be Planned." Harvard Business Review, v o l . 42 (January-February, 1963), pp. 49-66. Seashore, G.E. and Yuchtman, E. " F a c t o r i a l Analysis of Organi-zational Performance." Administrative Science Quarterly, December 1967, pp. 377-395. Schein, E.H. and Others. "Career Orientations and Perceptions of Rewarded A c t i v i t y i n a Research Organization." Administrative Science Quarterly, v o l . 9 (March, 196 4), pp. 333-349. Schein, E.H. "How to Break i n the College Graduate." Harvard  Business Review, November-December 1964, pp. 68-76. Scott, R.W. "Reactions to Supervision i n a Heteronomous Professional Organization." Administrative Science  Quarterly, v o l . 10, 1965, pp. 65-81. Shepherd, C l o v i s . "Status, Prestige and Esteem i n a Research Organization." Administrative Science Quarterly, v o l . 1, 1956, pp. 349-360. Terreberry, S h i r l e y . "The Evolution of Organizational Environments." Administrative Science Quarterly, March, 1968, pp. 590-613. 154 Terrien, F.W. and D.L. M i l l s . "The Ef f e c t s of Changing Size Upon the Internal Structure of Organizations." American S o c i o l o g i c a l Review, February, 1955, pp. 11-14. Wagner, Wesley L. "Leadership Style, H i e r a r c h i c a l Influence and Supervisory Role Obligations." Administrative  Science Quarterly, v o l . 9, 1964, pp. 391-420. Warner, Keith W. and Harens, A.E. "Goal Displacement and the I n t a n g i b i l i t y of Organizational Goals. Admin1strative  Science Quarterly, March 1968, pp. 539-555. GOVERNMENT DOCUMENTS Association of Universities and Colleges of Canada (AUCC). A B r i e f to the Senate Committee on Science P o l i c y . Ottawa: Queen's P r i n t e r , no. 44~TMay 1969)*,_ p. 5733. Bank of Canada. A Bri e f to the Senate Committee on Science  Po l i c y . Ottawa: Queen's P r i n t e r , no. 23 "(January 1969) , pp. 3435-3436. Canadian Labour Congress. A Bri e f to the Senate Committee on  Science P o l i c y . Ottawa: Queen's P r i n t e r , no. 61 (June 1969), pp. 7360-7372. Central Mortgage and Housing Corporation. A Br i e f to the Senate Committee on Science P o l i c y . Ottawa: Queen's Pri n t e r , no. 33 (February 1969), p. 4620. Committee of Presidents of Uni v e r s i t i e s of Ontario. A Brief  to the Senate Committee on Science P o l i c y . Ottawa: Queen^s P r i n t e r , no. 43 (May 1969), p. 5670. Department of Labour. A Bri e f to the Senate Committee on  Science P o l i c y . Ottawa: Queen's P r i n t e r , no. 27 (February 1969), pp. 3849. Department of Manpower and Immigration. A Bri e f to the- Senate  Committee on Science P o l i c y . Ottawa: Queen's Pr i n t e r , no. 2 8 (February 1969), pp. 4012-4013. Department of Regional and Economic Expansion. A B r i e f to the  Senate Committee on Science P o l i c y . Ottawa: Queen's Printer,, no. 40 (April 1969) , p. 4943. 155 Dominion Bureau of S t a t i s t i c s . A B r i e f to the Senate Committee on Science P o l i c y . Ottawa: Queen's Prin t e r , no. 24 "(February 1969) , ' pp7'"3T8l-3'4!9'8". Faculty of Science - York University. A Brief to the- Senate Committee on Science P o l i c y . Ottawa: Queen's P r i n t e r , no. 47 (May 1969), pp. 5 9 47-5950' Medical Research Council. A Br i e f to the Senate Committee on  Science P o l i c y . Ottawa: Queen's Pr i n t e r , no. 39 (February 1969),' p. 4143. National Research Council. A Brief to the Senate Committee on  Science P o l i c y . Ottawa: Queen's P r i n t e r , no. 3 (October 1968T, pp. 90-97. National Research Council. A Brief to the Senate Committee on  Science Pol i c y . Ottawa: Queen's Pr i n t e r , no. 21 (January 1969) , pp.. 3107-3113. Research Council of Alberta. A Br i e f to the Senate Committee  on Science P o l i c y . Ottawa: Queen's Pr i n t e r , no. 50 (June 1969) , p. 6282. Royal Society of Canada. A B r i e f to the Senate Committee on  Science P o l i c y . Ottawa: Queen's Pr i n t e r , no. 54 (June 1969), p. 6720. University of Alberta. A Br i e f to the Senate Committee on  Science P o l i c y . Ottawa: Queen's P r i n t e r , no. 48 (May 1969), pp. 6146-6150. University of Guelph. A B r i e f to the Senate Committee on  Science P o l i c y . Ottawa: Queen's P r i n t e r , no. 47 (May 1969) , p. 6041. University of Notre Dame. A Br i e f to the Senate Committee on Science P o l i c y . Ottawa: Queen's P r i n t e r , no. 48 (May 1969), pp. 6098-6099. University of Waterloo. A B r i e f to the Senate Committee on  Science P o l i c y . Ottawa: Queen's Pr i n t e r , no. 47 pp. 5985-5995. 