INFORMATION TO USERS This material was produced from a microfilm copy of the original document. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the original submitted. The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction. 1. The sign or “target" for pages apparently lacking from the document photographed is "Missing Page(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity. 2. When an image on the film is obliterated with a large round black mark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus cause a blurred image. You will find a good image of the page in the adjacent frame. 3. When a map, drawing or chart, etc., was part of the material being photographed the photographer followed a definite method in "sectioning" the material. It is customary to begin photoing at the upper left hand corner of a large sheet and to continue photoing from left to right in equal sections with a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until « complete. 4. The majority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints of "photographs" may be ordered at additional charge by writing the Order Department, giving the catalog number, title, author and specific pages you wish reproduced. 5. PLEASE NOTE: Some pages may have indistinct print. Filmed as received. Xerox University Microfilms 300 North Zeeb Road Ann Arbor, Michigan 48106 76 - 2 7 ,1 3 5 NELSON, Connie Holmes, 1945A STUDY OF INFORMATION SOURCES USED FOR INSTRUCTIONAL DECISION-MAKING BY MICHIGAN PUBLIC SCHOOL DISTRICT PERSONNEL. Michigan State University, Ph.D., 1976 Education, curriculum and instruction Xerox University Microfilms, Ann Arbor, Michigan 48106 > principals > teachers) used people sources more, and were higher in overall usage of information. 4. Use of literature sources was higher in rural areas; people and organizations dominated in urban areas. 5. The ranking of source use— people, literature, organizations— was matched exactly by the ranking of per­ ceived attributes of the information source. That is, people sources were ranked highest with regard to per­ ceived experience, with accessibility, quality, and ease of use. Literature followed, with organizations last. 6. Experience was the best single predictor of use for all three types of information sources. From these observations, several implications for both development and future research were presented. The most important of these follow. 1. A preference for people information sources was found, suggesting that any dissemination system must use a human intermediary. 2. This human intermediary should link users to producers of information, to the system's own and other data banks, and to other similar users. 3. Experience was the single largest predictor of useof a source, suggesting that educators should be given early— probably pre-service— experience with a variety of sources. DEDICATION This work is dedicated to Neil who lives the be­ lief that a marriage requires each of us to nurture and maintain ourselves; so that we may bring to the pairing a person of worth. Together we are more than two. ACKNOWLEDGMENTS There are a lot of persons who through the years have contributed to reaching this pinnacle in my educa­ tional career. Among those are my parents who taught me the value and joy of continually striving to better under­ stand the world. A special thanks to Dr. Robert Muth and the Middle Cities Education Association (MCEA) for providing the re­ sources necessary for data collection. At the time of this study, MCEA held a contract with the Michigan Depart­ ment of Education to develop an instructional management information system. Ms. Judy Pfaff dedicated many weekend hours and other times to helping program and run the data. Dr. Alexander Kloster encouraged my efforts throughout the study. The other three committee members, Dr. Louise Sause, Dr. Louis Romano, and Dr. George Myers, were helpful critics at needed times. Neil, David, Karen, Cailin, and sister Peggy were all very supportive of my work and helped to maintain our house as a home. TABLE OF CONTENTS Page ACKNOWLEDGMENTS ..................................... iii LIST OF TABLES .......................... vii LIST OF F I G U R E S ........................................ ix Chapter I . INTRODUCTION ............................ 1 Purpose. ................................. 1 N e e d ........................................ 2 Nature of the Study......................... 4 Hypotheses ................................. 7 Theoretical Framework. . . . ............... 9 Theoretical Basis for Information Source Selection...................... 11 Theoretical Basis for User Characteristics as Influential Variables on Information Sources Selection........................... 13 Theoretical Basis for Source Attributes as Influential Variables in Information Source Selection ......................... 17 Summary of the Theoretical Framework . . . 22 .................................... 23 Overview Assumptions.................................... 26 II. REVIEW OF RELATED L I T E R A T U R E ................... 27 Studies of Types of Information Sources. . . 28 Studies of Attributes of Users . . . . . . . 46 Studies of Attributes of Information Sources...................................... 53 Summary........................................ 61 III. METHODOLOGY..................................... 63 Introduction . . . . . ..................... 63 S a m p l e ........................................ 65 iv Page Population . . . . . ..................... 65 Overall Sample Description .............. 65 Regions........................ 66 Strata . . ................................68 69 Sample Selection ......................... Pre-Pilot and Pilot Samples. . . . . . . . . 75 Pre-Pilot Study. ......................... 75 Pilot S t u d y . ................................76 Instrumentation................................77 Questionnaire Description. ............ 78 Reliability..................................80 Administrative Details ................... 91 Design ................................92 Operational Definitions of Information Source Variables ................... 93 Organization of Information Sources for A n a l y s i s .................... 94 Operational Definitions of User Charac­ teristic V a r i a b l e s ......................... 94 Organization of User Characteristics for A n a l y s i s ................................95 Relationships Among Variables............... 96 Testable Hypotheses........................... 99 A n a l y s i s ..................................... 102 Summary.............. 107 IV. ANALYSIS OF F I N D I N G S .......................... 112 Introduction ............................... 112 Usage of Types of Information Sources. . . . 112 H y p o t h e s i s .................................113 D a t a ....................................... 113 D i s c u s s i o n .................................115 Effects of User Attributes on Information Usage....................................... 117 H y p o t h e s e s .................................117 Data .................................119 D i s c u s s i o n .................... 128 Effects of Source Attributes on Informa­ tion Choice........................... 130 H y p o t h e s e s ......................... 130 D a t a ....................................... 131 Summary....................................... 145 V. SUMMARY AND CONCLUSIONS........................ 151 Introduction ........................ 151 Summary............................... 151 Integrating Conclusions With Theory.......... 158 Conclusions on Information Source Selection................................... 158 v Page Conclusions on User Characteristics . . . 159 Conclusions on Attributes of Informa­ tion Sources.................. ...........160 Implications of Findings for Schools. . . . 162 Suggested Directions for Future Research. . 166 169 A Final Summary ........................... APPENDIX A .......... 171 APPENDIX B . . . . . . ............................. 175 APPENDIX C . . BIBLIOGRAPHY ...................................... 177 ............................. vi 188 LIST OF TABLES Table 3.1 3.2 3.3 3.4 Page SCHOOL DISTRICT CLASSIFICATION SYSTEM of the Michigan Department of Education .......... 70 Reliability Estimates on Responses to Infor­ mation Source Usage Based on Open-ended Q u e s t i o n .......... 89 Reliability Estimates on Responses of Infor­ mation Source Usage Based on Open-Ended and Closed Questions ........................... 91 Questionnaire Return Rates by District and Total..................................... . 93 3.5 Four Categories of Information Sources . . . . 95 4.1 Mean Use of Three Information Sources by Each of the Ten D i s t r i c t s ......................... 114 4.2 Analysis of Variance on Usage of Three Types of I n formation................................ 114 4.3 Mean Use of Three Information Sources Summed Over All D i s t r i c t s ............................115 4.4 Tukey Post Hoc Pair Comparisons.................116 4.5 Tests of Significance from Analysis of Var­ iance Related to the Interaction Hypoth­ eses . . . . . . . ......................... 122 4.6 Table of Significance from Analysis of Var­ iance Related to Each Statistically Sig­ nificant Hypothesis............................129 4.7 All Independent Variables Divided into Three G r o u p s ........................................ 131 vii Table 4.8 4.9 Page Comparison of Mean Scores on Perceived Attri­ butes of Information Sources with Mean Scores on District Usage of Information S o u r c e s ................................ Sample Correlation Matrix for Twleve Inde­ pendent Variables Describing Attributes of Information S o u r c e s ........ ............ .. 132 . 134 4.10 Multiple Regression Analysis: b weights or multiple regression coefficients. . .......... 2 4.11 Multiple Regression Analysis: R s, Beta Weight, and Squared Semi-partial correla­ tions, Attribute Variables for People Information Sources ......................... 2 4.12 Multiple Regression Analysis: R s, Beta Weights, and Squared Semi-Partial Correla­ tions, Attribute Variables for Organiza­ tional Information Sources. . . . .......... 2 4.13 Multiple Regression Analysis: R s, Beta Weights, and Squared Semi-Partial Correla­ tions, Attribute Variables for Literature Information Sources .......................... 138 140 141 142 4.14 Summary T a b l e ................................... 146 viii LIST OF FIGURES Figure Page 1.1 A Graphic Description of the Relationship Among Communication Research, Diffusion Research, and Information Science . ...........10 1.2 Hypothesized Subsets of Information-Seeking .............. Behavior. . 3.1 24 152 Regions Used in Sampling and in Analysis of Michigan Public School Districts. . . . . . . 67 3.2 Cluster Sampling by Region and Strata . . . . . 72 3.3 Questionnaire Format for Obtaining Ratings of Information Sources Used by Educators for Instructional Decision-Making .......... 79 3.4 Questionnaire Format for Obtaining Ratings of the Attributes (Ease of Use, Accessibil­ ity, Amount of Experience, and Content Qual­ ity) of the Information S o u r c e s ............... 81 3.5 Categories of Instructional Decisions Used to Establish the Reliability of the OpenEnded Question.............. 83 3.6 A Sample of a Closed Situation Used in Estimating the Reliability of the Infor­ mation Sources in Measuring Instructional Information-Seeking . ....................... 90 3.7 Operational Definitions of Information Sources for Instructional Decision-Making . . 94 3.8 A Graphic Interpretation of the Relationship Between the Three Types of Information S o u r c e s ........................................ 97 3.9 A Graphic Interpretation of the Relationship Between the Two Independent Variables— Region and Status— and the Dependent Variable— People Information Sources..................... 98 ix Figure Page 3.10 A Graphic Interpretation of the Relationship Between the Two Independent Variables— Strata and Status— -and the Dependent Var­ iable— People Information Sources. . . . . . 4.1 Observed Weighted Means for Teachers, Principals, and Central Administrators by Geographic R e g i o n .................. 99 120 4.2 Observed Weighted Means for Three Geo­ graphic Regions by Status of Professional Educators...................................... 121 4.3 Observed Weighted Means for Teachers, Prin­ cipals and Central Administrators by Dis­ trict S t r a t a .................................. 124 4.4 Observed Weighted Means for District Strata by Status of Professional Educators........... 126 5.1 Functions of a Human Intermediary in an Information System ......................... 164 5.2 Information-Seeking and Non-information Seeking by Professional Educators............. 165 5.3 Two Important Means of Improving the Educa­ tor's Experiences with Information Sources . x 166 CHAPTER I INTRODUCTION Purpose The educator of today is confronted with a vast array of information sources. among these sources. He must sort and choose This choice is based upon a combin­ ation of the nature of the information and certain char­ acteristics of the educator himself. The study deals with patterns of interaction between educators, information sources, and their respective attributes. The purpose of this investigation is four-fold. First, the information sources that are currently used for instructional decision-making in Michigan public schools are identified. Second, the extent of use of these infor­ mation sources by the professional school district person­ nel is determined. Third, the effects of the attributes of the information user— status, geographical location, and size of school district— on the type of information source selected are studied. Fourth, the effects of the attributes of the source itself— ease of use, accessibility, amount of experience, and source quality— on the type of information source selected are dealt with. 2 Potential information sources were identified by conducting a pilot study in which educators were asked to enumerate information sources presently being used to make instructional decisions of importance. A range of cri­ teria for information selection was identified by review­ ing previous research. First, this chapter explains the possible uses of this present study. discussed. Second the nature of the study is Hypotheses are presented in a general, non- testable form. Fourth, the theoretical basis for this study is established. Careful attention is given to sev­ eral theories; as there is no currently existing theory which singularly unifies the behavioral trends of school districts in instructional decision-making. Need The field of education does not lack quantitatively materials for instructional uses. Educators are over­ whelmed by volumes of data and ideas, but lack an adequate method of categorizing "which materials" for "which idea" under "what circumstances" and "for whom." Concurrent with this problem— of establishing what X material does for Z problem— is the desirability of developing a more systematic and less time-consuming method for the assess­ ment of this information by educators. Educational infor­ mation needs to be collected, organized, and transmitted. 3 The focus of this present study is one component of the total process necessary for a state department of education to develop an information system capable of handling pertinent instructional information for school districts. In Michigan, a feasibility study was conducted by the Middle Cities Education Association ( M C E A ) u n d e r contract with the Michigan Department of Education, to de­ termine "the technical and economic feasibility of alter­ native methods for gathering, storing, and retrieving instructional information. . ." 2 Information and its use is seen as the cornerstone to improving education in Michigan. A position paper pub­ lished by the Michigan State Board of Education states: . . . information and its use is the key to improvement in Michigan's system of public education: information which enables a district to improve the educational exper­ iences of its students; information which enables districts to replicate successful educational practices; and information which enables the Department of Education to pro­ vide services to improve the capacity of local and intermediate districts to meet their students' needs.3 1 . . Middle Cities Education Association is a consor­ tium of thirteen urban school districts in Michigan, co­ ordinating efforts to improve the educational environment of urban children. 2 Michigan Department of Education, Request for Proposal: Instructional Resource Support System, Fall of 1974, Part III-C Objective. 3 . Michigan State Board of Education, "An Intro­ ductory Statement of the Development of Management Infor­ mation Systems for Local School Districts: The Basic Frame of Reference." 1972, p. 4. 4 One of the specific tasks of the contractor was to identify alternative methods which could be used by the Michigan Department of Education in disseminating the information contained in the proposed information system labeled the Michigan Instructional Resource Support System (MIRSS). This particular task of the research contract provides a clear-cut need for focusing on an assessment of the current status of information sources being used by educators in making instructional decisions and the criteria by which these sources are chosen. Therefore, this present endeavor proposes to be of practical value in helping to determine successful methods of dissemina­ tion of instructional materials by beginning at step one— a look at the present state of affairs. The findings could be expanded in applicability to illuminate similar educational settings where an attempt is made to improve information dissemination techniques. Nature of the Study It seems appropriate at this time to relate some reasons for the particular approach of this study— an assessment of user habits. Some of the previous attempts to develop information systems support the decision to begin by viewing the present habits of recipients. For example, the early developers of the Educational Resources Information Clearinghouse (ERIC) became so engrossed in the needs (content) of the system itself, that they did 5 not anticipate the difficulty of disseminating informa­ tion from each of the clearinghouses. It was not until ERIC could be accessed from libraries; that use increased. ERIC managed to capture the audience of persons who reg­ ularly use libraries as a source of information. Thus, early user problems might have been diminished if the de­ velopers had initially addressed the question of what is the best way to handle information dissemination. It would be interesting to ask, in the histor­ ical perspective, why such an enormous invest­ ment was made without any empirical examinations of . . . the communication behaviors and infor­ mation needs of these potential users . . . .4 In contrast, the Classroom Teacher Support System (CTSS) is an example of a computer instructional system that achieved success by first studying actual behavior patterns and then basing the development of the system on this behavioral information. 5 For example, the developers of CTSS noted that teachers in Los Angelos Unified School District used tests as an instructional tool and that teachers spent large blocks of time on test preparation 4 Nan Lin and William D. Garvey, "Information Needs and Uses," Annual Review of Information Science and Technology, Vol. 7 (Washington, D.C.: American Society for Information Science, 1972), p. 27. 5 John F. Vinsonhaler and Robert D. Moon, "Infor­ mation Systems Applications in Education," Annual Review of Information Science and Technology, Vol.-8 (Washington, D .C .: American Society for Information Science, 1973), p. 287. 6 and scoring. The CTSS planners then focused their energy on providing teachers with computer assistance in testrelated clerical tasks. The question of how to best handle information dissemination was succinctly answered by Dr. Ted Ward, former director of the Learning Systems Institute and a consultant in the development of several large information systems. He stated "to conceptualize a system that re- quires the user to leap is to conceptualize a failure." 6 He emphasized the importance of not attempting to simul­ taneously change both the information and the habits of users. His approach has been summarized in the following steps: 1. Assess present user habits. 2. Use a known information source to introduce new material. 3. Cultivate use of this new material within a familiar context by periodically including the material within this context. 4. Make suggestions within the current context for the reader to try a new source. For example, if a teacher is presently obtaining ideas from a particular magazine, and one wishes to encourage the teacher to use a newly created teacher center, the g Interview, Ted W. Ward, Prof., Institute of In­ ternational Studies of Education, Education Department, Michigan State University, February 5, 1975. 7 "Ward approach" is to use the present source— magazine— as the context for disseminating new materials. Then gradually suggestions of new sources of information can be made within the context of the magazine. Thus, there can be a slow transition in information sources, if the material is really relevant to the educator. After analyzing previous attempts to design in­ formation systems it has become apparent that the most affectatious method is to first assess the habits of the user and then concentrate on the content of the system. Content is of little value if it lies unused. Hypotheses This study is concerned with the relationship of educators, information sources, and attributes of the sources. The task of understanding the connections be­ tween these three variables is approached via a three step process. The first variable which this study focuses up­ on is information sources. These sources are divided into three large groups— people sources, organizational sources, and literature sources. For example, people sources in­ clude principals, consultants from outside of the district; organizational sources include the state and intermediate district educational systems; and literature sources inelude professional journals, and commercial publications. The hypotheses for this variable are as follows: 7 . . . The entire list of information sources is given in Table 3, Chapter Three. 