70-9502 BROWNLEE, Richard Wiley, 1928COMMUNICATION NETWORKS AMONG THE PROFESSIONAL STAFF OF MICHIGAN'S STATE DEPARTMENT OF EDUCATION. Michigan State University, Ph.D., 1969 Education, administration University Microfilms, Inc., Ann Arbor, Michigan COMMUNICATION NETWORKS AMONG THE PROFESSIONAL STAFF OF MICHIGAN'S STATE DEPARTMENT OF E d J c ATION By Ri Wiley Brownlee A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1969 ABSTRACT COMMUNICATION NETWORKS AMONG THE PROFESSIONAL STAFF OF MICHIGAN'S STATE DEPARTMENT OF EDUCATION By R. Wiley Brownlee This field communication study traces communication networks among Michigan's State Department of Education professional staff regarding eight selected publications of the Department. A questionnaire gathered data from 246 subjects or 87 percent of the population from eight office work units, and the following hypotheses were explored: 1. Indicated communication regarding publications tends to be with subordinates rather than with superiors within the State Department of Education hierarchy: supported. un­ Despite an increase in average communication events as one moved up the hierarchal structure, subjects indicated more communication with superiors than with sub­ ordinates over all 2,777 communication events and in indi­ cated channels of first communication. 2. Some publications will be excluded from the communication networks of some office units: for five of eight office units. supported, R. Wiley Brownlee 3. Subjects Indicating mass media rather than interpersonal channels as first channels of communication will have earlier awareness of publications: unsupported, for seven of eight publications. 4. Subjects indicating interpersonal rather than mass media channels as channels of greatest communication will have a more positive attitude toward a given publication: supported, for five of eight publications (though none were significant in a one-tailed t test at the .05 level). 5-A. Greater average communication per subject in an office unit will correlate with greater consensus in atti­ tude within that office unit toward the publication: supported, with a Pearsonian correlation of .79* 5-B. Average communication per subject in an office unit will correlate positively with the degree to which all sub­ jects in that unit are included in its communication network: 6-A. supported, with a Pearsonian correlation of .62. Mean length of employment in the Department of Education of a unit's staff will correlate positively with that unit's using Interpersonal more than mass media communication channels: unsupported. The Pearsonian correlation was -0.^0, though only for employment within the Department. The correlation neared zero (-0.003) when considering total length of employment in the field of education. R. Wiley Brownlee 6-B. Mean length of employment in the Department of Education of a unit's staff will correlate positively with the degree to which all subjects in that unit are included in its communication network: unsupported. The Pearsonian correlation was -0.82, though only for employment within the Department. When considering length of employment in the field of education, the correlation was -0.19Communication patterns were identified with a Fortran-sociometric-distance-matrlx program with the following findings: 1. The 64 most-integrated subjects included relatively more supervisory personnel than consultants than the total population. 2. The 28 central-persons who were highest in both integration and participation in close cliques of four or more persons Included relatively more middle supervisory personnel than did the 64 most-integrated subjects. 3* A sociogram of the 28 central-persons showing their inner- and inter-office relationships revealed t h a t : (a) some of the 28 central-persons were soclometrlc Isolates in their own office units, (b) some office units had contact with as many as four other units, (c) some units had no representatives among the 28 central-persons or no contacts with any other office unit, and (d) some office units Included nearly all members within their communication network. R. Wiley Brownlee The following recommendations could help the Department better utilize identified communication patterns: 1. Conduct periodic and specific staff meetings which emphasize Interpersonnel communication. 2. Conduct orientation programs for advising all new personnel of major issues, duties, and obligations of the Department, and encouraging them to use the various formal and Informal channels of communication available. 3.' Furnish special workshops for middle management designed to Increase sensitivity and communication skills of unit, section, and division chiefs. Seek to involve all work units in the D e p a rtm e n t ’s communication structure through planned interaction between professional staff members of the several bureaus, empha­ sizing cross seeding of information, problem sharing, perspectives, and cooperative problem solving. ACKNOWLEDGEMENTS To Dr. John Suehr, my guidance committee chairman, I extend my deep appreciation. Through him, I have learned aspects of communication and human interaction which have helped me grow as an educator. To the other members of my doctoral Dr.Everett Rogers, Dr. Fred committee— Vescolani, and Dr. Sheldon Lowry— I am grateful for guidance and assistance. I am indebted to the cooperation of the profes­ sional staff of the State of Michigan's Department of Education for their suggestions and participation. Loving thanks are particularly expressed to my wife, Luan, and our son, Kenneth, for their patience, understanding, and unfailing encouragement during our doctoral program. ii TABLE OP CONTENTS Page L I S T OP T A B L E S ......................... v L I S T OP F I G U R E S ..........................................vii CHAPTER I. I N T R O D U C T I O N .................................. Problem ......... . . . . . . . A s s u m p t i o n s ............... Hypotheses .................................. D e f i n i t i o n s ............. Overview of Organization .................. L i m i t a t i o n s ............................... Summary .................................... II. III. IV. V. 1 1 2 3 *1 7 7 8 RELATED IDEAS AND L I T E R A T U R E ................ 11 PROCEDURES U S E D .............................. 25 P u b l i c a t i o n s ............................... S u b j e c t s .................................... I n s t r u m e n t a t i o n ........................... Collection of D a t a ............ *............ 25 27 30 32 R E S U L T S ....................................... 34 Hypothesis 1 ............................... Hypothesis 2 ............................... Hypothesis 3 ............................... Hypothesis 4 ............................... Hypothesis 5 - A ............................. Hypothesis 5 - B ............................. Hypothesis 6 - A ............................. Hypothesis 6 - B ............................. Sociometric Findings .................... 34 42 47 49 52 56 59 63 67 CONCLUSIONS, IMPLICATIONS, AND QUESTIONS . . 82 C o n c l u s i o n s ............................... I m p l i c a t i o n s ............................... Questions .................................. 82 85 86 lii CHAPTER VI. Page IMPRESSIONS AND RECOMMENDATIONS ............ 89 I m p r e s s i o n s ............................... Recommendations ........................... 89 94 BIBLIOGRAPHY ........................................ 99 APPENDIX A. B. METHODOLOGICAL N O T E S ....................... 102 Preparing Data for A n a l y s i s ............. Hypotheses Testing . .................... Soclometrlc Analysis ............. ... F o r m u l a e .................................. 102 103 106 108 THE I N S T R U M E N T .............................. 110 Base Data Q u e s t i o n n a i r e .................. Publication Recognition List ........... ......... List of Communication Channels 110 Ill 112 lv LIST OF TABLES Table 1. Page Distinguishing Characteristics of Interpersonal and Mass Media Commu­ nication C h a n n e l s ........................ 2. Number of Subjects by Office and Position 3. Number of Indicated Communication Channels by Category for Each P u b l i c a t i o n ...... 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. . 24 . 29 38 Distribution of Lateral, Downward, and Upward Communication ...................... Distribution of Subject Communication Which Was More Downward or More U p w a r d ...... 39 39 Comparison by Hierarchy Between Average Communication Per Subject and the Percent of Subjects Who Have More Communication ........... With Superiors or Subordinates 40 Distribution of Indicated First Channels of Communication by Hierarchal Position . . . 4l Members of Communication Networks (Non­ isolates) and Isolate Subjects by Office Units ............................... 44 Number of Members (Non-Isolates) of Office Communication Networks Recognizing Specific Publications ............................... 45 Number of Publications That Subjects Recognized or Knew Something About 46 . . . . Distribution of Communication by P u b l i c a t i o n .............................. 46 Subjects' First Channel of Communication and the Mean Time Elapsed from Printing to Subjects' Awareness by Publication . . . . 48 Subjects' Channel of Greatest Communication and Their Mean Attitude by Publication . . 51 v Table 14. 15. 16. 17- 18. 19. 20. 21. 22. 23. 24. Page Average Communication Per Subject and Standard Deviation of Attitude Toward Publications by Office U n i t ................ 54 Comparison of the Degree to Which All Subjects in an Office Unit Were Included in the Office Communication Network and the Average Communication Per Subject in the Office U n i t ......................... 57 Comparison of Mean Length of Employment and the Predominance of Interpersonal Over Mass Media Communication for Each Office U n i t ......................................... 61 Comparison of Mean Length of Employment and the Degree to Which All Subjects in a Unit Are Included in the Communication Network of That U n i t ....................... 65 Comparison of Profiles Between the 64 MostIntegrated Subjects and All Participants . 69 Profile of 64 Subjects With Highest SocioMetric Integration by Office and Position ............................. 70 Comparison of Distribution of 64 MostIntegrated Subjects and All Subjects by P o s i t i o n ................................ . 71 Comparison of Distribution of 64 MostIntegrated Subjects and All Subjects by O f f i c e ...................................... 72 Sociometric Cliques and Isolates Identified by P u b l i c a t i o n ............................. 75 Comparison of the Distribution of Sociometric Cliques and the Total Communication Events by P u b l i c a t i o n ............................. Comparison of 28 Central-Persons With Total Subjects by Office and Hierarchal P o s i t i o n .................................... vl 76 81 LIST OP FIGURES Figure 1. 2. 3. 4. 5. 6. Page A Portion of the Line Staff Organization in M i c h i g a n ’s Department of Education . . . . 16 Scattergram Showing Pearson r Correlation Between the Average Communication Per Subject and the Standard Deviation of Attitude Toward Publications by Office U n i t ......................................... 55 Scattergram Showing Pearson r Correlation Between the Degree to Which All Subjects in an Office Unit Were Included in the Communication Network and the Average Communication Per Subject in the Office U n i t ........................... 58 Scattergram Showing Pearson r Correlation Between Mean Length of Employment and the Predominance of Interpersonal Over Mass Media Communication For Each Office U n i t ......................................... 62 Scattergram Showing Pearson r Correlation Between the Mean Length of Employment in the Department of Education and the Degree to Which All Subjects in a Unit Are In­ cluded in the Communication Network of That U n i t .................................. 66 Sociogram Showing Inneroffice and Inter­ office Communication Between Twenty-Eight C e n t r a l - P e r s o n s ............... 80 vii CHAPTER I INTRODUCTION To understand the behavior of an organization, one must understand the channels by which information is commu­ nicated between Its different parts. Communication is a basic process of organizations; it Is the nervous-system through which organizations cohere, which permits their members to cooperate and coordinate activities. secondary or derived aspect of organizations, Not a communication is a basic function, as is shown in the researches of Baker (19^8), Baveles (1950), Deutsch (1952), Davis Cohen and Bennis (1953), (1962), Guetzkow (1965), and Ellon (1968). Problem This study sought to Identify the patterns by which the collective professional membership of M i c h i g a n Ts State Department of Education utilize mass media and interpersonal communication channels when they interact about certain Department publications. In addition, the study sought for relationships between the professional employees and their office units, their length of employment, and their atti­ tudes toward the publications. 1 2 Assumptions This research made the following underlying assumptions: 1. That publications could be considered one end product of the activities of Michigan's State Department of Education. 2. That these products (publications) of the organiza­ tion represented or reflected the interests of the organi­ zation. 3. That these interests were communicated by 3taff members through identifiable mass media and interpersonal communication patterns. H. That knowing the communication networks regarding selected publications which individuals and office staffs used within the Department could help the Department design better orientation and other inservice training programs. 5. That the Department gained if its professional staff were aware of not only the products (publications) of their immediate Job assignments but also those of the Department as a whole. That is, with this greater awareness, State Department of Education professional employees might have some grasp of the "total picture" regarding the whole D epartment. 3 Hypotheses After identifying sociometric networks among the professional staff members, the study explored the follow­ ing hypotheses: 1. Subjects will indicate greater communication regarding publications with subordinates than with superiors within the State Department of Education hierarchy. 2. Some publications will be excluded from the c ommu­ nication networks of some office units. 3* Subjects indicating mass media channels as first channels of communication will have earlier awareness of the publications than subjects indicating interpersonal channels as first channels of communication. Subjects indicating interpersonal channels as channels of greatest communication will have a more positive attitude toward a given publication than subjects indicating mass media channels as their channels of greatest communica­ tion . 5. The larger the average communication per subject found in an office unit, a. the greater will be the consensus in attitude within the office unit toward the publication, and b. the higher will be the degree to which all subjects in that unit are included in its communication network. 4 6. There will be a positive relationship between the mean length of employment of a u n i t ’s staff in the Depart­ ment of Education and: a. the unit's predominate use of interpersonal rather than mass media communication channels, and also b. the degree to which all subjects in the unit are included in the communication network of that u n i t . Definitions The following terms are defined as they will be used in the study: 1. Subject: A professional staff member in a unit of Michigan's State Department of Education which participated in this study. Prom a population of 28*J, there were 246 C873C) who participated. 