156 MISCELLANEOUS Andrews, P.M. and D.C. Pelz. Analysis Memos from a Study of  S c i e n t i f i c Personnel. Ann Arbor, Michigan: Survey Research Center, I n s t i t u t e for Soc i a l Research, University of Michigan, no. 7 and no. 8 (March 1961), p. 83. Boyer, Susan and Others. U.B.C. MVTAB. Vancouver: The University of B r i t i s h Columbia Computing Center, 1966. Brown, D.J. "The Productivity of University Educators." Unpublished thesis for the degree Master of Arts . Vancouver: University of B r i t i s h Columbia, 1968. Patchen, Martin. Supervisory Methods and Group Performance  Norms. Ann Arbor: Survey Research Center, Ins t i t u t e for S ocial Research, University of Michigan, 1962. A P P E N D I C E S 158 Appendix (i) Random Sampling Procedure for Selection of University Sample Rank order a l l s o c i a l science departments according to si z e , on separate l i s t s for each type of department (psychology, sociology, anthropology, s o c i a l work, edu-cation psychology, education administration, economics and psychiatry). Calculate the number of: (a) s o c i a l science departments i n Canadian u n i v e r s i t i e s and (b) s o c i a l s c i e n t i s t s i n Canadian u n i v e r s i t i e s . From the information i n #2, calculate the average size of Canadian university s o c i a l science departments. Given the l i m i t a t i o n of 550 questionnaires for the university sample, divide t h i s by the average siz e of s o c i a l science departments, to get the number of depart-ments that can be included i n the sample. Calculate the percentage of the t o t a l number of s o c i a l science departments that t h i s sample represents, i . e . no. of departments i n sample _ ^ t o t a l no. of s o c i a l science departments -Consider the l i s t s of departments. Note natural breaks i n size and divide each l i s t into groups of departments that are of small, medium, and large s i z e . Take the percentage figure from #5 and multiply t h i s by the number of departments i n each small, medium and 159 large group, to get the number of departments allowed for the sample (nearest whole number i n each case, for the number of departments). 8. Number each group on each l i s t from 01-N. 9. Take a table of random numbers and sel e c t a s t a r t i n g place, with eyes closed. 10. Move down the columns of numbers u n t i l a number correspon-ding to that i n the f i r s t group, on the f i r s t l i s t , i s found. Mark that department. Continue down the random number rows u n t i l the allowed number of departments has been selected from that group. 11. Repeat #9 and #10 for each group of departments on each l i s t . 12. This procedure should provide a representative random sample of univer s i t y s o c i a l science departments. Note: On completing the procedure, a double check was made by adding up the t o t a l number of departments and the t o t a l number of university s o c i a l s c i e n t i s t s — t h e figures checked out very c l o s e l y with those i n #4. Appendix ( i i ) Four Types of Organizations: Their Sample Size and Questionnaire Return Data Question- Sample Question- Responded but Total No.of post naires size naire res- did not return response cards re-returned ponse rate o. a questionnaire rate o. turned 1. Business o • '" ™ (9 departments) 47 76 61.75 0 61.75 47 2. Government (40 departments) 137 243 56 .40 20 64.6 133 3. Soc i a l Development (27 departments) 109 198 54.75 16 62.8 106 4. University (42 departments) 230 625 36.80 119 55.8 230 Totals 523 1142 45.75 155 59.3 516 5. Unidentifiable hand-writing on post cards 6 6. Post cards returned but no name 12 7. Questionnaires returned a f t e r cut-off date and not included i n sample 11 Totals 534 1142 45.75 155 60.2 534 H ov o 161 Appendix ( i i i ) Reasons Given for Not Responding to the Questionnaire • • , ( Reason : Number • 9-"o 1. Not involved.in.research (not enough to j u s t i f y completing the questionnaire--administration, teaching etc.) 57 36. 8 2. Please exclude—no reason given 49 31. 6 3. No time to respond 9 5. 8 4. On leave--informed by secretary of research department 7 4. 5 5. Questionnaire not applicable to type of research being carried out 10 6. 45 6. Moved to new research department and not active i n research yet 8 5. 16 7. Questionnaire c r i t i c i s m — l a c k s c o n f i d e n t i -a l i t y , questions not appropriate etc. 9 5. 8 8. Plans to answer questionnaire, but.... 1 • 645 9. Objects to questionnaires i n general 3 1. 93 10. Prank responses 2 1. 29 11. Questionnaires returned (not completed), address unknown 155 17 100. 0% 16 2 . Appendix (iv) The Questionnaire THE CANADIAN RESEARCH SCIENTIST His Research Role and Organizational Environment Introduction To A l l Respondents: The following questionnaire items attempt to obtain information about the Canadian research s c i e n t i s t , his re-search a c t i v i t i e s and the research environment i n which he carries out his work. Since Canadian research s c i e n t i s t s work i n many kinds of organizations, an attempt was made to word questionnaire items i n such a way that they would be applicable to most s i t u a t i o n s . Please respond to a l l items, but i f some items do not apply to your p a r t i c u l a r s i t u a t i o n , please stroke them out. Please sign your name and the name of your research unit or department, on the back of the post card provided. Please mail the questionnaire and post card separately. This procedure w i l l ensure your anonymity and permit me to maintain a record of returns. A summary of the findings of t h i s study, keeping the i d e n t i f i c a t i o n of individuals and research units or departments c o n f i d e n t i a l , w i l l be made available to your research unit or department when the analysis i s completed. Your cooperation i n helping us