GENEflAL VS: SPECEHC ENS‘FWMENT CONTENT EN EVAWEAHE‘EG COWNECATEON AMONG WWAQE magmas “masts {:09 flu Deg?“ a)? M. A. MECE’EBGM STATE GE‘YERSK‘E' Merz‘is Sc Spier £966 TH ESlS LIBRARY Michigan State University ABSTRACT GENERAL VS. SPECIFIC INSTRUMENT CONTENT IN EVALUATING COMMUNICATION AMONG AEROSPACE ENGINEERS by Morris S. Spier The problem was to explore the extent to which, in evaluating communication among aerospace engineers, a self-administered, paper and pencil instrument containing relatively general content could be substituted for a coded interview method concerned with content relatively specific to the organization under study. Interviews (MI) sampling engineers' knowledge of management-control policies and procedures and various aspects of general departmental operations were administered to engineers serving manage- ment functions as part of a longitudinal study (Wickert, 1965) designed to evaluate the management-control communi- cation system in a medium—sized, engineering-research firm producing aircraft and aerospace components. Similarly, interviews (PI) sampling engineers' knowledge of various technical and Operational aspects of projects on which they were currently working were administered to subjects work- ing on projects. Morris S. Spier Sprecher's (1964) "Creativity and Productivity Survey" (CPS), originally developed to sample engineers' perceptions of behaviors characteristic of "creative re- search scientists," was modified for use as the general instrument in the company studied. The CPS was adminis- tered to the same sample of engineers (N = 5M) interviewed in the 1965 phase of Wickert's research. Measures of agreement were obtained by correlating the interview re- sponses of each engineer with those of every other engineer. Similarly, the CPS responses of each engineer were cor— related with those of every other engineer. Three analyses were performed on the data. The first compared the results obtained by means of the general and specific instruments in an attempt to determine the ex- tent to which the respective measures were parallel. Both the CPS—MI and CPS-PI comparisons indicated that the two sets of measures were essentially uncorrelated. It was con- cluded that the general instrument could not be directly substituted for the specific instrument as a short-cut in evaluating communication among aerospace engineers. Another conclusion was that the two instruments were each measuring something distinctive. From one point of View each could then be correlated with a number of different measures in an attempt to better ascertain the nature of these two instruments. From another point of view, that of the Wickert (1965) study, the interviews were not merely picking up Morris S° Spier common stereotypes among the engineers that could have been measured by means of some general verbal instrument. In other words, the present study served as a kind of control for the larger study's basic instruments, namely, the quan- tified interviews. Second, since the two types of measures were es- sentially uncorrelated and measuring different things, an attempt was made to determine the extent to which the re- spective types of measures could differentiate between departments and projects characterized as having a high degree of internal communication (as identified by the formal Organization Chart) and a control department and project artificially set up to have low levels of internal communication (Criterion l). The results of the CPS data did predict management communication. The CPS data did not, however, discriminate meaningfully between actual and control projects. The CPS measures were, however, capable of isolating individual deviant cases in both the manage- ment and project settings. It was also found that the MI and PI measures did not differentiate meaningfully between actual and "artificial" departments and projects. In the third analysis, an analysis parallel to the second analysis, the CPS and interview data were re— lated to a second criterion (Criterion 2). Criterion 2 consisted of estimates of goodness-and-frequency of com- munication between all possible engineering pairs made Morris S. Spier independently by two raters. A check of these ratings showed that inter-rater agreement was only moderate for both the management and project estimates. Findings regarding the relation between these ratings (Criterion 2) and the communi- cation measures showed that the CPS data were statistically associated with the management estimates but not with the estimates of project communication. Both the MI and PI data were significantly associated with the management and pro- ject estimates, respectively. The strengths of all the rela- tionships were rather weak. One conclusion was that introducing the CPS as a general measure into the framework of the broader com— munications measurement research effort, of which the present research was a part, might help to overcome several technical problems of specific instruments (MI and PI), namely, limited applicability and the need for constant revision to meet changing organizational conditions. The CPS had the added practical advantages of ease of administration and scoring. In addition, the content of the instrument was sufficiently general to allow application to all areas and levels of operation within the organization studied, yet it was inter- esting and meaningful to the type of personnel involved without arousing the fear of management evaluation of the subject's performance or knowledge. Approved; M ”2, WM Date: C}u/\4 34h la 6 L U r GENERAL VS. SPECIFIC INSTRUMENT CONTENT IN EVALUATING COMMUNICATION AMONG AEROSPACE ENGINEERS By Morris 8. Spier A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1966 To my wife, Alice ACKNOWLEDGMENTS The author wishes to express his deepest appre- ciation to Dr. F. R. Wickert, chairman of his committee and major professor, for his guidance in this research and for his assistance in the develOpment of this manuscript, and to Drs. Alice H. Eagly and J. E. Hunter, whose stimu- lating ideas were of immeasurable value. Thanks are also extended: To the engineers and management of the company studied, who, notwithstanding their considerable contribution, must re- main anonymous; To Dr. T. B. Sprecher for permission to use his "Creativity and Productivity Survey" in the present research; To John P. Calicchia, friend and fellow graduate student for his encouragement; and To the author's wife, Alice T. Spier, who contributed to the well-being of both this thesis and its author. iii TABLE OF CONTENTS ACKNOWLEDGMENTS LIST OF TABLES LIST OF APPENDICES . . . . . . . . . . . . . INTRODUCTION . . . . . . . . . . . . . . . . BACKGROUND OF THE PRESENT STUDY . . . . . . . PROBLEMS AND PROCEDURES . . . . . . . . . . . . Sampling . ... . . . . . . . . Distribution of. the Surveys . . . . . . Description of the CPS . . . . ‘ . Scoring of the CPS and Data Processing . . Analysis of the Interview Data . . . . . Comparison of the Results Obtained Using ‘ the General and Specific Instruments Criteria . . . . . . . . . . . . . . . . . The Organization Chart . . . Estimates of Communication . . . Inter-rater Agreement . Comparison of Estimates with CPS and Interview Data . . . . . RESULTS 0 O O O O O O O O O O O O O 0 O O I Comparison of the Results Obtained Using the General and Specific Instruments Management Communication . . . . . . Project Communication . . . . . . . Comparison of the Results Obtained with Both Criteria . . . . . . . . . . . The Organization Chart . Management Communication Project Communication . . Estimates of Communication Management Communication . Project Communication . . . . iv Page iii iv viii 26 26 29 32 37 42 AH A8 50 5A 56 56 57 6O 63 6H 6H 67 7O 7O 73 TABLE OF CONTENTS--(Continued) Page DISCUSSION . . . . . . . . . . . . . . . . . . . . . . 76 SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . 90 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . 96 APPENDICES . . . . . . . . . . . . . . . . . . . . . . 