2. Attitude: specifically, A manner showing of one's disposition; a positive, negative, or neutral feeling about a publication as indicated by a subject's assigning a numerical value from a five point scale to each sentence completion response (item L) on the instrument. Scale: +2 ** very positive, +1 * somewhat positive, 0 ** neutral, -1 - somewhat negative, -2 ■ very negative. 3. Publication: Printed matter; specifically, certain materials originating from the Department of Education and selected by respective bureau chiefs as reflecting signifi­ cant interests of the Department about which all 5 professional staff members should be aware. Eight different publications were used in the study. 4. Communication; The giving and receiving of infor­ mation; an event resulting in awareness or interaction about a publication through mass media or Interpersonal channels by a subject. 5. Netw o r k ; A system; or communication patterns t identified by constructing a sociogram between individuals and office units. 6. Unit; The collective professional membership of each of eight offices used in the study. 7. Upward-communlcatlon: Communication between a subject and a person in a Job position at a superior hierarchal level. 8. Downward-communlcation; Communication between a subject and a person in a Job position at a subordinate hierarchal level. 9. Hierarchy; Graded ranks; specifically, the formal line-staff structure found within the State Department of Education Job classification system. 10. Ecco-analysis: An operational technique used in communication studies which takes a piece of Information and traces it thru time and space. Communication about eight selected publications was traced here through this technique. 6 11. Channel: A source of Information communicated; specifically, by a subject on the instrument under one of the categories A thru H, the name of one person or place involved in any awareness or interaction by the subject about one of the selected publications. 12. Interpersonal: Between persons; specifically, those channels of communication Involving word-of-mouth or verbal interaction (responses to items P, (1, and H on the Instrument). 13. Mass m e d i a : Vehicle for information transfer to a general population; specifically, such channels of commu­ nication about a publication resulting in awareness or Interaction as radio, television, or written materials (responses to items A, B, C, D, and E on the Instrument). 14. Member: One who belongs; specifically, any subject (a person choosing or a person chosen) who has had communica­ tion with other subjects within his own office unit and is thus considered a part of that unit's communication network. 15. Isolate^ Set a part from the others; specifically, any subject who has no Indicated communication with other subjects within his own office unit and is thus not con­ sidered a member of that unit's communication network. 16. Integration: To be together as a whole; specifi­ cally, the degree to which an office unit includes all of its staff in the communication network. That is, high integration means a low percentage of isolates. 7 Overview of Organization In this research, communication regarding interaction about eight publications of the State Depart­ ment of Education was traced by the "ecco analysis" tech­ nique, a method which focuses on a unit of information and follows it through time and space. The data were obtained from written responses by 2H6 subjects on an open-ended questionnaire which asked them to recall communication channels used in interactions about the publications. Analysis of the data revealed communication patterns among the professional staff, as individuals and as collective office units. Limitations Several important factors limit this study. 1. Since ecco analysis data were secured from many individuals, the technique depended on positive rapport with the respondents. 2. In view of the variety of channels of communica­ tion, the unique significance and impact of each channel was largely ignored. For each publication only two channels (the first and the perceived channel of most communication) were considered, beyond numerical counting and pattern di s t r ibution. 8 3* Organizational products and activities other than publications may be Just as significant or even more sig­ nificant indicators of subject awareness of Department of Education Interests. *1. This study made no estimate of how often individuals used mass media and Interpersonal channels. 5. It is always dangerous to assume that any correla­ tions which exist between identified communication patterns and other data imply cause and effect relationships. 6. The channels of communication recalled was limited by the employee's memory as well as by his degree of sincerity while completing the project instrument. 7. The validity of publications selected by bureau and division chiefs in representing significant interests of the Department was limited by their Judgement. Summary This field communications network study using an ecco analysis technique has looked at staff interaction in regard to eight selected publications of the State Depart­ ment of Education. data. A questionnaire was used to gather the An analysis of the data sought to identify: 1. patterns of communication, 2. tendencies in direction of communication, 3. Isolated units in the communication network, 9 iJ. relative degrees of utilization of mass media and Interpersonal communication channels, 5. the frequency with which mass media and inter­ personal channels were ranked as first sources of communication, 6. the frequency with which mass media and inter­ personal channels were indicated as the channels of greatest communication, 7. the relationship between the first source of commu­ nication (mass media or interpersonal) and the elapsed time from publication date to awareness date, 8. the relationship between the channel of greatest communication (mass media or interpersonal) and the subject's attitude toward the publications, 9. the relationship between the number of communication channels used and consensus in attitude and integration within an office unit, 10. the relationship between the length of employment and a unit's predominate use of mass media or Interpersonal communication channels, and 11. the relationship between length of employment and the degree to which all subjects in the unit were Included in the communication network of that unit. Besides generating a body of empirical knowledge regarding communication in the State Department of Education, this research contributed information on the specific 10 abilities of an ecco analysis technique for a field study In communication within a hierarchal organization such as the State Department of Education. CHAPTER II RELATED IDEAS AND LITERATURE Communication within an organization is always structured. As Starbuck (1965) concludes, "The overall pattern of relations among people, tasks, roles, and bits of Information can be bewilderingly intricate, yet pattern there is" (p. 3^7). Furthermore, it is fascinating to note that after systematic communication develops, the system becomes regularized (Getzkow, 1965, p. 539). A study of communication can take place in an ex­ perimental laboratory setting or through observation and analysis of a communication process in its normal opera­ tional surroundings. The later technique, called a field study, describes this research. Field communication analysis has used at least five different approaches (Davis, 1953)* The techniques are not mutually exclusive though they do in fact fre­ quently combine and overlap. These five approaches do not include the purer forms confined to the experimental laboratory: 11 12 1. In "living-in," groups are Interviewed and observed by close association over long periods of time: see Roethlisberger (1939) and Whyte (19^3). 2. In "indirect analysis," such as Baker (19^8) and Jacobson (1951) used, an analysis of attitudes and inter­ personal associations revealed communication patterns. 3. In "duty study," communications are studied as they pass any particular spot. These studies are similar to traffic surveys in which the observer sits beside a partic­ ular road and records data about traffic as it passes him. *1. In "cross section analysis," communications events in process at any point in time are recorded and analyzed. 5. In "ecco analysis," the observer follows bits of information from their beginning to their destinations. This technique focuses on a unit of information and follows it through aspects of time and space. It is the fifth approach, ecco analysis, which this field study in communications utilized. Davis used this method in a study of 67 management personnel of the Tex Tan Company in 1953- His basic approach was to obtain data from each communication recipient by means of a question­ naire. The resulting "Jig-saw puzzle," when assembled, showed the pattern of how the information spread within the organization. Davis argues that ecco analysis permits using a simple questionnaire without expensive interviewing. points out that many of the criteria of effective survey He 13 methods are met such as low cost, short response period, simplicity, and adaptability. The method has negligible effects on the communication being studied and interferes little with organizational activities (Davis, 1953, p. 308). The relationship between the researcher and sub­ ject is very important. For as Scott (1968) emphasized, power is unevenly distributed so that some members of an organization have more control than others over whether or not the researcher is allowed to conduct his investigation (p. 263). Sensitivity is important to insure the success and cooperative efforts of the organization under study. The ecco analysis type of communication field observation used in this study promised less disturbance to the organization than did other methods. Theoretically, communications within an organiza­ tion can be analysed according to either flow or content. The present study investigated aspects of the flow of information, and looked somewhat at content. The follow­ ing modification of a model for information flow proposed by Shannon and Weaver (19*19) conceptualizes the basic structure of a general communication system: Information origin— Message— Transmitter— Signals— A B C ^ D — ¥ Receiver— Treatment— Begin new cycle or termination E F —¥ G — > m Components of this study can be substituted in the model as follows. Information origin: State Department of Education Publications, Messages: interests of the department, Transmitters: Signals: people and paper, mass media, written and interpersonal, oral communications, Receivers: professional staff members, Treatment: reading or interaction with other persons, and Termination: further interaction or termination of .interaction. Publications (B). (A) of the Department contain interests The Interests are transmitted (C) as messages by mass media or interpersonal (D) signals to the professional staff members (E) who in turn treat (F) the information by read­ ing or interacting with other persons until a termination (Q) is reached where further interaction begins a new cycle or there is no more interaction regarding the information. Note that a treatment which is an interpersonal interaction may result in the staff member being a new information source and thus beginning a new cycle; A, B, C, etc. One of the major concerns of this study was to discover the networks or patterns used in the transmission 15 of messages (B) to receivers signals (E), and the nature of the (D) which were used In the transmission. First, let us examine some basic patterns In communication net­ works. Following this, we will look at some research findings regarding the nature of mass media and Inter­ personal communication signals as they apply to the hypotheses proposed. The classical concepts of communication networks refer to vertical and lateral patterns. In recent years, the generalized terms of line, branch, and circle patterns have been used. a. Vertical networks are those patterns which convey messages up and down the hlerarchal ladder of an organization. The State Department of Education has a line-staff structure typical of civil service bureaucracies which function to perform service, administrative, and advisory tasks for the public. Peabody (1965) points out that all organizations beyond a minimum size and fleeting existence are characterized by hlerarchal relationships (p. 8 1 3 ). The line staff organization of the SDE is partially shown in the "tree" diagram, Figure 1. Ellon (1968) states that the organization chart provides a model of the structure of an organization, but almost invariably it Is found to represent an idealized concept of what the organization Is like, or what it should be like. This is why students of organization theory have LEVEL V State Superintendent of Schools A PORTION OF THE LINF. STAFF ORGANIZATION TN MICHIGAN'S DEPARTMENT OF EDUCATION LEVEL IV (Bureau) Educational Services Research LEVEL III (Division) Curriculum X SpeMal Education Higher Education Administrative Services School for 'School for HTn3 DeaT-------- State Library Vocational Education o LEVEL II (Section) Curriculum •Student Services \ — — p \ Compensatory Education LEVEL I (Unit) ^ \Title I (Fed, ftpgrams) \\\Title III (Fed.Programs) \\ ^Section IV (State Programs) \Section III (State Programs) 17 been searching for more accurate methods to record the Informal structure— for example, by drawing sociograms to Indicate the lines of communications In the system (p. 267). Research concerning communication In the hierarchal tree structure of organizations points to tendencies relevant to hypothesis 1 in this study. Both Burns (1954) and Davis (1953) noted a marked tendency for interactions to be initiated downward rather than upward. speaks of frequent blocks between levels Cartwright Davis also (p. 307). (1959) observed that the typical organizational chart is a tree, and that every point in a tree which is not an end point is an articulation point. He argues that the absence or malfunctioning of an articulation unit has widespread repercussions for the organization, and that the high vulnerability of these structures may account for the frequency of informal or unapproved communication channels. Starbuck (1965) furthers the argument by in­ sisting one should be cautious about ever assuming only a linear relationship (p. 345). b. Lateral communication networks within a system are not as rigidly determined as the traditional line and staff channels, but are nonetheless generally stable. Increases in organizational size affect the vertical dimension but not the horizontal dimension. This phenomenon was observed by Donald (1959) in her study of the League of 18 Women Voters. She found that as the organization size Increased, the formal vertical dimension absorbed the additional communication burden, while the informal hori­ zontal channels remained saturated. In a study among supervisors in a textile mill, Simpson (1959) found the communication for the first line foremen was mainly horizontal. c. In recent years, the classical concepts of line-staff and lateral communication patterns have been included In more generalized concepts of communication nets. Baveles (1950) Identifies four types of communica­ tion networks: (1) circle patterns— where all Individuals communicate with only two other persons (one on each side). (2) chain patterns— where there are two types of positions; relay positions where men can communicate with one person on each side and end positions where men can communicate with only one person. 