obtain t h i s information w i l l be greatly appre-ciated. Sincerely, T.R. Pelton, Research Assistant Center for the Study of Administration i n Education University of B r i t i s h Columbia Vancouver 8, B.C. 163 A. BACKGROUND INFORMATION 1. Date of B i r t h : Place: 2. Degrees: Year University (a) Bachelor's (b.) Master's (c) Doctorate . . . (d) I f no degree, number of years post secondary education.. (e) If (d) i s applicable, where: Number of Years i . University or College i i . I n s t i t u t e of Technology i i i . Other (please specify) *3. In describing your s c i e n t i f i c d i s c i p l i n e s or professional f i e l d s , t r y to break down broad f i e l d s such as psychology or sociology into more s p e c i f i c areas such as c l i n i c a l psychology or sociology of r e l i g i o n , etc. Part I. In what s c i e n t i f i c d i s c i p l i n e s or professional f i e l d s did you s p e c i a l i z e i n your graduate or technical studies? • ( a y — : . . : ; T — (b) .... — — (c) .... (d) No formal s c i e n t i f i c , technical or professional education . Part IT. L i s t the above s c i e n t i f i c d i s c i p l i n e s or professional f i e l d s i n order of t h e i r importance to your present work. (a) (b) '.. (c) (d) None of s p e c i a l importance: * Items added to Pelz-Andrews Questionnaire items. 164 4. Previous research experience: ! Job t i t l e s of positions Employer (Univ.etc. No.of years held: Please specify) held: i . i i . i i i , i v . v. *5. (a) Name of research unit or department i n which are are currently employed; (b) name of organization of which the research unit or department i s a part: , (c.) Length of your employment i n th i s organization: Yrs Mths (d) Length of employment i n the research uni t or department i n which you are now working: Yrs Mths (e) Your o f f i c i a l job t i t l e : , (f) Your immediate supervisor's job t i t l e : .... :6(a) Your current annual salary: under 6,000 12,000 - 13,999 6,000 - 7,999 14,000 - 15,999 8,000 - 9,999 16,000 - 17,999 10,000 -11,999 18,000 or more B. THE NATURE OF YOUR RESEARCH ACTIVITIES The following l i s t of statements describes some of the a c t i -v i t i e s of research s c i e n t i s t s (supervisors should consider both the research work they supervise, and the work they carry out themselves). Please consider your "average" working month, i f possible. Rating Scale 1 none 2 not very much 3 some 4 much 5 very much * Items added to Pelz-Andrews Questionnaire items. 165 7. How much of your working time do you spend on the following a c t i v i t i e s i n your present research unit or department? Please c i r c l e the appropriate number (a) Discovery of s p e c i f i c knowledge for the solution of p a r t i c u l a r problems (applied research) (b) Discovery of new knowledge, not immediately applicable to p a r t i c u l a r problems (basic research) 8 *(d) *(e) 10, Please c i r c l e the appropriate number Rating Scale 1 t o t a l l y unim-portant 2 unimportant 3 neither impor-tant nor unimportant 4 important 5 very important How important are the following c r i t e r i a i n your selection of research projects? (a) Projects that are i n t e r e s t i n g i n themselves (b) Projects that seek new knowledge—because the search for new knowledge i s i n t r i n -s i c a l l y good *(c) Projects that may contribute to the well-being of some s o c i a l group Projects that are delegated by some higher authority 1 2 3 4 5 Projects that contribute to the organi-zation's f i n a n c i a l well-being 1 2 3 4 5 How much emphasis do members of your research R a t i n 9 Scale unit or department place on the coordination 1 of t h e i r e f f o r t s f or some common objectives? Please c i r c l e the appropriate number How much time pressure do you work under? ( i . e . r esults needed i n a hurry, deadlines to be met, etc.) none 2 not very much 3 some 4 much 5 very much * Items added to Pelz-Andrews Questionnaire items. 166 11. How much influence do you f e e l the following people have i n deciding your work goals and objectives? (a) Subordinates (b) Colleagues .. -(c) Yourself (d) Immediate supervisor of. department head (e) Higher l e v e l supervisor(s) (f) Management o f f i c i a l ( s ) (g) Other people or groups (please specify) 1 2 3 4 5 1 . 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 1 2 3 4 5 12. Some individuals are completely involved i n t h e i r research work, absorbed by i t night and day. For others, t h e i r work i s simply one of several i n t e r e s t s . To what extent are you involved, or how much involvement do you f e e l for your present research work? 1 2 3 4 5 13. Which of the following best describes your research unit or department? Please c i r c l e the appropriate number. Highly centralized control of a c t i v i t i e s Centralized control Neither centralized nor decentralized Decentralized control Highly decentralized control of a c t i v i t i e s 14. Over the past f i v e years, about how many of the following have you completed? (1) Approxi-mate number i n l a s t 5 years; (2) approximate number while employed i n t h i s research unit ( i f less than 5 years) i . Speeches i i . A r t i c l e s accepted by professional journals i i i . Books completed i v . Books accepted for publication ...... v. Patents or patent applications v i . Monographs completed v i i . Monographs accepted for publication v i i i . Project completion reports 1 2 3 4 5 167 Appendix (v) Questionnaire V a l i d i t y and R e l i a b i l i t y The r e l i a b i l i t y of the Pelz-Andrews questionnaire was checked by the authors. They had a group of fity-two scien-t i s t s respond to the questionnaire twice, af t e r a two month i n t e r v a l . The second time they responded, they were t o l d to respond as i f they were seeing the questions for the f i r s t time. Mean response scores were calculated i n each case and the two sets of mean scores correlated .97, i n d i c a t i n g that the r e l a t i v e standing of the group on these items at the f i r s t administration was an almost perfect predictor of the r e l a t i v e standing at the second administration. Since t h e i r analysis involved cimparisons among groups rather than i n d i v i d u a l s , as w i l l be the case i n the present study, an examination of the s t a b i l i t y of group means seemed an appropriate way to assess r e l i a b i l i t y . When i n d i v i d u a l scores on the two administrations were compared, the median r e l i a b i l i t y c o e f f i c i e n t was 0.62; 83% of the correlations were 0.5 or higher. 1 The v a l i d i t y of the questionnaire performance scores was checked against laboratory records. For patents, the Spearman rank-order c o r r e l a t i o n , between lab records and res-pondent claims, was 0.91. For published papers the c o r r e l a t i o n was 0.82. D.C. Pelz and F.M. Andrews, S c i e n t i s t s i n Organizations (John Wiley and Sons, 1966), p. 291. 168 The r e l i a b i l i t y , a f t e r a two-month period, between the two administrations of performance items was (Pearson product moment correlations) 1.00 for patents and 0.91 for papers. These authors concluded that research s c i e n t i s t s reported 2 research outputs with high and consistent accuracy. Pelz and Andrews, op. c i t . , p. 272. The authors speculated that discrepancies i n performance scores could be accounted for by s c i e n t i s t s writing papers for journals, and so on, without leaving a record i n the organization's f i l e s . 1 6 9 Appendix (vi) Questionnaire Cover Letter CENTER FOR THE STUDY OF ADMINISTRATION IN EDUCATION The University of B r i t i s h Columbia Vancouver 8, B.C. To Canadian Research S c i e n t i s t s : Canadian Research S c i e n t i s t questionnaires are now being mailed to two thousand research personnel i n government departments, unive r s i t y departments, business firms and s o c i a l development agencies across Canada. This study i s sponsored, i n part, by the Canada Department of Labour and the Alberta Human Resources Research Council. The primary focus of the study i s on the i d e n t i f i c a t i o n of the kinds of organizational environments and conditions which provide the most s a t i s f a c t o r y settings for the various types of research. In order to obtain adequate information for thi s study, the questionnaire i s somewhat longer than usual. May we ask for your patience and co-operation, therefore, i n completing the questionnaire and returning i t as soon as possible? To compensate you, i n some small way, for your time i n completing the questionnaire, we w i l l send a summary of our findings to your research department for your perusal. Individuals and departments w i l l not, of course, be i d e n t i -f i a b l e i n any reports. Yours sinc e r e l y , "T.R. Pelton" (Dr.) Harold J . Dyck, Project Director Terrance R. Pelton, Research Assistant 170 Appendix (vii) Questionnaire Follow-Up Letter Number One CENTER FOR THE STUDY OF ADMINISTRATION IN EDUCATION The University of B r i t i s h Columbia Vancouver 8, B.C. Dear Respondent, This i s an urgent plea for help. In September a Canadian Research S c i e n t i s t questionnaire was mailed to you as part of a study of Canadian research organizations. The primary focus of the study i s on the i d e n t i f i c a t i o n of the kinds of organizational environments which provide the most sat i s f a c t o r y settings for various types of research. A summary of the findings of the study w i l l be made available to research s t a f f i n each organizational s e t t i n g , when the study i s com-pleted. While i t would be f o o l i s h to assert that the results of the study w i l l have far reaching consequences, i t i s re-garded as s u f f i c i e n t l y important to warrant Canada Department of Labour and Alberta Human Resources Research Council support. Perhaps more to the point, i t i s regarded as extremely impor-tant to those of us who have a heavy personal investment i n the study. The p l a i n truth of the matter i s that i f there i s no change i n the present rate of return, the study w i l l be a washout. Not only w i l l whatever value the study may have not be r e a l i z e d , but a sizeable investment of funds, time and personal commitment w i l l have been wasted. Hence, the plea for