7 8 People information sources are used more fre­ quently than either literature or organizational sources. Organizational sources are used more frequently than literature sources. Second, this study considers what attributes of the educator (user) influence the selection of information sources. Characteristics of the user that are of interest include the size and geographic location of the school district and the status of the professional personnel withO in the school' district. The hypotheses for this set of independent variables are as follows: Status of educational personnel affects choice of information source. Size of school district affects choice of information source. Geographical location of the school district affects choice of information source. Third, this study is interested in determining whether certain attributes of the information sources affect information-seeking behavior. Thus, the third group of hypotheses is concerned with the set of four variables that describe the attributes of the source it­ self. These attributes are the perceived ease of use of an information source, the perceived accessibility of the source, the perceived quality, and the reported degree of experience in using the source. Definitions of these four Q These attributes are defined in detail in Chapter III. 9 criteria, as they were presented to questionnaire respond­ ents, are presented in Chapter II. The hypotheses for this set of independent variables are as follows: A meaningful amount of the variance in choice of information source is explained by these four criteria. There is a difference in the predictive ability among the four criteria regarding the use of information sources. Theoretical Framework Simply stated, the concern of this study is in­ formation usage of professional staff in Michigan school districts. Specifically, three types of information sources are identified— people sources, organizational sources, and literature sources. The direction of this study is influenced by both behavioral science theories and the "real world." Probing information usage in instructional de­ cision-making implies a look at communication between source and user. Two subfields of communication research that have significance for this study are diffusion re­ search and research in information science. These two bodies of research are not mutually exclusive in their con­ ceptual focus as is illustrated in Figure 1.1. Diffusion research is concerned with the functions that affect the adoption or rejection of a new idea or innovation. Infor­ mation science research is "the discipline that is con­ cerned with the processes by which individuals communicate 10 with other individuals over time and distance." g Thus, information science research deals with both innovations and more routine tasks. Communication Research Diffusion Research Information Science Research Figure 1.1— A Graphic Description of the Relation ship Among Communication Research, Diffusion Research, and Information Science. 9 Charles W.N. Thompson, "Technology Utilization," Annual Review of Information Science and Technology, Vol. 10, (Washington, D.C.: American Socxety for Information Science, 1975), p. 385. 11 Theoretical Basis for Infor­ mation Source Selection Historically, there are six distinct disciplines that have dealt with diffusion research. The most pertin­ ent for this study are anthropology, rural sociology, and education. In the 1960's Everett M. Rogers promoted the concept of a field of diffusion research, independent of any specific discipline, through his leadership at the Dif fusion Documents Center at Michigan State University. In Communications of Innovations, he demonstrated the like­ nesses in the basic concepts of the processes of adoption regardless of the innovation or the academic discipline of the investigator.'*'® Through the merging of this research from a variety of disciplinary lineages, Rogers developed the concept of the S-Curve of Adoption, a testable gen­ eralization for explaining the rate of application of a new idea.11 The implications of diffusion research for this present study is that the implied traditional theoretical model of the communication path for new ideas was people to people. That is, when a person learns a new idea it is communicated via people. 1®Everett M. Rogers and F. Floyd Shoemaker, Communication of Innovations (New York: The Free Press, 1971). “ i;LIbid., p. 177. 12 The innovation diffuses through a social system, defined as a population of individuals . . ..12 Thus, people sources should certainly be explored in this study as a potential means of gaining insight in­ to the process of instructional decision-making. The broad base of diffusion research also indicates that people are the most frequently sought sources for informa­ tion. This traditional theoretical base— communication within a social system of individuals— has been challenged as an inadequate explanation. Research indicating that other sources of information are also used in the adoption of an innovation are explored in Chapter II. These chal­ lenges to the traditional theoretical model of people-topeople communication of ideas imply that this study must be concerned with more than just direct contact with human resources. Thus, it becomes eminent to look at the use of people, organizational and literature sources in instruc­ tional decision-making. 12 Gerhard Eichholy and Everett M. Rogers, Resis­ tance to the Adoption of Audio-Visual Aids by Elemen­ tary School Teachers: Contrasts and Similarities to Agricultural Innovation," in Innovation in Education, edited by Matthew B. Miles (New York: Teachers College, Columbia University, 1964), pp. 312-313. 13 Theoretical Basis for User Characteristics as Influ­ ential Variables on Information Sources Selection Furthermore, this study explores the question of whether variations in the type of information source chosen depend on the attributes of the user. The traditional model of diffusion research assumes that people to people dialogue is the correct vehicle regardless of the attri­ butes of the user. Differences in user characteristics are only seen as affecting the rate of adoption. The field of information science is of importance to this study in constructing a theoretical framework which relates to user attributes. As previously mentioned, the field of information science is concerned with a broad range of information— from new ideas to routine tasks. This study is concerned with the attributes of educators which allow them to deal with different types of informa­ tion. Thus, the inquiries in this field have direct im­ plications for this study. The field of information science also abounds with studies that empirically approach the question of what sources will be used for what purposes. Thus, given a user of information, what attributes affect information usage and which ones seem to have no influence on the re­ ceptivity of the user? The research in this area which was deemed significant for this study is reviewed in Chap­ ter II. 14 A brief resume of the development of this field helps to establish its significance to the theoretical design of this study. The original concepts of informa­ tion science can be attributed to two main sources: the operations research methods developed during World War II and the processes of system engineering as taughts in the 1950's at MIT. 13 The genesis of operations research is attributed to the discovery by a civilian research team that one member of each British gun crew was assigned to an obsolete function. . . . they were told that he was there to handle the horses. Horses? But there were no horses! Well, perhaps not., but there surely had been horses' the ones that drew cannon for British troops in the days up into World War 1.14 From this modest beginning, operations research grew into a process for systematically studying the relationship between man and machines. With the growth in the technological applicability of the computer to solve human problems emerged the need to understand the bridge between technology and benefac­ tor. Early emphasis on understanding the gap focused on how the user (man) could be changed to be compatible with the computer (machine). In the last decade, a substantial number of studies emerged on information needs and uses 13 Vinsonhaler and Moon, "Information Systems Ap­ plication in Education," p. 8. 14 Paul Dickson, Think Tanks (New York: Books, Inc., 1971), p. 22. Ballantme 15 with the reverse intent— to better understand how the computer could be adapted to meet man's information needs. Regardless of future assumptions about the role of the computer, the portion of the discipline of information science that is of interest to this study deals specifi­ cally with what has been recently discovered about user habits. No general theory of information use exists yet, but the parameters which will confine the theory have been defined. According to Paisley: . . . we now urgently need theories of informa­ tion-processing behavior that will generate propositions concerning channel selection, amount of seeking; effects on productivity of informa­ tion quality, quantity . . . we cannot interpret data on information needs and uses without recog­ nizing that the scientist/technologist stands at the center of many systems that touch every aspect of his work. An understanding of these systems is essential if we are to improve information transfer.16 These systems can be configurated by a series of concentric circles. Beginning with the innermost circle, one can place the user himself as an information processor. The other five can then be consecutively built as the indi­ vidual in a work team, the individual in a formal 15 16 Thompson, "Technology Utilization,11 p. 398. William J. Paisley includes educators among the technologists. As the educator applies the science of learning and behavior in "Information Needs and Uses," An­ nual Review of Information Science and Technology, Vol. 3, (Chicago: Encyclopedia Britannica, Inc., 1968), p. 10. 16 organization, the individual in a professional society, the user in an "invisible college," and the user in a formal 17 information system. This present study deals with some of these pa­ rameters that must be encompassed by a theory of informa­ tion usage. The individual can be viewed as a member of a work team by analyzing the influence that status has upon the use of information. Specifically, the difference in information usage among teachers, principals, and cen­ tral administrative staff in Michigan school districts is investigated. This study also deals with the individual as he/she interacts within a formal organization by con­ sidering the variable of district size. In this case, school districts in Michigan were divided into six cat­ egories delineated by student population. Aside from the theoretical implications discussed so far, the task of releasing information to the entire state dictates that one begins with an inductive process. One must first determine what sources under what condi­ tions will initiate macro-differences. Michigan encompas­ ses a wide and varied geographic area from the megalopolis 17 Thomas J. Allen uses the concept of "invisible college" to explain the loosely structured social system that exists between professionals of like interests, as opposed to status or propinquity. The exact boundaries are difficult to determine as a member is usually not aware of all other members. In "Information Needs and Uses," Annual Review of Information Science and Technology, Vol. 4, (Chicago: Encyclopedia Britannica, Inc., 1969), p. 21. 17 belt around southern Lake Michigan to the remote and iso­ lated peninsula of northern Michigan. Thus, the question is complicated by the possibility that geographic differ­ ences might affect information behavior among educators. To this point, three attributes of users have been selected for investigation: status of user, size of users' school district, and geographical location of users' school district. But user attributes are only part of the present concern. Theoretical Basis for Source Attributes as Influential Variables in Information Source Selection The field of information science also demonstrates the effect that attributes of the source, as opposed to the user, can have on information selection. A more com­ plete review of these studies in channel selection are presented in Chapter II. At this point it is sufficient to note that the discipline of information science is attempting to develop theory encompassing attributes of sources. There are, however, several behavioral theories that give direction to the types of source attributes that influence information-seeking behavior. In the late 1950's Herbert A. Simon revised his book on Administrative Behavior in which he more clearly integrated his thoughts on his proposed theory of human 18 choice in decision-making. 18 According to Simon, people satisfice— make decisions without knowledge or awareness of all choices. The impact of satisficing behavior is that people necessarily ignore examining all possible al­ ternative solutions, disregard the interrelatedness of the environment, and base decisions on simplified models. Satisficing behavior implies limitations on rationality, which is defined within the context of decision-making as "the selection of preferred behavior alternatives in terms of some systems of values whereby the consequences of behavxor can be evaluated." 19 Simon feels that rationality is more closely achieved by an individual within the framework of an organization than by an isolated person; as an organization is better able to broaden the knowledge base beyond the capacity of a single individual. Organizations and institutions provide the general stimuli and attention-directors that channelize the behaviors of the members of the group . . .20 In accordance with Simon, an educator makes a de­ cision with less than perfect knowledge. The educational environment defines the parameters of stimuli available to its members. 18 York: Herbert A. Simon, Administrative Behavior (New The Free Press, 1965). ^Ibid., p. 75. 20Ibid., p. 100. 19 In accepting this premise that organizations determine the stimuli or criteria by which members make decisions, it becomes necessary to investigate those cri­ teria that the educational environment imposes on educa­ tors. Specifically, this endeavor looks at what influ­ ences the organizational environment might have on the particular attributes of information sources that deter­ mine which of those sources are used by educators in de­ cision-making. The prevailing influence of an organizational en­ vironment, according to Simon, is the requirement that an organization seek to maximize its use of resources. There­ fore, members must make decisions on the basis of effi­ ciency. The criterion of efficiency dictates that choice of alternatives which produces the largest result for the given application of resources.21 Further support for the efficiency criteria as a key to the influence that an educational organization has on information source selection is found in the works of James D. Thompson. His theoretical thoughts on organiza­ tional behavior help in defining which source attributes affect information-seeking behavior of educators. Within his general theory, Thompson presents a proposition that an organization, in the absence of the ability to measure true effectiveness, measures efficiency. ^Ibid., p. 179. 20 Neither the public school nor the university has proof positive that the student it produces is well educated. The proof is in a lifetime, and the effects of the educational organization can hardly be separated from other influences on that individual. But the fellowships won by its stu­ dents, and occupational-placement statistics can be compiled q u i c k l y . 2 2 Thus, when cause-effect knowledge is incomplete, or be­ lieved to be so, an organization judges itself on efficien­ cy rather than effectiveness. There is certainly a lot of historical evidence to support Thompson in his claim that educators are pri­ marily concerned with efficiency. Callahan presents a thorough review of how educators became obsessed with measuring inputs to the extent of almost totally ignoring their effectiveness m attaining product goals. 23 For example, Callahan points out that Ayres in 1901 in Laggards in Our School, drew an analysis between the school and the factory. His work emphasized the cost of the repeater to the taxpayer, rather than the socio-economic forces that could explain some of the reasons for the large num­ ber of repeaters. Bobitt, another early twentieth century captain of educational efficiency, attempted to apply scientific management principles to education by developing specific achievement standards. 22 Louis: James D. Thompson, Organizations in Action (St. McGraw-Hill Book Company, 1967), p. 92. 21 . . . the ability to add at a speed of sixty-five combinations/minute with an accuracy of ninetyfour per cent . . . A teacher who fell short of the standard was "unmistakably shown to be a weak teacher," and the supervisor would have "incontest­ able evidence of inefficiency" against the weak teacher who cannot or refuses to improve.24 How does the criterion of efficiency affect the information-seeking behavior of educators? It is suggested here that the choice of sources is based on the perceived efficiency of use. here. use. Three specific criteria are considered First, educators prefer sources which are easy to Second, they prefer sources that are accessible. Third, they prefer sources with which they have experi­ ence. All three of these are simply operationalizations of the criterion of efficiency as expressed by both Simon and Thompson. Thus, this study looks at the attributes of ease of use of a source, accessibility of a source, and the de­ gree of experience in using a particular source. Research studies that are described in Chapter II also lend support to the investigation of these attributes as potential ex­ planations for why educators prefer certain sources. If Simon is accurate in his analysis of decision­ making within an organization, then these three efficiency criteria should have priority in determining information source selection. sidered. Yet, the alternative should be con­ If the educational environment reinforces a more 24Ibid., pp. 81-82. 22 rational approach to decision-making (maximizing), then a wider spectrum of source attributes may affect informa­ tion-seeking behavior. Thus, the inclusion of the cri­ teria of content quality. The addition of this attribute sets up a potentially competing set of arguments for why educators select certain information sources when making instructional decisions. One answer to the degree of con- gruity among these four attributes lies in the analysis of the data in this study. Summary of the Theoretical Framework The theoretical framework of this endeavor has emerged from viewing theories of information usage from a variety of orientations— behavioral science, information science, diffusion research, and the real world. The con­ vergence of ideas about information usage from such di­ verse angles is perceived to be a source of strength for the particular approach of this work. Thus, this kalei­ doscopic theoretical framework sets the stage for looking at choice of information source as it relates to both the attributes of the user and the attributes of the sources. In examining the information-seeking behavior of educators, three types of information sources are used as the framework for this study— people sources, literature sources, and organizations as sources. 23 Next, in relation to these three types of informa tion sources this study looks at three attributes of users. First, this study focuses on the impact of status on usage of an information source. the effect of district size. Second, it examines Third, it views the effect of geographic location. In addition, this study examines the influence of the attributes of a source on information source selec­ tion. First, this study looks at the attribute of ease of use. Second, this study views the attribute of acces­ sibility. ence. Third comes the attribute of degree of experi­ Fourth, this study focuses on the attribute of quality of the information source. Overview In the first chapter, the three basic elements of this study are presented— information sources, attributes of information users, and attributes of the information sources. The conceptual framework for each of these ele­ ments of the study is developed, drawing on theoretical ideas from several vantage points. The entire framework is best visualized in Figure 1.2. Each element and sub-element is viewed as a part of the set of all information-seeking behavior. In the second chapter, the literature sources are systematically reviewed as they relate to each of the basic elements of this study. First, those studies that 24 Information Sources Attributes of Users -People -Organizations -Literature -status -district size -geographic location V = (set of) All Information Seeking Behavior Attributes of Sources Other -Accessibility -Ease of Use -Experience -Quality / Figure 1.2— Hypothesized Subsets of InformationSeeking Behavior. examine types of information sources used in decision­ making are discussed. Second, the body of literature deal­ ing with user characteristics that affect informationseeking behavior is presented. Third, the research related to pertinent characteristics of information sources is 25 reviewed. Research material in the field of education is discussed separately from those studies in other fields that have implications for this present endeavor. In the third chapter, the survey research method­ ology of this study is explained. techniques are described. First, the sampling This includes a discussion of both the telephone pilot sample used to develop the mail questionnaires and the sample for the written question­ naire. Second, the nature and content of the question­ naire is discussed. in testable form. Third, the hypotheses are presented Fourth, the research designs to be used in interpreting the data are introduced. Fifth, the models for analysis are presented. In the fourth chapter, the data results are ex­ amined in relation to each of the hypotheses under study. In the fifth chapter, three sets of conclusions are drawn. First, are those conclusions about types of information sources appropriate to the information-seek­ ing behavior of educators. Second, are those conclusions about user characteristics that influence source selection. Third, are those conclusions about source criteria that affect the selection of an information source. Finally, the implications of this study for the design of informa­ tion systems for educators is discussed. Suggestions for the possible direction of future research into the infor­ mation-seeking behavior of educators are offered. 