19 (3) branch patterns— where all communicatio must go through one central person at a node position. (4) line-branch patterns— where communicati nets have combined properties of both line and branch systems. Cohen and Bennis (1 9 6 1 ) used a modification of the branch pattern called a wheel in their studies on leader­ ship in communication networks. They found that former network experience did affect problem solving patterns in latter tasks. When the communication networks changed from a wheel to a circle, the groups organized themselves into the more efficient problem solving chains rather than the less efficient circle. 20 As we pointed out previously, systems become regularized after communication patterns are developed (Guetzkow, 1965). When messages are systematically sent and received, patterns of interrelations develop among the nodes. The work of Cohen and Bennis (1961) related sig­ nificantly to the potential use of this study's findings in the State Department of Education. They demonstrated experimentally that the regular ties involved in utiliza­ tion of a communication net may be transferred from one situation to another. Using a Bavelas model, they changed five-man communication task groups from a wheel to a circle interaction pattern and vice versa. They found that the ten groups which had been shifted from a wheel to a circle network midway in the experiment were able to transfer their communication patterns and in the end were superior to the group that had been kept in circle networks. Tendencies pertinent to hypotheses 1 and 2 have been revealed by other communication research. Social communication.processes in small groups studied by Bales (1951) indicated regular initiation-response tendencies. Those who "put in" more acts tend to be recipients of more "out put." This concept was generalized to larger organi­ zations by Habbe (1952), who found that supplying more information to employees creates the desire for more information. In a questionnaire study of two plants, Habbe 21 found that in the plant in which 55 percent of the workers agreed that the company did a good Job of telling them what was going on and what was being planned, there was a corre­ sponding greater "wish to know more" than in another in which only 18 percent of the workers felt this same company was communicating adequately with its employees. The principal interest of hypotheses 3 and 4 r e ­ garded oral and written signals which are being transmitted. Referring to the information-flow model above, the trans­ mission (C) of messages (B) to receiver (E) takes place by the use of both written and verbal methods. Research has shown advantages and disadvantages associated with both written and oral communication techniques. Guetzkow (1965) pointed out how oral messages compete with written messages; telephones have the advantage of demanding complete momentary attention of the receiver as well as the caller, for example, but they fail to provide a permanent record unless a tape recorder is attached. Written messages, although relatively permanent, must compete not only with oral forms but with each other for reading time. Baker (19*18), in a study of industrial relation policies in over *10 manufacturing offices, stressed that larger organizations had to rely more on the written word, with the result being greater difficulty in insuring accuracy and undistorted communication. Habbe (1952) studied the sources from which employees in two plants of a large manufacturing concern 22 had gotten most of their information about the company. He found 60 percent (of the 273 employees responding) said they obtained their information from reading while the other 40 percent received such information by talking with people. Habbe's main conclusion was that both the oral and the written forms were important. Other research has shown verbal communication pre­ dominating. Pace to face communication was observed to be dominant by Davis (1953). In a study presenting printed and spoken word forms to subjects, Cox (1965) concluded that auditory presentation alone was more effective than combined audio-visual presentation (p. 40). Helen Baker found that in the dissemination of information policies, the chief emphasis had been on the written word, but that employees like to get information from word-of-mouth. She observed that the more carefully planned programs took into account both types of communications and used one to sup­ plement the other (p. 3 6 ). Research conducted by Everett Rogers and his associates related to hypotheses 5 and 6. Rogers (1968) uses the more generalized terms "mass media sources" and "interpersonal sources" of communication. These terms indicate the same basic dichotomy as written vs. verbal sources. In this study, written communication was con­ sidered as part of the more generalized term mass-media. Likewise, any verbal or oral communication was considered 23 a part of the more generalized term inter-personal communication. Tn 1962 Rogers (p. 311), used the terms impersonal and personal in a similar context. He notes that impersonal information sources are most important at the evaluation stage in the process of adopting Innovations The concept of stages in the diffusion of innovations, as developed by Rogers, also related to hypothesis indicates three basic . Rogers stages in the adoption period: awareness, trial, and adoption. In the present study, a given publication was considered as an innovation into the existing system. The awareness stage is the point of first interaction or first awareness of the publication’s exist­ ence. The adoption stage is the point of an encouraging or positive attitude in regard to the publication. Rogers (1968) points out some important distinguish ing characteristics between interpersonal and mass media channels (Table 1). In context, this study clearly depicts the important differences between written/radlo-andtelevislon and verbal/word-of-mouth communication sources. The ecco analysis methodology of this study used publications as the innovative ’'bits" of information traced The publications were produced by various bureaus for con­ sumption by both the public and professional educators in and out of the department. The publications included various operational guidelines, policy statements, and reports of research activities. These publications had 24 different audiences. A study of 104 researchers by Murray (1965) found that multiple audiences are associated with multiple publications. Moreover, and perhaps most important, was how publications reflected the interests of the organization which produced them. Table 1. DISTINGUISHING CHARACTERISTICS OP INTERPERSONAL AND MASS MEDIA COMMUNICATION CHANNELS Communication Characteristic Interpersonal Channels Mass Media Channels 1. Direction of message flow two-way one-way 2. Speed to a large audience slow rapid 3. Message accuracy to a large audience low high 4. Ability to select receiver high low 5. Ability to overcome the selectivity processes (selective exposure perception, and re­ tention) high low 6. Amount of feedback high low 7. Possible kinds of effect attitude change Increased knowledge CHAPTER III PROCEDURES USED Publications Early In the planning for this study, It became clear materials must be found about which professional employees would talk openly. Materials which the subjects perceived as threatening or issues which might be politi­ cally sensitive from the subjects' view would not do. While consulting with the various department bureau and division chiefs, furthermore, I sensed that many theoretically useful materials were exceptionally sensitive from a departmental aspect. Announcements of policy changes within the department or educational Issues before the public would have filled the requirements for an ecco analysis field communication study. But they were much too sensitive for a study which demanded open and honest re­ sponses of communication recall. Publications of the State Department of Education were found to have fewer drawbacks and more assets than any other item considered. Besides, the publications had exact printing dates for establishing precise times of 25 26 entrance of the innovations into the system. The publications chosen came from various parts of the Depart­ ment and represented a broad spectrum of the interests of the State Department of Education. Moreover, the bureau and division chiefs involved were, without exception, willing to suggest specific publications of which they felt their professional staff should be aware. These chiefs were also willing to support and encourage their staffs in a communications study about publications. Out of this background, then, eight publications were chosen: 1. Michigan School Finance Study by Thomas, printed in July, 1968; 2. May, Education in Moral Values in M i c h i g a n , printed in 1968; 3. The Michigan State Plan for Vocational Education, revised printing in April, *1. 1967; State Plan for Higher Education In M i c h i g a n , by Harold Smith, printed in September, 1968; 5. Excerpts from the Michigan Education Feasibility S t u d y , printed in October, 1968; 6. General Information Pamphlet for Michigan School for the D e a f , printed in 1958; 7. Rules of the Civil Service Commis s i o n , printed on February 15, 1968; and 27 8. Analysis of Michigan Public School Revenues and Expenditures 1967-68; Bulletin 1 0 1 1 , printed In December, 1968. Subjects Figure 1 represented part of the line staff organi zational chart for the State Department of Education (see page 16). Of the five bureaus of the Department, three operated predominately in consulting service activities. The Administrative Service Bureau and the State Library functioned in a considerably different role, so they were not included in the research. Along with the professional staff from the State Superintendent's Executive Offices, the following three bureaus were used: 1. the Bureau of Educational Services, 2. the Bureau of Research, and 3. the Bureau of Higher Education, Within these, eight naturally distinct work units were identified: 1. The State Superintendent and his Assistants, five persons, located in the Michigan National Tower Building, Lansing. 2. The Bureau John Qafner of Research, l*i persons, located in the Building on South Pine Street, Lansing. 28 3. The Bureau of Higher Education, 4l persons, located in the Pruden Building, Lansing. 4. The Curriculum Division of the Bureau of Educa­ tional Services, 46 persons, located in the Board of Water and Light Building, Lansing. 5. The Special Education Division of the Bureau of Educational Services, 13 persons, located in the Davenport Building, Lansing. 6. The Division of Vocational Education in the Bureau of Educational Services, 34 persons, located in the Bauch Building, Lansing. 7. The Michigan School for the Blind in the Bureau of Educational Services, 64 professional educators, located on West Willow Street, Lansing. 8. The Michigan School for the Deaf in the Bureau of Educational Services, 60 professional educators, located on West Court Street, Flint. The research was to study the total population. Finally it included 87 percent of the population. A com­ plete description of the population and the respective percentages ultimately included in the study, by office, are shown in Table 2. The eight hlerarchal levels repre­ sented in the Department are also shown on Table 2 along with their respective populations and the percentage of each population participating. 29 Tabio 2 Section Chief Division Chief Total Subjects Potential Population 1 50 1 1 1 56 60 95* School for the Blind 5 1 b5 1 1 1 5b 6b fib* 1 13 13 100* Special Education Division 12 Curriculum Division 3 Bureau of Higher Education 3 Bureau of Research 2 b 31 5 1 1 1 b2 b6 91* 17 2 6 3 1 36 bl 88* 1 11 lb 78* 30 3b 88* 3 b 12 33* 28b 87* 8 ' Divsion of Vocational Education 2? Executive Offices 1 5 ? 1 Total Subjects 15 6 186 lb 11 8 3 3 2b6 Potential Population 18 6 211j 15 11 11 b 5 28b 73* 75* 60* Percent of Population Participating as Subjects 83* 100* Percent of Population Participating Unit Chief 2 OFFICE UNIT I State Supt.ft Assistants Admin. Assistant Consultant or Cert.Teach. School for the Desf POSITION— } Bureau Chief Intern NUMBER OF SUBJECTS BY OFFICE AND POSITION 87* 93% 100* 875* 30 In all 2*16 out of the population of 28*1 professional employees were Included In the research. The distribution of that fraction of the population not Included was r e l a ­ tively even over the entire population. Persons absent from the staff meetings at which the data was collected were often out of town consulting or were ill. A follow-up was attempted when contact could be made immediately upon their return. Where contact had to be delayed, on the other hand, the individuals were omitted for fear their responses might be contaminated by discussing the survey with other staff members. Only 38 persons or 13 percent of the popu­ lation were omitted from the research. Instrumentation The data was collected through the administration of an instrument (see Appendix B) which consisted of: 1. a Base Data Questionnaire, 2. a Publication Recognition List, and 3. a List of Communication Channels. The Base Data Questionnaire requested the following information: present position, name, age, office location, time in present position, total time in the Department, time worked in the field of education, and time worked In the field of education in Michigan. The Publication Recognition List was compiled from choices suggested by bureau and division chiefs in the 31 De p a rtment. Each chief was asked to select those publications produced by his division which reflected sig­ nificant interests of the Department of Education and of which he felt professional staff members should be aware. The List of Communication Channels requested the subject to do the following: 1. Identify the various communication categories of employee awareness and interaction regarding the eight SDE p u b l i cations. The List of Communication Channels form asked for recall of channels of awareness and interaction regarding specific publications. (This technique of open ended questions was utilized by Rogers and Meynen (1 9 6 5 ) in their study on diffusion in 2-lJD weed spray innovation among Columbian peasants.) The following categories of communication channels are separated in the instrument: a. Mass Media Channels— (1) radio and television; (2) written memos or letters from person(s) In the employee's office unit, person(s) not In the employee's office unit but still in the SDE, and person's outside of the SDE; and (3) written publications of the SDE and of other outside organizations. b. Interpersonal Channels— including spoken commu­ nication with person(s) in the employee's office unit, 32 person(s) outside of the office unit but still in the SDE, and person(s) outside of the SDE. 2. Identify his recall of the first communication channel for awareness of the publication and the date of this first awareness. 