help. Our records indicate that the questionnaire mailed to you has not been returned. I f you have not as yet completed the questionnaire, w i l l you please complete and return i t at your e a r l i e s t convenience? Yours sinc e r e l y , "T.R. Pelton" (Dr.) Harold J . Dyck, Project Director Terrance R. Pelton, Research Assistant 171 Appendix ( v i i i ) Q u e s t i o n n a i r e Follow-Up L e t t e r Number Two CENTER FOR THE STUDY OF ADMINISTRATION IN EDUCATION The U n i v e r s i t y o f B r i t i s h Columbia Vancouver 8 , B.C. Dear Respondent, Canadian Research S c i e n t i s t q u e s t i o n n a i r e s were m a i l e d i n June and September, t o one thousand r e s e a r c h p e r s o n n e l i n government, u n i v e r s i t y , business and s o c i a l development type r e s e a r c h o r g a n i z a t i o n s as p a r t o f a study o f Canadian r e s e a r c h o r g a n i z a t i o n s . During the h o l i d a y season some of these q u e s t i o n n a i r e s went a s t r a y and a number of r e s e a r c h people have requested a d d i t i o n a l copies i n order to p a r t i c i p a t e i n the study. In the event t h a t your q u e s t i o n n a i r e d i d not reach you, or was l o s t , I have taken the l i b e r t y of sending you an a d d i t i o n a l copy. T h i s study i s sponsored, i n p a r t , by the Canada Department of Labour and the A l b e r t a Human Resources Research c o u n c i l . The primary focus of the study i s on the i d e n t i f i c a -t i o n of the kin d s of o r g a n i z a t i o n a l environments which p r o v i d e the most s a t i s f a c t o r y s e t t i n g s f o r v a r i o u s types of r e s e a r c h . A summary of the r e s u l t s w i l l be made a v a i l a b l e to the per-sonnel of r e s e a r c h o r g a n i z a t i o n s p a r t i c i p a t i n g i n the study. A rough p r e l i m i n a r y a n a l y s i s o f the data c o l l e c t e d so f a r would i n d i c a t e some i n t e r e s t i n g d i f f e r e n c e s i n the r e -search environments p r o v i d e d by d i f f e r e n t o r g a n i z a t i o n a l s e t t i n g s . In order t h a t g e n e r a l i z a t i o n s can be made, however, a h i g h r e t u r n r a t e from the sample p o p u l a t i o n i s e s s e n t i a l . We appeal t o you to h e l p us o b t a i n t h i s h i g h r e t u r n r a t e . I f you have not as y e t completed the q u e s t i o n n a i r e , w i l l you , p l e a s e complete and r e t u r n i t as soon as p o s s i b l e ? Yours s i n c e r e l y , (Dr.) H a r o l d J . Dyck, P r o j e c t D i r e c t o r Terrance R. P e l t o n , Research A s s i s t a n t P.S. I f f o r some reason you wish t o be excluded from the sample p o p u l a t i o n of t h i s study, would you k i n d l y i n d i c a t e t h i s d e s i r e on the e n c l o s e d s e l f - a d d r e s s e d postcard? Thank you. 172 Appendix (Ix) S c i e n t i s t Output: Rationale for Its Use The "output" items used i n t h i s study were the same as those used by Pelz and Andrews. They defined "output" as any "unpublished, written, or o r a l presentation." 1 They had con-cluded that these items were a good representation of "research output." Among the items they used i n t h e i r questionnaire were the items referred to as "output number 1," and "output number 2" used i n t h i s study. Items included i n output number 1 were number of: 1. a r t i c l e s published i n journals; 2. formal reports (Project Completion Reports); and 3. unpublished o r a l presentations (speeches). Items included i n output number 2 were number of: 1. books completed; and 2. monographs, completed. The National Research Council (NRC) seemed to support the items selected to obtain some measure of s c i e n t i s t output. An important c r i t e r i o n i n assessing excellence i n the s c i e n t i s t i s the number and qua l i t y of his s c i e n t i f i c papers, patents, products and processes. S c i e n t i f i c achievement can be evaluated on the basis of a r t i c l e s published i n competently refereed journals and such papers available for further scrutiny by grant se l e c t i o n committees.2 D.C. Pelz and F.M. Andrews, S c i e n t i s t s i n Organizations (John Wiley and Sons, 1966), p. 272. Appendix (v) discussed the r e l i a b i l i t y and validity of these items. 2 NRC B r i e f to the Special Senate Committee on Science Policy, no. 3 (October 23, 1968), p. 92. 