98 Table 10 11 LIST OF TABLES Frequency of engineers administered MI and/or PI 0 O O O O O O O O O O O O O O O O O O O 0 Frequency of engineers with given job title administered MI 0 O O O C O O O O O O 0 Frequency of engineers with given job title administered PI . . . . . . . . . . . . . Frequency of engineers with given job title administered MI and/or PI and frequency of engineers who returned CPS . . . . . Frequency of engineers administered MI and/or PI and who returned CPS . . . . . . . . . . Illustration of the scoring of the CPS using responses of a hypothetical subject . . . Frequency of engineers with given job title for whom both CPS and MI data were available Frequency of engineers with given job title for whom both CPS and PI data were available Contingency table. Communication estimates of Rater 1 vs. communication estimates of Rater 2 for management interviewees . . Contingency table. Communication estimates of Rater 1 vs. communication estimates of Rater 2 for project interviewees . . Agreement between Rater l and Rater 2 in estimating the degree of communication between engineers administered: (a) MI, and (b) PI. Chi-square test of associ- ation, N, degrees of freedom, significance level and coefficient of contingency . . . . vi Page 16 l6 16 28 28 33 “3 “3 51 51 52 Table 12 13 14 15 l6 17 18 19 LIST OF TABLES--(Continued) Contingency table. CPS vs. MI data . . . . . Comparison of results obtained with CPS and (a) MI, and (b) PI. Chi—square test of association, N, degrees of freedom, sig- nificance level, and coefficient of contingency . . . . . . . . . . . . . . . Contingency table. CPS vs. PI data . . Average correlations, variances, N's, and standard deviations for actual and control departments for: (a) CPS, and (b) MI data Average correlations, variances, N's, and standard deviations for actual and control projects for: (a) CPS, and (b) PI data . Comparison of (a) CPS and (b) MI data with estimates of communication. Chi-square test of association, N, degrees Offifreedom, significance level, and coefficient of contingency . . . . . . . . . . . . . . . Comparison of (a) CPS and (b) PI data with estimates of communication. Chi-square test of association, N, degrees of freedom, significance level, and coefficient of contingency . . . . . . . . . . . . . . . . Summary of CPS-criteria and interview- criteria analyses . . . . . . . . . . vii Page 58 60 62 65 69 71 7A 79 Appendix A ’1JL'IIUOCI1 LIST OF APPENDICES Management Interview Schedule Project Interview Schedule . The CPS O O O O O O O O O O O O 0 O O 0 Management Code Sheet . . . . . . . . . Project Code Sheet . . . . . . . . . . . Matrix of CPS Correlations between Pairs of Engineers Administered MI . . . . . Matrix of MI Correlations between Pairs of Engineers . . . . . . . . . . . . . Matrix of CPS Correlations between Pairs of Engineers Administered PI . . . . Matrix of PI Correlations between Pairs of Engineers . . . . . . . . . . . . . Contingency Tables for Regrouped Data: (a) CPS vs. MI; (b) CPS vs. PI . . . . Matrix of Management Communication Estimates--Rater 1 . . . . . . . . . Matrix of Management Communication EStimateS--Rater 2 o o o o o ‘o o 0 Matrix of Project Communication Estimates—— Rater l O O O O O O O O O O O ‘ O O ‘0 0 0 Matrix of Project Communication Estimates-- Rater. 2 O O O O O O O O I O O O O O O 0 Matrix of Combined Management Communica- tion Estimates . . . . . . . . . . . . viii Page 99 105 108 112 115 118 119 120 121 122 123 12A 125 126 127 Appendix P LIST OF APPENDICES--(Continued) Matrix of Combined Project Communication Estimates . . . . . . . . . . . . Contingency Tables. CPS (Management) Data vs.: (a) Rater l; (b) Rater 2; (c) Combined Estimates . . ., . . Contingency Tables. MI Data vs.: (a) Rater 1; (b) Rater 2; (c) Combined Estimates . . . . . . . . . . . . . . Contingency Tables. CPS (Project) Data vs.: (a) Rater 1; (b) Rater 2; (0) Combined Estimates . . . . . . . Contingency Tables. PI Data vs.: (a) Rater 1; (b) Rater 2; (0) Combined Estimates . . . . . . . . . . . . . . ix Page 128 129 132 135 138 INTRODUCTION The present study is an attempt to determine the extent to which, in evaluating intraorganizational communi- cation, an instrument containing relatively general content can be substituted for an instrument containing content relatively specific to the organization being studied. The importance of intraorganizational communication in industry has been discussed by several writers, although they differ somewhat on the values ascribed to good communi— cation, or to a particular kind of communication. Hall (1951) suggested that management and employees may come to appreciate each other's problems and difficulties through effective communication. Davis (195A) felt that good com- munication within management itself is essential as a pre- requisite to good managerial decision making. According to Otis and Treuhaft (19A9) and Speroff (195A), teamwork attitudes may be developed by good communication. Effective communication may also be considered a means toward increas- ing productivity (Gracey, 1956) and improving morale (De La Ossa, l9U7). In light of the many aspects of the industrial environment affected by the quality of intraorganizational communications, it is not surprising that the available literature contains numerous articles concerned with develop- ing theories of communication and with devising techniques for improving communication. But strangely enough, no one appears to have come forward with an objective technique whereby a company can quickly and easily evaluate its com- munications network if it has reason to believe that a par— ticular problem of production or morale, for example, may be due to communication difficulties. The communications research methods suggested to date have several drawbacks which would prevent their use for this purpose. The following studies provide examples of some of the existing techniques and their limitations. Maier, Hoffman, Hooven, and Read (1961) and Maier, Hoffman, and Read (1963) used interviews to evaluate the quality of management communications. They relied on highly specific content, namely, the nature of the subordinate's job duties and responsibilities to measure agreement between superior and immediate—subordinate supervisors. Read (1962) Operationally defined accuracy of upward communication as the degree of agreement between superior and subordinate about the relative degree of dif- ficulty that various problems cause the subordinate, i.e., the degree to which certain aspects of the subordinate's job were problems to the subordinate. The aspects of their jobs which subordinates considered to be problems were iden- tified by interviewing the subordinates involved. The problem areas identified by an analysis of the interview content were presented as a list to both the superior and the sub- ordinate and both were asked to rank the items to reflect the degree to which they were problems to subordinates. Thus, Read's approach, too, relied on the interview tech- nique for data collection, and on content which was highly specific to the company in which he was conducting research. One approach which did not use the interview method but still relied on highly specific content was that of Funk and Becker (1952). These researchers suggested the use of a multiple-choice test to evaluate communications based on policy manuals, memoranda, letters to the super- visory staff and rank-and-file employees, as well as all verbal instructions in a given organization. In this case, the effectiveness of intraorganizational communications was Operationally defined in terms of an individual's agreement with the established company policies and procedures. These techniques of evaluating intraorganizational communication may be criticized on two grounds. In the first place, they rely heavily on data collection techniques that can be costly in terms of time, effort, and dollars from the points-of—view of both the researcher and the com- pany. The Maier, g£_a1. (1961), the Maier, gt_a1. (1963), and the Read (1963) methods are limited by the tremendous amount of time and effort that must go into developing and administering an effective interview and the large number of man—hours involved in the content analysis of the com— pleted interview protocols. The Funk and Becker (1952) technique, too, is limited by the tremendous volume of material that may have to be reviewed in order to get material to write questions for the multiple-choice test and by the time and effort involved in constructing a re- liable and valid multiple-choice test. In the second place, the existing methods are limited by virtue of their content specificity. This limi- tation means, essentially, that after all the work involved in constructing the measure, the completed instrument may not be generally applicable from one company to another or even from one department to another in the same company. In addition, generalization may be difficult from one organi- zation (or segment of a company) at one point in time to the same organization (or segment) at another point in time since, as Flanagan (1959) pointed out, highly specific measures rely heavily on a stable work environment and will need drastic revision if the work setting is changed in any important respect. Thus, each of the methods discussed above is limited to use in the company in which it was de- veloped. If the authors wish to apply their techniques in another company, or even, in some cases, in other segments of the same company, the whole instrument may have to be revised. It is possible that more than one instrument may have to be develOped for situations in which departments within a particular company differ greatly from each other in function or set-up or for evaluation of communication in different companies. Any multiple—choice test of specific company policies and procedures (Funk and Becker, 1952) or any list of specific problems faced by subordinates in a particular company (Read, 1962) will require major revision in the event of any major change in organizational structure or in job content or structure. In summary, it can be seen that the problem of content generality-specificity is an important one in psycho— logical research. Psychologists are often called upon to answer questions about specific communication problems in a specific company. Research methods to date have tended to rely on the construction of specific indicators or scales to fit the situation under investigation. While these specific instruments may yield staisfactory results for the purposes intended, they may not be generally applicable from one research situation to another and any important change in the situation studied may require extensive revision of the specific