26 Assumptions There are a number of assumptions operative here in supporting the parameters of this study. First, that information-seeking behavior is ex­ plainable. Second, that attributes of information users and attributes of information sources help to define informa­ tion-seeking behavior. Third, that there are a finite number of both user and source characteristics that can be described and ex­ plained. Fourth, that the relationship between information source sielection and user characteristics and the relation­ ship between information source selection and source attributes can be identified and described. CHAPTER II REVIEW OF RELATED LITERATURE The intent of this study is to seek definitive answers to the information-seeking behavior of Michigan educators when involved in the task of making instruction­ al decisions. identified. First, the major information sources are Second, the possible characteristics of the information user that affects choice of sources is exam­ ined. Third, four attributes of sources are explored for their potential effects on information choice. This chapter is conceptually divided into three distinct subsections. The first section reviews those literature sources that were helpful in determining the range of information sources to be examined in this study and the hypothesized direction for frequency of use. The second section discusses those literature sources that in­ fluenced the choice of characteristics about users to be studied. The third section contains the body of litera­ ture that summarizes research that helped in determining what information source criteria might influence informa­ tion-seeking behavior. 27 28 Studies of Types of Information Sources There are numerous investigations both within the field of education and in other behavioral science disci­ plines that gave focus to the information sources chosen for this study. As mentioned earlier, the work of Rogers in diffusion research was a synthesis of the findings, of many studies in six distinct disciplines. Diffusion re­ search is based on the assumption that communication occurs within a social system. Thus, people to people communica­ tion is the basic vehicle for transmission of knowledge. Early diffusion studies in rural sociology laid the foundation for viewing people as the key means for transmission of new ideas.'*' One study in particular was credited with the establishment of the assumptions on which future diffusion research was based. In 1943, Ryan and Gross investigated the diffusion process of hybrid seed corn. 2 Their study concluded that people were the most influential source in persuading farmers to use hy­ brid seed. Specifically, they found that the average farmer first heard of hybrid seed from a salesman; but neighbors were the strongest factor in actual adoption of hybrid corn. ^"Everett M. Rogers and F. Floyd Shoemaker, Com­ munication of Innovations (New York: The Free Press, 1971), Chapter 2. 2 Bryce Ryan and Neal C. Gross, "The Diffusion of Hybrid Seed Corn in Two Iowa Communities," Rural Sociology, 7 (March, 1943), 15-24. 29 The field of information science also abounds with research that supports this work and focuses on three basic information sources— people, organizations, and literature« The major impetus began in the fifties when attempts were made to categorize user needs with user characteristics. The majority of studies support the importance of people information sources in decision­ making . First, information science research outside the field of education is examined. Most attention is devoted to the annual reviews (1968-1975) of information usage contained in Annual Review of Information Science and Technology. This journal contains the most prestigeous works in terms of research quality in the field of informa­ tion science. There is a heavy emphasis on investigations of users of scientific and technological information as this was the area of most concern until the late sixties when comprehensive studies of users in social sciences and education emerged. Allen and Cooney conducted a major study of the communication patterns of more than twelve hundred re­ search and development scientists and engineers in Ireland. 3 One of the relevant findings for this study was 3 Thomas J. Allen and S. Cooney, "Institutional Roles in Technology Transfer: A Diagnosis of the Situa­ tion in One Small Country," R and D Management, 4 (October, 1973), p. 50. 30 that personal contact was an effective means of exchang­ ing technological information. In another study, Goldhar sent questionnaires to winners of the annual contest for the one hundred signif­ icant technological products sponsored by Industrial Re4 search. He asked for details on the types, sources, and channels of information used during the idea generation stage. One conclusion relevant to this study was that in­ ternal and informal people sources were most important for stimulating ideas for new products. Rothwell and Robertson reviewed thirty-two studies on information usage and its relation to technological innovation between 1957-1973. 5 The authors summarized their findings by stating that personal contact and personal experiences were important communication channels. In an additional review, Clarke reviewed over one hundred and seventy references on the decision-making process with technologically-oriented firms. He noted that formal com­ munication models are seldom used by decision-makers.^ 4 Joel D. Goldhar, "Information, Idea Generation and Technological Innovations," Paper presented at the NATO Ad­ vanced Study Institute on Technology Transfer, Every, France, (June 25 - July 6, 1973) , p. 47. 5 R. Rothwell and A.B. Robertson, "The Role of Com­ munications in Technological Innovation," Research Policy, 2(October, 1973), p. 223. 6 Charles W.N. Thompson, "Technology Utilization," Annual Review of Information Science and Technology, Vol. 10, (Washington, D.C.: American Society for Information Science, 1975), p. 396. 31 Bernard, in examining the transfer of information between marine scientists and U.S. policymakers, demonstrated that again personal contact is the most important means 7 for obtaxnxng information. Two Department of Defense studies also support the theme of this work in recognizing the importance of people as sources for decision-making. For example, in one study over fifteen hundred interviews were conducted with defense industry personnel. The results focused on the significance of colleagues as the first source of in­ formation .8 Most of the studies discussed to this point relied on structured interviews or self-administered question­ naires. There are two additional distinctive works— one a memoir and the other a field observation— that lend weight to the hypothesis that people sources will be used most frequently in educational decision-making. James D. Watson, the Nobel Laureate who discovered the molecular structure of DNA, described his use of in­ formation in the following way. Conversation among half a dozen people in London and Cambridge provided most of the pieces for 7 Russell H. Bernard, "Scxentxsts and Polxcy Makers: An Ethnography of Communication," Human Organization, Vol. 33 (Fall, 1974), p. 272. 8 William J. Paisley, "Information Needs and Users," Annual Review of Information Science and Technology, Vol. 3, (Chicago: Encyclopedia Britannica, Inc., 1968) , p. 10. 32 the DNA puzzle, yet one long-forgotten journal article and a few handbook references filled in pieces that conversation alone could never supply .... Each information channel was essential, but scores of hours were spent in conversation for each hour of reading.9 Peter Newman, the editor of Canada's leading news magazine Macleans, is in the process of writing a three volume series on the analysis of the power-holders in Canada.^ Volume one was just recently released. His research is relevant to this study in that it provides an insightful look at information flow within the business world, adding strength to the existing diverse research on the importance of people in decision-making. Newman in­ troduced his chapter on working the system with this state­ ment. It sometimes seems as if members of Canada's Establishment spend most of their waking hours on one mammoth conference call . . . . At the stratospheric financial altitudes in which they operate, what matters is not what they do but whom they know, what entrees they can provide. The idea is always to be extending your reach, consolidating your contacts, knowing something that somebody else doesn't know. At this level, it is knowledge, not money, that creates power. The network of private schools, clubs, weddings, funerals, receptions, board meetings, country weekends, and other rounds of encounters pro­ vides an effective bush-telegraph for the bar­ ter system according to which the Establishment game is played. Deals are made, information is traded, tips are exchanged, recommendations ^Ibid., p. 26. "^Peter C. Newman, The Canadian Establishment, Vol. 1, (Toronto: McClelland and Stewart Limited, 1975). 33 quickly granted or withheld. Each time, personal debits and credits are carefully noted.11 Snatches of antecdotal converstions of several of the establishment members lend additional confirmation to the hypothesis that people sources are most important in decision-making. For example, the president of the Can­ ada Packers, W.F. "Bill" McLean divulged, I feel that if I get out of my depths in some­ thing, I can go to one of these guys I know and ask him. We consult a lot.12 Another member of the identified establishment, Douglas Fullerton, former chairman of the National Capital Com­ mission, corporate director and pivotal business-govern­ ment coordinator, admitted You build your network through personal charm, exchanging favors and money, but most of all by having information to exchange.13 Another revealing antecdotal account is that of Donald C. "Ben" Webster, president of Helix Investments Limited. You operate by knowing people, meeting them in various ways, in common sorts of things.1^ And finally the thoughts of Gordon Sharwood, chief execu­ tive officer of Guaranty Trust Company, helps to make clear the extensive use of people in decision-making. i:LIbid., p. 175. 12Ibid., p. 178. 13Ibid., p. 179.. 34 You see somebody in the washroom of the Toronto Club and you say, "Did you hear that so-and-so is doing this-and-that?" and you become a part of this sort of informal network of intelligence that gets put together through snatches of con­ versation all over the place from all sorts of people.15 Besides the studies reviewed so far, there are several major studies in the field of education that set the stage for this present study on information-seeking behavior. First, extensive planning of educational in­ formation systems lead two nationally-known systems men, Vinsonhaler and Moon to conclude . . . greater attention needs to be paid to human factors in the design and evaluation of many types of educational information system applications . . . the need seems to be less for additional technical developments than for persons capable of relating existing tech­ nical tools to the human progress that is education.15 Cormier contributed a study that has direct appli­ cation. He approached the question of information-seek­ ing behavior by means of the critical-incident technique. Respondents, school superintendents, described informa­ tion usage in terms of an important decision they had made in the previous week. He found the main source of ■^Ibid., p. 181. 16 John F. Vinsonhaler and Robert D. Moon, "In­ formation System Applications in Education," Annual Re­ view of Information Science and Technology, Vol. 8 (Washington, D .C .: American Society for Information Science, 1973), p. 307. 35 information to be people, specifically the internal dist n c t teaching staff. 17 Rittenhouse and associates conducted an extensive study into the processes of information utilization in decision-making at the local school district level. work makes two points of relevance here. His First, the range of educators sampled in each district was similar to this work. Each district sample included the superintendent, an assistant superintendent, two members of the district staff, four principals, and four teachers. Second, he concluded that the most frequently used information sources were "colleagues in one's own school system, principals and vice-principals, contacts at professional meetings, superintendents, and curriculum specialists." 18 Thus, people sources tend to be the preferred source. According to another study by McCracken, state directors of vocational education were more likely to seek information through personal contacts than by searching the literature. 19 These results were obtained through telephone interviews, using a structured questionnaire. 17 Roger A. Cormier, "Information in Each Opera­ tional Area for Important Decisions of School Superintend­ ents: A Summary," Journal of Educational Data Processing, 1971, pp. 30-32. 18 Carl H. Rittenhouse, "Educational Information Uses and Users," AV Communication Review, Vol. 19 (Spring, 1971), p. 81. 19 J. David McCracken, Information Needs of State Directors of Vocational Education (Washington, D.C.: Government Printing Office, March 1973), p. 103. 36 Telephone interviewing was also the methodology chosen for the pilot study for this present work. Al­ though a thorough review of this study occurs in Chapter 3, it is worth mentioning now. First, the pilot study was used to identify a meaningful range of information sources for use on the self-administered questionnaire. Educators were asked to relate an instructional decision that they had made during the last week. They then were asked to relate the information sources used in arriving at a de­ cision. The results of this pilot study gave an insight­ ful look at the amount of reliance educators have on people sources for instructional decision-making. Out of 174 responses mentioned by thirty randomly-selected edu­ cators— ten central administrative staff, ten principals, and ten teachers— sixty-eight per cent were people sources. To this point the review of literature has fo­ cused on research that supports people as the most impor­ tant source of information in decision-making. First, material outside of the field of education was summarized. Second, a look was taken at the education studies on in­ formation-seeking behavior. The following discussion lends support for the inclusion of literature sources as a feasible base of information for decision-making. First is a sketch of studies outside the field of education. 37 As mentioned previously, early agricultural studies of the diffusion process had immense influence in supporting the assumption that people sources were the underlying means of communication in decisions about adopting new innovations. Later diffusion research in agriculture lead to the development of a competing hy­ pothesis on information-seeking behavior. These more re­ cent studies are relevant to this research as they support the use of sources other than people for obtaining infor­ mation on which to base decisions. Through mainly the impetus of the U.S. Department of Agriculture and the agriculture extension units, many more studies of the adoption process in rural technology occurred. Investi­ gations were made on the adoption process of fertilizers, weed sprays, vaccinations, antibiotic drugs, and sanita­ tion. For example, one result of these studies was the refinement of the adoption process as a function of time. Thus, agricultural researchers discovered that different information sources are used at varying stages in the acceptance of an innovation. Evidence as to which infor­ mation source for what stage is still inconclusive. Studies by Rogers and Beal, by Wilkening, and by Copp, Sill and Brown agree that mass media use (printed materlals, radio) was most important at the awareness stage. 20 20 Robert G. Mason, "Securing Clientele Acceptance and Cooperation," Rural Sociology, 29(March 1964), p. 41. 38 Personal sources (extension agents and peers) were stated as most important during the interest and acceptance stages. All source types, including the addition of com­ mercial sources were cited as used during the experimental stage. Some later work by Mason contradicted the sequen­ tial pairings of sources with adoption stages. relevance to this study is clear. Yet, the Empirical evidence exists for the use of literature sources in addition to people for seeking information. The question should there­ fore be raised here. These rural sociology studies are the intellec­ tual ancestors to the early diffusion work in education. These studies had a diffuse influence on the methodologies, focus, and interpretation of research in the acceptance of educational innovations. Perhaps, it would have been better for the understanding of educational communication processes if the influence of traditional agricultural diffusion research had been defused. For the emphasis in agricultural studies is mainly on the establishment of new "things" as opposed to new ideas of processes. And there are probably important differences in the nature of diffusion of physical and biologi­ cal "hardware" innovations versus social science "software" items.21 The fact that education and agriculture are not inclusive­ ly concerned with transmitting the same type of knowledge 21 Rogers and Shoemaker, Communication of Innova­ tions, p. 79. I 39 again increases the justification for looking at more than people sources. There is also a collection of studies on informa­ tion source selection from the field of information science that lends some support to literature sources as crucial to the decision-making process. In one study by Johnston and Gibbons, the pattern of information flow in the development of technological innovations was demonstrated. - They interviewed all the personnel involved in the conception of thirty innovations. The authors con­ cluded that printed matter and personal contact are equal contributions, and that the manager "should promote diversity and flexibility to insure access to as wide a range of information as possible.22 Another study suggested that the specific purpose for the information affected the choice of information source. For example, if usage is divided into the two functions of idea generation and problem definition, dif­ ferent types of information are shown to be initially sought. Baker and Utterback found that for idea-genera- tion problems, literature sources were essential; while interpersonal interactions were important for problem def. . . 23 m i t i o n functions. 22 23 Their work makes two points of Thompson, "Technology Utilization," p. 406. Thomas J. Allen, "Information Needs and Uses," Annual Review of Information Science and Technology, Vol. 4, (Chicago: Encyclopedia Brittanica, Inc., 1969), p. 6. 40 particular relevance to this study. One, this study may find results differing from other information science studies, as the specific purpose for which information is sought by the participants in this work is for instruc­ tional decision-making. Second, their work gives support for some proportional relationship between people and literature as sources for instructional decision-making. In another study, Wood and Ronayne introduced the idea that literature sources play a secondary role in decision-making. These researchers used recorded dis­ cussions about readership of certain target journals as the basis for analysis of information-seeking behavior of members of the Royal Institute of Chemistry and the Chem­ ical Society. Their concluding stance was the following. These journals do not make a very marked contri­ bution to direct work activity, but they provide useful ancillary information, help maintain some awareness of research programs outside the readers' own areas, and are sometimes a stim­ ulating source of ideas . . ..24 Another study by Langrish et al. is interesting to this present endeavor as it points to a plausible ex­ planation for why literature sources appear to play such a minor role in daily decision-making. These authors studied the stimuli for the development of fifty-one in­ novations by industrial firms. 24 These writers purport: John Martyn, "Information Needs and Uses," Annual Review of Information Science and Technology, Vol. 9, (Washington, D.C.: American Society for Information Science, 1974), pp. 12-13. 