3. Identify his perception of his channel of greatest communication regarding each publication. M. Complete an open ended sentence which would show how he felt about the publication. The degree of positive or negative attitude toward the specific publication should be revealed by the sentence. Collection of Data The several bureau and division chiefs granted opportunities for the instrument to be administered during one of their regular staff meetings. The staff members being surveyed were given a brief outline of the project pointing up its purpose and me t h o d ­ ology. The three parts of the instrument were explained. After filling out the Base Data Questionnaire, the partici­ pants' attention was directed to the Publication Recognition List. On this list each was asked to circle the number in front of every publication he recognized in any way or had ever heard about. A copy of each publication was held up to facilitate recall or recognition. Upon completion of t h e ’recognition list, the participant was asked to take 33 several copies of the List for Communication C h a n n e l s . He was requested to complete one form for each of the publications he had circled on the Publication Recognition List, After answering the three questions at the end of the form and completing the sentence showing how he felt about the publication, all forms were handed in. When all participants had handed in their completed forms, the researcher left the room with his materials and the staff continued with the business on their agenda. CHAPTER IV RESULTS The data were analyzed, first, for information regarding each of the six hypotheses, and then, for socio­ metric material. In general, the data were analyzed by publications and by office units. Computer program limi­ tations and natural divisions of the Department’s work units made such an analysis more manageable and more useful than a treatment of all subjects and materials at one time. Hypothesis 1 That indicated communication regarding publications tends to be with subordinates rather than with superiors within the State Department of Education hierarchy was not supported from any of three different perspectives. Table 3 shows the distribution of all 2,777 commu­ nication events measured. A listing of lateral, downward and upward communication generated by the 246 subjects appears In Table 4. And Table 5 lists the number of subjects who individually Indicated more downward or more upward communication. The communication events which could not be classified as lateral, downward, or upward Included 34 35 channels which did not reveal any specific Individual. Channels such as other publications, radio, TV, meetings, or unsigned memos and letters were not classifiable as downward, upward, or lateral relative to the subject. Finally, 1,663 or 60 percent of the communications were unclassifiable. or 18 percent. The lateral communications numbered 382 Downward communications numbered 297 or 10 percent, while total upward communications were 435 or 16 percent. Thus, of the total of communication events, 16 percent minus 10 percent, or 6 percent more communica­ tions, were directed toward superiors rather than toward subordinates. In a second perspective, Table 6 examined the percentage distribution of predominately downward over upward communication by individual subjects at the same hlerarchal level. The top and bottom members of the hierarchy were omitted because the subjects from those levels had only one direction to communicate other than lateral. Among the remaining six hlerarchal levels, five showed a larger percentage of subjects' communication with superiors than with subordinates. Table 6 also details the general rise in the number of average communications per subject as one goes up the hierarchy from Interns to the State Superintendent and his Assistants. Since the number of people below you rises further up the hlerarchal ladder, the general growth In the number of communications per person as you go up the hierarchy would favor downward communication. But despite both of these factors, commu­ nication with subordinates exceeded communication with superiors in only one of the six applicable levels. The hypothesis was not supported for any of the following levels: administrative assistants, consultants and certified teachers, unit chiefs, section chiefs, and division chiefs; it was supported for bureau chiefs. Since there were only three bureau chiefs in the study, however, and these very close to the top, the other five cases probably represent the true trend. The third perspective, summarized in Table 7, analyzes the distribution of indicated first channels of communication. Again, in five out of the six applicable levels, more subjects indicated first channels of communica­ tion which were above rather than below their own hlerarchal level. Apparently there was in fact more upward communica­ tion. Recent discussions with several of the subjects who helped generated the data, however, raised the possibility that a psychological factor may have interfered with the subjects' recall. Subjects may have honestly recalled persons of higher status with whom they had interaction more easily than those who are below them in status. The testing of this possibility pertains significantly to the study of communications in hlerarchal organizations where 37 status or rank plays a large and recognized role. It could well be that an "ecco analysis" study of field communica­ tions study should avoid posing questions about direction of communication by recall questionnaires. Judging from the data collected in this study, In any case, the hypothe­ sis that communication events about publications tends to be with subordinates rather than with superiors in the Department of Education was rejected. 30 Table 3. NUMBER OF INDICATED COMMUNICATION CHANNELS BY CATEGORY FOR EACH PUBLICATION > MASS MEDIA Radio or Television Written nemo or letter from person in your office X 1 CO• • S c s Pam.Sch. For Deaf PUBLICATION— CHANNELS^ . £S i£ R & ££ 57 11 3 5 26 h *4 * H 0 u a M£ % o • r-J r-4 > £ s0 4 i3 *■«O g O Cl 5 7 3 111 b* * 58 16 27 U6 23 10 72 lb 266 9* 39 23 2b bb 28 10 5b 16 238 9* 17 b lb 20 9 1 16 b 85 3* 129 53 61 70 b7 33 121* 35 552 20* 200 37 08 90 L6 23 178 3b 696 25* Spoke with or heard about it from person not 90 in your office but still in SDE 16 66 59 38 7 7h 22 boo lb * Spoke with or heard about it from person out­ 100 side of SDE 20 b9 149 31 11 21 16 297 11* 8 22 20 9 17 35 5 132 5* 37li I42I4 235 112 581 Hi9 2777 100 * b* 21* 5* 100? Written memo or letter from person not in your office but still in SDE Written memo or letter from person outside of SDE Other Publications INTERPERSONAL Spoke with or heard about it from person in your office Unident.Chan. TOTAL Percent 16 7UU 180 26* . 7* Hi* 15* 8* 39 Table J4.. DISTHIBUTTON OF LATERAL, DOWNWARD, AND UPWARD COMMUNICATION Number of Communications Percent of Total Lateral Coftinunlcati ons 382 1h% Downward Communications 297 10% Upward Communications U35 16% Communications whict could not be classified as lat­ eral, downward or upward 1,663 60% Total 2,777 100% Table 5. DISTRIBUTION OF SUBJECT COMMUNICATION WHICH WAS MORE DOWNWARD 0 MORE UPWARD Times Indicated More downward than upward communications indicated Equal downward and upward coranunication indicated Less downward than upward communication Indicated Total Percent of Total 9$ 13* 3hO h6% 297 kl% 732 100% 1*0 Table 6 . COMPARISON BY HIERARCHY BETWEEN AVERAGE COMMUNICATION PER SUBJECT AND THE PERCENT OP SUBJECTS WHO HAVE MORE COMMUNICATION WITH SUPERIORS OR SUBORDINATES Average Subjects Comm.per indicating No. of Subject more conon. with subor. Subject! than super. Interns Subjects indicating more comm, with super than subor Subjects indicating equal com. Suport with super, Hyp. and subor. # 1 1? 7.7 0* 63* 37* N/A 6 6.2 29* 1*2* 29% No Consultants or Cert.Teacl .186 8.7 1% 1*2* 51* No Admin.Asst. Unit Chief H* 16.1* 2X56 1*1* 38* No Section Chief 11 20.3 30* 1*0* 30* No Div.Chief 8 30.3 29% 1*2* 29* No Bureau Chief 3 27.7 $6% 16* 28* Yes Supt.& Assts. 3 27.3 11% 0* 83* N/A Total 2t*6 Hyp.# 1.: Coununicatlon events between staff members regarding publications tends to be with subordinates rather than with superiors within the State Department of Education hierarchy was not supported by £ out of 6 applicable positions. la Table 7. DISTRIBUTION OF INDICATED FIRST CHANNELS OF COMUNICATICN BY HIERARCHAL POSITION POSITION No. of Indications Percent superior Percent Percent Subordinate Lateral Supporting Hyp.# 1 20 100* N/A 0* N/A 6 83* 0* 17* No 223 *>8* 2* 1*0* No Unit Chief 23 61* 30* 0* No Section Chief 32 59* 111* 0* No Division Chief 28 GU% 11* 25* No Bureau Chief 1$ 27* 53* 20* Yes Supt. & Asst. 8 N/A 37* 63* N/A Intern Admin. Assistant Consultants & Cert. Teach. k2 Hypothesis 2 That some publications will be excluded from the communication networks of some office units was supported in five out of eight office units. As Table 8 shows, the percentage of subjects who were members (a chooser or a chosen by at least one other subject in the same office unit) varied considerably from office to office: from 100 percent Integration in the Bureau of Research to 6k per­ cent integration in the School for the Deaf. Prom among the subjects of each office the number of those who indicated awareness or recognition of ea c h publication appears in Table 9. There were 1*1 cases where 3 percent or less of the member-subjects recognized a publication. And in nine instances, no member-subject of an office unit recognized a publication. was supported in five offices: So the hypothesis the School for the Deaf, the School for the Blind, the Bureau of Higher Education, the Bureau of Research, and the Executive Offices. When considering publications recognized by 3 p e r ­ cent or less of the member-subjects, awareness followed this descending order: the Rules of the Civil Service Com­ mission, the Finance Study, Bulletin 1011, State Plan for Vocational Education, Education TV Study, Education in Moral Values, State Plan for Higher Education, and the Information Pamphlet for the School for the Deaf. Indeed, as Table 11 notes, the Finance Study and the Rules of the 43 Civil Service Commission received 26 percent and 21 percent of all communication events concerning all publications. A certain degree of provincialism characterized awareness of publications. Greatest awareness of a publica­ tion comes from the office of its origin in several cases: Education in Moral Values, originating from the Curriculum Division; The State Plan for Vocational Education, from the Division of Vocational Education; The State Plan for Higher Education, from the Bureau of Higher Education; The Education TV Feasibility Study, originating from the Cur­ riculum Division; and the Information Pamphlet for the School for the Deaf, from the School for the Deaf. Along with the bureau or division of origin the Executive Offices had the highest percentage of awareness for five of the eight publications. In conclusion, there appears to be significant gaps in employee awareness about publications of the Department of Education. If we are to assume that these publications do represent major interest areas of the Department, it might be prudent to find the manner by which such publica­ tions as the Rules of the Civil Service Commission and the Finance Study by Thomas became so well known and the subjects of so many communication events. The techniques of disseminating information about these publications might well suggest how the Department could promote other major Interests. Wi Tablo "i. Bureau of Hieher Ed. °fi s fc 0 U V ID ££ Voc. Ed. Div. Executive Offices 13 u? 36 n 30 ii 33% 8* ?.% 3% 0* 17* ?.*% 67% 92% 9B? 97% 100* 83* 79* Offiee Units v * School for Deaf School for Blind 1 Subjects in Office Unit 96 Percent Isolates 36* Percent of Subjects in Office Com­ munication network a f1 ft •H C. S. VI 1I ' Total Subjects Curr. Div. | MEMBERS OF COMMUNICATION NETWORKS (NON-1SOLATES) AND ISOLATE SUBJECTS BY OFFICE UNITS ?J|6 b5 Table 9 NUMBER OP MEMBERS (NON-JSOLATFS) OF OFFICE COMMUNICATION NETWORKS RECOGNIZING SPECIFIC PUBLICATIONS School for the Blind Special Ed. Divis ion Bureau of Higher Ed. «m c xt: . p t< ftt CD 0> Q) m 0) (h rti Hnance Study by Thomas 3* 3* 75* 95* 69* 100* 8b* 100* Educ. in Moral Values o* 3% 17% Uh% 111* 18* 2b* 67* State Plan for Vocational Educ. 6* 6* 33% 29% 17* ll5* 92* 33% State Plan for Higher Educ. 0* 3% a* 15* 9b* 9* hh* 100* Educ. TV Feasibility Study 0* 3% 25* 56* 11* 18* 12* 100* Info, Pamphlet for School, for the Deaf 53* 0* 8* 0* 0* b* 0* Rules of Civil Service Comm, 78* 100* 33* 5b* 71* 10* ua* 100* Rev.& Exp. 67-68 Bulletin 1011 0* 0* 17* 10* 20* 73* 16* 67* 36 36 12 bi 35 11 25 3 yes no no yea yes no yes Office Unit - - Total membersubjects in office unit Supporting Hyp.# 2: Some publications yes will be excluded from the communi­ cation networks of some office units 5* ic -\ J*iHn C *rt O > • ■CD *H»Q (J • Executive Offices Publication School for the Deaf o § EJ fj t. O 'u *£> £ o53 a U6 Table 10. NUMBER OF PUBLICATIONS THAT SUBJECTS RECOGNIZED OR KNEW SOMETHING ABOUT Number of Publications Indicated 0 1 2 3 h Number of Subjects 2 Iifl UB 3$ 30 Fereont of Subjects \% 20% 20% l.h% 1??! 17% Table U2 11. 6 7 8 Total 23 111 Ij ?ii6 9% $% 2% 100* - DISTRI 13(11ON OF COMMUNICATION BY PUBLICATION Publication Finance Study by Thomas Rules of the Civil Service Commission State Plan for Higher Education State Plan for Vocational Education Education T.V. Feasibility Study Education in Moral Values in Michigan Revenue and Expenditures Public Schools 1967-68 Information Pamphlet for the Srhool forthe Deaf Total Communication Events Indicated ?,777 “ Percent 26% 2\% 15£ 1h% 8T. 7% U% 100?! 47 Hypothesis 3 That subjects indicating mass media as first channels of communication will have earlier awareness of publications than subjects indicating interpersonal channels as first channels of communication was not supported for seven out of the eight publications. The hypothesis was also rejected for the mean across all publications, all as summarized in Table 12. In six out of eight cases, in fact, the mean time from printing to awareness was shorter for subjects indicating Interpersonal channels as first channels of communication. These results were quite contrary to R o g e r s 1 find­ ings (1 9 6 8 ) where communication by mass media channels reached farmers more quickly than Interpersonal channels, in the early stages of awareness. In every case except for the Rules of the Civil Service Commission, the present study found that interpersonal communication resulted in earlier awareness of publications. Surveying the distribution of all communication, Table 3 revealed that total mass media communications nearly equaled interpersonal communications 45 percent and 50 per­ cent respectively. And since Table 3 also found that half of the interpersonal communications occurred in the subjects' offices, we might expect most awareness to come from In­ formal intra-office contacts. W Tablo 12. 3.3 1.2 19 h 1*6 19 1.5 1.2 33 3.6 1.1 3.0 No No No Yes No No a. o p M tO fc . CD 1.7 Ih 16 27 ?.li h.U No No 5.o 2.6 Number of subjects indicating inter­ personal channels as first channels of awareness Ii8 Mean time in months from pub, printing to subjects' awareness 1.7 Mean time all subjects Bull. 1011 120 St,Plan for Higher Ed. 12 e: * o 3 s It.Li 1.6 9 Supporting Hyp,# 3: Subjects indicating mass media channels as first channels of No communication will have earlier aware­ ness of the publicat. than subjects indica­ ting interpersonal channels as first channels of comm. 52 ** -V■A « g tn c rp i~ 3.1) * Attitude scale: 1B very negative; 2a somewhat negative; 3* neutral; 1)H somewhat positive; 5* very positive Note: even though the absolute value showed the hypothesis supported by J> publications, none of the differences were significant in a one-tailed t test at the .0$ level. 