instrument. Also, the task of having to de- velOp new measures and new methods of analysis to meet each new problem (even though the problems may occur in somewhat similar settings) can be both costly and time consuming. It would be an advantage to the psychologist to be able to use a general indicator which would do the job of many specific indicators constructed to measure the same things in different situations. The purpose of the present research is to deter- mine the extent to which a self-administered, paper-and— pencil type of survey containing relatively general content (yet which is meaningful and interesting to the engineers involved) can do the job of an interview schedule containing relatively specific content in evaluating the quality of communications effectiveness among aerospace engineers. Such an instrument would be a useful tool in communications research in industry. It is, of course, possible that some general measure exists which crosses over all occupational lines, but the identification of such an instrument is be- yond the sc0pe of the present study. This research will be concerned only with finding some general measure for use in engineering-research organizations in the aerospace industry. It is a first step in exploring how generalizable a more general instrument may be. BACKGROUND OF THE PRESENT STUDY The opportunity to compare the results obtained with an instrument containing relatively general content with those of an interview containing relatively specific content arose in connection with a study by Wickertl. In- 1964, Wickert (1965) began an investigation of a social psychological problem in industry, namely, the study of just how well the change from a relatively informal to a relatively formal system of management-control communica- tion had been accomplished in an engineering-research or— ganization producing aircraft and aerospace instruments and components. Since the present research is based on Wickert's study, it will be of value to review Wickert's study in some detail. In the early 1950's, the company studied had in effect a relatively informal system whereby the information required for company Operation was communicated. The com— pany's accounting functions, for example, an important source of management-control information needed for planning and coordination purposes, were highly decentralized and paper work was kept to a minimum. During the 1950's and 1Study in preparation, 1966. early 1960's, however, the company was experiencing a general,overa11 physical growth and business expansion as a result of a merger with another aerospace company and of the general maturing of the aerospace industry as a whole. At the same time, the Federal government was be- ginning to require stricter financial accounting in con- nection with its defense and aerospace contracts. These factors, taken together, resulted in a gradual formali- zation of the company's communications set-up with more stringent and more detailed accounting and management- control procedures. This, in essence, demanded a more highly centralized accounting Operation with greater re- liance on formal written documents for the communication of management-control information. Thus, the situation within the company up until 1964 had been one of a gradual formalization of the organization's communications system. Some employees, primarily the engineering-research per- sonnel, felt that the situation was beginning to require too much paper work and was thus not allowing them enough time to be "creative." It was at this point, in the early part of 1964, that the researchers were invited to evaluate the effectiveness of the division's communications set-up. In the year following the 1964 phase of the re- search, the company and the industry in general continued to grow. During this period, too, the company reviSed its management-control system to the extent that its accounting operations became even more highly centralized and the com- pany began to place even heavier emphasis on formal written documents for the communication of management-control infor- mation. In the early part of 1965, the researchers were fortunate enough to be invited back to the company to evalu- ate the effectiveness of the communications set—up in light of the changes that had recently taken place. (The present study is based on the results of the 1965 phase of Wickert's research.) Exploratory interviewing in early 1964 suggested that two types of communication were important in the organi- zation under study: management communication and project communication. "Management communication" in this creative environment consisted of: the communication of information about general departmental Operations, for example, the origin of projects and programs, employment and placement policies, problems encountered in day-tO-day work, and so on; and the communication of typical management-control in- formation: the upward flow of communications necessary for management's planning and coordination purposes, and the downward flow of orders and authorizations. The flow of information content characterized as "management communi- cations" occurred mainly between individuals in middle- management positions at the departmental level, i.e., de- partment heads and their immediate-subordinate section heads, and between individuals in upper-management positions, 10 i.e., division heads and the Vice-President of Engineering. "Project communication," on the other hand, consisted mainly of the communication of technical matters that fellow en- gineers need to pass on to each other in order to work co- operatively on common projects and programs. Thus, Wickert confined himself mainly to specific aspects of the job environment in evaluating communication among aerospace engineers. Two standard interview schedules were developed to get at this content in a systematic way: one set of questions was designed to measure management communication and the other project communication. The "management" interview (MI) was made up of a set of standard, open-ended questions sampling communications content in the areas of general departmental Operations and management-control policies and procedures. The interview was designed to take about one hour to administer. In the general area of project development, the interviewees were asked to consider the possible sources of impetus for new ideas, projects, and programs and then to list the obstacles to the develOp— ment of new ideas. The next set of questions was concerned with the planning stages of project develOpment and con— tained questions about the use, adequacy, and value of the "OCR" (Opening and Closing Report), a document used as a source of management information for decision-making pur- poses. Once a decision has been made to proceed with a 11 project, the project passes from the planning to the "EPA" (Engineering Project Authorization) stage. The interviewees were asked to consider the effectiveness of the EPA (an authorization and scheduling document) in helping the proj- ect engineer control his project, and to consider also the steps a project engineer could take to keep down production costs on his project, i.e., what resources in the company were available to him. Still in the area of controlling an on-going project, the interviewees were asked to list the benefits the project engineer gets from the Monthly Engineer- ing Forecast (an accounting document designed to supply time, man-power, and financial information for a project to date). Interviewees were then asked to list the benefits the proj— ect engineer and the department head obtained from the Monthly Engineering Forecast based on a number of projects. The interviewees were then asked to provide their idea of specific key figures that were indicative of the volume of business the division was handling. First they were asked, "About how much money do you think Engineering spent last month in the accomplishment of its engineering programs?" A second question was concerned with the ratio of company money to customer money that had been spent the previous month. The next question asked the interviewees to list the methods by which an individual's work assignments were kept adjusted to his special qualifications and interests. The interviewees were then asked to list the matters that 12 the department manager (even if the interviewee was not one himself) would like to feel free to decide for himself rather than refer up to higher authority. The last question asked for the opposite side of the coin, so to speak, namely, to list matters that the department head would like to refer to higher levels rather than decide for himself. The "project" interview (PI) as with the MI de- signed to take about one hour to administer, consisted of a standard set of Open—ended questions designed to get at each interviewee's common understanding of a specific proj- ect on. which he was currently working. They began with an inquiry about the purpose of the project and how the project gotstarted. They then asked the interviewee to identify his role in the project and to identify the other persons concerned with the project along with their specific roles. The next set of questions dealt with company and customer interest in the project and were designed to tap the interviewee's contact with the business aspects of the project. Next the interviewees were asked