41 At first sight, the thought that only 9h of 102 ideas came from literature, which is basically the information worker's raw material and stockin-trade, is deflating and possibly discouraging, but there are two points worth noting. The first is that a number of ideas come from education— from information, that is, acquired possibly a long time before the innovative idea came into being. Information is not always a 'now' thing, deployed within a short time after re­ ceipt; it may often be brought into use many years after its distribution.25 It is obvious from the discussion of both research in diffusion processes and within the field of informa­ tion science that literature does not seem to be the main source of information for decision-making. Thus, the re­ search reviewed so far provides substantive content to support the hypothesis of this study that people sources are used most frequently in educational decision-making. Attention is now directed specifically to educational re­ search on literature as a viable source of information. No studies were found in the field of education that argued for literature as the main source for decision­ making. In spite of this minor role that literature seems to play in decision-making processes, the real world demonstrates the voluminous effort put forth to publish more and more printed materials. Furthermore, person to person communication is often time-wise impractical. Thus, reality is a sound reminder of the need to understand what role literature does play in educational instructional de­ cision-making . ^Ibid., p. 18. 42 The most extensive research on the use of litera­ ture sources by educators comes from ERIC— Educational Resource Information Center. ERIC offers a number of bibliographic and publication services. ERIC's founding objectives called for gathering, indexing, abstracting, and making available important literature in the field of education? for preparing reviews and synthesis of the educational literature and; for making available to local practitioners the literature collected by ERIC. The first two objectives are being successfully accomplished. third objective has eluded most efforts to date. The In an opinion paper published in the journal of American Society for Information Science, Paisley offered this summative comment on professionals orientation to literature sources. No scientific or professional group has ever been found to prefer print sources over the informa­ tion and advice that colleagues provide . . . . Scientists can be described as "print-tolerant," and that is more than can be said of physicians, educators, engineers, etc. Of course, every educated person in this society can and does read a great deal, but the latter groups rely heavily on colleagues and co-workers to keep them informed on developments within their professions.26 Paisley offered two alternative models for in­ creasing ERIC's effectiveness in reaching practitioners. Paisley models are relevant to this study as they 26 William J. Paisley, "Improving a Field Based (Eric-Like) Information System," American Society for Information Science, Vol. 22, No.- 6 (November-December, 1971), p. 403. 43 presented two practical means of dealing with the dual reality of people as the preferred information source and of the existence of immense quantities of printed educa­ tional information. One model was a continuing education program whereby ERIC information would be packaged into inservice programs. By using this framework, people would be the vehicle for introducing written materials. The second model had a conceptual base similar to that of the County Agricultural Extension Program. Educational ex­ tension agents would act as the relayer and interpreter of ERIC's knowledge base. It is worth reviewing Rittenhouse1s study as ad­ ditional support for the consideration of literature as a plausible information source for educational decisionmaking. 27 This author noted that at times educators used texts and curriculum materials from outside the district. The pilot study for this present endeavor helped to make clear that literature sources are used by educa­ tors in making instructional decisions. Precisely, lit­ erature sources accounted for thirteen per cent of the responses. In summary, it is evident that literature sources are at critical times vital; at other times helpful and supportive of people sources. 27 Rittenhouse, "Educational Information Uses and Users," p. 81. 44 What about organizational sources? This present study hypothesized on the basis of the pilot study that organizational sources would be used more frequently than literature sources. In that pilot, nineteen per cent of the response to information sources used in instructional decision-making were attributable to organizations. This included the Michigan Department of Education, interme­ diate school districts, school boards, community libraries, and community organizations. The literature lends an indecisive tone to any attempts to determine the significance of organizations in instructional decision-making. lack of any studies to review. The main problem is There were only two pieces of research found that seemed to have any relevance to this study— one in diffusion research and one in education. Rogers, the leading diffusion research expert, con­ ducted a study on the characteristics of innovators. Rogers asserted that the results enlarged the focus on information sources. He concluded that innovators use broader and more impersonal sources of information, such as universities, conferences, and laboratory schools. 28 Thus, this study lends support to the need to better under­ stand the role of organizational information sources. og Everett M. Rogers, "What are Innovators Like?" in Change Process in the Public Schools, ed. by R.O. Carlson (Eugene, Oregon: Center for the Advanced Study of Educational Administration, 1964). 45 Rittenhouse1s study into the processes of informa­ tion utilization at the local school district level has relevance here as he included some measures of organiza­ tional sources. He discovered that federally funded re­ search and development programs and information services were the least used sources. Empirical evidence on the frequency of use of organizational sources in educational decision-making is limited. Several plausible explanations are offered. One, education is traditionally a local endeavor. It is only in the last two decades with district consolidation and an increase in federal and state funding that any real need to communicate with organizations has emerged. With changes in funding structure have come changes in instructional programs. 29 Thus, the need to interact with organ­ izations when making instructional decisions has grown. Second, it is interesting to note that while many educational decisions are collective decisions, rather than individual decisions, most of the diffusion research was based on the assumption that individuals adopted edu­ cational innovations. The startling fact is that even though research has taken the school systems as the adopting 29 Robert Neil Nelson, "A Field Study of the Means by Which the Michigan Department of Education Reduces Uncertainty in its Environment," (Ph.D. dissertation, Michigan State University, 1975). 46 unit, very limited attention has been paid to concepts related to organizational theory . ... The fact that school systems are organizations has been generally overlooked.30 Furthermore, the fact that for fiscal year 76 the Michigan Department of Education received 207 million dollars from the federal government for instructional programs is sufficient reason for inclusion of organizational sources in this study. 31 It is time to recognize that dealing with a bureaucracy is different than dealing with a principal, teacher, or curriculum salesman. Thus, this study distinguishes information-seeking from people in a large bureaucracy from other informal people sources. Studies of Attributes of Users The next major section of Chapter Two reviews the literature that is concerned with the effects of user characteristics on information-seeking behavior. Liter­ ature sources helped to operationalize several specific user characteristics that seemed to influence user be­ havior. Ideas from the research articles and books were then custom-designed to fit the uniqueness of Michigan. The final selection included status (central 30 . Rogers and Shoemaker, Communication of Innova­ tions , p. 61. 31 Quoted in Michigan Department of Education Fact Sheet, February, 1,9V5. Irregular single page publication of the Michigan Department of Education. 47 administrative staff, principals, and teachers), school district size, and geographic location of users. First is a review of the literature outside the field of education. Allen's studies on information usage by personnel employed in research and development proj­ ects is relevant as his conclusions support this study's use of educational status as a variable. Specifically, he found that "scientists lean more heavily on literature sources; while technologists depend on oral sources." 32 It seems that what one does may affect what sources he uses. Other research demonstrates that not only do per­ sons in different disciplines use information channels differently, but that different groups within a discipline also vary in their usage of information sources. For ex­ ample, one study of graduate students and faculty members in psychology departments in the United States provided evidence that In general, the faculty used journals and pro­ fessional meetings as information sources more frequently than did the graduate students, who often had to rely on interpersonal channels and textbooks to get their i n f o r m a t i o n . 33 32 33 Paisley, "Information Needs and Uses," p. 10. Nan Lin and William D. Garney, "Information Needs and Uses," Annual Review of Information Science and Technology, V o l . 1~, (Washington, D.C.: American Society for Information Science, 1972), p. 23. 48 Thus, this study adds additional support to consider the effect of educational status (central administrative staff, principals, and teachers) on information-seeking behavior. An additional study by Slater contributes further evidence for the inclusion of status. Slater chose grad­ uate and skilled nongraduate chemists for his sample. His results on the information-seeking behavior of these chem­ ists was different from the other studies of on-the-job scientists and technologists. He asserted that his sample was "highly dependent on documentary sources, particularly those that are wide in coverage, up-to-date and readily accessible." 34 Another study of research and development employ­ ees lends additional justification for the inclusion of the characteristics of status, district size, and geo­ graphical location. Rosenbloom and Wolek conducted a questionnaire study of the information sources used by two thousand engineers and scientists in thirteen divisions of four large corporations and from twelve hundred members of the Institute of Electrical and Electronics Engineers. They found that information sources were differentially used according to work setting, task, professional activ­ ity, and professional orientation. 34 35 Martyne, "Information Needs and Uses," p. 13. Paisley, "Information Needs and Uses," p. 12. 35 49 Next, a look at the educational research on user attributes as they affect information-seeking behavior. Diffusion research proved to be the most helpful in the selection of potentially relevant user characteristics. In the 1920 and 1930's Paul Mort and his col­ leagues at Columbia University had the monopoly on diffu­ sion research in education. They emphasized the effect that local school characteristics had on innovativeness. The typical variable of the adopting unit that was used was level of per pupil expenditure. Mort and his research assistants, through the use of mail questionnaires, deter­ mined a school's financial expenditures and then related this cost factor to the rate at which the school adopted a new educational technology like kindergarten, the "platoon school," and driver's training. In summarizing the approximately two hundred studies carried out under the direction of Mort, Donald H. Ross said, "If but one question can be asked, on the basis of the response to which a prediction of adaptability (adoption of innova­ tions) is to be made, the question is: How much is spent per pupil?"36 Although Mort and his co-workers contributed to an in-depth understanding of time lag in educational adop­ tion, their prolific undertakings did little to further 36 Rogers and Shoemaker, Communication of Innova­ tions, p. 61. 50 the understanding of why high expenditure districts adopted new xdeas first. ceptualized. 37 Their approach was noo narrowly con- For example, they failed to go beyond per pupil cost to discover what the characteristic links be­ tween cost and adoption were. Mort's work at Columbia University demonstrates four points of relevance to this study. First, it pro­ vides an historical context for educational efforts at understanding the relationship between users and informa­ tion. Second, the relationship between users and informa­ tion is more complex than can be adequately explained by a single variable— per pupil cost. Third, it makes clear the need to operationalize factors that affect per pupil cost. Fourth, Mort's endeavors are objected to as a basis for this study as the funding source for most new educa­ tional innovations has changed from solely local initia­ tive to state and federal funding. . . . this relationship of wealth to innovativeness was not supported by the results of Carlson's (1964) study of the diffusion of modern math, perhaps be­ cause much of the cost of adoption was provided by change agencies rather than the adopting s c h o o l s . 38 Later efforts in education have operationalized other user characteristics. For example, Carlson did a 37 Richard 0. Carlson, Adoption of Educational In­ novations (Eugene, Oregon: The Center for the Advanced Study of Educational Administration, 1965), p. 9. 38 Rogers and Shoemaker, Communication of Innova­ tions , p. 59. 51 study in Allegeny County, Pennsylvania and the state of West Virginia on the effects of the status of superintendents among their peers on rate of adoption of modern math. He concluded that there was a direct relationship between the two variables. The higher the superintendent's position in the social structure, the sooner he adopted modern math in his district. It should be noted that these results imply a relationship and not causality. Carlson's work is relevant to this study for two reasons. First, he directed attention to the influence that status has on seeking information. In order to adopt modern math, superintendents need information. status superintendents sought information first. information was sought from people. Higher Second, For the underlying assumption of Carlson's work follows the traditional basis of diffusion research— information flows through a social network. In addition, Carlson's study supports the use of the variable-geographic location. His data also suggested that regionalism may be a more influential variable than status. He studied the patterns of advice-seeking among the superintendents. . . . the advice-seeking contacts in the two geo­ graphical areas differed quite markedly. In Allegheny County, a county covering relatively few square miles and having a relatively high 39 p. 27. . Carlson, Adoption of Educational Innovations, 39 52 density of school superintendents, distance be­ tween advisees and advisors did not regulate the superintendents’ advice-seeking contacts . . . . In West Virginia, a state covering relatively many square miles and having a relatively low density of superintendents, distance between advisees and advisors did regulate the superin­ tendents 1 advice-seeking contacts . . .. When superintendents in the state asked one another for advice they tended to ask the superintend­ ents in the neighboring c o u n t i e s . 40 The differences between Allegheny County and the state of West Virginia parallel closely the differences between the lower and upper peninsulas of Michigan. Thus, the unique physical make-up of the state of Michigan dictates a par­ ticular need to include geographic location. Again, reference is made to the study by Ritten­ house into information utilization by local school dis­ tricts. He indicated "that the only district character­ istic that might be related to information need and use variables was enrollment size." 41 He postulated that dis­ trict size was an important variable as it affects the number of resources available to educators. The lack of educational research on the effects of user characteristics on information-seeking behavior alone justifies the macro-approach of this study. Those attributes that are most obvious as potential differences must be studied first. Thus, according to the available 40Ibid., pp. 42-44. 41 Rittenhouse, "Educational Information Uses and Users," p. 86. 53 literature, status, district size, and geographic loca­ tion are good beginning characteristics to explore. Furthermore, real world practicality must be con­ sidered. How much customizing of information channels for educational decision-makers can be adequately handled? Time-wise, economically, and politically it may be pos­ sible to individualize channels for educators in different status groups, districts, and geographical regions. But on the other hand, to provide all teachers who are musiclovers with audiotapes and all principals who like watch­ ing television with videotapes is an impossibility. The sheer complexity of user characteristics analysis could inundate us! holds. This may be an exaggeration, but the point There is a limit to the fine-tuning that can be practically handled in providing information for decision­ making . Thus, this study attempts a macro-analysis— a look at the user attributes of status, district size, and geo­ graphic location. Studies of Attributes of Information Sources This third major subsection of Chapter Two re­ views the literature that deals with the question of what attributes of the information source influence informationseeking behavior. Four criteria were selected as possible indicators of why Michigan educators might prefer one 54 source over another. A review of the research from dif­ fusion studies, information science, and the field of education indicated that four promising criteria for a better understanding of information source selection are accessibility, ease of use, technical quality, and amount of experience. Each of these criteria were defined for the respondents in the following way. Accessibility: A criterion on which information sources are judged according to the amount of time and/or distance to obtain the source. Ease of Use: A criterion on which information sources are judged according to the amount of difficulty in using the source once it is obtained or reached. Amount of Experience: A criterion on which infor­ mation sources are judged according to the frequency of use. Technical Quality: A criterion on which informa­ tion sources are judged according to the content charac­ teristics. The reader will recall that the theoretical basis for interest in these attributes was provided in part by Herbert Simon. 42 According to Simon, decision-makers satisfice; they choose the first adequate alternative from a limited number of perceived options. 42 York; If educators Herbert A. Simon, Administrative Behavior (New The Free Press, 1965). 55 behave in this way they choose the source which supplies adequate information, enough to satisfice, at least cost in time, energy, and money. Thompson's work also sug­ gested that the educational (organizational) environment encourages pursuit of efficiency in choosing sources. 43 Thus, it is hypothesized that educators will choose information sources which are easy to use, acces­ sible, and familiar (experience of user). Recall in Chap­ ter One the statement that "to conceptualize a system which requires the user to leap is to conceptualize a 44 failure." Yet Simon— and innumerable advocates of rational planning and decision-making— do state that maximizing is preferred. One should seek the best solution, not just the first one. Such users are primarily concerned with the quality of an information source. esis emerges. So a competing hypoth­ If educators maximize, they are more con­ cerned with the technical quality of a source than with either accessibility, ease of use, or prior existence. Several studies contribute to one side or the other of this conflicting picture of user behavior. First, a look at the literature outside the field of education. 43 The most influential work for determining James D. Thompson, Organizations in Action (St. McGraw-Hill Book Company, 1967), p. 92. 44 Interview, Ted. W. Ward, Prof., Institute of In­ ternational Studies of Education, Education Department, Michigan State University, February 5, 1975. Louis: 56 source attributes that might affect information usage habits was done by Gerstberger and Allen. 45 They studied criteria affecting information channel selection by engin­ eers. Each of the thirty-three participants were given a questionnaire each week for four weeks. They were asked to describe the order in which they sought different types of information in solving a problem during the week. In addition the engineers were given periodic questionnaires asking them to rank nine different information channels on the basis of accessibility, ease of use, technical quality, and degree of experience. Gerstberger and Allen's conclusions are support­ ive of this present study's attempt to describe relevant source attributes as accessibility, ease of use, degree of experience, and quality. First, they concluded that accessibility was the single most important determinant of use of a particular information source independent of the expected value of the information it provides. Second, both accessibility and perceived technical quality influ­ enced choice of source. Third, perception of accessibil­ ity was influenced by the amount of experience in using a source. 45 Peter G. Gerstberger and Thomas J. Allen, "Cri­ teria Used by Research and Development Engineers in the Selection of an Information Source," American Psychological Association, Vol. 52, No. 4, (August, 1968), pp. 272279. 57 The relevance of these results to this study are many. First, all four criteria need to be studied. Sec­ ond, these four criteria may be very interrelated in their affect on actual selection of an information source. Thrid, choice of information source reflects the respond­ ents perceptions; not an academic definition of source criteria. In another study, Friedlander found that sources that were perceived to be more accessible were used more often by members of the faculty of a medical school. 