52 Hypothesis 5-A That greater average communication per subject in an office work unit will correlate with greater consensus in attitude within that office unit was supported, with a Pearsonian correlation coefficient between the two varia­ bles of +0.79. The findings are summarized in Table 14, and Figure 2 presents the correlation as a scattergram. The variation in standard deviations in each office's attitude toward all publications measured the degree of consensus in attitude on a five point scale. The Bureau of Higher Education with 1.06 and the School for the Blind with 0.79 had the largest and smallest standard deviations respectively. Considering all the publications, the School for the Blind and the Division of Vocational Education had the most positive mean attitudes, with the least positive attitudes in the Bureau of Research and the Executive Offices. But all mean attitudes across the eight offices fell in the relatively positive range. As with hypothesis this narrow range in average attitudes raises questions of significance. In spite of this small range, however, the relation­ ship between the average communications per subject in each office and the mean attitude revealed a substantial correla­ tion of +0.79. Clearly the more communication there is per subject in an office, then, the more likely it is that , 53 the subjects will have similar attitudes toward the eight publications under consideration. The average communication per subject increases in order from the School for the Deaf, to the School for the Blind, Special Education Division, Vocational Educa­ tion Division, Bureau of Higher Education, Curriculum Division, Bureau of Research, and the Executive Offices. The variation in average communications in office units Jumps a dramatic 800 percent from the School for the Deaf, with 3-12 average communication events per subject, to the Executive Offices with its 2*1.75 average communications per subject. 5Ji Tablo Ll|, Bureau of Research Voc.Ed. Div. Executive Offices Total subjects in office Bureau of Higher Ed. Total communicatior events in office 178 163 762 627 222 385 99 b « r» o S c 0> fj, to Curriculum Division > Sch.for 31ind Office Unit ■ Sch.for Deaf AVERAGE COMMUNICATION PER SUBJECT AND STANDARD DEVIATION OF ATTITUDE TOWARD PUBLICATIONS Br OFFICE UNIT 56 5b 13 U2 36 11 30 Average communica­ tion per subject in office unit 3.12 1.70 12.5b 18. lb 17.bl 20.18 12.83 2b. 75 Mean attitude * toward all publications 3.22 3.6l 3.53 3-32 3.3? 3.23 3.59 3.26 0.82 0.79 0.82 0.90 1.06 1.01 0.86 1.29 Standard devia­ tion of attitude toword all publications Pearson r correlation between the average coiranunionions per subject and the standard deviation of attitude n +0.788 * Attitude scale: 1" very negative; 2“ somewhat negative; 3* neutral b* somewhat positive; 5“ very positive Supporting hypothesis 0 5-A: Greater average communication per subject in an office owrk unit will correlate with greater consensus in attitude within that office unit toward the publications. Average communications per subject 30 25 20 15 10 § CO CO CM Os 8 CM rH o CM CM 00 CM CM Standard deviation of attitude Figure 2. SCATTERSRAM SIKWING PEARSON r CORRELATION BETWEEN TOE AVERAGE COMMUNICATION PER SUBJECT AND TOE STANDARD DEVIATION OF ATTITUDE TOWARD PUBLICATIONS BY OFFICE UNIT . 56 Hypothesis 5-B That the average number of communications per subject In an office unit will correlate positively with the degree to which all subjects In that unit are included in Its communication network was supported, with a Pearsonian correlation between the two variables of +0.62. The findings were summarized in Table 15, and Figure 3 presents the correlation as a scattergram. Table 8 graphed the degree to which all subjects in an office unit were Included in its communication network: in order to be included in the network a subject had to be a chooser of, or a chosen by at least one other subject in the same office unit. subject thus categorized was called a member-subject. Any The average communication per subject in each office unit, how­ ever, was calculated on the basis of all subjects and not Just the member-subjects in the communication net, although all other subjects in the office unit were labeled isolatesubjects. As the scattergram demonstrates, the more the subjects in an office communicate, the more likely it is that office will include its subjects in a communication network having more member-subjects in relation to offices with fewer average communications per subject. The results of the executive office with its small n of four were spurious. Omitting this office from the calculations would have resulted in an even larger correlation coefficient in support of the hypothesis. 57 Table 1$, COMPARISON OP THE DEGREE TO WHICH ALL SUBJECTS IN AN OFFICE UNIT WERE INCLUDED IN THE OFFICE COMMUNICATION NETWORK AND THE AVERAGE COMMUNICATION PER SUBJECT IN THE OFFICE UNIT Office Unit Percent of subjects Included in the of­ fice comm, network Average communica­ tions per subject School for the Deaf 6b* 3.12 School for the Blind 67* b.70 Speical Educ. Division 92* 12.5b Curriculum Division 96* 10.1b Bureau of Higher Educ. 91% 17.bl 100* ?0.10 Division of Vocational Ed. 63* 12.83 Executive Offices 19% 2b. 75 Bureau of Research Pearson r correlation between the percent of subjects included in the of­ fice communication network and the average communications per subject “ -*0.6l6. Supporting Hypothesis # 5-B: The avorage communication per subject in an office unit will correlate positively with the degree to which all subjects in that unit are included in its communication network. SB Percent of subjects included in office communication network 20 10 Average communications per subject Figure 3. SCATTERGRAM SHOWING PKARSON r CORRELATION BETWEEN THE DEGREE TO WHICH ALL SUBJECTS IN AN OFFICE UNIT WERE INCLUDED IN THE COMMUNICATION NETWORK AND THE AVERAGE COMMUNICATION PER SUBJECT IN THE OFFICE UNIT 59 Hypothesis 6-A That the mean length of employment in the Department of Education of a unit's staff will correlate positively with that unit's using interpersonal more than mass media communication channels was unsupported. correlation was -0.^0. The Pearsonlan The findings were summarized in Table 16, and Figure 4 presents the correlation as a scattergram. The relative use of interpersonal over mass media communication channels was determined by subtracting an office unit's total indicated mass media channels from its total indicated interpersonal channels. Hence a predominate use of interpersonal channels assumed the form of a positive difference score. This difference score was correlated with the mean time of employment in years that the subjects worked (a) in the Department, (b) in the office of present position, (c) in the field of education in Michigan, and (d) in the field of education as a whole. The results show that the longer a person was employed by the Department, the more likely he was to use mass media rather than interpersonal channels of communica­ tion. Such probabilities dropped near zero when considering time employed in the field of education outside of the De­ partment; the relationship seems to hold for the Department of Education as an organization and not for educators as a group. 60 Prom this perspective, the Department might fruitfully capitalize upon existing organizational patterns to utilize the strengths of both the long tenured and short tenured employees. If these two groups were encouraged to interact more, awareness of all publications should in­ crease. In other words, with no improvement in present efforts to disseminate information about publications, a conscientious effort to increase interaction between the short and long tenured staff members would stimulate greater awareness of major Departmental interests. 61 'i'u'nlo U :i. a «; 01 a Bur. Higher 5d. orrice Unit Sch.for Blind COMPARISON OP MEAN LENGTH OP EMPLOYMENT AND THE PHKDOMI NANCE OP JNTKRPF.RSGNAL OVER MAOS MEDIA COMMUNICATION FOR EACH OFFICE UNIT £ a.!? 7.3 3.3 3.0 2 .7 0 .8 8 .0 9.7 6 .6 3.5 ll.a 3.3 0 .9 11.0 11.0 1 ? .0 1 6 .8 1Li-1< 9 .1 1.8 2 1 .7 17.K n .i 1 3 .^ 17.1 16.2 1 0 .2 ? .6 23.6 1 8 .7 +6 -5 +100 +1? +38 '■IS* (0 « r.i f. 0 JVC to Mean time in years employed in office of present posi­ tion Mean time in years employed by Mich. State Dept.of Educ. Mean time in years in the field of edtic. in Michigan Mean time in yearn in the field of educ. Total Inter­ personal comm. minus total mass media comm within office -3? • rs « f* •r* Cl U V, SJ u x‘ (; *. «P 0> 10 & IS• c £• > •rl Q •H • *» (0 (1) 3t> T Ol a> *-• iKS5 2 .3 01 10.9 Pearson r correlation between mean time in present position and inter­ personal-mass media difference snore ■ -O.Ji72 Pearson r correlation between mean time in Dept, of Bduc. and intcrperoonal-mnss media difference score » -OJ4OI4 Pearson r correlation between mean time in Michigan and interpersonaImass media difference score *» -0.096 Pearson r correlation between mean time in field of education and intorpersona 1-macs medio difference score «*• +0.003 Not supporting Hypothesis H 6-A: The mean length of employment in the SDE of a unit's staff will correlate positively with that unit's using interpersonal more than mans media communication channeln. 62 Interpersonal cCTm™micat^>rwT!nu!!£^™a8 medi communication within office unit +100 • + 90 +60 +50 +20 +10 -20 Mean time in years employed by Dept, of Edur. Figure 1*. SCATTERGRAM SHOWING PEARSON r CORRELATION RE1WEW KEAN LENGTH OF EMPLOY­ MENT AND THE PREDOMINANCE OF INTERPERSONAL OVER MASS MEDIA COMKUNIOATION FOR EACH OFFICE UNIT 63 Hypothesis 6-B That the mean length of employment In the Department of Education of a unit's staff will correlate positively with the degree to which all subjects in that unit are in­ cluded In Its communication network was unsupported. Pearsonian correlation coefficient was -0.82. The The findings were summarized in Table 17, and Figure 5 presents the correlation as a scattergram. The percentage of subjects included in each office communication network was determined under hypothesis 5-B. This percentage was correlated with the mean time of employ­ ment In years that subjects worked (a) in the Department, (b) in the office of present position, (c) in the field of education in Michigan, and (d) in the field of education as a whole. The results show that the longer a person was employed by the Department, the more likely he was to become isolated in his office unit, while employees with shorter tenure tended to be involved in the communication network of their office unit. Such probabilities dropped near zero when considering time employed in the field of education outside of the Department; the relationship 3eems to hold for the Department of Education as an organization and not for educators as a group. From this perspective, given as well the results of hypothesis 6-A, the greatest strength for communication 6H seems to come from short tenured employees through interpersonal channels, while the greatest barriers to communication stem from long tenured employees who prefer mass media channels for communication. If the long tenured and short tenured employees were encouraged to Interact, then, isolation should begin to yield to cooperative utilization of the respective strengths in mass media and interpersonal communication of each. 6$ T a M e 17 . Office Units - — ^ Mean tine in yearn employed in office of present position Mean time in years employed by Dept, of Ed. Mean time in years in the field of cduc. in Michigan Mean time in years employed in the rteld or education Percent of subjects includ­ ed in the office unit comm. net. (D n u «8• Xti l to Sch.for Blind COMPARISON OF MEAN LENGTH OF EMPLOYMENT AND 'INK DEGREE TO WHICH ALL SUBJECTS IN A UNIT ARF. INCLUDED IN THE COMMUNICATION NETWORK OK THAT UNIT • • ms w £ ■ > a * 0) tn o> •s • 1 >I •( n in 0 u1 *ri •V. 8 c 01 rtf jt t. : t-. tt> 0 i} n Cl cr; « 8 u 8 I n. 25 * i. o 8.5 7.3 3.3 3.0 2.7 0.8 8.0 2.3 9.7 8.6 3.5 li.8 3.3 0.9 11.0 10.9 11.0 13.0 1,6.0 llj.Ili 9.1 1.8 21.7 17.5 11.1 13.5 17.7 16.2 10.2 5.6 23.8 18.7 fill* 67% 9?* 98* 97* 100* 83* 75* Pesroon r correlation between mean length of employment in present position and the percent of integration within office units ** -0.750 Pearson r correlation between mean length of employment in Dept, of Edur, and the percent of integration within office units * -0.817 Pearson r correlation between mean length of employment In Michigan and the percent of integration within office units » -0.381 Pearson r correlation between mean length of employment in field of Edur. and the percent of integration within office units j -0.19li Not supporting Hypothesis 0 6-B: Hie mean length of employment in the SDE of a unit's staff, will correlate positively with the degree to which all subjects in that unit arc included in its communication network. 66 Percent of subjects included in office comunication network 70> 60; 10 11 Mean time in years employed by the Department of Rdur. Figure 5. SCATTERGHAM SHOWING PEARSON r CORRELATION BETWEEN 71TE MEAN LENGTH OF EM­ PLOYMENT IN THE DEPARTMENT OF EDUCATION AND THE DEGREE TO WHICH ALL SUB­ JECTS IN A UNIT ARE INCLUDFD IN THE COMMUNICATT ON NETWORK OF H U T UNIT 67 Sociometric Findings 64 Most-Integrated; The 64 most-integrated subjects were chosen from a list of the 20 most-integrated subjects for each publication. These 64 subjects had the most chains of communication of all 246 subjects in the study. To compare this group with the total number of participants, several profiles of comparison were prepared. A comparative profile between the 64 most-integrated subjects and the 246 subjects appears in Table 18. The two groups were compared by average attitude toward all publica­ tions, average publications recognized, indicated first channel of communication, indicated channel of greatest communication, average age, average years in the Department, and average total communications. There were no remarkable differences found on any of these items. Table 19 shows the actual distribution of the 64 most-integrated by office and hierarchal position. Table 20 presents a comparison of the distribution of the 64 most-integrated with the total 246 subjects by hierarchal position. The Spearman rank order correlation between the two groups was .33, not a strong correlation. But all the hierarchal positions above consultant had a greater per­ centage representation in the 64 than in the 246 subjects. The results showed that supervisory positions included more soclometric-integrated-persons than would be expected. 68 This finding agrees with those shown in Table 6 regarding the same supervisory positions: supervisory personnel have more individual communication events per subject than those not in supervisory positions. Thus, the chiefs have not only the most communication events per person but also the most chains of communication in the sociometric network. Table 21 presents a comparison between the distri­ bution among the 64 most-integrated and the 246 total subjects by office unit. The Spearman rank order correla­ tion between the two groups was tion. .55, again a weak correla­ But if a unit was among the offices with the higher or lower percentages of total subjects, about half the units fell among the same higher or lower offices in the percentage distribution of the 64 most-integrated subjects. 