to state their idea of the most significant technical breakthrough to date on the project and to explain its significance. Then they were asked their idea of the most important remaining tech- nical obstacle and to give their Opinion of what was re- quired to overcome the obstacle. Following this, the inter— viewees were then asked to indicate whether or not the project was progressing in accordance with its established 13 schedule and to account for its being ahead or behind schedule. Finally, in this same general vein, they were asked how they thought the technical progress on the proj- ect was going compared with the dollars spent to date. (COpies of the management and project interview schedules appear in Appendix A and Appendix B, respectively.) Persons in the organization who knew the company well and who had some knowledge of sampling procedures helped in the selection of 60 engineers from a population of work groups to be interviewed in the 1965 phase of Wickert's research. The basic sampling plan was affected by the decision to Operationally define effectiveness of communication as the degree of agreement on aspects of the job environment between individuals who should be communi- cating Well with each other, e.g., in the case of the MI superior supervisors and their immediate-subordinate super- visors and in the case of the PI engineers working on the same project. The sampling plan was further influenced by the findings of the exploratory interviews that two kinds of communication were important in the organization under study: "management" and "project" communications. Table 1 shows the number Of management and/or project interviews administered. It should be noted that even though only 60 engineers took part in Wickert's study, a total of 70 inter- views had to be administered because 10 engineers served both management and project functions and were interviewed l4 separately in each capacity.2 The sampling procedure for selecting "management" interviewees differed as follows from the sampling method used to select "project" inter- viewees. The company's Organization Chart was used in selecting the engineers who were interviewed using the MI set of questions. The chart was used first to identify the various upper-and middle-management job titles that were in effect in the organization at the time the 1965 sample was chosen. Table 2 shows the number of engineers with a given job title who were administered the MI. All positions in the upper-management net were included in the sample: the Vice-President of Engineering and two of the three immediate-subordinate Division Managers. According to the Organization Chart, the Division Managers were the immediate superiors of the Department Heads. The Organi— zation Chart was then used to identify the population of departments Operating in the company at that time. Eight departments out of 11 existing departments were chosen to be included in the sample. The heads of each of these departments as well as each of their immediate-subordinate 2In actuality, while only 60 engineers took part in this study, 71 interviews were conducted because one engineer was working on two projects. He was interviewed separately in each of his roles. For purposes of compari— son Of the results of the general and specific instruments, however, one of his interviews was randomly chosen to be omitted for ease of analysis. This will be more fully ex— plained in the Procedures section of this paper. 15 Section Heads were interviewed. The average number of Sec- tion Heads under each Department Head was approximately 3. As Table 2 indicates, 34 management interviews were adminis- tered. Each engineer was interviewed individually. The projects included in the sample were, so far as possible, representative of the type of work going on in the organization at the time. It has already been pointed out that Wickert's measuring system was applied in 1964 under relatively loose management controls which had, how- ever, been subject to gradual formalization since the 1950's. The 1965 research was to be conducted under what were to have been tighter controls in light of the changes that had taken place in the organization during the intervening period (which possibly would make working conditions even less satisfactory for the traditionally freedom-from-controls- loving, creative man). The sampling plan required, there- fore, that the sample of projects chosen from the approxi- mately 50 programs existing at the time, be weighted in the direction of those projects which demanded the more creative output. The sample finally consisted of 10 projects ranging from the "pure" research type project to the more narrowly "applied" cost reduction and product-modification type proj— ects. In addition, the projects were chosen to represent programs at various stages of completion. The two or three projects engaged in classified work were not considered for inclusion in the sample. 16 DESCRIPTION OF THE SAMPLE ENGINEERS IN WICKERT'S (1965) STUDY Table l.——Frequency of engineers administered MI and/or PI. Interview Frequency Management 34 Project 26 Management and Project ‘10 70 Note: See footnote on page 11. Table 2.--Frequency of engineers with given job title administered MI. Job Title Frequency Vice-President of Engineering Division Manager Department Head Section Head LION UUCDNI-J Table 3.——Frequency of engineers with given job title administered PI. Job Title Frequency1 Division Manager2 1 Department Head3 6 Section HeadLl 5 11 Project Engineer 18 "3? Note: See footnote on page 11. lTen engineers served both management and project functions and were interviewed separately in each capacity. 2This Division Manager was interviewed separately in both his management and project capacities. 3Three Department Heads served both management and project functions and were interviewed separately in each capacity. “Six Section Heads served both management and proj— ect functions and were interviewed separately in each capacity. 5One project engineer served on two projects and was interviewed separately for each project. 17 Table 3 shows the number of engineers with a given job title who were interviewed using the "project" set of interview questions (PI). Persons who acted simply as con- sultants to a project were not interviewed. The sample of engineers interviewed on each project finally consisted of the key men who had been working on the project for some appreciable part of its life and who were considered to be regular members of the project team. This always included the project head and an average of about 3 other persons. The PI's were administered individually to each of the en- gineers working on each project. Table 3 also indicates that several of the engineers served both management and project functions and that one engineer served on two proj- ects. These individuals were interViewed separately in each of their capacities. In general, the subjects were cooperative in responding to both types of interview. Following the interviewing phase of the research, content analysis was performed on the completed interview protocols, and separate master code books were develOped to reflect the content items found in the content analysis of the management and project interviews, respectively. The protocol content was subsequently coded to reflect the number of times each content item actually occurred in the protocols. The prOposed scoring system would systematically compare the number of content items in common between en— gineers who should be communicating well with each other, 18 e.g., department heads and section heads in the same depart— ment and engineers working together on the same research project. Thus, Wickert operationally defined communication among aerospace engineers in terms of the extent of agree— ment between them on various aspects of their job environment. As research progressed, the investigators became aware of some of the difficulties arising from the use of data collection techniques which rely on highly specific content (i.e., specific to the company being studied) and unwieldy methods of analysis. One drawback, of course, from the point of View of both the experimenter and the com- pany, was the high cost, in terms of time and manpower needed for interviewing, content analysis, and coding. These prob- lems are especially serious from the company's standpoint. First, an average interview may consume as much as one hour Of an employee's time. This takes on added significance in cases in which the respondents hold important managerial positions or are comparatively highly paid specialists, e.g., engineers or scientists. A second serious consideration is the delay in feedback to the company due to the time-and effort-consuming character of the content analysis and coding procedures. The specific content of the interview questions suggested other limitations. In the first place, it neces— sitated the development of two interview schedules to measure management and project communications, respectively. In the 19 second place, the specific content of Wickert's interview limits the applicability of the instrument to the company in which it was develOped. Also, as was noted in the introduction to this paper, Flanagan (1959) has pointed out that a specific measure will require extensive revision if the work environment is changed in any important aspect. This type of change was noted shortly after the completion of the Wickert study when this researcher accompanied a group of students on a tour of the company studied. During a talk by one of the division accountants, it was noted that several of the management-control documents which had been in use while the data were being collected only a few weeks before, were now no longer in existence and several others had had their names and functions changed. The problems and limitations encountered in using the specific instrument suggested the need for some instru- ment of more general content and one that could be more easily administered and scored. In this sense, the present study is a kind of "control" on Wickert's study since it is concerned with the question of whether results parallel to those obtained by means of the specific instrument could have been Obtained using an instrument containing more general content. Several of the existing techniques for evaluating intraorganizational communication, along with some of their limitations, were described in the introduction to this paper. 