46 As previously mentioned, different information sources were preferred by graduate students and faculty in departments of psychology. 47 The question is raised as to whether source attributes played a role in information choice. For example, graduate students preferred inter­ personal sources. Yet, faculty members should be more accessible to each other than to graduate students. Per­ haps, the psychological cost is too high and thus ease of use plays a more influential role than accessibility. Furthermore, faculty members preferred information from professional meetings more than graduate students. Is this because the meetings are more accessible to them in terms of cost, time and distance? 46 47 pp. 22-23. Or is it because the Martyn, "Information Needs and Uses," p. 9. Lin and Garney, "Information Needs and Uses," 58 professors have more experience with these sources? As these examples illustrate, the ramifications of the in­ fluence of source attributes on information choice are large. Again reference is made to Allen's study of re­ search and development personnel. His work is particular­ ly relevant to this endeavor as he determined that the criterion of accessibility was more influential than the criterion of ease of use in the choice of information source. Allen demonstrated that internal sources were more extensively used than external sources. Allen's re­ sults seemed to surprise him as he intrepreted external sources as easier to use. He explained that the vendor, outside paid consultant, or the customer were anxious to receive responses to their needs. On the other hand, Allen recognized "there is a psychological cost in confessing an information need to internal sources, especially the tech­ nical staff."^8 According to another study by Trippi and Wilson, personal experience was an important indicator in deter­ mining type of information usage. Their conclusions were based on a study of 418 industrial firms in San Diego 49 County. 48 49 Paisley, "Information Needs and Uses," p. 11. Thompson, "Technology Utilization," p. 406. 59 There are three educational studies that have al­ ready been cited that need to be recalled at this time. Each of these studies provided some support for the im­ portance of the criteria of accessibility, ease of use, technical quality, and experience in information selec­ tion. Cormier, in his study of the decision-making habits of school superintendents, found that seventy per cent of the used information sources were rated as "very easily or easily accessible." 50 His work supported accessibil­ ity as a determining characteristic of user habits in in­ formation-seeking. Rittenhouse's study on decision-making at the local school district demonstrated that ease of use was a determining factor in information selection. "Important problems in the utilization of educational information include . . . understanding procedures for using informa­ tion systems and obtaining precise, structured information from school systems where change is occurring." 51 Carlson's study of the adoption process of modern math among school superintendents also is relevant to the present discussion of criteria that influence choice of information source. He noted that superintendents in West Virginia sought advice from fellow colleagues in their 50 Cormier, "Important Decisions of School Super­ intendents," pp. 30-32. 51 Rittenhouse, "Educational Information Uses and Needs," p. 81. 60 "immediate locality." 52 . . Thus, accessibility may at least be an important factor in determining source selection in sparsely populated areas like the upper peninsula in Michigan. As is evident from the above discussion, no edu­ cational research was found that supported experience with an information source or the quality of the content of the source as contributing factors in determining source selection. for three reasons. The decision was made to include these First is the empirical support. The findings from research in information science— Gerstgerger and Allen, and Trippi and Wilson— were influential. Sec­ ond, is the theoretical support from the works of Thompson and Allen. Third, is the logical support. Educators seem to be concerned with increasing one's experiences; as is evident from teacher-training programs, inservice meetings, workshops, and conferences. Common sense would indicate that educators should be interested in the quality of con­ tent. Thus, the final group of criteria for study are the source attributes of accessibility, ease of use, de­ gree of experience, and content quality. Whether the hunches used in selecting these criteria are correct is only determinable as the data of this study is analyzed. CO p. 41. , Carlson, Adoption of Educational Innovations, 61 Summary Instructional decision-making is dependent on (1) information sources, (2) characteristics of the decision­ maker, and (3) the relationship between the information sources and the decision-maker. Each of these areas— types of information sources, user attributes, and source characteristics— was described and discussed in relation to both literature outside education and from within the field. The literature dramatically emphasized the large use of people as sources for decision-making. Studies of literature sources gloomily admitted to the scanty depend­ ency on this source for information-seeking. Arguments for the inclusion of organizational sources were discussed. The next section of Chapter Two presented the literature that suggested the user characteristics of status, school district size, and geographical location as pertinent attributes for study. The emphasis is on sim­ plistic but relevant characteristics. The discussion im­ plied that too often research falls into the trap of creat­ ing a marvelously efficient means of predicting and con"X trolling absolutely trivial phenomena. This downfall is hopefully avoided in this research. The final section presented the literature that demonstrated the relevance of the criteria of accessibil­ ity, ease of use, experience, and content quality in 62 determining information source selection. It was noted that most evidence comes from the field of information science, as few studies in education have yet to focus on source attributes. CHAPTER III METHODOLOGY Introduction The intent of this study is to examine the current information-seeking behavior of professional educators. Within this general construct the specific purposes are three-fold. First is to identify and quantify the infor­ mation sources used by educators. The second is to exam­ ine and describe the influence of attributes of the pro­ fessional educator on choice of information source. The third is to identify and describe the effects of the char­ acteristics of the information sources on use of these sources. This chapter presents a description of the proces­ ses used to conduct this study. interest is defined. First, the population of Next the sample is identified. This section includes a description of the sampling techniques and a justification for the delineation of three sampling variables— status, region, and strata. Third, the nature of the prepilot and pilot studies are explained. Both of these studies served as major guides for the final presentation of items in the question­ naire. 63 64 Next, this chapter contains a description of the development of the questionnaire used as the measuring in­ strument for collecting data to answer the specific ques­ tions raised by the study. 1. What information sources do professional educators use? To what extent do they use these information sources? 2. Do certain characteristics of the professional educator influence the selection of informa­ tion sources? 3. Do attributes of the information sources affect information-seeking behavior? In addition, the specific approaches that were taken to strengthen the reliability of the questionnaire are am­ plified. The next portion of this chapter describes the de­ sign of the present study. The operational definitions of the variables in all hypotheses are given. The rela­ tionship among these variables is illustrated. The next step is to present the hypotheses in testable form. This presentation sets the stage for iden­ tifying the statistical procedures used to arrive at prob­ ability statements about the significance of each hypoth­ esis. Therefore, the ensuing procedure is to designate the models used to test the hypotheses. A discussion is included of the appropriateness of the models for the pur­ poses of this study. are stated. The assumptions for these models 65 Thus Chapter III provided the framework for analy­ zing the data of this study. Sampling techniques, instru­ mentation, reliability, design, testable hypotheses, and models— all are the necessary adjuncts to the process of analysis. Sample Population The population of interest was the five hundred thirty public school districts in the state of Michigan. These districts were exclusively kindergarten through grade twelve districts. Results attained from randomly sampling this population were generalizable to this en­ tire population as defined. Overall Sample Description The sample for this study was ten public school districts from the above defined population. size was arbitrary. The sample Cost limitations dictated that the sample be as small as possible while maintaining the op­ timal minimal size necessary to insure that analyses were not jeopardized. As stated earlier, this study was interested in determining the effects of characteristics of the profes­ sional educator on information-seeking behavior. In Chap­ ter II, it was reported that potentially influential user characteristics were status, geographical location, and 66 district size.^" Two of these variables— geographical lo­ cation and district size— were dealt with before the sample was chosen. These two variables were systematically handled to insure that they could be appropriately repre­ sented in the final sample selection. For example, before sample selection occurred, the state was divided into four geographical regions. Within each of these regions the school districts were stratified by district size. Thus, the school districts used in the sample for this present study were obtained through a multi-stage sampling process. This procedure pulled out these two variables for the purpose of quantification. This process also helped to reduce variability within the selected sam­ ple. Error variance was thus reduced and the sample size needed to test for significance was decreased. Regions Figure 3.1 illustrates the four regions used as the operational definition of the variable geographic lo­ cation. These regions were not unique, although their exact demarcations are thought to be original to this study. These regional boundary lines were the result of a synthesis from maps of the economic regions from the Gov­ ernor's Office, of the REMSES districts of the Michigan Department of Education, of regions of the state library ^Above, pp. 45-51. 67 REGION I REGION II REGION III REGION IV Figure 3.1— Regions Used in Sampling and in Analysis of Michigan Public School Districts. 68 system, of regions covered by commercial educational dis­ tributors, and of population density figures. Region I encompassed the Upper Peninsula of Mich­ igan. If information was to be distributed to an entire state, then the user habits of the professional educators of the Upper Peninsula of Michigan needed identification. Region II encompassed the upper half of the lower penin­ sula; Region III covered the lower half of the lower pen­ insula. A division was made between these parts of the lower peninsula on the basis of a hunch that the distinc­ tion in population density between the two areas made information habits different. 2 Region IV was exclusively the metropolitan city of Detroit. a unique entity. Detroit was treated as A large proportion— about 15 per cent— of the professional educators of Michigan were included within the jurisdiction of this city's school district. In this instance randomization proved absurd, as it was unjustifiable to combine Detroit with other areas of the state and take the chance of deleting it in the selection process. Strata District size was operationalized by using a mod­ ified version of the classification system developed by the Michigan Department .of Education. 2 Above, p. 50. School districts 69 fell into one of five population strata as defined in Table 3.1. For the purpose of this study Detroit is pulled out of the Metropolitan Core Stratum to create a sixth stratum, Detroit. This Michigan Department of Edu­ cation stratification system was used because it divided districts according to population size, 3 it was handy, and other research has effectively used these categories. This latter reason could encourage the analogies between aspects of this present study with findings of others. Sample Selection On the basis of these two variables— region and strata— sampling clusters were artificially created. Each of these possible clusters is illustrated in Figure 3.2. Some cells are empty because of the non-existence of cer­ tain population strata within certain regions. For exam­ ple, there were no metropolitan core or urban fringe dis­ tricts in either Region I or II. Furthermore, Region IV has empty cells as it contains only Detroit itself. From each of the "full cells," a school district was randomly selected. It should be noted that for choosing districts for strata classifications of city and town, Regions I and II were combined. final sample. This meant two less districts in the This decision was made because of the neces­ sity to compromise within each randomly selected school 3 Above, p. 50. Table 3.1— SCHOOL DISTRICT CLASSIFICATION SYSTEM of the Michigan Department of Education. Criteria for Selection Type of District Must meet one of the following criteria: Metropolitan Core Cities a. The community is the central city of a Michigan Standard Metropolitan Statistical Area. b. The Community is an enclave within the central city of a Standard Metropolitan Statistical Area, c. The community was previously classified as a Metropolitan Core City. Must meet one of the following criteria, regardless of size: Urban Fringe Towns a. The mailing address of the community is a Metro­ politan Core City or a City unless it is on a RFD Route. b. The community is within ten miles of the center of a Metropolitan Core City. c. The Community is within five miles of the center of a city. A community must have a population of 2,500 to 9,999 Rural communities impacted by large military instal­ lations nearby are also classified as towns. Table 3.1— Continued. Type of District Criteria for Selection Cities A community must have a population of 10,000 or more and not have been classified as a Metropoli­ tan Core City or Urban Fringe. Rural A community 2,500 or if City, Urban outside the Source: must have a population of less than the address is an RFD Route of a Town, Fringe, or Metropolitan Core, and lies perimeter defined under urban fringe. Classification system used by the Michigan Department of Education for the Educational Assessment Program. 72 Regions Upper Michigan Lower Michigan Detroit 4 1 2 3 Strata 4 A cluster from which a school district is selected. An empty cell. Figure 3.2— Cluster Sampling by Region and Strata. district and the fixed district, Detroit. The sheer com­ plexity of developing a list of professional educators of similar status within a given region-strata cluster was simply overshelming. In addition, in many areas of the state, political factors made the task of stratifying by status before selection an impossibility. Many districts 73 would not release names of teachers without special per­ mission from the local board of education. It should be noted that by nesting status within district, the school districts became the basic unit of analysis. This nesting has implications for the analysis procedures, as described later in this chapter. As the review of literature in Chapter II indi­ cated, people with different jobs tend to do things dif­ ferently. This study operationalized the variable of status by dividing professional educators into three clas­ sifications, which were defined as follows. Central Administrative Staff (CAS); For the pur­ pose of this study, central administrative staff included those professional educators who worked out of the central district office in a capacity that directly impacted on instruction. (e.g. curriculum coordinators, reading con­ sultants, directors of elementary principals . . .). Principal (P): A professional educator who man­ aged the instructional climate of one or more district buildings. Teacher (T); A professional educator who was directly in charge of instruction K-12. Selection of persons within each of these status classifications was quite complicated because of the range in size of these levels among the ten districts. The ex­ pedient outcome was ten persons in each status level— 74 central administrators, principals, and teachers— -per district. This totalled a potential sample of three hun­ dred . In all ten school districts, there were more than ten teachers in each. Thus, teachers in all districts were randomly selected. This made a final sample of one hundred teachers to whom questionnaires were sent. Likewise, in districts containing more than ten principals each, random selection was possible. In the five districts that had less than ten principals, the en­ tire group was used. The composite sample of principals to whom questionnaires were sent numbered seventy-three. Because of the range of responsibilities of the central administrative persons, selection was handled dif­ ferently than for either teachers or principals. For ex­ ample, central staff personnel like the business manager or the director of transportation were removed before se­ lection occurred. This was to avoid contaminating the sample with persons who had little direct involvement in instructional decision-making. In only two districts were there more than ten central administrators with appro­ priate titles. In these two districts persons were ran­ domly chosen. In the other districts, the entire group was selected. Questionnaires were sent to a total of forty-five central administrators. 75 It should be noted that Detroit was a sampling anomaly. This city had eight regions of its own. One of these regions was randomly selected and then ten persons from each of the three status levels were randomly chosen. Questionnaires were sent to a total of two hundred eighteen persons in ten districts in Michigan. This group formed the sample upon which this study is based. Pre-Pilot and Pilot Samples A pre-pilot and pilot studies were used to achieve a more reliable measuring instrument for this present study. The intent here is to describe the nature of the samples in the pre-pilot and the pilot. The use of these samples for increasing the reliability of the questionnaire is discussed in the next section on instrumentation. Pre-Pilot Study The pre-pilot consisted of semi-structured inter­ views with two central administrators, two principals, and two teachers in a purposefully selected school district. Sample pages from the interview schedule are contained in Appendix A. The major purpose of this pre-pilot was to gain some ideas on how to operationalize the concept of infor­ mation-seeking behavior for instructional decision-making. This study was helpful in three major ways. First, it helped to circumscribe the possible sources of information 76 used by professional educators in making instructional decisions. Second, it served to sharpen the focus of this study and to suggest possible hypotheses. Third, the pre­ pilot provided a learning experience in how to word indi­ vidual questions. Pilot Study The pilot study provided additional "fine tuning 11 for the present study. and relationships. It helped to identify variables The procedure for acquiring the dis­ tricts for the pilot study was identical to the process used to obtain the ten districts for the present study. Although sampling with replacement was used, no districts were the same for the pilot and the present endeavor, ex­ cept for Detroit which was purposefully selected for both studies. Within each of the ten districts used for the pilot, one person from each of the three status levels— central administrator, principal, and teacher— was inter­ viewed by telephone. by telephone. Thus, thirty persons were surveyed The selection of those persons was left to the discretion of the district contact person. A semi­ structured interview, as shown in Appendix B, was used. The interviews ranged in length from fifteen to forty min­ utes. The pilot study was particularly helpful in two major ways. First, the ninety instructional decisions re­ ported were used to form eleven major categories of 77 instructional decisions. Examples are community involve­ ment, evaluation/assessment, and building level curriculum. Second, the reported information sources were used to form a separate classification system of three information sources. For example, it was determined that the sources best fell into three categories— people sources, literature sources, and organizational sources. The more frequently given responses in each of these categories served to operationalize these three categories of information sources for the final questionnaire. For instance, oper­ ational definitions of organizational sources included the Michigan Department of Education and the Intermediate School Districts. Instrumentation The self-administered questionnaire was the basic instrument used for examining the major questions in this study. Thus, the questionnaire provided data on informa­ tion sources, user attributes, and information source char­ acteristics as they related to instructional decision­ making. The entire questionnaire is included as Appendix C. This section on instrumentation first delineates each major portion of that questionnaire in order to show how data was obtained on sources and attributes. Secondly, this section of Chapter III addresses the question of the reliability of the instrument. Ideas for constructing 78 the questionnaire were gathered during the pre-pilot and pilot studies. 