173 Appendix (x) Rationale for Using the Chi-Square S t a t i s t i c According to Si e g e l , nonparametric s t a t i s t i c s have cer-t a i n advantages over parametric s t a t i s t i c s . A parametric s t a t i s t i c a l test i s a te s t whose model s p e c i f i e s certain conditions (given on page 19) about the parameters of the population from which the research sample was drawn. Since these conditions are not o r d i n a r i l y tested, they are assumed to hold. The meaningfulness of the results of a parametric test depends on the v a l i d i t y of these assumptions. Parametric tests also require that the scores under analysis r e s u l t from measure-ment i n the strength of at least an i n t e r v a l scale. A nonparametric s t a t i s t i c a l test i s a te s t whose model does not specify conditions about the parameters of the population from which the sample was drawn. Certain assumptions are associated with most non-parametric s t a t i s t i c a l t e s t s , i . e . , that the obser-vations are independent and that the variable under study has underlying continuity, but these assump-tions are fewer and much weaker than those associ-ated with parametric t e s t s . Moreover, nonparametric tests do not require measurement so strong as that required for the parametric test s ; most nonparametric tests apply to data i n an ordinal scale, and some apply to data i n a nominal scale.1 Siegel strongly advised against using parametric s t a t i s t i c s when "ordinal" measurement i s used. At the r i s k of being excessively r e p e t i t i o u s , the writer wishes to emphasize here that parametric s t a -t i s t i c a l t e s t s , which use means and standard deviations ( i . e . , which require the operations of arithmetic on the o r i g i n a l scores), ought not to be used with data i n an ordinal scale. The properties of an ord i n a l scale are not isomorphic to the numerical system known as arithmetic. When only the rank order of scores i s known, means and standard deviations found on the S. Siegel, Nonparametric S t a t i s t i c s (New York: McGraw-H i l l Book Co., 1956), pp. 30-31. 174 scores themselves are " i n error" to the extent that the successive i n t e r v a l s (distances between classes) on the scale are not equal.2 Since most of the questionnaire data i n thi s study were i n the form of "ordinal" measurements of s c i e n t i s t s ' percep-tions (rating scale responses) and since only one of the samples was selected using random sample procedures, i t was decided to use nonparametric s t a t i s t i c s . The most appropriate s t a t i s t i c a l t e s t appeared to be the chi-square test for "k independent samples," which was described by Siegel i n his text, Nonparametric S t a t i s t i c s . In the analysis of research data, the i n v e s t i -gator often needs to decide whether several inde-pendent samples should be regarded as having come from the same population. Sample values almost always d i f f e r somewhat, and the problem i s to determine whether the observed sample differences s i g n i f y differences among populations or whether they are merely the chance variations that are to be expected among random samples from the same population.3 According to Siegel, when the data of research consist of frequencies i n discrete categories, the chi-square t e s t may be used to determine the significance of differences between two or more groups. The measurement involved may be as weak as nominal s c a l i n g . Siegel stated that: The hypothesis under te s t i s usually that the two (or more) groups d i f f e r with respect to some ch a r a c t e r i s t i c and therefore with respect to the Siegel, op. c i t . , p. 26. 3 I b i d . , p. 10 4. 175 r e l a t i v e frequency with which group members f a l l i n several categories. To test t h i s hypothesis, we count the number of cases from each group which f a l l i n the various categories, and compare the proportion of cases from one group i n the various categories with the proportion of cases from the other group(s). For example, we might test whether two p o l i t i c a l groups d i f f e r i n t h e i r agreement or disagreement with some opinion. . .4 The following method of computing chi-square for "k independent samples" was described by Siegel. To apply the chi-square t e s t , one f i r s t arranges the frequencies i n a k x r table. The n u l l hypo-thesis i s that the k samples of frequencies, or proportions, have come from the same population or from i d e n t i c a l populations. This hypothesis, that the k samples do not d i f f e r among themselves, may be tested by applying formula (6.3) : 2 v v (0.... - E-...)2 ^ = -12 ±J__ ( 6 . 3 ) i = l j=l E. . zed. where CK.. = observed number of cases categories i n the i^-h row of the j*-* 1 column. E^j = number of cases expected under Ho to be categorized i n i 1 " * 1 row of j*"* 1 column, as determined by method presented on page 105. r k E E d i r e c t one to sum over a l l c e l l s . i = l j = l Under Ho, the sampling d i s t r i b u t i o n of chi-square as computed from formula (6.3) can be shown to be approximated by a chi-square d i s t r i b u t i o n with df = ( k - 1 ) ( r - l ) , where k = the number of columns and r = the number of rows. Thus the p r o b a b i l i t y associated with the occurrence of values as large as an observed chi-square i s given i n Table C of the Appendix. I f an observed value of chi-square i s equal to or larger than that given i n Table C for a p a r t i c u l a r l e v e l of si g n i f i c a n c e and for df = ( k - 1 ) ( r - l ) , then Ho may be rejected at that l e v e l of s i g n i f i c a n c e . ^ 4 I b i d . , p. 10 4. ^Ibid., p. 175. 