20 It was also noted that a review of the available literature failed to turn up any attempts to develOp more general instruments for use in this research area. The present study must, therefore, be considered an exploratory one since the literature failed to suggest any guide lines for the direction that research in this area should take. The preliminary literature search did, however, turn up an instrument develOped by Sprecher (1964) to explore the vari- ability of meaning assigned to the word "creativity" in a variety of settings. Sprecher's (1964) "Creativity and Productivity Survey" (CPS), while not originally develOped for use in communications research, seemed promising for use as the general instrument in this research. Sprecher (1959) set out to explore the variability of the meaning of the term "creativity" by allowing engineers in a company producing aircraft components to read into the word whatever meanings they chose. The procedure was that of selecting work groups whose members were sufficiently well acquainted with each other so that the men could rate each other. Nine such groups were identified and 12 men were randomly selected from each group to take part in the study. Each engineer was asked to rank in terms of creativity the other 11 engineers in his group. No definition of crea- tivity was supplied. Following the ranking procedure, each engineer was asked to give reasons why he had chosen the tOp two men as more creative than the bottom two men and to give recent incidents as supporting evidence. 21 Each of the engineers participating in the study had also been asked to solve three brief Open-end engineering problems. The engineers were asked to rank the creativity of the answers to the problem from the most creative to the least creative. Each man was then asked to give reasons why the answers he had ranked at the two extremes differed in their creativity. Again, no definition of creativity was supplied. The reasons given by the engineers in explaining why men were judged creative and the reasons given for why answers to the engineering problems were rated as creative were categorized by content analyzing the data. Sprecher reports that the results were both expected and surprising. In the first place, the results, according to Sprecher, supported previous findings of researchers in this area that novelty and worth of ideas are important factors in creativity. On the other hand, Sprecher notes that the results of the content analysis brought out other factors which engineers considered important in judging creativity of both men and solutions to problems. These, according to Sprecher, seemed to be "work—habit" variables such as independence in problem solving and the achievement of com- prehensive answers. Sprecher reports that these factors may have been overlooked by previous researchers. In a paper presented at the Third University of Utah Conference on "The Identification of Creative Scientific Talent" in 1959, Sprecher (1964) considered the criterion 22 problem involved in identifying the creative individual in an organizational setting. The problem, as Sprecher saw it, was in part one of developing a definition of the term crea- tivity in light of the variability of meaning assigned to the word. In his report, Sprecher traced the background of his thinking which eventually led to the develOpment of the CPS as a "criterion analysis form," i.e., for use in determining the meaning of the word creativity as it is used in some specific setting. Sprecher noted that the results of the study re- ported above (Sprecher, 1959), in which engineers emphasized the importance of work habits when giving incidents and reasons telling why they had rated particular men or par- ticular solutions to engineering problems as creative, led him to search for other sources to support the view that work habits are important. Sprecher's review of the liter- ature, and the results of a content analysis carried out by him on a series of critical incidents describing the charac— teristics of successful engineers in a steel company (no reference provided) convinced him of two things: first, that creativity involved both ideas and work habits; and second, "that there was a sufficient similarity between the generalized kinds of characteristics needed in different types of engineering or research activity that a criterion analysis form could be developed which would have broad ap— plicability" (1964, p. 82). 23 Sprecher developed the CPS to be used in describ- ing the overall meaning of creativity in specific situations. Moreover, Sprecher (1964) noted that the same form, with different instructions, could be used to describe a particu— lar individual. Sprecher's (1959) research had provided him with lists of reasons engineers had given for rating par- ticular men as creative and the reasons they had given for rating particular answers to engineering problems as creative. Twelve different work habits and characteristics of creative ideas were chosen from these lists in such a way as to com- prehensively represent the factors mentioned by the engineers. The factors chosen include: independence, novel ideas, liking for problems, valuable answers, analysis, number of solutions, technical competence, planning, energy, persever- ance, communication, and personal relations. These are presented in the completed Survey as a list of statements worded as behavior descriptions. Each factor is represented six times in six different statements. Sprecher (1964) noted that some of the factors he chose were highly unlikely to be involved, but they were included to insure complete coverage. A more detailed description of the way in which raters use the CPS in actual practice will be offered in a later section of this paper. Although exception can be taken to the use, in communications research, of an instrument designed primarily to determine the meaning of creativity, two points are worth 24 considering. First, one of the characteristics of a general measure is that it must be broadly applicable. The problem then becomes one of being able to identify some quality which engineering-research organizations may have in common. An answer is a creative environment where jobs require new ideas to meet new and complex situations. Second, creativity is an aspect of the work environment which is general enough to the nature of the organization and the work performed by the organization to be used in the present study yet still be meaningful to the engineers and, thus, still be useful for research purposes. In terms of the present study, a look at the type of work done by the company studied and at the nature of the aerospace industry itself revealed that the company could best be described as a "creative" engineering—research organization. The emphasis on creativity and creative out- put can, perhaps, best be described as inherent in the com- pany's Operations. Without the ability to perform in this type of environment an engineer simply will not make the grade, and without the personnel with the ability to per- form creatively, an aerospace company simply will not be able to compete in the industrial marketplace. So the creative aspect of the situation can certainly be considered a basic part of each aerospace engineer's work environment just as are the management-control policies and procedures and the technical aspects of particular projects as measured 25 by the specific instrument. Another consideration touching on the applicability of the CPS has already been noted in describing the develOpment of the scale. It will be re- called that Sprecher considered the types of characteristics listed in the survey to be generally applicable ones, that is, they are work habits which would be characteristic of individuals doing engineering or research work in a variety of settings. A no less important rationale for the use of the CPS lies in the fact that the work habits making up the scale were collected in a company somewhat similar, at least in function, to the one studied in the present re— search. Thus, the behaviors described in the CPS are those which would be meaningful to the engineers who served as subjects in the present study in terms of their job environ- ments, i.e., they are the types of behaviors which an aero- space engineer is