69 'Iv ib I o ] <1, CflMPAftJSON OK PROFILES BETWEEN THE 61j MOST-INTEGRATED SUBJECTS AND ALL Regarding data from 6h-most-integrnted Regarding data from all subjects Prof1le ]tom Average attitude toward nil publ. 3-h somewhat pos. 3.5 somewhat pos Average publica­ tions recognized Ij.8 publications 3.li publications Indicating mass media chanels as lot channels of corm. Ii9? U2% indicating inter­ personal channels as lot. chan, of communion t1 on 51f 56? Indicating mass media chan, aa chan, of most communication 36? 3BJT 62% 6?% Ijli.Pyears 39.6yeors 7.9ycaro 7. lyenrs Indicating inter­ personal chan.as chan, of most conmunication Average age Average years in Dept, of Educ. Average comm, per subject 6.66 com/subj. 10.69 com/subJ. 70 'Vnblo 19 PROFILE OF 6U SUBJECTS WITH HIGHEST SOCIOMETRIC INTEGRATION BY OFFICE AND POSITION T< £ C OJ V) 1 1 School for the Deaf i; l School Tor the Blind i i Special Educ, D.1vlnior. ? Curri rulum Division 13 !> 3 ? Bureau of Higher Educ. I Bureau of Reneareh 3 Division of Vocational Ed. S' Executive Offices 1 Total 1 Percent 1/ 33 6 ®y * +■» t£n»* » +o) cc £ 1 1 HCP ♦» Percent tv 0) •r< Bureau Chief a> *+-i j r cj Division Thief Adntin. Asst s. Consultants or Cert. Teach. OTficn 1 Intern n Cf) n 6 13/ 2 3/ 3 ?/ 1 1 ?o 31/ 3 1 10 1.6/ L 1 $ 7 1 1 i? 5 7 ? 1/ 19?! 8/ 113! 3?! 6/ 13 VO/ 3 h 6/ 3 6li 100/ 100/ 71 Table e ’L0. COMPARISON OF DISTRIBUTION OF 6b* MOST-INTEQRATED SUBJECTS AND ALL SUBJECTS BY POSITION Posi ti on Consultant or Cert. Teach Percent distribution of 6b-most-integrated Percent distribution of all ?b6 subjects #2 762 12 62 102 62 Section Chief 856 52 Ditfeion Chief 112 32 Admin. Assistant 12 2% Bureau Chief 32 12 State Supt. it Assistants 52 12 Intern Unit Chief Total 1002 6b subjects 1002 ?b6 subj ects The Spearman rank order correlation between the two distributions was -*0,33 72 Table PI, COMPARISON OF DISTRIBUTION OF 61* MOST-INTEGRATED SUBJECTS AND ALL SUBJECTS BY OFFICE Office Percent distribtuion of 6)i-most-integrotcd Percent distribution of all. ?M& subjects School for the Denf 135K 23* School for the Blind 3% 22% Curriculum Division 31% \7% Bureau of Higher Educ. 16£ l£S Vocational Ed. Division 20% \?i Bureau of Research 6% hi Executi vn Offices m. n Special Educ. Division Toto.1 The 'ipy.'i r m a n war. +0,J;J|.G 100?' 6)i subjects ronl: e -ri'olf-1 Io n 1 o t w o e n 100?! tho subjects two d I.stri bu t i o n a 73 Sociometric Isolates and Cliques: The computet* program analyzed the distribution of sociometrio for each publication. isolates These Isolates and sociometric cliques are compared in Table 22. The Spearman rank order correlation for the comparison was a predictable -0.98. The fewer people interacting about publications, the fewer the cliques that would be formed by the remaining subjects. This sociometric analysis of cliques agreed with the per­ centage distribution of all 2,777 communication events, when compared across office units. this comparison. Table 23 summarizes The top four publications in percentage of total communication also led in percentage of socio­ metric cliques. The Rules of the Civil Service Commission, Finance Study by Thomas, State Plan for Vocational Educa­ tion, and,the State Plan for Higher Education collectively accounted for 74 percent of the 407 identified cliques and 76 percent of the total 2,777 communication events. Those who wish to promote awareness of their publications might well review Table 3$ which details just which channels resulted in awareness or interaction about publications having the most indicated channels and cliques. The top two in this category were the Finance Study by Thomas and the Rules of the Civil Service Commission. Two channels were among the top three channels utilized by both publications; office and (1) interpersonal communication within the (2) other publications from the mass media 74 category. These two channels were also the most utilized across all 2,777 communication events in the study. So deliberate efforts by the Department to encourage staff m e m b e r s ’ interaction within their office units might help Other frequently mentioned mass media publications were public newspapers, SDE News release memos, SDE Michigan Education News, SDE Education Reports, and professional journals. 75 Table 22. SOCICMETRIC CLIQUES AND ISOLATES IDENTIFIED BY PUBLICATION Publication Cliques Isolates Buies of Civil Service Conn. 106 136 Finance Study by Thomas 79 178 State Plan for Vocational Educ. 62 222 State Plan for Higher Education 52 227 Educ. in Moral Values in Mich. 38 256 Rev. & Exp. 67-68 Bulletin 1011 37 26b Education TV Feasibility Study 22 258 Information Pamphlet for Sch.for the Deaf 11 292 Total 1*07 The Spearman rank correlation between the cliques and isolates was >0.98 76 Table 23. COMPARISON OF THE DISTRIBUTION OF SOCICMETRIC CLIQUES AND THE TOTAL COMMUNICATION EVENTS BY PUBLICATION Publication Distribution of 1*07 Sociometric cliques Distribution of 2,777 total com­ munication events Rules of Civil Service Comm. 26?t 21% Finance Study by Thomas 20* 26% State Plan for Vocational Edcu. 19% llj* State Plan for Higher Education 13* 19% Educ. in Moral Values in Mich. 9% 7% Rev. and Exp. 67-66 Bulletin 1011 9% 9% Educ. Television Feasibility Study 9% 8* Info. Pamphlet for School for the Deaf 3* h% Total Jj07 cliques 2,777 communications 77 20 Central-Persons: From among the 64 most- integrated subjects, those who were involved in cliques of four or more persons at the one and two distance levels were identified as the 28 "central-persons." A sociogram, Figure 6 showed that 24 of these 28 were mutually involved with one or more of the other central-persons. A comparison of the 28 central-persons with the 246 subjects by office and hierarchal position is shown in Table 24. The Curriculum Division, Bureau of Higher Educa­ tion, Division of Vocational Education, Bureau of Research, and the Executive Offices have still higher percentages among the 28 central-persons than among the 64 mostintegrated subjects. Also, unit, section, and division chiefs accounted for significantly higher percentages of the 28 central-persons than of the 64 most integrated subjects. The sociogram of the 28 central-persons and their inner- and inter-office relationships reveals no other cliques, though it certainly does not exclude that possibility. Any other cliques, however, would not Include persons from the 64 most-integrated in the Department. The sociogram points up several Interesting patterns which could be helpful to persons who would like to improve the overall interaction within their work units or with other work unitB. The Curriculum Division had not only the most members, but also the most complete interaction among its 78 members and with other units. The Division of Vocational Education, the Bureau of Higher Education, the Executive Offices, and the School for the Deaf each had one or more of the 64 most-integrated persons in the Department who were also involved in small cliques of four or more persons, but who nevertheless were isolated from the other centralpersons in their own units. This should not imply that the 64 most-integrated persons ought necessarily to communicate more with each other. They have already, by definition, communicated extensively with other people and are involved in other cliques. The sociogram definitely suggests, however, that existing resources within office units are not being utilized by all members. Each office unit, then, should review the opportunities it provides for persons to inter­ act with each other, and to what degree these opportunities are designed or accidental. 79 Large circles refer to office units* Small circles refer to individuals* Connecting lines are between individuals who have mutual involvement in one or more of each o t h e r s cliques* Numbers refer to the individualfs hierarchal position: 3 - Consultants k - Unit Chiefs £ - Section Chiefs 6 - Division Chiefs 7 - Bureau Chiefs 8 - Assistants to the State Superintendent Figure 6 * SOCIOGRAM SHOWING INNEROFFICE AND INTEROFFICE COMMUNICATION BETWEEN TWENTY-EIGHT CENTRAL-PERSONS Division Vocational Educat^&n /rw // \ School for the Blind Bureau of Higher Education Figure 6, Executive Offices Bureau of Research Curriculum \Division School for the Deaf Special Education Division 81 Table 2l\. COMPARISON OP 28 CENTRAL-PERSONS WITH TOTAL SUBJECTS BY OFFICE AND HIERARCHAL POSITION Percent of 28-centralpersons Office Percent of 614-most integrated Percent of 2b6 total subjects School for the Deaf U% 13* 23* School. for the Blind 0% 3* 22* Curriculum Division 32% 31* 17* Higher Education 21% 16* 15* Vocational Education 29% 20* 12* Special Education Div. 0% 5* 5* Bureau of Research 1% 6* h% 7% 6% 2* 100% 100* 100* U3% 0* 18* 1$% 18* 0* 3% 3% 100* 52* 1* 10* 8* 11* 1* 3* 5* 100* 76* 6* 6* 5* 3* 2* 1* 1* 100* Executive Offices Total Hierarchal Position Consultants Interns Unit Chiefs Section Chiefs Division Chiefs Admin. Assists. Bureau Chiefs State Supt.& Assts. Total CHAPTER V CONCLUSIONS, IMPLICATIONS, AND QUESTIONS Conclusions Exploring six hypotheses, about how 246 professional staff members in eight office work units of Michigan's State Department of Education communicate about eight publications of the Department, has suggested the following conclusions: 1. Subjects tended to communicate less with sub­ ordinates than with superiors. 2. Most office units (five out of eight) excluded at least one of the eight publications from their communica­ tions network. 3. Subjects indicating mass media channels as first channels of communication had no earlier publication aware­ ness than subjects indicating interpersonal channels as first channels of communication (for seven out of eight publications). 4. Subjects indicated interpersonal over mass media channels as the channels of greatest communication for five out of eight publications. Even so, none of the mass media channels was exceeded significantly on a one-tailed t test at the .05 level. 82 83 5-A. The more communications per subject in an office unit, the greater was the average consensus of attitude toward all publications. A Pearsonian correlation co­ efficient of .78 was found. Still, the general attitude toward all publications varied little, with mean attitudes in all eight office units falling in the somewhat-positive range. 5-B. The more communications per subject in an office unit, the more likely it was that all. members of the office were included in its communications network. A Pearsonian correlation coefficient of .62 was found. 6-A. No significant relationship emerged between the mean length of employment in an office unit and that unit's use of Interpersonal rather than mass media communication channels. In fact, a Pearsonian correlation coefficient of -0.40 was found for employment within the Department; and when length of employment in the field of education was considered, the correlation was nearly zero (.003). 6-B. N o significant relationship emerged between the mean length of employment in an office unit and the degree to which that unit included all staff in it3 communications network. A Pearsonian correlation coefficient of -0.82 was found for employment within the Department of Education; but when length of employment in the field of education was considered, the correlation approached zero (-0.19). 84 A sociometric analysis by a Fortran-Sociometricdistance-matrix program of the 2^46 subjects, plus 38 non­ subjects (also professional staff members of the Department), yielded the following conclusions: 1. The 64 most-integrated subjects included relatively more supervisory personnel than consultants than did the general distribution of the 2k6 subjects. 2. The office distribution of the 6k most-integrated subjects correlated only slightly with the general office distribution of the 246 participants, with a Spearman rank order coefficient of +0.55. 3. The number of sociometric isolates in an office unit correlated negatively with the number of sociometric cliques, with a Pearsonian correlation coefficient of -0.98. 4. The 28 central-persons who were highest both in integration and participation in close cliques of four or more included relatively more middle supervisory personnel (unit, section, and division chiefs) than did the 64 mostintegrated subjects. 5. A sociogram of these 28 central-persons showing their inner- and inter-office relationships revealed: a. that some of the 28 central-persons were sociometric isolates in their own office units, b. that some office units had contact with as many as four other units, 85 c. that some units had no representative subjects among the 28 central-persons or no contacts with other office units, and d. that some office units excelled In their In­ clusion of all members within their unit's communications network. Implications The next chapter will discuss those practical implications of this study which might help the Department Improve general communication within the organization. But, several implications of this research are relevant to the general field of communication studies. 1. Despite the possible unrealiabllity of staff recall of communications with subordinates and superiors, the ecco analysis type of field communication study still generated considerable data about communication in a formal hierarchal organization such as the Department of Education. 2. The classical communication patterns of wheels, chains and branches are identifiable with the aid of the Fortran-sociometrlc-distance-matrlx program authored by Dr. Nan Lin of John Hopkins University. 3. Habbe's and Bailey's conclusions (1952 and 19^8), that both oral and written communication forms were important, were supported in this research. 86 . Word-of-mouth or Interpersonal communication was found to predominate over mass media communication 50 per­ cent to 45 percent, supporting the results observed by Davis (1953). Rogers' (1 9 6 5 ) findings in communication studies of rural populations, that mass media channels gave faster communication in the early awareness stage, were not supported in this study. 5. Murray's (1965) findings that multiple audiences are associated with multiple publications was supported in this research. Questions Several questions raised by this research suggest the need for more information about specific aspects of the communication within the Department of Education. 1. Was the recall of a greater number of channels of communication with superiors affected by a psychological tendency to remember interaction with persons of higher status more easily than interaction with subordinates? Is a questionnaire like the one used adequate for measuring direction of communication within a hierarchal organization? 2. Do interpersonal channels of communication produce earlier awareness than mass media channels in other hier­ archal organizations? 3. Would other differences between long and short tenured employees suggest new ways of promoting communication? 87 Analysis of the specific mass media and interpersonal channels used by these two groups might tell us how best to disseminate new materials within the Department. 4. Why were some subjects and some office units isolated, both sociometrically and in their general aware­ ness of Department publications? 5. Since other publications (a mass media channel) accounted for 20 percent of all communication, what were these other publications? Are they mostly from within, or from outside of the Department? Are staff members finding them by design or by accident as far as the D e p a r t m e n t s dissemination program is concerned? Are there better ways to use the publications identified? 6. What deliberate techniques of dissemination were previously used for those publications about which there was the greatest awareness? 