176 Appendix (xi) The Contingency C o e f f i c i e n t : C According to Siegel, the contingency c o e f f i c i e n t C i s a measure of the extent of association or r e l a t i o n between two sets of a t t r i b u t e s . To be able to use the contingency c o e f f i c i e n t , i t i s not necessary that we be able to assume under-ly i n g continuity for the various categories used to measure either or both sets of a t t r i b u t e s . In f a c t we do not even need to be able to order the cate-gories i n any p a r t i c u l a r way. The contingency co-e f f i c i e n t as computed from a contingency table, w i l l have the same value regardless of how the categories are arranged i n the rows and columns.1 To compute C, Siegel stated that data from chi-square computations could be incorporated into the formula. The degree of association between two sets of att r i b u t e s , whether orderable or not and irrespec-t i v e of the nature of the variable ( i t may be either continuous or discrete) or of the underlying d i s t r i b u t i o n of the att r i b u t e (the population d i s -t r i b u t i o n may be normal or any other shape), may be found from a contingency table of the frequencies by ^ _ C = / X 2 (9-1) V N + y2 2 where x = formula (6.3). In other words, i n order to compute C, one f i r s t computes the value of chi-square by formula (6.3) and then inserts that value into formula (9.1) to get C. 2 S. Siegel, Nonparametric S t a t i s t i c s (New York: McGraw-H i l l Book Co., 1956), p. 196. 2 Ibid., p. 197. Also on p. 199, Siegel points out that "we may tes t whether an observed C d i f f e r s s i g n i f i c a n t l y from chance simply by determining whether the chi-square value for the data i s s i g n i f i c a n t . " 177 Limitations of C The upper l i m i t for the contingency c o e f f i c i e n t i s a function of the number of categories. When k = r, the upper l i m i t for C, that i s , the C which would occur for two pe r f e c t l y correlated variables, i s k-1 k For instance, the upper l i m i t of C for a 2 x 2 table i s | = .707 For a 3 x 3 table, the maximum value which C can att a i n i s 2 3 The li m i t a t i o n s of C, therefore are: (a) i t does not a t t a i n unity when there i s a perfect association, (b) comparisons cannot be made unless the tables compared are the same s i z e , (c) C i s not d i r e c t l y comparable to any other measure of c o r r e l a t i o n . Siegel, op. c i t . , p. 201. 178 Appendix (xii) Definitions of Research A c t i v i t i e s As stated by the N.R.C., the c l a s s i f i c a t i o n of the re-search a c t i v i t i e s of s c i e n t i s t s i s usually into either basic applied, or developmental research a c t i v i t i e s . These terms, according to the N.R.C., are commonly used i n a general way by both s c i e n t i s t s and others as broad loose categories. 1 They stated further that: In p r a c t i c e , however, when one gets down to s p e c i f i c s , i t i s almost impossible to get agree-ment on any s p e c i f i c project as to which category i s correct. In f a c t , l i k e beauty, the nature of the categorization depends a great deal on the eye of the beholder, and every beholder sees d i f f e r e n t l y . 2 General agreement from several sources, however, per-mitted the following general c l a s s i f i c a t i o n of research a c t i -v i t i e s . Basic research. Discovery of new knowledge, not immediately applicable to p a r t i c u l a r problems. Applied research. Discovery of s p e c i f i c knowledge for the solution of p a r t i c u l a r problems. Developmental research. The development or improve-ment of products and/or processes using research information.^ National Research Council of Canada (N.R.C), B r i e f to the Special Senate Committee on Science Po l i c y (Ottawa: Queen's P r i n t e r , no. 21, Jan. 29, 1969), p. 3113. 2 I b i d . Science Council of Canada, Towards a National Science Po l i c y , B r i e f to the Special Senate Committee on Science P o l i c y (Ottawa: Queen's P r i n t e r , Report No. 4, Sec. 2, 1969), p. 7; also, University of Alberta, B r i e f to the Special Senate Committee on Science Policy (Ottawa: Queen's Pr i n t e r , no. 48, May 29, 1969), p. 6150; also, University of Guelph, B r i e f to 179 The Science Council, l i k e the N.R.C., implied that these research a c t i v i t i e s have no d i s t i n c t boundaries, but merge into each other and are part of what could be considered 4 a "spectrum of s c i e n t i f i c a c t i v i t i e s . " the Special Senate Committee on Science Policy (Ottawa: Queen's P r i n t e r , no. 47, May 28, 1969), p. 6041; also, York University, B r i e f No. 47, op. c i t . , p. 5947. 4 Science Council of Canada, l o c . c i t . 180 Appendix ( x i i i ) Levels of Research Output #2 versus Levels of S c i e n t i s t Influence i n Deciding Work Goals and Objectives (Combined Samples) How much influence do you f e e l the following people have i n deciding your work goals and objectives? Yourself 1 2 3 4 5 Levels of Levels of Influence Research Low Medium High N Output #2 1,2 "3 4,5 0 f 16 63 307 386 H 4 16 80 T 3 12 59 f 5 "4 112 121 Low 1-4 H 4 3 93 T 1 1 3 f 0 2 14 16 Medium 5-8 H 0 12 88 T . 