likely to encounter or which are likely to be demanded of him in his day-to—day work. In a sense, then, an engineer, in filling out the CPS, may not simply be describing the "creative aerospace engineer"; he may be describing a code of behavior for aerospace engineers within his own occupational frame—of—reference as reflected in the norms of his work group and communicated to each engineer through both the formal and informal interaction of the members Of the work group. PROBLEM AND PROCEDURES It will be recalled that the purpose of this re— search was to determine the extent to which an instrument containing relatively general content could be substituted for an interview schedule containing content relatively specific to the organization being studied in evaluating communication among aerospace engineers. As a test of this, the main thesis of this paper, the results obtained in the 1965 phase of an interview study conducted by Wickert (1965) were compared with the results obtained by means of an in— strument containing relatively general content (CPS, Sprecher, 1964). This was done in an attempt to determine the extent to which the results obtained using the general instrument were parallel to those obtained using the more specific measure. In-a test of a subordinate issue, an attempt was made, using two outside criteria, to determine the extent to which the two instruments were, in fact, measuring com- munication. The following discussion will describe these procedures in greater detail. SAMPLING The present research was set up and carried out in the engineering division of a medium-sized engineering research firm producing aircraft and aerospace components. 26 27 The basic decision affecting the sampling plan of this study was the decision to attempt to obtain CPS data from each of the engineers who had been interviewed in the 1965 phase of Wickert's research. The sampling plan required that the CPS be administered to the same sample of engineers (N = 60) described in connection with Wickert's interview study in the previous section of this paper. The subjects were males, and without exception were performing engineering work at the time of the study. Each of the engineers had been work- ing in the division for at least 2 years. Table 4 shows the actual number of surveys returned as compared with the total sample of engineers interviewed in the 1965 phase of Wickert's study. Of the 60 question— naires distributed, 54 were returnedmaking for a response rate of 90%. The pressures of the practical situation made it impossible to obtain the 6 unreturned forms. Table 5 indicates that CPS data were obtained for 22 of the 34 engineers who were administered only the MI as well as for 24 of the 26 engineers administered only the P1. In addition, the completed CPS was returned by 8 of the 10 engineers who were interviewed in both their management and project capac- ities. 28 DESCRIPTION OF THE SAMPLE OF ENGINEERS IN THE PRESENT STUDY COMPARED WITH THE SAMPLE OF ENGINEERS IN THE 1965 PHASE OF WICKERT'S STUDY Table 4.--Frequency of engineers with a given job title administered MI and/or PI and frequency of engineers who returned CPS. Frequency Job Title Interviewed CPS Returned V.P. of Engineering Division Manager Department Head 11 9 Section Head 27 26 Project Engineer l9 18 6o 54 ‘1? * Note: The reader is referred to Tables 1, 2, and 3 for a more comprehensive description of the sample of engineers interviewed in Wickert's (1965) research. Table 5.--Frequency of engineers administered MI and/or PI and who returned CPS. Frequency Type of Interview Interviewed CPS Returned Management 34 22 Project 26 24 Management and Project 10 8 70 54 29 DISTRIBUTION OF THE SURVEYS The surveys were distributed by the personnel staff man for the engineering division of the company under study with an attached memorandum indicating management's support of the research. The questionnaires were distrib— uted as; inter—office correspondence. The memorandum con- tained a list of the names of each of the engineers partic- ipating ill the study and the name of the man to whom the particular CPS was being forwarded was underlined. In addition, the following paragraph appeared on the sheet followed by the personnel man's signature: In order to provide Dr. Wickert and his associates with all information necessary to finalize the 1965 Management Communication Study project, please com- plete the attached survey and return it to me . . . Your early and conscientious attention to this re- quest will be beneficial to both our division and the MSU team. If you have any questions, please contact me. The 60 questionnaires were distributed in April, 1965, when approximately half of the interviews had been completed. All of the surveys completed (N = 54) were returned within 4 weeks after their distribution. The interviewing phase Of Wickert's (1965) study was still in progress at the time the questionnaires were returned. DESCRIPTION OF THE CPS The CPS, as used in the present research (see Appendix C) was a 4—page mimeographed reproduction of the original form develOped by Sprecher (1964) with only minor modifications to fit the situation under study. 30 The first page, in addition to providing space for the identification of the particular subject (e.g., name, department, etc.), contained the following explanation: This is a survey to ask your help in describing the characteristics of creative aerospace engineers. You are asked to judge carefully the performance of men you believe to be creative, and who have been pro- ductive in this sense. By giving your impression of the importance of various activities or characteristics you can describe such men. On the next page is a sample of the pro- cedure we will use as it might be applied to the field of office management. Please turn the page. Reference to the Psychological Service of Pittsburgh was deleted from Sprecher's (1964) original form and replaced by the present investigator's name followed by "Michigan State University, Management Communications Study Extension, April, 1965." In addition, the original survey asked for a description of the "creative research scientist"; this phrase was changed for use in the present situation to read "creative aerospace engineer." The second page consisted of a sample set of characteristics that could be used to describe successful office managers. This single set of characteristics served to illustrate the procedure to be used by the engineers in completing the CPS. It had been filled out in 3 successive steps with a description Of each step contained in a short paragraph to the left of each illustration. The third and fourth pages each contained 3 sets of characteristics to be used by the engineers in the present study in describing an individual in their creative 31 work settings. Twelve different factors identified by Sprecher (1959, 1964) as being those used to describe the creative person or product were represented in these six sets of characteristics: independence, novel ideas, liking for problems, valuable answers, analysis, number of solutions, technical competence, planning, energy, perseverance, com- munication, and personal relations. The factors were pre- sented as a list of statements worded as behavior descriptions. Each factor was represented 6 times in 6 different statements. In actual use, the engineer (rater) was required to mark each statement with either a plus (+), minus (-), or zero (0). A plus rating indicated that the rater considered this factor to be characteristic of the creative individual and a minus rating that he did not consider it characteristic. Thus, each rater established an indifference point at those items (if any) which he characterized as zero with the items rated plus above this zero point and those rated minus below it. After each set of statements was marked either (+), (0), or (-), the rater was required to rank, in the order of their importance, those items which he had marked plus and rank separately those which he had marked minus, essentially ig- noring the statements rated as zero. The following set of instructions appeared at the tOp of the third page immediately preceeding the first set of statements: In the following sets of items you have no re- strictions as to the number of statements you may mark as characteristic or as not characteristic of creative 32 aerospace engineers. You may mark all of the state- ments as (+), all as (-), or all as (O), or any com- bination of (+), (-), and (0). However, be sure to rank all of the (+) statements, and then all of the (-) statements, before going on to the next set. Do not rank the (0) statements. Below are six sets of statements, possible charac- teristics of creative aerospace engineers. Considering your background and experience, describe what you mean by creativity from your observations. Tie this in as closely to yourgpresent work situation as possible. After placing a (+), (0), or (-) before each statement, be sure to rank the (+) statements, and then the (-) ones. There is no time limit. Go ahead. Two modifications were made in these instructions as they appeared in the original CPS (Sprecher, 1964). First, where the phrase "creative research scientists" appeared in the original form, the phrase was again changed to read "creative aerospace engineers." Second, the phrase, "Tie this in as closely to your present work situation as possible" was underlined in the present set of instructions. SCORING OF THE CPS AND DATA PROCESSING The example contained in Table 6 below is meant to illustrate the procedure a hypothetical engineer might have used in filling out the CPS. Moreover, it illustrates the "final ranking" procedure performed on the data by this investigator for purposes of data processing. The "Rating'column to the immediate left of the statements in Table 6, represents the thinking of a hypo- thetical engineer about which of the statements he felt were characteristic, neutral, or not characteristic of an 33 Table 6.