7. Why were subjects* indicated attitudes toward all publications so nearly homogeneous? Are staff members really in accord with the major Interests of the Department to the degree that their attitude toward the publications used in this study indicate? 8. Large apparent differences emerged between the professional educators* use of different communication categories or their integration within the communication networks between their employment in the Department of 88 Education and their earlier educational work. Department is causing this? What in the CHAPTER VI IMPRESSIONS AND RECOMMENDATIONS Impressions My impressions while doing this study are given here to present a broader base for understanding and interpreting the other, more objective data. Even though they were quite subjective, these impressions of the organization under study may be relevant. Peelings and attitudes were "in the air" throughout the planning and data gathering which gave insights into possible alterna­ tives for utilization of the identified communication patterns. When I first came to the Department to help write abstracts for Title I projects during the summer of 1966, I remember that a problem survey had just been completed. The first priority deriving from that survey concerned communications. The staff members In the particular section to which I was assigned were new to the Department. They had come in with the recently available funds from the Elementary and Secondary Education Act (ESEA). The general feeling among this new staff was that they did not know the 89 90 i "big picture" in regard to the Department of Education. They feared that there was duplication of efforts among the various units. They wanted orientation meetings which would give them insight information of the general activi­ ties of the Department. We were located in a long open room without partitions. Lively conversation took place back and forth across the expansive area all during the day. Questions asking where to find this bit of information or who should be contacted for that problem were constantly being shouted out. Frequently, the answer was found, but just as often no one present knew where to go to resolve the specific question at hand. place. A healthy exchange of ideas was taking Camaraderie rose. The feeling that "we are all in this together" predominated. Staff meetings were needed; orientation programs would have been welcome. But informa­ tion was interchanged openly and a lot of communication took place. I moved from the Title X office at the close of the summer of 1966, and did not return to visit for two years. In the summer of 1968, I took a contract in the Department with the Bureau of Research. While visiting some of my old friends in Title I, X was struck by the different atmosphere in the office area. been rearranged. The rooms had There were partitions throughout the once spacious room, and 12 to 15 little offices ranged 91 up and down the sides of the room. Secretaries were set into little private corners with filing cabinets separating them from each other. The glass partitions gave good visability, but you could not talk with anyone unless you went into his immediate office cell. Something was missing as I went down the rows greeting old friends and stepping into this office or that for a short talk. We would talk about the political and logistical issues of the day, but there was a new atmosphere. Suspicion was in the air. A lack of trust came through in the conversations. People talked as if they could not trust the man in the next cubicle. Even though we were in a closed room, voices ■ were lowered and furtive glances to the exterior preceded any discussion of "hot issues" or current problems in the Department or with Michigan education in general. Back in my new office, however, in the recently re-organized Bureau of Research, a new staff was pooled in one large room. Five or six new consultants were talking back and forth across the room. They were asking questions that struck me as very similar to those I had heard during the summer of 19 6 6 . Who do you see about this problem? What does the Department think about? information about? Research. Who do I see for Camaradarie was high in the Bureau of It was missing in the old Title I office. had happened? What Were there reasons for the change in climate? Had the people changed? Was communication going on in the 92 Title I unit that I had failed to recognize? Was there more written or more verbal communication taking place? The seeds of this dissertation were even then in my mind. While developing the design for this study, I went to the various division and bureau chiefs and asked for their approval and advice. varied. Reactions to my inquiries Without exception the chiefs liked the idea of a communication study. were different. However, reasons for their interest Political implications dominated their minds, as the following typical reactions make clear: yes, do the study in my bureau, and if there are repercussions I can handle it; it is not advisable to go to other bureaus; what are you intending to do with the find i n g s ; I am not sure that you w o n ’t be told to stop by people up the ladder, for it looks threatening; do you think it will do any good Just to talk about communication in the Department; I'll be glad to help; maybe the head office will see some of my problems this way; they treat us like we were not a part of the Department of Education; yes, do the study, only say that it is a study of organizations in general and that the Department of Education Is being used because it Is handyl yes, do the study, but tell my people that the head office Is conducting it and don't tell them it Is for your dissertation; yes, do the 3tudy and I'll ask for my s t a f f ’s cooperation. motives. All finally cooperated, if from various 93 The majority of the participants were quite receptive to the study. Only in one instance did someone speak out in a hostile tone after the completion of the instrument. The Individual said, form, what a waste of time." "Here is your blasted That meeting was in the unit where the chief had insisted that I say the study was being made by the head office and asked me not to mention my doctoral dissertation. In most meetings, the subjects were positive and happy about participating. They would joke and say things like; we sure need communication around here; do you really think this will help; at least you'll get your degree so it won't be a complete waste; I'm glad to see this study; please come back at another staff meeting and tell us about the results. By enlarge the mood was friendly and the participants sincere in their contribution. After the data was tabulated, I went to several chiefs and participants and asked them to look.at the m a ­ terial and contribute their thinking to an analysis. Several impressions came from this. Many immediately sought to Justify and give a rationale for the patterns identified. I found it very difficult to present the material in a way that would encourage people to use it constructively, toward real and specific goals. I was always left feeling that the job of overall communications belonged to "someone else," that there was little one * t person could do in such a large organization spread out in 94 so many places. Still, several of the per3on3 contacted showed interest in utilizing the findings. On two o c ­ casions, I was invited to prepare a summary and present it at a staff meeting. With this background and in light of the findings of the study, the following recommendations are submitted to the Department. A department-wide program will have to be initiated if the general communications and morale are to be improved. The following suggestions are presented as possible alternatives which would enable the Department to utilize the findings of this study for improvement of communications within Michigan's State Department of Education. Recommendations In the following recommendations there is a conscious leap from the specific to the general. The data and findings specifically refer to the communication n e t ­ works among the professional staff in regard to communica­ tion about eight selected publications. The general communication patterns of professional staff members is assumed to be similar for most areas of Interaction w ithin the Department, a view supported by most participants w h o contributed their thinking to the data analysis. 1. Channels. Increase the Use of Interpersonal Communication The Department and its professional staff 95 members would both profit from increased use of interpersonal or word-of-mouth communication, because: a. The use of interpersonal channels of commu­ nication actually resulted in earlier awareness for six out of eight publications (even though the findings were not statistically significant). b. The use of interpersonal channels resulted in a more positive attitude toward publications in five out of eight cases. c. Interpersonal channels were the most used channels for both of the publications which had attained greatest subject awareness. Bureaus could conduct periodic and specific staff meetings which will emphasize interpersonal communication. Task oriented or information sharing workshops are recom­ mended, so structured that small group interaction will take place between staff members in different sections and units within the bureau. 2. Emphasize the Use of All Communication Ch a n n e l s . Assuming that the Department profits when its professional staff members give a unified interpretation of the major departmental interests to the public, an increase in all types of communication would tend to give a greater con­ sensus in attitude toward these interests. This study showed that the larger the number of average communications 96 per subject the greater was the consensus of subjects1 attitudes within an office unit. Increased use of all communication channels would also help integrate more professional staff members into the office communications network, which would reduce isolated staff members. More integration would tend to include more long and short tenured staff and allow the total membership to capitalize on the natural communica­ tion strengths of each group. The longer tenured staff tended to be stronger in mass media communication resources and the shorter tenured staff tended to be stronger in the interpersonal communication resources. All new personnel should participate in an orientation program which will advise them of the major issues, duties, and'obligations of the Department. Further, the orientation should de­ scribe the various formal and informal channels of commu­ nication available for their use. The new personnel should be periodically encouraged to use them and to feel it is a part of their professional duty to be aware of all major interests of the Department's operation. 3. Furnish Special Programs for Middle Management. The single outstanding characteristic of the 6*1 mostintegrated persons in the Department was that they were from the middle management ranks, i.e., unit, section, and division chiefs. As Cartwright (1959) pointed out, these persons are at node points or articulation points in a 97 hierarchal tree. Cartwright warns that any malfunction of such articulation points will leave widespread repercussions for the organization. In view of this, it will profit the Department to take steps to see that the middle management personnel have a more balanced interaction with those above and below them. An unbalanced emphasis of communication to superiors over subordinates will not be in the best interest of the Department. Further, when these same key position persons become isolates, as was shown to be the case in the sociogram presentation, large areas of the organizational tree could well be isolated also. middle management personnel, Special programs for the such as briefings on major departmental issues and "hot11 issues, and workshops to develop skills in encouraging communication between and with subordinates, would help the Department as a whole. The Department should involve their middle manage­ ment personnel in special work shops designed to increase their sensitivity and communication skills. Help with such programs is available from collegiate Education a n d Commu­ nication departments. . Promote Efforts to Include All Work Units in the Department Communication Struc t u r e . Some units studied were glaringly aloof from the general mainstream o f commu­ nication and communication patterns. A n attitude of isolation and even resentment about the isolation was expressed by the staff in some units. Many, voiced their 98 desire to bring their units into the mainstream of the Department. Periodic interaction between professional staff members of the several bureaus should be planned, emphasizing cross seeding of information, problem sharing, Joint perspectives, and cooperative problem solving. Such departmental inservice programs as new employee orientation and inner- and inter-office work­ shops, which will permit and encourage more interpersonal interaction among the professional staff members of the State Department of Education, in fact, any programs which will help build a "we-ness" or primary group feeling, will profit the whole Department in their leadership, service, and guidance of education in Michigan. BIBLIOGRAPHY BIBLIOGRAPHY Baker, Helen. "Company-wide Understanding of Industrial Relation Policies; A Study in Communications." Princeton: Princeton University, Department of Economic and Social Institutions, Industrial Relations Section, 19^8. Baker, Helen; Ballentine, John W.; and True, John M. "Transmitting Information Through Management and Union Channels." Princeton: Princeton University, Department of Economic and Social Institutions, Industrial Relations Section, 19**8. Bavelas, Alex. "Communication Patterns in Task Oriented G r o u p s ." Journal of the Acoustical Society of America, V o l . 22 (November, 1956). Broadbent, D. E. Perception and Communication. Pergaman P r e s s , 1 9 5 b . New York: Burns, Tom. "The Directions of Activity of Communication in a Departmental Executive Group." Human Relations, Vol. 7:1 (195*0, pp. 73-79. Cartwright D. "The Potential Contribution of Graph Theory to Organization Theory." Modern Organization Theory; A Symposium of the Foundation for Research on Human Behavior. Edited by M. Halre. New York: Wiley, 1959, pp. 25^-271. Cohen, Arthur M., and Bennis, Warren G. "Continuity of Leadership in Communication Networks." Human Relations, Vol. 14:4 (November, 1961), p p . 351-367• Cox, Keith D. "An Investigation of the Interaction of the Printed and Spoken Word." Unpublished Doctoral Thesis, Michigan State University, 1965. Davis, Keith. "A Method of Studying Communication Patterns in Organizations." Personnel Psychology, Vol. 6 (1953), pp. 301-312. « 99 100 Deutsch, Karl W. "On Communication Models In the Social Sciences." The Public Opinion Quarterly, Vol. 16 (19527, P"-' 367. Donald, Marjorie Nancy. "Some Concomitants of Varying Patterns of Communication in a Large Organiza­ tion." Unpublished Doctoral Thesis, University of Michigan, 1959. Eilon, Samuel. "Taxonomy of Communications." Admlnlstra- . tive Science Quarterly, Vol. 13:2 (September. 