4 The final product was an original ques­ tionnaire designed to be responsive to the specific pur­ poses of this present study. Questionnaire Description Information Sources. The first area of concern was the identification of information sources used by pro­ fessional educators. Data were obtained by asking the re­ spondents to report information sources used in seeking an answer to a current or pending instructional decision. The information sources were a fixed set that were to be rated on a scale from one to four. shown in Figure 3.3. The exact format is Each respondent was asked to follow this format twice. This list of information sources was randomly ordered to assure no biasing as to which sources were near the beginning of the list and what were near the end. In addition an "other" category was provided to the respond­ ent as an alternative to the use of the eleven given in­ formation sources. Attributes of Professional Educator. The second area of concern was the attributes of the professional educators. It is recalled that these user characteristics 4 In addition, the studies by Gerstgerger and Allen; McCracken; and Rittenhouse were particularly germane to the creation of this questionnaire. Above, p. 35 and p. 54. 79 Think about a decision that you have made recently or are in the process of making that affects the stu­ dents in your classroom or in your district. Please summarize this decision in the space below. Have any of the following information sources been used in making this decision? Please indicate your opinion on each source by circling the appropriate number. m OBI mb OBBB nxmt USED sonm u OSZD b e x b b x v e lt teachers fran district............... 2 3 4 professional Journals ............... 2 3 4 laterandlate district staff ........... 2 3 4 colleagues outside district ........... 2 3 4 special professional staff within district ....... 2 3 4 ............. 2 3 4 2 3 4 consultants from outside district . . . . 2 3 4 coonerclal publications 2 3 4 M)B personnel . . ..................... 2 3 4 publications by MBE, lntemediate . . . . district, local district 2 3 4 other _ _ _ _ _ _ _ _ _ _ 2 3 4 principals from district . central administrative staff ..... . . .... Figure 3.3— Questionnaire Format for Obtaining Ratings of Information Sources Used by Educators for In­ structional Decision-Making. aMDE: Michigan Department of Education. 80 — status, district size, and geographic location— were accounted for in the construction of the sample. Thus, each questionnaire was marked according to region, strata, and status levels. Attributes of Information Sources. The third area of interest was to obtain data on the perceived attributes of the information sources. Each respondent was asked to rate a fixed set of information sources on each of four source attributes. It is recalled from Chapter II that these criteria were defined for the respondents as follows: Accessibility: a criterion on which information sources are judged according to the amount of time and/or distance to obtain the source. Ease of Use: a criterion on which information sources are judged according to the amount of difficulty in using the source once it is obtained or reached. Amount of Experience: a criterion on which in­ formation sources are judged according to the frequency of use. Technical Quality: a criterion on which infor­ mation sources are judged according to the content characteristics. In Figure 3.4, a sample page of the questionnaire illus­ trates the exact format in which these four attributes were presented to the respondents. Reliability It was crucial to the meaningfulness of the data in this present study to establish the reliability of the 81 AMOUNT OF EXPERIENCE Please rate each of the following Information sources according to the amount of experience you have had as an educator In using each source. Please circle the number that corresponds to how often you have used each source. era u se s teachers fro® district ............... b a re ly USED s o o t nos USED 1 .2 3 professional journals ................. 1 2 3 lntaraedlste district staff ........... 1 2 3 colleagues outside district ........... 1 2 3 special professional staff within district 1 2 3 2 3 ........... principals from district.......... l central administrative staff ......... 1 2 3 consultants from outside district . . . . 1 2 3 cmasarcial publications............... 1 2 3 M B personnel.................. . . . l 2 3 publications by M>E, Intermediate . . . . 1 district, local district 2 3 other ..... ru q u z m r USED 1 Figure 3.4— Questionnaire Format for Obtaining Ratings of the Attributes (Ease of Use, Accessibility, Amount of Experience, and Content Quality) of the Information Sources. questionnaire as the measuring instrument. Relationships among variables have little meaning if one can not depend on the results of the measurement of the variables. Thus, concern for reliability comes from the need to establish 82 the accuracy of dependability in measurement. The subject is best approached by discussing the reliability estimates for the three major parts of the questionnaire— -open-ended question, fixed set of information sources, and the four attributes of the information sources. Reliability of Open-ended Question. First, the reliability of the open-ended question in eliciting re­ sponses termed instructional decisions must be determined. The pilot study was useful here. As mentioned earlier, the ninety instructional decisions reported during the telephone interviews were categorized into eleven types of instructional decisions as shown in Figure 3.5. Into these categories were placed the two hundred forty-nine instructional decisions from the open-ended question on the questionnaire. It was possible to place all but ten of these responses into these categories. The ability to place ninety-six per cent of these decisions into cate­ gories established by in-depth interviews with thirty on­ site professional educators indicated that this question­ naire was measuring information source responses to in­ structional decisions. Examples of the types of instructional decisions reported on the questionnaire are given on the following pages. 83 Category of Decisions Student/Class-Level Curriculum Changes Building-Level Curriculum Changes District-Level Curriculum Changes Student Evaluation and Assessment Program-Level Evaluation and Assessment Counseling-Discipline (students and/or parents) Community Involvement in Educational Programs Educational Equipment and Facilities Inservice Personnel/Labor Relations/Staffing Routine Administrative Instructional Decisions Figure 3.5— Categories of Instructional Decisions Used to Establish the Reliability of the Openended Question. STUDENT/CLASS-LEVEL CURRICULUM CHANGES "Whether I should individualize my math program." "Reading groups— just how far do you go in split­ ting them up? Also, what about supplementary materials?" "Do I want to set up interest groups?" BUILDING-LEVEL CURRICULUM CHANGES "Implement a career education program for all classrooms and grade levels in my building." "Decision to investigate the multi-age individual­ ized learning structures for possible implementa­ tion." 84 "Discontinue the use of commercial spelling texts." DISTRICT-LEVEL CURRICULUM CHANGES "Use of Title I monies (I was a member of a com­ mittee but had to give my viewpoint on whether we should have summer school or save money for use in the fall)." "I have been looking for ways to increase the vo­ cational experiences of the EMI classes." "I have tried to develop ways to humanize educa­ tion in the elementary schools." STUDENT-EVALUATION AND ASSESSMENT "To drop letter grades for 6 th grade students. Keeping them for 7th and 8 th grade, but phasing them in next two years." "Create an Objective Referenced Test in reading for 3rd grade." "How children are to be properly evaluated during parent-teacher conferences." PROGRAM-LEVEL EVALUATION AND ASSESSMENT "The decision concerned the evaluation of a pro­ posed positive action learning program for stu­ dents who do not function well in the classroom It • • • "We have decided to have a systematic analysis of the curriculum in each elementary school and junior high school." COUNSELLING-DISCIPLINE (students and/or parents) "The decision which results in having a pupil taken from a regular classroom and put in a special education room." "A parent was reluctant to send a fifth grade child on to sixth grade." 85 "Have students treat each other with more kindness." COMMUNITY INVOLVEMENT IN EDUCATIONAL PROGRAMS "Should we have workshops for parents on how to help children at home now or wait until the fall. What direction should these workshops take?" "Involvement of parents in data feedback on im­ plementation of a new up coming program for fall, 1976." "Establishing mini-courses one hour per week for sixth grade students using outside personnel." EDUCATIONAL EQUIPMENT AND FACILITIES "To build a green house for our science program." "To create outdoor learning centers." "Establish a room as a learning center in the building." INSERVICE "Helping native American educational aide for Indian children." "I had to write an article for the teacher's club newsletter for which I needed information, so I turned to a reliable source. This article would hlep teachers to in turn help students." "In process of deciding how to use new science materials." PERSONNEL/LABOR RELATIONS/STAFFING "Effective use of roving teachers such as art, music, physical education scheduling problems." "Decided to have my new student teacher begin teach­ ing one period a day from the very beginning." 86 ROUTINE ADMINISTRATIVE DECISIONS "The music teacher wanted the students in my class who make up 90% of the glee club to practice when city wide testing was scheduled for that time." "Whether to have 6-45 minute classes or five 55 minute classes changing each quarter and only re­ quiring three of four quarters of academic sub­ jects." "Have made the decision that kindergarten round­ up would be uniform in all elementary schools." Reliability of Definitions. Second, it was neces­ sary to establish whether each of the definitions of the attributes of information sources— ease of use, content quality, accessibility, and experience— were consistently understood by the informants. An ambiguous definition could be interpreted in several ways. This was a definite concern as ambiguity leads to higher error variance (error 5 due to chance), which decreases reliability. The pre­ pilot study was the greatest help in improving on the def­ initions of each of these criteria. For example, as a result of the pre-pilot interviews it was deemed necessary to better clarify the distinction between the attributes of accessibility and ease of use. The concept of acces­ sibility as defined in this study seemed easily understood until the concept of ease of use was introduced. Thus, in the final questionnaire the directions for the attribute "ease of use" read as follows: ^r = 1 _Ve//vin(j r = reliability estimate, Vg = error variance, V. , = systematic variance between indi­ viduals. lncu 87 Please rate each of the following sources according to how easy you feel each is to use once you have obtained the source. (Accessibility = how easy it is to reach the source. Ease of use = how easy the source is to use once you have obtained it). The pre-pilot interviewees had little difficulty under­ standing either the definition of the attribute of exper­ ience or content quality. Thus, no change was made in the definitions of these two variables for the question­ naire. Reliability of Responses. Third, it is important to establish whether the responses to the information sources (Figures 3.4 and 3.6) were measuring "true" indi­ vidual differences due to lack of clarity in interpreting each information source. Measurement aims at the "true" scores of individ­ uals. The only way that we can know that this aim is accomplished is by learning the "true" differ­ ences between individuals. When we say that re­ liability is the accuracy of a measuring instru­ ment, we mean that a reliable instrument more or less measures the "true" scores of individuals, the "more or less" depending on the reliability of the instrument. That "true" scores are meas­ ured can be inferred only from the "true" differ­ ences between individuals, although neither of these can be directly measured, of course. What we do is to infer the "true" differences from the fallible, empirical, measured differences, which are always to some extent corrupted by errors of measurement . . . Then the proportion of the "pure" variance to all the variance, "pure" and "impure," is the estimate of the reliability of the measuring instrument . 6 Fred N. Kerlinger, Foundations of Behavioral Research (Chicago: Holt, Rinehart and Winston, Inc., ±573), pp. 449-450. 88 Ambiguity in either distinguishing between the eleven in­ formation source categories or in interpreting the meaning of any source leads to random response# and " . . . liability of random numbers is approximately zero." the re7 Thus, what was desired to obtain a meaningful measure of infor­ mation source usage was accurate and consistent responses. Reliability estimates of the responses to informa­ tion source usage were obtained on two different sets of data by using an analysis of variance technique for obtaing ing the error variance and variance between individuals. First, a comparison was made between the two different sets of responses to the open-ended question (Figure 3.3) on instructional decision-making. Table 3.2 shows the re­ liability estimates on the usage of information sources when grouped as people sources, organizational sources, and literature sources. These estimates were obtained by subtracting the ratio of error variance to between indi­ vidual variance from one— the index indicating perfect re­ liability. V. , ind. These high reliability estimates were interpreted as indicating that professional educators used the same sources in responding to the situations they selected in the open-ended question. Thus, the individuals were ^Ibid., pp. 447-449. 89 Table 3.2— Reliability Estimates on Responses to Informa­ tion Source Usage Based on Open-ended Question. Information Sources People Sources .80 Organizational Sources .76 Literature Sources .82 consistent in their use of information sources. Individ­ ual information source items were therefore easily deter­ mined . The second procedure used to test the reliability of information sources in obtaining "true" individual dif­ ferences in information-seeking behavior was more complex, but more powerful. The former process could be objected to on the basis that there was not enough time lag between the first and second administration of the open-ended ques' tion. Thus, the assumption of independence would be vio­ lated. In the following procedure a comparison was made between the responses on information source usage between the situations in the open-ended question and given de­ cision-making situations. Professional educators were asked to indicate which of the same fixed set of information sources they would use in five given instructional decision-making sit­ uations. A sample page is provided in Figure 3.6. 90 SITUATION TWO: As a Teacher Imagine that you are the teacher in the following situation. Two volunteer parents have been assigned to your room five mornings per week. As a teacher, you are unsure as to how to most effectively make use Of their additional and needed assistance. Please circle the number which Indicates the extent to which you would use each of the following sources of Information to help make your decision. RARELY USE SOMETIMES USE 2 3 4 1 2 3 4 .1 2 3 4 1 2 3 4 . . 1 2 3 4 NEVES USE teachers from district . . professional Journals ......... 1 ....... .... intermediate district staff ......... colleagues outside district special professional staff within district ......... ....... FREQUENTLY USE ....... 1 2 3 4 central administrative staff ......... 1 2 3 4 consultants from outside district . . . . 1 2 3 4 commercial publications ............... 1 2 3 MDE personnel ......... principals from district . . . ............. 1 2 3 4 publications bp MDE, intermediate . . . . district, local district 1 2 3 4 2 3 4 other ....... 1 Figure 3.6— A Sample of a Closed Situation Used in Esti mating the Reliability of the Information Sources in Measuring Instructional Infor­ mation-Seeking . When the individual1s responses to these closed situations were compared with the respondents replies to the open-ended question, the reliability of these informa­ tion sources were again determined. These reliability 91 estimates are given in Table 3.3. The high degree in re­ liability of the information sources in measuring "true" individual differences provided the required basis for an accurate analysis of the data in the present study. Table 3.3. Reliability Estimates on Responses of Infor­ mation Source Usage Based on Open-ended and Closed Questions. Information Sources Reliability Estimates People Sources .89 Organizational Sources .76 Literature Sources .50 Administrative Details Permission to send questionnaires to each of the ten districts was obtained by calling the superintendent or the assistant superintendent in charge of research and/or curriculum. In a few cases, follow-up letters to boards of education were also necessary. There were no refusals by the original ten districts selected, although some districts had reservations about participation. The most common concerns were questionnaire overload and anonymity of district responses. A contact person was established with each of the ten districts. administrator. In all cases, the person was a central 92 The questionnaires were mailed separately to cen­ tral administrators and principals. In about half the dis­ tricts the questionnaires were also sent directly to the teachers. In the other districts, local policy did not permit the release of teacher lists. In these cases ran­ dom numbers were sent to the district contact person and from an alphabetized payroll list of all teachers, numbers were matched with names. Questionnaires were then de­ livered to individual teachers via intradistrict mailing service. A stamped, return address envelope was included with each questionnaire. In addition, each respondent was asked to mail a return-addressed post card with his/ her signature at the same time that he/she mailed the questionnaire. This procedure assured the anonymity of the respondent and yet maintained the ability to discover who had not returned questionnaires. In those districts where teachers' names were not released, the postcards were identified by the random number used for selection. The overall return rate for questionnaires was sixty per cent. The district total return rates are given in Table 3.4. Design In this section, the operational definitions of all the variables that are of interest to this study are 93 Table 3.4— Questionnaire Return Rates by District and Total. District Number Per Cent Return 34.5 53.8 34.3 83.3 71.4 76.0 73.9 01 02 03 04 05 06 07 08 09 88.2 60.0 54.5 10 Total (131 of 218) discussed. 60.0 The organization of these variables for analy­ sis is discussed. Operational Definitions of Information Source Vari­ ables The first set of variables that was of interest to this study was the information sources. It is recalled that the three major categories of information sources— people, organizations, and literature— were determined in part by interviewee responses in the pre-pilot and pilot studies. It is further remembered that the most frequently used information sources according to the pilot responses provided the operational definitions for these three infor­ mation source categories. Figure 3.7 is a list of the operational definitions of the information sources used for instructional decision-making. 94 — teachers from district — professional journals — intermediate district staff — colleagues outside district — special professional staff within district — principals from district — central administrative staff — consultants from outside district — commercial publications — MDEa personnel — publications by MDE, intermediate district, local district Figure 3.7— Operational Definitions of Information Sources for Instructional Decision-Making. aMDE: Michigan Department of Education Organization of Information Sources for Analysis For purposes of making the data more concise and interpretable, these twelve sources are group for analy­ sis into four groups as shown in Table 3.5. Operational Definitions of User Characteristic Variables The next set of variables— status, school district size, and geographic location— described characteristics of information users. The operational definitions were 95 Table 3.5— Four Categories of Information Sources. Teachers from district Colleagues outside district Special professional staff within district Principals from district Central administrative staff Consultants from outside district People Sources Organizational Sources Intermediate district staff MDE personnel Literature Sources Professional journals Commercial publications Publications by MDE, intermediate district, local district Sources filled in by respondents Other Sources detailed in the section on sampling. It is recalled that status included three professional levels— teacher, prin­ cipal, and central administrator. The variable geographic location was operationalized by dividing the state of Mich­ igan into four regions which are illustrated in Figure 3.1. Population was the criterion that determined six school district strata— rural, town, city, urban fringe, metropolitan core, and Detroit. Organization of User Charac­ teristics for Analysis The problem of design has been referred to as data discipline. "The design of research and the analysis of 96 data spring from the demands of research problems." 