--Illustration of the scoring of the CPS using re- sponses of a hypothetical subject. Ranking Final Ranking (+) (-) Rating Statement 10 3 — 1. Shows his enthusiasm for new assignments 7.5 O 2. Keeps up a quite vigorous activity level 4 4 + 3. Gives out with a number of ideas readily 3 3 + 4. Knows what he is talking about technically 5 Thinks of a problem's prac- tical requirements 6 Shows he can get accepted by others he meets 2 2 + 7. Is able to reach a decision on his own 8. Tends to develOp new ideas & relationships 9 7.5 O . Considers every angle of an individual problem 11 2 - 10. Reports his findings concisely to others 7.5 O 11. Carried out a task thoroughly & carefully 7.5 O 12. Sees what may happen and is ready for it. individual in the hypothetical creative setting. The (+) and (-) columns represent the engineer's rankings of the statements which he had previously rated as plus or minus. Items 3, 4, 5, 7, and 8 were considered to be characteristic of the creative individual and were then ranked in the order of their "importance": item 8 was considered "most character- istic" and was ranked "1"; statement 7 'was considered next in order of importance in this particular work environment 34 and received a ranking of "2," and so on. Statements 2, 9, ll,and 12 were seen as "neutral" (0) and were not ranked by the subject. Finally, items 1, 6, and 10 were felt to be not characteristic and were again ranked in the order of their importance. Statement 6 was considered by the subject to be the "least characteristic" of the behaviors listed in the set of statements and was ranked "1." Itemiulwas also considered to be not characteristic as evidenced by the minus rating, but the subject did not consider item 10 to be as unimportant a behavior as item 6. Thus, statement 10 was ranked "2." Furthermore, statement 1, while still considered not characteristic, was not felt to be as un- important as either statement 6 orll)and was given a rank of "3." The resulting data reflect essentially a continuum of response from "characteristic" through "not characteris— tic" with the item rated and ranked as +1 having been con- sidered the most characteristic behavior of a creative individual in the imaginary work setting, the items rated zero seem as "neutral" behaviors, and with the item rated and ranked as -1 having been considered the least character- istic of the items in this particular set of statements. The completed CPS for the hypothetical engineer, then, would consist of 6 sets of statements of this type filled out in a similar manner. The column labeled "Final Ranking" in Table 6 did not appear on the CPS as it was administered to the en- gineers in the present study. It serves only to illustrate 35 an operation carried out by this investigator on each set of statements filled out by each of the engineers who took part in the present research. This Operation was performed in an attempt to retain each engineer's response continuum (i.e., his evaluation of the statements in terms of their "most characteristic" to "least characteristic" standings). The example in Table 6 again makes use of the hypothetical data. In the final ranking procedure, the ranked position of each statement still reflected the ordering of the items as they were rated and ranked by the hypothetical engineer. Since there were 12 statements in the set, the final ranking represented a continuum of response from 1 ("most character— istic") through 12 ("least characteristic") with each of the statements rated as zero (0) receiving tied ranks. The final ranking data for each of the 6 sets of 12 statements appearing on each subject's completed CPS were placed on computer punch cards in order to allow analysis of the data by the CDC 3600 computer facilities available through the Computer Institute for Social Science Research (CISSR) at Michigan State University. A program was written which compared the CPS responses of each en- gineer with that of every other engineer. Thus, the extent of agreement between any pair of engineers (i.e., agreement as to which behaviors listed in the CPS were characteristic of creative aerospace engineers in the company under study) 36 was Operationally defined as the degree of similarity be- tween their respective CPS response patterns as measured by Pearson product-moment correlation coefficients. Since CPS data were available for 54 engineers, the resulting data output consisted of a 54 x 54 correlation matrix con- sisting of the correlations obtained by comparing the CPS response pattern of each of the 54 engineers who had com- pleted the survey with that of each of the other 53 en- gineers for whom CPS data were available. These 1431 correlation coefficients constituted the basic CPS data upon which all subsequent analyses were performed. It should be recalled that the general and specific instruments were administered to the same sample of engineers in the same company at approximately the same time. Thus, two measures existed for each individual: first, his responses to the items of the CPS; and second, his responses to either the M1 or PI sets of questions. The brief outline of the method given at the beginning of the Procedures section of this paper noted that the results Obtained using the two instruments would be compared to determine the extent to which their results were parallel. The methods of scoring and of analysis of the CPS data have already been described. The following section will describe the method used to analyze the interview data and a subse- quent section will deal with the manner in which these two sets of data were compared. 37 ANALYSIS OF THE INTERVIEW DATA The management and project sets of interview questions were described in an earlier section of this paper. It was also pointed out at that time that the re- spective interviews were administered individually to each engineer depending on his function (i.e., management or project) in the division. Several engineers served both management and project functions and were interviewed separately in each of their capacities. One project en- gineer was working on two different projects at the time the interviews were conducted. He was, therefore, inter- viewed separately in terms of his membership in each proj- ect. Since, however, only one CPS was required of him regardless of the number Of projects he was involved with, one of his PI's was randomly chosen to be omitted from the analysis. Thus, Tables 2 and 3, in indicating that 70 interviews were conducted, reflect the number of interviews that were available for use in terms of the design of the present study; 34 different engineers were interviewed using the MI and 36 different engineers were interviewed using the PI. Following the interviewing phase of the research, content analysis was performed on the completed interview protocols by two graduate students working independently. Separate Master Code Books were develOped with the help of a mediator (to resolve disputes involving categorization) 38 to reflect the content items found in the content analysis of the management and project interviews, respectively. Code sheets were constructed using these categories. Two different graduate students, again working independently, went back to the original interview protocols and coded the content into the categories (which had been developed by the content analyzers) in such a way as to reflect the number of times each content item actually occurred in the completed protocols. The coders then met together with a mediator to develop a Master Code Sheet to reflect the con- tent of each interview in terms of the content analysis categories. When they were done, they had one Master Code Sheet which reflected the categories of response for each interviewee. The Master Code Sheet for the MI consisted of 326 possible response categories. If an engineer had responded in a particular category, the coder simply placed a check mark in that space on the Master Code Sheet. The Master Code Sheet developed for the PI content categories had 169 possible response categories. Again, if an en- gineer had responded in a particular category, the coder simply placed a mark in the space provided for that cate— gory on the Master Code Sheet. (COpies of the "management" and "project" code sheets may be seen in Appendices D and E, respectively.) Since the method of analysis of results called for a comparison of the response patterns of each engineer with those of every other engineer (this will be explained 39 in greater detail below) a review of the interview questions and of the content categories was made to determine cases in which such a comparison would not be meaningful for the pur— poses of the present research. Engineers' responses to questions 3.1 ("What do you see as your role with respect to