196B), pp. 2^6-288. ’ Guetzkow, Harold. "Communications in Organizations." Handbook of Organizations. Edited by James March. Chic a g o : Hand McNally and Company, 1965, pp. 53^-573. Habbe, Steven. "Communicating with Employees." in Personnel Policy No. 129* New York: Industrial Conference Board, 1952. Studies National Jacobson, E., and Seashore, S. "Communication Practices in Complex Organizations." Journal of Social Is s u e s , Vol. 7 (1951), pp. 28-40. Leavitt, H. L. "Some Effects of Certain Communication Patterns on Group Performance." Readings in Social Problems, 3rd Edition; also in Journal of Abnormal Social Psychology, Vol. 46 "(ijJ5l), pp. 3b-5d. Murray, John Stephen, "Influences of Self Concepts, Significant Others and Interactive Patterns of the Publication Activity of Agricultural Scientists." Unpublished Doctoral Thesis, Michigan State University, 1965. Peabody, Robert L., and Rourke, Francis E. "Public Bureaurcies." Handbook of Organizations. Edited by James March. Chicago: Rand McNally and Company, 1965, pp. 802-837. Roethlisberger, F. J., and Dickson, W. T. Management and the W o r k e r . Cambridge: Harvard University Press, : 1539 Rogers, Everett M. "Communication Channels in the Dif­ fusion of Technology: Combining Mass Media and Interpersonal Channels." Unpublished Essay, Michigan State University, 1968. 101 Rogers, Everett M. Diffusion of Innovations, The Free Press, 1902, p. 307. New York: Rogers, Everett M., and Meynen, Wlcky L. "Communication Sources for 2, 4-D Weed Spray Among Colombian Peasants." Rural Sociology, Vol. 30:2 (June, 1965), pp. 2i’3=STff. ---- Scott, W. Richard. "Field Methods In the Study of Organiza­ tions." Handbook of Organizations. Edited by James March. Chicago: Rand McNally and Company, 1965, pp. 261-30*1. Shannon, Claude, and Wearver, Warren. The Mathematical Theory of Communlcation. Urbans! University of Illinois Press, 19^9. Simpson, R. L. "Vertical and Horizontal Communication In Formal Organizations." Administrative Science Quarterly, Vol. 4 (September, 1959), pp. lbb-196. Starbuck, William H. "Mathematics and Organizational Theory." Handbook of Organizations. Edited by James MarcfT! Chicago: Rand McNally and Company, 1965, pp. 335-386. Travers, Robert M. "The Transmission of Information to Human Receivers." Audio Visual Communication Review, Vol. 12 (Winter, 19bl), pp. 373-385. Whyte, William Foote. Street Corner Society. University of Chicago Press, 19^3. Chicago: APPENDICIES APPENDIX A METHODOLOGICAL NOTES METHODOLOGICAL NOTES Preparing Data for Analysis The responses on the Instrument forms were numerically coded in preparation for punching on IBM data processing cards. Two decks of cards were made, as follows Deck 1 contained approximately 2,600 cards. Eight cards (one for each publication) were made for each subject and each non-subject whose name appeared as a channel chosen by any participant. The data coding on .each card specified the individual's name, Job and office unit, the publication under consideration, and'every channel of commu nication indicated on the Instrument for the specific pub­ lication. The coding for each channel indicated Included the name or place of the channel, its category (A, B, C, D, E, P, G, or H on the Instrument) and the Job position when appropriate. Preparation was geared to use on the Michigan State University's Control Data 3600 computer of an exist­ ing Fortran-Distance-Matrix-Sociometric program authored by Dr. Nan Lin of Johns Hopkins University and furnished by the MSU Communication Research Services. Deck 2 contained approximately 1,000 cards. A card was made for each publication to which each subject 102 103 had responded. The data coding on each card In Deck 2 specified the s u b j e c t s name, Job position, office unit, sex, age, years In present Job, years In SDE, years in the field of education, years in the field of education in Michigan, total number of publications responded to, the names of those publications, the specific publication under consideration, the first channel of communication and its category, the number of months elapsed from printing of the publication to the indicated awareness, channel of most communication and its category, the rating of the subject's attitude toward the publication, the total number of mass media channels indicated, the total number of interpersonal channels indicated, and finally the total number of channels indicated which were above, below, and lateral to the subject's hierarchal position. In assigning code numbers to the open-ended re­ sponses with which the subjects indicated their attitudes toward the publications, a five point scale was used: 1 - very negative, 2 « somewhat negative, 3 - neutral, 4 ■ somewhat positive, and 5 *» very positive. A relia­ bility check for scoring was run on a random sample of 30 sentences. A Pearson r correlation coefficient of +0.934 was found between this researcher and an independent scorer. Hypotheses Testing To explore individual hypotheses, the following procedures were, followed: ion Hypothesis 1: Using Deck 2, for each subject, the number of indicated channels upward from the s u b j e c t s hierarchal position were subtracted from the number of indicated channels downward from the subject. The d i s ­ tribution of subjects who had predominately downward or predominately upward communication was recorded by hierarchal position. And the distribution of subjects who indicated first channels of communication which were upward from the subject's hierarchal position was compared with the distribution of subjects who indicated first channels of communication which were downward. Finally from Deck 1, the total number of lateral, upward, downward, and nonclassifiable (as to direction) communications were recorded along with each subject's hierarchal position. Hypothesis 2: office units. Deck 1 was sorted into the eight For each unit, the indicated channels of communication were screened to find those subjects who had mentioned or had been mentioned by any other member of their office unit. "members." "isolates." These were removed and labeled All others from that unit were considered as The distribution of members and isolates were recorded as degrees of integration both by office unit and by publication. Hypothesis 3: Deck 2 was sorted by publication. For each publication the subjects were sorted as to whether they had indicated a mass media or an interpersonal channel 105 as their first channel of communication. Only subjects who had been employed in the Department at the time of the printing were considered in computing the mean time from printing to subject awareness for each publication. Hypothesis 4: Deck 2 was sorted by publication. The subjects were then separated as to whether they had indicated a mass media or an interpersonal channel as their channel of greatest communication. The mean attitude ratings for both the mass media and the interpersonal groups were recorded for each publication. Hypothesis 5-A: office units. Deck 2 was sorted into the eight The average number of total Indicated commu­ nication channels per subject was computed for each office unit. The standard deviation of attitude toward all publications was computed for each office unit. Then a Pearsonian correlation coefficient was computed between the average communications per subject and the standard deviation of attitude across all eight office units. Hypothesis 5-B: A Pearsonian correlation co­ efficient was computed between office units' percent of integration (computed for hypothesis 2) and the communica­ tion ratios (computed for hypothesis 5-A) across all eight office units. Hypothesis 6-A: office units. (in years) Deck 2 was sorted into the eight The mean length of subjects' employment (a) in present office, (b) within the SDE, 106 (c) In the field of education in Michigan, and (d) in the field of education was computed for each office u n i t . And for each office unit, the number of indicated mass media channels was subtracted from the number of indicated inter­ personal channels, yielding a positive or negative measure of the predominate use of interpersonal over mass media communication channels by each office unit. A Pearsonian correlation coefficient was computed between the index of predominate use of interpersonal oyer mass media channels of communication and the mean time of employment across all eight office units for each of the four length-ofemployment categories. Hypothesis 6-B: A Pearsonian correlation coef­ ficient was computed between the mean length of employment for each office (computed for hypothesis 5-A and the percent of integration computed for hypothesis 2) across all eight office units. 1 Sociometric Analysis In the sociometric analysis, the following procedure was used. Deck 1 was sorted by publication. The limita­ tions of the computer program precluded a sociometric analysis of all publications simultaneously. The available Sociometric-Distance-Matrix program had a capacity of 400 subjects (I.e. a 400 x 400 matrix), with a maximum of 25 choices per subject. Thus the raw data for each 107 publication consisted of one card for each subject and each non-subject employee. Taking one publication at a time, the raw data was processed through the Michigan State University Control Data 3600 computer using a Fortran language Distance-Matrix Sociometric program authored by Dr. Nan Lin of John Hopkins University. Some items on the computer print-out were not pertinent to the present research; however, the index of integration for each subject was. This index was defined as the number of different persons with whom a given subject is connected directly or indirectly. For example, if subject 100 was mentioned by subject 101 as a channel of communication, the computer counted their connection as a direct connection at a distance of 1. If subject 100 was mentioned by subject 101 and subject 101 was mentioned by subject 102, the computer counted subjects 100 and 102 as indirectly connected at a distance of 2. As more and more subjects were between any given pair, the computer showed the indirect connection as a larger and larger distance level. of 283, The maximum distance was n-1 or a distance since there were 284 persons (246 subjects plus 38 non-subjects) in the study. For each publication, the 20 subjects with the highest integration indices were recorded. The 20 names from each of the eight publications were pooled and the 64 subjects with the highest sociometric Integration were 108 found. A profile of these 64 subjects was compared with a profile of the 246 total participants. Another output which was applicable to this study was found in the "clique print out." Using each subject as a central person, the print out showed the clique of persons connected to each subject and the distance of each connec­ tion. Using the 64 most-integrated subjects as central persons, a search was conducted to find out how many of the 64 most-integrated were also involved in cliques of four or more persons at the 1 and 2 distance levels. Twenty-eight subjects out of the 64 most-integrated were found to be involved in such cliques. These 28 were labeled "central- persons" because they participated doubly in communication: among the 64 most-integrated and in cliques of four or more persons with short chains (distances of 1 or 2). Prom among the various cliques Involving the 28 central-persons, a search was conducted to find those who were reciprocally included in each o t h e r ’s cliques. There were 24 of the 28 central-persons who were mutually Included in one or more of each others cliques. A formal sociogram showing the interaction between these 28 centralpersons was then constructed. Formulae The following formulae, used in statistical calculations, were taken from William Hays, Statistics for 109 Psychologists (New York: Holt, Rinehart and Winston, Inc., 1963). 1. Pearson product moment correlation coefficient (p. 506): NEXY - EXEY Xy / T N EX 2 - (EX)2 ] [NEY2 - (EY)2 ] 2. Spearman Rank Order coefficient 6ED‘ r s 0 1 N(N - 1) (p. 6 ^ ) APPENDIX B THE INSTRUMENT THE INSTRUMENT Base Data Questionnaire Your n a m e .................... Sex: 1. 2. 3. 4. 5. M P (circle one) Your Office Building..... A g e ..... Date today ........ Circle the letter representing your present position. a. non-teaching certificated professional (doctor, nurse, house parent, social worker) b. consultant (certificated teacher) c. Intern d. unit chief (dept, head, dean, asst, principal) e. section chief (principal) f. division chief (superintendent) g. bureau chief (SDE associate superintendent) h. State Superintendent and assistants I. state board member J. other (write in)...................... Years Months How long have you worked in the office of your present position? _____ ______ How long have you worked for the Michigan State Department of Educ.? _____ ______ How long have you worked in the field of education?______________________ _____ ______ How long have you worked in the field of education in Michigan? ______ 110 _____ Ill Publication Recognition List Instructions: Circle the number In front of every publication that you recognize at all or have heard something about. 1. Michi^an^School Finance Study by Thomas; Mich. Dept. 2. Education in Moral Values in Michigan; Mich. Dept. Ed.. May, 19b« 3. The Michigan State Plan for Vocational Educa t i o n ; M i c h . D e p t . E d ., R e v . A p r . r2>7 4. State Plan for Higher Education in Michigan. Smith; D e p t . E d ., S e p t . 19f>b 5. Excerpts from the Mich. Educational Television Feasibility S t u d y ; M.S.D.E.. 196b 6. General Information Pamphlet for Michigan School for the D e a f ; M.S.D.E.. 1958 7. Rules of the Civil Service Commission; February 15. i m ---------------: — 8. Analysis of Mich. Public School Revenues and Expendi­ tures 1967-68. Bulletin 1011; State Dept, of Education, December, 1968. 112 List of Communication Channels Publication Your name Directions: If you have heard of this publication, write In the names of all persons or places that were Involved in any of your communications about the publication. Place the name and its category in the appropriate section below. Some possible categories of channels of communica­ tion are listed here to help your recall. Extra Space on Reverse. Radio, television, read the publication itself, SDE News Release Memo, SDE Mich. E d u c , News, public newspaper, professional or non-professional magazine, professional or non-professional book, heard of it but d o n ’t remember the channel at all, remember the channel but don't remember the place, o t h e r .......... Secretary, clerk, intern, professional with no teaching certificate, certificated teacher, student, principal, department chairman, superintendent, s u p t . staff, consult­ ant, section chief, unit chief, division chief, bureau chief, state supt., state board member, college or university personnel, school secretary, remember the channel but d o n ’t remember who, o t h e r .......... I. Written or Mass Media C hannels: II. Verbal or Interpersonal Channels: A. Radio or Television F. Spoke with or heard about it from person(s) in your office B. Written memo or letter from person(s) in your office G. Spoke with or heard about it from person(s) not in your office but still within the SDE H. Spoke with or heard about it from person(s) outside of the SDE C. D. E. Written memo or letter from person(s) not in your office but still within the SDE Written memo or letter from person(s) outside of the SDE Publications Other: 113 I. Name the one person or place which you recall as your first channel of communication about this publication J. To the best of your recollection, write in the date of this first communication K. Name the one person or place which you feel was the channel for most of your communication about this publication. L. Write one sentence which reflects your feelings about the publication. EXTRA SPACE: ' EXTRA S P A C E : Use the space below if necessary to list the persons involved in the items A thru K above. Place the appropri­ ate category and section letter by each name.