9 Thus, due to the small sample in Regions I and II, they were combined for analysis. Regions I and II were then referred to as Upper Michigan. Region III was referred to as Lower Michigan, and Region IV remained Detroit. Furthermore, because of the poor response from Detroit and the other large metropolitan core city, the strata Detroit, Metropolitan Core, and Urban Fringe were combined and referred to as the urban strata. The other three strata remained uncombined— city, town, and rural. The third set of variables— ease of use, acces­ sibility, degree of experience, and content quality— des­ cribed the attributes of the information sources themselves. It is recalled that these four characteristics were de­ fined in previous sections.^ Relationships Among Variables The design of a study implies that the relation­ ships among the variables be described. The first rela­ tionship of interest to this study was that which deter­ mines any quantitative differences in information usage by type. Thus, the average usage of information, organ­ izational information, and literature information was cal­ culated and then tested for any significant difference 9 Ibid., p. 628. ■^Above, p. 9. 97 among the three average use figures. This relationship is illustrated in Figure 3.8. Information Sources People________ Organization________ Literature na = 10 n = 10 n = 10 Figure 3.8— A Graphic Interpretation of the Relationship Between the Three Types of Information Sources. a n - means of ten school districts for each type of source. Secondly, this study was concerned with determin­ ing any significant relationship between attributes of professional educators and the selection of information sources. It is recalled that these characteristics of pro­ fessional educators were operationalized as status levels, geographic regions, and student population strata. Region and strata were interdependent variables, as certain strata tended to cluster in certain regions. Thus, these two independent variables needed to be tested separately, as the statistical procedures for analysis were not robust to the violation of the assumption of independence. There­ fore, this study considers user attributes as two sets of two independent variables— status and region, and status and strata. These two sets of independent variables were tested over three dependent variables— people sources, 98 organizational sources, and literature sources. Figures 3.9 and 3.10 illustrate the relationship between these two sets of independent variables and the dependent variable, people information sources.^ STATUS Regions Teacher Principal Central Administrator Upper Michigan Lower Michigan Detroit Figure 3.9— A Graphic Interpretation of the Relationship Between the Two Independent Variables— Region and Status— and the Dependent Variable— People Information Sources. Thirdly, this study was concerned with investigat­ ing the effects and the magnitude of the effects of four characteristics of information sources on the selection of information sources for instructional decision-making. Two specific relationship- were considered through the use of multiple regression analysis. First was the ■^The same two sets of relationships— status and region, and status and strata— are analyzed for the other two dependent variables— organizational information sources, and literature sources. 99 STATUS Regions Teacher Principal Central Administrator Urban City Town Rural Figure 3.10— A Graphic Interpretation of the Relationship Between the Two Independent Variables— Strata and Status— and the Dependent Variable— People Information Sources. relationship that indicated how much of the selection of information sources was presumable due to all four attri­ butes— ease of use, accessibility, experience, and content quality. Second, estimates were given of the relative influence of each of these independent variables on infor­ mation source selection. Testable Hypotheses This section of Chapter III simply states all of the hypotheses to be tested in this present study. The alternate hypotheses are stated in both word and symbolic form. As previously described, this study focused upon the types of information sources used by school districts. The hypothesis for this single variable— information source — is expressed as follows. Alternate Hypothesis: There will be found a differ­ ence in information source usage by Michigan school districts as measured by an average rating of usage of people sources, organizational sources, and literature sources. Symbolically: Legend: H1 : ^ f u3 yi^ = People source mean Vi2 - Organizational source mean Vi3 = Literature source mean This study also considered what attributes of pro­ fessional educators influenced the selection of informa­ tion sources. These attributes were operationalized as status levels, geographic regions, and student population strata. The following hypotheses express the effects of the two sets of fixed independent variables— status and region, and status and strata— on the dependent variable people information sources. 12 Alternate Hypothesis: Usage of people (organization, by literature) information sources by Michigan public school districts does vary according to the status level of the user— teacher, principal, and central administrator. Symbolically: Legend: H2 : ^ ^ 013 ^ 0 = Teacher status group mean 012 = Principal status group mean otg = Central administrator status group mean 12 The other two dependent variables— organizational information sources and literature information sources— are written into these hypotheses in parentheses to indi­ cate that the same relationship was tested for both of them. 101 Alternate Hypothesis: Usage of people (organizational, literature) information sources by Michigan public school districts does vary according to the user's geographic region— Upper Michigan, Lower Michigan and Detroit. Symbolically: Legend: : 3^ / 32 ^ ^3 f ® 3-^ = Upper Michigan group mean 32 = Lower Michigan group mean 33 = Detroit Alternate Hypothesis: Usage of people (organizational, literature) information sources by Michigan public school districts does vary according to the user's school district stratum— urban, city, town, and rural. Symbolically: Legend: H4 : yx / Y 2 1* Y 3 ? Y 4 / 0 y^ = Urban stratum group mean Y 2 = City stratum group mean Y 3 = Town stratum group mean Y 4 = Rural stratum group mean Alternate Hypothesis: Usage of people (organizational, literature) information sources by Michigan public school districts does vary according to the status level of the user interacting with his geographic region. Symbolically: Legend: H^: aS / 0 a = Status level means 3 = Geographic regions means Alternate Hypothesis: Usage of people (organizational, literature) information sources by Michigan public school districts does vary according to the status level of the user interacting with his/her school district stratum. Symbolically: Hg: ay / 0 102 Legend: a = Status level means Y = Strata level means Alternative Hypothesis: Usage of people (organization­ al, literature) information sources by Michigan public school districts does vary according to the perceived attributes of the sources— accessibility, ease of use, experience, and content quality. Symbolically: Legend: : 3^0 3 = the perceived attributes of the sources Alternate Hypothesis: The ability of the information source attributes to predict usage of people (or­ ganizational, literature) sources by Michigan public school districts varies among the four attributes. Symbolically: H g :M^^ ^ M 2 £ Mg ? M4 f 0 Legend: M^ = Amount attributed to experience M2 = Amount attributed to ease of use M g = Amount attributed to accessibility M^ = Amount attributed to quality Analysis This section of Chapter III presents the three models used for analysisof the questionnaire data. Reasons forthe choice of these models is presented, along with a discussion of the assumptions made about the data. Because these statistical formulas derived from these models are all commonly accepted, they are not included in this present study. 103 The first model of interest was the linear equa.. 13 tion. x = u + a = e Analysis of variance (ANOVA) is based on this model. This technique measures the significance of differences between units or levels of an independent variable. It was of interest to this present study to measure the difference in usage among three information sources— people, organ­ izations, and literature. Thus,, one way ANOVA was appro­ priate as it measured any meaningful differences among these three types of information sources. The model for one way ANOVA is based on three assumptions about the individual responses to information source usage. (1) Normality (2) Equality of Variance (3) Independence between and within cells There was no reason to assume that the assumption of nor­ mality was not met. Thus, for ANOVA the standard proce­ dure is to assume normality; as violations have minimal adverse effects on data results. Equality of variance can be assumed to be met if equal cell sizes are used. For this part of the study the cell sizes were equal, as 13 . . . These components of the linear equation are in­ terpreted in the following way. x = the individual score on information usage; u = the grand mean for information usage; a = usage effect; and e = individual sampling error. 104 each cell was composed of the ten district means. 14 In­ dependence within groups was controlled by sending indi­ vidual questionnaires to each respondent, most of whom were randomly selected. Independence between groups was achieved by having the cell sizes equal. The second model of interest was an extension of the linear equation. 15 Both a repeated measure analysis and a two way analysis of variance (ANOVA) were based on this model. x = u + a + 3 + a$ + e This technique measured the individual and inter­ action effects of independent variables on dependent vari­ ables. As explained in the design section of Chapter III, it was of interest to this present study to look at the effects of status, geographical region, and school dis­ trict strata on the usage of information sources. It is recalled that because of the redundancy in the concepts of region and strata the independent variables status and region, and status and strata were separately grouped. This model was appropriate as it provided the necessary format for measuring any significant effects of a single 14 15 Above, Figure 3.7. a = the effect of one of the independent vari­ ables, g = the effect of the second independent variable and a$ = the effect of the interaction of these two variables. 105 independent variable or the interaction of the independent variables on the dependent variables. This model is based on the same assumptions as a one way ANOVA— normality, equality of variance, and in­ dependence between and within cells. Normality can again be assumed, as there was no reason to doubt a normal dis­ tribution of scores. Furthermore, fifty-five per cent of the respondents who returned questionnaires were randomly selected and ninety-three per cent of them were from ran­ domly selected districts. As with a one-way ANOVA, a two way ANOVA is robust to minor deviations. The assumption of the equality of variance was met by using a finite correction factor to reduce the sampling variance. Because the population of districts from which the sampling occurred was finite, and because the sample from this finite population was small, a finite correction factor (1-a/A) 16 was used to reduce the sample variance for each district. Without the use of this finite cor­ rection factor, the sample variance in a small sample overestimated the population variance. Independence within groups was controlled by send­ ing individual questionnaires to each selected respondent within each of the nine randomly chosen districts, and Detroit. Independence between groups was achieved by a second weighting factor. Because the random sampling was "^a = number in sample, A = number in district. 106 by district and the unit of analysis of this part of the study must be the individual, proportionality between the cells was needed. This weighting was a series of complex steps carried out over the mean scores and the variance scores for each status level within a district. For the analysis of region by status, the weighting factor was the number of persons in each status level over the total num­ ber of persons in the same status level in all districts composing the same region. For the analysis of strata by status, the weighting factor was the number of persons in each status level over the total number of persons in the same status level in all districts composing the same strata. The third model of interest was an extension of the linear regression model. y = 30 + 3 ^ 17 + 32 X 2 + 33 X 3 + 34 X 4 + e Multiple regression analysis was based on this model. Mul­ tiple regression measured how much of the dependent vari­ able was due to the combined and singular influence of the independent variables, and it gave some idea of the rela­ tive amount of influence of each independent variable. It was of interest to this study to learn how much of 17 y = dependent'variable, 3g = multiple correla­ tion coefficient, $^X^* * * £4 X 4 = raw regression co­ efficient for each independent variable (the conditional distribution of Y given X. • • • X.), e = error of predic­ tion. 107 information usage can be predicted by the four defined criteria— ease of use, accessibility, experience, and quality. It was also of concern to establish the relative influence of each of these four criteria. Therefore, multiple regression was very appropriate for this analysis. Multiple regression is based on the assumptions of normality and equality of variance. There was no reason not to assume that the dependent variable— information usage behavior— was normally distributed and had equal variance for each value of the independent variables. In this case, violations of the assumptions are not extremely hazardous to distorting data as the technique of multiple regression is a robust technique that resists violations of both assumptions. Summary The sample for this present study was ten kinder­ garten through grade twelve public school districts in Michigan. The.sample was selected by means of a multi­ stage cluster sampling technique that insured accurate rep­ resentation of the variables— geographic region, status of professional educators, and school district size. With one exception, the school districts were randomly selected. i Within school districts, individual respondents were ran­ domly selected wherever, "real world" practicalities allowed it. 108 Pre-pilot and pilot studies helped to operation­ alize the concept of information-seeking behavior for in­ structional decision-making. These studies also provided the base line data for determining the reliability of part of the measuring instrument in this present study. An original self-administered questionnaire was the basic instrument used for measuring information-seek­ ing behavior of Michigan public school districts. An originally constructed set of information sources was used for determining what types of information were used by educators, and what attributes of information sources influenced educators in choosing among sources. Region, status, and strata were marked on each questionnaire sent, so that the data would also be analyzed on the basis of these three independent variables. The reliability of the questionnaire was strength­ ened by the use of the pre-pilot and pilot for refining definitions and instructions, and for determining a range of information sources used for instructional decision­ making. Furthermore, an analysis of variance technique was used to test the reliability of the responses to use of information sources. The obtained reliability estimates were extraordinarily high for an original questionnaire on its "first run." The administrative details for distributing and collecting the questionnaires were given. It was 109 necessary to insure that respondents answered the ques­ tions independently of other respondents. Too, it was important to maintain the anonymity of the respondents, but at the same time provide a means for keeping track of the number of returned questionnaires by region, strata, and status. An overall return rate of sixty per cent was achieved. The operational definitions of all the variables to be studied were given. In Figure 3.7 the operational definitions of the three types of information sources— people, organizations, and literature— were listed. The variable status was defined as three different profession­ al levels— teacher, principal, and central administrator. The variable geographic region was defined by dividing the state into four regions as shown in Table 3.1. The variable, school district size, was operationalized by using a modified form of the school district classifica­ tion system developed by the Michigan Department of Educa­ tion. The defined strata were rural, town, city, urban fringe, metropolitan core, and Detroit. The description of the information sources themselves included the attri­ butes of ease of use, accessibility, experience, and con­ tent quality. In two instances, the operational definitions for region and strata were clustered for data analysis. Re­ gion I and II, as shown in Table 3.1, were combined and 110 referred to as Upper Michigan. Region III remained Lower Michigan, and Region IV was Detroit. For analysis of strata effects, Metropolitan Core, Urban Fringe, and De­ troit strata were combined and labeled the urban stratum. The other definitions of strata remained the same— city, town, and rural. Next, this chapter looked at the relationships be­ tween these defined variables. First was the relation between the three types of information sources. Second was the relationship between the independent variables region and status, and strata and status; and the depend­ ent variable information sources. Third was the relation­ ship between the descriptive characteristics of informa­ tion sources and these sources themselves. All eight of the hypotheses to be tested in this study were written in both verbal and symbolic form. Chapter III presented the models to be used for analyzing the data. First was the linear equation model, second was the model based on an extension of the linear equation, and third was a modified form of the linear re­ gression model. The appropriateness of these models to the analysis of this study was discussed. Furthermore, the assumptions upon which these models are based were identified. Explanations of how these assumptions were met was included. Ill Thus, this chapter discussed in detail all of the vital processes necessary to make the analysis of the data in this present study meaningful. Sampling tech­ niques, pilot studies, instrumentation, reliability, de­ sign, testable hypotheses, and models— are all the impor­ tant adjuncts for introducing Chapter IV, the data analy­ sis. CHAPTER IV ANALYSIS OF FINDINGS Introduction In Chapter IV, each of the eight testable hypoth­ eses of the present study are restated. The analysis of each hypothesis or cluster of hypotheses is presented. When appropriate, ANOVA tables, multiple regression data, and mean scores are given for those hypotheses that attain statistical significance. Furthermore, the probability levels for acceptance or rejection are also mentioned. Finally, this chapter includes a discussion of the data findings. Consideration of these results is reserved for Chapter V. Usage of Types of Information Sources One major focus of this study was on the types of information sources used by school districts. The intent is to determine the extent to which districts seek infor­ mation for instructional decisions from people, organiza­ tions, and/or literature. It is recalled that sample se­ lection was done by randomly choosing school districts within predefined geographic regions and school district 112 113 strata. This sampling procedure defined the correct unit of analysis to be the school district, rather than the in­ dividual professional educator. Hypothesis Alternate Hypothesis: There is a difference in in­ formation source usage by Michigan school dis­ tricts as measured by an average rating of usage of people sources, organizational sources, and literature sources. Symbolically: H^: y^ £ ^ ^3 ^ 0 Legend: y^ = People source mean y 2 = Organizational source mean y^ = Literature source mean Data People information sources, organizational infor­ mation sources, and literature organizational sources were the three independent variables. The intensity of infor­ mation usage by the ten sample school districts was the dependent variable. Thus, the average usage scores of each district for people, organizational, and literature sources were the thirty means used for the univariate an­ alysis of variance. These are given in Table 4.1. Table 4.2 illustrates that there was a statistical­ ly significant difference in type of information usage among the districts. Therefore the null hypothesis was rejected in favor of the alternative. 114 Table 4.1— Mean Use of Three Information Sources by Each of the Ten Districts. Districts People 1 2.05 1.20 1.41 2 1.58 1.16 1.27 3 2.13 1.27 1.39 4 2.00 1.19 1.58 5 2.43 1.38 1.68 6 2.38 1.45 1.82 7 2.27 1.24 2.00 8 1.89 1.13 1.67 9 2.12 1.22 1.56 2.30 1.33 1.54 10 Organizations Literature Table 4.2— Analysis of Variance on Usage of Three Types of Information. F Variable Between Source Types Within Source Types *p < 2 1.854 27 0.040 46.626* .0001 Table 4.3 shows.the means over districts for people, organizational, and literature information sources. People sources were used more than either organizational 115 Table 4.3— Mean Use of Three Information Sources Summed Over All Districts. Mean Type of Information People Sources 2.1106 Organizational Sources 1.2560 Literature Sources 1.5922 or literature sources. Organizational sources were used less than either people or literature sources. To deter­ mine which of these mean differences were statistically significant, it was necessary to use a post hoc comparison procedure. The Tukey post hoc method was used as narrower confidence intervals are achieved for pair comparisons. At a confidence level of .05, the difference be­ tween all possible pairs among people, organizational, and literature sources were statistically significant. For if the value of zero was excluded from the computed confidence intervals, then the null hypotheses were rejected and the contrasts were statistically significant. These confi­ dence intervals are presented in Table 4.4. Discussion The data for this part of the present study shows that there were statistically significant differences in information usage among the three defined types of infor­ mation sources. The observed order of preference was 116 Table 4.4— Tukey Post Hoc Pair Comparisons. Contrast Tukey Value Confidence Interval People vs Organizations (.8546) ± .1842 .6704 People vs Literature (. 