0 .5 3 21 69 433 523 X =14.43 f = frequency; H = horizontal percentage f; C = .17 T = t o t a l percentage f. 181 - Appendix (xiv) Levels of Research Output #2 versus Levels of Supervisor Influence i n Deciding S c i e n t i s t Work Goals and Objectives (Combined Sample) How much influence do you f e e l the following people have i n deciding your work goals and objectives? Immediate supervisor or department head 1 2 3 4 5 Levels of Levels of Supervisor Influence Research Low Medium High N Output #2 . 1,2 3 4,5 f 144 112 130 386 0 H 37 29 34 T. 28 21 2 5 . f 76 17. 28 121 Low 1-4 H 63 14 23 T. 15 3 5 f 8 4 4 ••16 Medium 5-8 H 50 25 25 T 2 • 1 1 228 133 162 523 X = 25.51 (.01) f = frequency; H = horizontal percentage f; C = .215 T = t o t a l percentage f. 182 Appendix (xv) Levels of Research Output #2 versus Levels of C e n t r a l i -zation/Decemtralization (Combined Samples) Which of the following best describes your research unit or department? , Please c i r c l e the appropriate number. Highly centralized control of a c t i v i t i e s 1 Centralized control 2 Neither centralized nor decentralized 3 Decentralized control 4 Highly decentralized control of a c t i v i t i e s 5 Levels of Levels of Centralization/Decentralization Research Centralized Neither Decentralized N Output #2 1,2 3 4,5 f 116 120 150.7 386 0 H 30 31 39 T 22 23 29 f 27 28 - - 66 121 Low 1-4 H 22 23 55 T 5 5 13 f 6 3 7 Medium 5-8 H 38 19 44 T 1 .5 1 149 151 223 523 X = 10.33 (.05) f = frequency; H = horizontal percentage f; C = .139 T = t o t a l percentage f. 183 Appendix (xvi) Levels of Research Output #1 Versus Levels of Coordination (Combined Samples) How much emphasis do members of your research unit or department place on the coordination of t h e i r e f f o r t s for some common objectives? 1 2 3 4 5 Levels of Research Low and Medium High Output #1 Coordination Coordination N 1,2,3 4,5 f 173 119 292 Low 0-4 H 59 41 T 33 23 . f 146 85 231 High 5+ H 63 37 T 28 16 319 204 523 X 2 = -84 184 Appendix (xvii) Research Output versus Levels of S c i e n t i s t Decision-Making Influence How much influence do you f e e l the following people have i n deciding your work goals and objectives? Yourself 1 2 3 4 5 Research Output (1+) Organization Low and Medium High Levels of Type Influence 1,2,3 Influence 4,5 0 1 1 Bus . 0 100 Gov. 4 15 22 85 26 S.D. 4 17 23 83 27 Univ. 3 4 80 96 83 11 8 125 92 X 2 = 5.7 (.10) C = *Research Output = 1 or more books or monographs completed Chi-square Values Separate Samples vs. Separate Samples Gov. vs S.D. ^2 = 00 Gov. vs.Univ.x? = 4.56 (.05) Univ.vs. S.D.x2 = 4.28(.05) Separate Samples vs.Combined Samples Univ. vs. Gov + S.D. X 2 = 5.73 (.02) S.D. vs. Univ.+ Gov.X 2 = 2.05 (.20) Gov. vs. Univ.+ S.D. X 2 = 2.30 (:;2G) 185 Appendix ( x v i i i ) Research Output versus Levels of Supervisor Decision Making Influence How much influence do you f e e l the following people have i n deciding your work goals and objectives? Immediate supervisor or department head 1 2 3 4 5 Research Output (1+) Organization Low and Medium High Supervisor Type Supervisor I n f l . Influence 4,5 1,2,3 B U S . 1 100 . . . . o . . , 0 1 Gov. 7 27 19 73 26 S.D. 19 70 8 30 27 Univ. 78 94 5 6 83 104 77 32 23 136 X = 49.42 (.0 01) Research output = 1 or more books or C = .515 monographs completed. Chi-square Values Separate vs.Combined Samples Separate vs. Separate Samples 36.27(.001) Un.vs.Gov x 2 = 51.18(.001) .69 Un.vs.S.D. x 2 - 10.89(.001) 43.85(.001) Gov.vs.S.D.x2 = 10.00 (.01) Un.vs.S.D.+Gov. xt = S.D.vs.Un.+Gov. X = Gov.vs.Un.+S.D. x 2 = 186 Appendix (xix) Research Output versus Levels of Centralized and Decentralized Control Which of the following best described your research u n i t or department? Highly centralized control of a c t i v i t i e s 1 Centralized control 2 Neither centralized nor decentralized 3 Decentralized control 4 Highly decentralized control of a c t i v i t i e s 5 Organi zation Type Research Centralized 1,2 Output Neither 3 (1+) Decentralized 4,5 N Bus. 0 0 .1 100 0 0 1 Gov. . 14. . . 54 7 27 5 19-26 S.D. 10 37 7 26 10 37 2 7 Univ. 9 16 58 83 11 19 70 33 24 30 23 73 53 136 X = 29.30 (.001) Research output = 1 or more books or C = .42 monographs completed. Chi-square values Separate vs. Combined Samples Separate vs. Separate Samples Un.vs.Gov.+S.D. X 2 = 26.97(.001) Un.vs.Gov. X 2 = 26.68(.001) S.D.vs.Gov.+Un. X 2 = 4.23(.20) Un.vs.S.D. X 2 = 12 ..07 (.001) Gov.vs.Un.+S.D. X 2 = 19.04(.001) S.D.vs.Gov.X 2 = 2.31 187 Appendix (xx) Research Output #2 versus Levels of Coordination How much emphasis do members of your research uni t or department place on the coordination of t h e i r e f f o r t s for some common objectives? 1 2 3 4 5 Organization Type Research Low and Medium Coordination 1,2,3 Output (1+) . High Levels of Coordination 4,5 . N Bus. 0 0 1 -', 100 1 Gov. 11 42 . 15 58 26 S.D. 10 37 17 63 27 Univ. 73 88 10 12 83 94 69 42 31 136 X = 35.56 (.001) C = .455 Research Output = 1 or more books or monographs completed Chi-square values Separate vs. Combined Samples Separate vs. Separate Samples Un.vs. S.D.+Gbv. S.D.vs.Un.+Gov. Gov.vs.Un.+S.D. X 2 = 35.39(.001) Un.vs.Gov. X2 = 16.24(.001) Un.vs.S.D. X 2 = 10.82(.01) S.D.vs.Gov. X 2 = 21.71(.001) X 2 = 28.5K.001) X 2 = .15 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0102274/manifest

Comment

Related Items