5184) ± .1842 .3342 In-service Experiences with Information v Sources > PROFESSIONAL EDUCATOR Figure 5.3— Two Important Means of Improving the Educator's Experiences with Information Sources. Suggested Directions for Future Research We enter upon an enterprise in most cases without full knowledge of all the factors that will enter into it, or all of the possible phases which it may develop. It is therefore of the utmost import tance to be prepared, to rightly comprehend the nature, bearings, and influence of such unforeseen elements when they shall definitely present them­ selves as actualities. If our vision is narrowed 167 by a pre-conceived theory as to what will happen, we are almost certain to misinterpret the facts and to misjudge the issue. If, on the other hand, we have in mind hypothetical forecasts of the various contingencies that may arise, we shall be the more likely to recognize the true facts when they do pre­ sent themselves.... The mind, having anticipated the possible phases which may arise, has prepared itself for action under any one that may come up, and it is therefore ready-armed, and is predisposed to act in the line appropriate to the e v e n t . 5 This study documented the information-seeking be­ havior of kindergarten through grade twelve school dis­ tricts. One observation is that school districts minimally use information of any type that is external to themselves. This is the documented state of affairs in spite of the given complexity of the instructional process. Chamberlin in 1897 aptly stated the instructional task of the educa­ tor. As there is an endless variety of mental processes and combinations and an indefinite number of orders of procedure, the advantage of different methods under different conditions is almost axiomatic. This being granted, there is presented to the teacher (educator) the problem of selection and of adaptation to meet the needs of any specific issue that may present it­ self. It is important therefore, that the teacher shall have in mind a full array of possible condi­ tions and states of mind which may be presented, in order that, when any one of these shall become an actual case, he may recognize it and be ready for the emergency .6 Future research may explore the question of why educators use so little information by looking at the influence that the school (organization) environment has on information5 Chamberlin, p. 18. 6Ibid., p. 23. 168 seeking behavior. Operational definitions of possible in­ fluential factors should be identified and then measured for their influence. This study illustrated that persons of higher educational status used all types of information more. Is this a result of their position? Does the organiza­ tional environment reward and/or expect central adminis­ trators to seek information? On the other hand, is facil­ ity with using information a prerequisite to advancement up the educational ladder? The research question is simply one of whether the position of administration requires to seek more information or whether educators who seek infor­ mation are advanced to administrative positions. Research on this question would help to define the influence that the educational (organization) environment has on informa­ tion-seeking behavior. Regardless of status, professional educators tended to use people information sources the most. This study also showed that experience was the best single predictor of use for all three types of information sources. This observation generates additional questions to be studied. Do people use people sources more frequently because they have more experience with people? If so, then given sim­ ilar experiences with all three types of sources would they then use the sources interchangeably? 169 The conclusions of the present study also point out that persons who use literature sources are very likely to use people and organizational sources as well. Further­ more, educators who use organizational sources are very likely to use people and literature sources as well. In contrast to these findings is the observation that use of people sources does not predict use of literature or organ­ izational sources. One potential research question is the determination of why persons using literature or organiza­ tional sources are more prone to use all three types. One set of hypotheses to be studied should deal with the criterion of experience. Does the "all type" user group have more experiences will all three sources? A competing set of hypotheses would look at user charac­ teristics. Do people who use literature, for example, have better developed reading skills, such as excellent skimming ability, speed or better comprehension? Do per­ sons who prefer to use only people sources feel more com­ fortable with the additional sensory stimuli provided by interacting with people? The above are some examples of the kinds of ques­ tions that could be investigated based on the present information-seeking behavior of professional educators. A Final Summary This study described the current information seek­ ing behavior of professional educators when making 170 instructional decisions. Through this description, the study hopefully added to the understanding of human be­ havior . APPENDIX A 172 Case # _ ______ A. State in one sentence, if possible, a tnijor professional problem or concern in the area of Instruction requiring a decision. Limit this problem or concern to one that has been pressing this month. B. To solve this problem or concern did (do) you need information? Yes C. ____ __ No Please place a check beside one type of information that you used (are using). demographic studies (annual report reports of data)' • .expert opinion _____ experimental research (published) _____ survey "how to do it" information _____ community opinion ____ "why" information legislation _____ district, building policy _____ experimental research (unpublished) 173 D # reference to the type of information that you checked above, please recall the search order of all information channels consulted even though some or possibly all yielded little or no relevant information. (Place a 1 in the space provided in front of the source you consulted first; place a 2 in front of the source you consulted second; and so forth.) in “ person filling similar professional role (e.g., teacher to teacher or principal to principal) professional organization consultant (MEA, MSBA, MAS A, . . . ) school board member librarian _ teacher(s) ■ ... . principal central administrative staff Michigan Department of Education parents ■ external expert (a person employed in a job other than in schools or related agencies who does not fall into an above category) committee meetings seminar, conference, university class literature published by professional organization to which you belong literature of education (books, professional pamphlets, etc. that are publlcally accessible) external literature (a book, pamphlet, etc. that was not expressly written for educators) experimentation (tried a new mode of organization, idea, etc.)hire "outside” expert E. For those sources you have ranked, please indicate the method of contact using the code below: A B C - person to person conversation by phone by letter 174 Accessibility I am interested in ascertaining how you would rank (from 1-17) the following information channels (i.e., sources of information) in terms of their accessibility to you (time and/or distance to obtain the source of information). .1 ■ most accessible 17 ■. least accessible ' ’ person filling similar professional role (e.g., teacher to teacher or principal to principal) professional organization.consultant (MEA, MSBA, MASA, . . . ) ‘ _____ school board member librarian ' teacher(s) _____ principal • central administrative staff _____ Michigan Department of Education ' ' _____ parents external expert (a person employed in a Job other than in schools or related agencies who does not fall into an above category) committee meetings • seminar, conference, university class _____ literature published by professional organization to which you belong ' literature of education (books, professional pamphlets, etc. that are publlcally accessible) ' external literature (a book, pamphlet, etc. that was not expressly written for educators) experimentation (tried a new mode of organization, idea, etc.) _____ hire "outside" expert APPENDIX B 175 r Person role: Decision #1 Decision #2 Decision #3 Type Type Sources Sources Sources: c. c. c. 176 Type: d. d. e. 1. Type: 1. Type: 1. Type: 2. Sources: 2. Sources: 2. Sources: a. a. a. b. b. •>. c. c. c. d. d. d. e. e. e. APPENDIX C A SAMPLE SURVEY of Michigan School Districts on Current Use of Information Sources for Educational Decision-Making 178 179 -i- Dear Educator: Thank you for your cooperation in filling out this questionnaire on decision-making. The purpose of this survey is to determine ways to share educa­ tional materials within Michigan. This survey will help to determine information sources educa­ tors currently use to make educational decisions for their students. Please approach this task from the perspective of what you currently do; not what you would like to do given more time, money, etc. Also, please keep this questionnaire your opinion and not a group consensus. When you return your questionnaire in the envelope provided, please drop the post card in the mail also. That way we will know your questionnaire is coming, but will not be able to identify any specific one as yours. The only identification is by region of Michigan and by population strata (rural, town, city, urban, metropolitan). Ano­ nymity at both district level and individual level is insured. Thanks again! Looking forward to your helpful and candid comments. 180 The following list of information sources is used throughout the questionnaire. If ever you need to mention additional information sources to answer a question, please indicate under the heading "other." Please take time to look this list over as it will be referred to in an abbreviated form throughout the rest of the questionnaire. Information Sources teachers publications by professional educational organizations intermediate school district (personnel and/or workshops' colleagues outside of district (phone calls, letters, committee meetings, etc.) special professional staff within district (speech, librarian, counselor, etc.) principals central administrative staff (superintendent . . . ) consultants (outside district) commercial educational publications (Instructor, Leamin Michigan Department of Education (personnel and/or workshops) publications by MDE, ISD, or school district other 181 Think about a decision that you have made recently or are in the process of making that affects the stu­ dents in your classroom or in your district. Please summarize this decision in the space below. Have any of the following information sources been used in making this decision? Please indicate your opinion on each source by circling the appropriate number. D D HOT IB B USED nzm,r OSED OSD esibbxvbly bonhhax teachers fro* district • • • • ....... 2 professional journals ............... 2 3 d intermediate district staff . . . . . . . 2 3 d colleagues outside district ........... 2 3 d special professional staff within district ........... 2 3 d ............. 2 3 d 2 3 d consultants from outside district . . . . 2 3 d caenerclel publications ............... 2 MDE personnel . . ..................... 2 publications by MDE, intermediate . . . . district, local district 2 other 2 principals from district central administrative staff ....... • • • • 3 d 3 3 d d 3 3 4 d 182 Please repeat what you have just done on Page 1 but with another decision that you have recently made. Please summarize the decision below. Indicate again your reaction to each of the following information sources with regard to the decision you have just mentioned. Please indicate your opinion on each source by circling the appropriate number. D m EOT dsb teachers fro* district............... 1 USED OSES SOMEWHAT EXTBMIVBI 2 3 A obbd BUBBLY professional journals ............... 1 2 3 A Internedlace district staff ........... 1 2 3 A colleagues outside district . . . . . . . 1 2 3 4 ........... 1 2 3 A ............. 1 2 3 A 1 2 3 4 1 2 3 4 coanercial publications ............... 1 2 3 4 ISE personnel ........................ 1 2 3 4 publications by MDE, Intermediate . . . . district, local district 1 2 3 4 other 1 2 3 4 spaelal professional staff withIs district principals from district central administrative staff ....... consultants fro® outside district . . . . .... SECTION TWO 183 184 ACCESSIBILITY Please rate each of the following information sources according to how accessible you feel the source is, i.e., accessible in terms of time and/or distance to obtain the source. Please circle the number that corresponds to how accessible each source is for you. NEVER ACCESSIBLE HOI VERY ACCESSIBLE SOMEWHAT ACCESSIBLE ALWAYS AT MI FINGERTIPS teachers from district • • • 2 3 professional journals . . . 2 3 Intermediate district staff 2 3 colleagues outside district 2 3 special professional staff • within district 2 3 principals from district 2 3 • • 2 3 consultants from outside district 2 3 4 commercial publications ....... 2 3 4 MDE personnel . . . 2 3 4 2 3 4 .... central administrative staff ........... publications by MDE, Intermediate district, local district other ________________________ AMOUNT OF EXPERIENCE Please rate each of the following information sources according to the amount of experience you have had as an educator in using each source. Please circle the number that corresponds to how often you have used each source. SEVER USD RARELY OSD SOMETIMES USD teacher* from district • • • 2 3 professional Journals • • • 2 3 Intermediate district staff 2 3 colleagues outside district 2 3 special professional staff within district 2 3 principals from district ............. 2 3 central administrative staff ......... 2 3 consultants from outside district • • • • 2 3 cosnerclel publications ............... 2 3 KDE personnel ........... • ......... 2 3 publications by MDE, Intermediate . . . . district, local district 2 3 other............... ....... FKBQUUTLY USD 186 EASE OF US1HG THE SOURCE Please rate each of the following sources according to how easy you feel each is to use once you have obtained the source. (Accessibility = how easy it is to reach the source. Ease of use = how easy the source is to use once you have obtained it.) Please circle the number that corresponds to how easy each source Is to use once you have the source RAKELY USE SOMETIMES USE teachers from district ........ 2 3 4 professional journals 2 3 4 intermediate district staff . . - 2 3 4 colleagues outside district . . 2 3 4 . . . 2 3 4 principals from district . . . . 2 3 4 central administrative staff . . 2 3 4 consultants from outside district 2 3 4 commercial publications ....... 2 3 4 2 3 4 publications by MDE, intermediate district, local district 2 3 4 other 2 3 4 NEVEH USE ' ....... special professional staff within district MDE personnel ................. .... 1 FREQUE USE TECHNICAL QUALITY Please rate each of the following sources according to the quality of the information you received from it. (Does the information work for you?) Please circle the number that corresponds to your assessment of the level of quality of each source. NEVER HIGH QUALITY RARELY HICK QUALITY SOMETIMES HIGH QUALITY ALWAYS HIGH QUALITY teachers from district • • • 2 3 4 professional journals , . . 2 3 4 Intermediate district staff 2 3 4 colleagues outside district 2 3 4 3 4 special professional staff vithin districc principals fron district .... ' 4 . . 4 consultants from outside district 4 commercial publications ....... 4 MDE personnel ................. 4 publications by MDE, intermediate . district, local district 4 central administrative staff other ____________ BIBLIOGRAPHY BIBLIOGRAPHY Allen, Thomas J. "Information Needs and Uses." Annual Re­ view of Information Science and Technology. Vol. 4. Chicago: Encyclopedia Britannica, Inc., 1969. Allen, Thomas J., and Cooney, S. "Institutional Roles in Te’ chnology Transfer: A Diagnosis of the Situa­ tion in One Small Country." R and D Management 4:1(October, 1973), p. 44-51. Bernard, H. Russell. "Scientists and Policy Makers: An Ethnography of Communication." Human Organization. Vol. 33 (Fall, 1974), 261-275. Callahan, Raymond E. Education and the Cult of Efficiency. Chicago: The University of Chicago Press, 1962. Carlson, Richard 0. Adoption of Educational Innovation. Eugene, Oregon: Center for the Advanced Study of Educational Administration, 1967. Cormier, Roger A. "Information in Each Operational Area for Important Decisions of School Superintendents." Ph.D. dissertation. University of Alberta. 1970. Dickson, Paul. 1971. Think Tanks. New York: Ballantine Books. Goldhar, Joel D. "Information, Idea Generation and Tech­ nological Innovations." Paper presented at the NATO Advanced Study Institute on Technology Trans­ fer. Every, France. June 25-July 6, 1973. Gerstberger, Peter G. and Allen, Thomas J. "Criteria Used by Research and Development Engineers in the Sel­ ection of an Information Source." Journal of Applied Psychology. Vol. 52 (August, 1968), 272279. Kerlinger, Fred N. Foundations of Behavioral Research. Chicago: Holt, Rinehart and Winston, Inc., 1973. 189 190 Lin, Nan and Garvey, William D. "Information Needs and Uses." Annual Review of Information Science and Technology 1 Vol. 7. Washington, D.C.: American Society for Information Science, 1972. Martyn, John. "Information Needs and Uses.'! Annual Review of Information Science and Technology. Vol. 9~. Washington, D.C.: American Society for Infor­ mation Science, 1974. Mason, Robert G. "Securing Clientele Acceptance and Co­ operation." Rural Sociology. Vol. 29 (March, 1964) pp. 4o-inn Miles, Matthew B., ed. Innovation in Education. New York Teachers College, Columbia University, 1964. Nelson, Robert Neil. "A Field Study of the Means by Which The Michigan Department of Education Reduces Un­ certainty in its Environment." Ph.D. dissertation Michigan State University, 1975. Newman, Peter C. The Canadian Establishment. Toronto: McClelland and Stewart Limited. 1975. Paisley, William J. "Information Needs and Uses." Annual Review of Information Science and Technology. Vol 3. Chicago: Encyclopedia Bntannica, Inc., 1968. Paisley, William J. "Improving a Field Based (ERIC-Like) Information System." Journal of the American Society for Information Science. (November-December, 1971), 399-408. Rittenhouse, Carl H. "Educational Information Uses and Users." AV Communication Review. Vol. 19 (Spring 1971), pp. 76-88. Rogers, Everett M. and Shoemaker, F. Floyd. Communication of Innovations. New York: The Free Press. 1971. Rothwell, R. and Robertson, A.B. "The Role of Communica­ tions in Technological Innovation." Research Policy. 2 (October, 1973), 204-225. Ryan, Bryce and Gross, Neal C. "The Diffusion of Hybrid Seed Corn in Two Iowa Communities." Rural Sociol ogy. 7 (March, 1943), 15-24. Simon, Herbert A. Administrative Behavior. The Free Press, 1957. New York: 191 Thompson, Charles W. "Technology Utilization." Annual Review of Information Science and Technology. Vol. 10. Washington, D.C.: American Society for Information Science. 1975. Thompson, James D. Organizations in Action. McGraw-Hill Book Company. 1967. Toronto: Vinsonhaler, John F. and Moon, Robert D. "Information Systems: Applications in Education." Annual Re­ view of Information Science and Technology" Vol. 9. Washington, D.C.: American Society for Infor­ mation Science. 1973. Ward, Ted W. Professor at Michigan State University, In­ stitute of International Studies of Education. Private interview held on February 5, 1975. General References Baker, Frank B. "Computer-Based Instructional Management Systems: A First Look." Review of Educational Research. Vol. 41 (February 1971), 51-70. Burchinal, Lee G. "The Educational Resources Information Center: An Emergent National System." The Journal of Educational Data Processing. Vol. 7 (April, 1970), pp. 55-67. Carlson, Richard 0.; Gallaher, Art. Jr.; Miles, Matthew B.; Pellegrin, Roland J.; and Rogers, Everett M. Change Processes in the Public Schools. Eugene, Oregon: Center for the Advanced Study of Educa­ tional Administration. 1965. Carlson, Richard 0. School Superintendents Careers and Performances. Eugene, Oregon: University of Oregon. 1972. Havelock, Ronald G. Planning for Innovation. Ann Arbor, Michigan: Institute for Social Research. 1971. Havelock, Ronald G. The Change Agent's Guide to Innova­ tion in Education. Englewood Cliffs, New Jersey: Educational Technology Publications. 1973. Lippey, Gerlad; Toggenburger, Frank; and Brown, C. Douglas. "Computer-Based Exercises Aid Classroom Teachers." Association for Educational Data Systems Journal (AEDS). 4 (March, 1971), 75-84. 192 Mathies, Lorraine. "The Educational Resources Information Center: An Agent of Change." The Journal of Educational Data Processing. Vol.”7 (April, 1970), pp. 123-129. Michigan Department of Education. "Request for Proposal: Instructional Resource Support System." Fall, 1974. Murdick, Robert G. and Ross, Joel E. Information Systems for Modern Management. Englewood Cliffs, New Jersey:Prentice-Hall, Inc., 1975. Oppenheim, A.M. Questionnaire Design and Attitude Measure­ ment. New York: Basic Books, Inc. 1966. Vinsonhaler, John F., and Moon, Robert D. "Information Systems: Basic Concepts." Handbook of Computers and Information Systems in Education and Social' Science (yet to be published). Course material Education Department, Michigan State University. 1975.