this project?"), 3.2("About what proportion of your working day have you spent on this project during the past month?"), and 3.3 ("About what prOportion [of your time] do you expect to spend [on this project] during the coming month?") were subsequently omitted from the analysis of the PI data. It is obvious that an individual engineer's responses to these questions would depend entirely on his personal situation and any similarity of response between two engineers would only reflect similarity in their poSitions in the company. These questions would be useful in a communications research project in which an engineer's responses could be compared with some established company standard (for example, a job analysis or description) to determine his agreement (or knowledge) of the established policy. This was neither the purpose nor design of the present research. Accordingly, these 3 PI questions were omitted from analysis in the present research. A similar problem was encountered in.a review of the content analysis categories developed for both the "management" and "project" interview responses. Since one of the purposes of Wickert's study was to measure changes 40 in communications content from 1964 to 1965, the Master Code Books were develOped in such a way as to accurately and comprehensively reflect all of the engineers' responses. This again meant that some response categories should be eliminated from analysis in the present study. The types of categories omitted were primarily those which: (1) re— flected responses which were not responsive to the inter- view question and had thus been coded into an "other" category; (2) questions which for one reason or another had not been answered and for which responses had been coded in "no answer" or "not asked" categories; and (3)cate- gories which had been developed by the investigators not primarily as a reflection of the engineers' responses but, rather, with some separate analysis in mind. Of the original 326 possible response categories developed for the MI, 84 were omitted from the analysis for the above reasons. Similarly, 36 of the original 169 possible response cate- gories developed for the PI were omitted from analysis. Thus, what remained were 242 categories, any number of which might or might not have been used by an engineer in responding to the management set of interview questions; and 133 categories, any number of which might or might not have been used by an engineer in responding to the project set of interview questions. Each engineer's responses were placed on computer punch cards to allow for computer analysis. Two separate data decks were punched: one for the MI data and one for 41 the PI data. For example, all of the remaining 242 possible response categories developed in the MI protocol analysis were punched on cards for each engineer. If the engineer had responded in a particular category it was punched "1." If the engineer had not responded in the category it was punched "0." This same procedure was used in connection with the P1 to analyze responses to the 133 remaining cate- gories. A prepared program (BASTAT) was used to correlate (product—moment correlations) the response pattern of each engineer who had been administered the MI with that of every other engineer who had been administered the MI. Similarly, the response pattern of each engineer who had been adminis- tered the PI was compared with that of every other engineer who had been administered the PI. Thus, the extent of agreement between: (1) any 2212 of "management" interviewees (agreement on aspects of general departmental operations and management—control procedures); and (2) any Bag; of "project" interviewees (agreement on aspects of common projects), was operationally defined as the degree of similarity between their respective "management" or "project" interview response patterns as measured by Pearson product—moment correlations performed between each pair of engineers. The data output consisted of two correlation matrices: a 34 x 34 matrix re- flecting the values of r obtained by correlating the response patterns of the "management" interviewees; and a 36 x 36 matrix reflecting the values of r obtained by correlating the response patterns of the "project" interviewees. 42 COMPARISON OF THE RESULTS OBTAINED USING THE GENERAL AND SPECIFIC INSTRUMENTS The comparison of the results obtained by means of the general and specific instruments called for a com— parison of the correlation matrices resulting from the analysis of the two sets of data. Tables 7 and 8 show the number of engineers with a given job title in 1965 for whom both interview and CPS data were available. Table 7 indicates that CPS data were available for 30 of the 34 engineers administered the MI. Similarly, Table 8 indicates that CPS data were available for 32 of the 36 engineers ad- ministered the PI. This meant that both CPS and MI data were available for 30 engineers and that both CPS and PI data were available for 32 engineers. A new, 30 x 30 CPS matrix was developed from the original 54 x 54 matrix by pulling out the correlations between pairs of engineers who had been interviewed using the MI set of questions. The four engineers who had been administered the MI but had not returned the CPS were drOpped from the MI (34 x 34) corre- lation matrix. What remained were two 30 x 30 correlation matrices: the one consisting of the correlations between pairs of engineers based on their responses to the CPS (see Appendix F), and the other consisting of the correlations between the same pairs of engineers based on their responses to the management interview (MI) set of questions (see Ap— pendix G). The range of correlations within each of these 43 Table 7.--Frequency of engineers with given job title for whom both CPS and MI data were available. Management CPS Job Title Interview Returned V.P. of Engineering 1 1 Division Manager 2 0 Department Head 8 6 Section Head 23 23 34 30 Table 8.--Frequency of engineers with given job title for whom both CPS and PI data were available. Project CPS Job Title Interview Returned Division Manager 1 0 Department Head 6 5 Section Head 11 10 Project Engineer l8 17 36 32 matrices was determined and divided into 10 equally sized intervals. matrices by square test thesis that two sets of computed to ciation (if This allowed for a comparison of the two means of a 10 x 10 contingency table. A chi- of association was performed to test the hypo- no statistical association existed between the data. A coefficient of contingency was then describe the strength of the existing asso- any). 44 In a similar manner, a 32 x 32 CPS matrix was de- veloped from the original 54 x 54 matrix by pulling out the correlations between pairs of engineers who had been inter- viewed using the PI set of questions. The 4 engineers who had been administered the PI but had not filled out the CPS were drOpped from the original 36 x 36 project interview correlation matrix. What remained were two 32 x 32 cor- relation matrices: the one consisting of the correlations between pairs of engineers based on their responses to the CPS (see Appendix H), and the other consisting of the cor- relations between the same pairs of engineers based on their responses to the PI (see Appendix I). The range of cor— relation coefficients was again determined for each matrix and divided into 10 equally sized intervals allowing a com- parison of the two matrices by means of a 10 x 10 contingency table. A chi-square test of association was carried out to test the hypothesis that no statistical association existed between the two sets of data. A coefficient of contingency was computed to describe the strength of the association (if any) between the two sets of data. CRITERIA In an attempt to determine whether the general and/or specific instrument(s) were actually measuring "com— munication," the obtained data were compared with two ex- ternal criteria. First, an attempt was made to determine 45 the extent to which the general and specific instruments could, in fact, differentiate between "high" and "low" com— munication groups (as defined by the formal organizational structure of the company being studied). Second, an attempt was made to determine the relationship between the obtained measures of communication (i.e., product-moment correlations between pairs of engineers) and independent estimates of the amount of communication between these pairs of engineers made by 2 different raters. The Organization Chart The formal Organization Chart in effect at the time the present research was conducted was used to identify the engineers who, according to the existing orgnaizational structure, should have been communicating well with each other: this meant, essentially, in terms of "management communication," engineers at the department level, i.e., a particular department head and his immediate-subordinate section heads; and in terms of "project communication," engineers who were working on a common project at the time the study was carried out. Engineers working in the same department or on common projects ("high" internal communi- cation groups) should show greater agreement with each other than should engineers simply chosen at random and not working in the same department or on the same project ("low" internal communication groups). One test of the degree to 46 which the general and/or specific instruments(s) were, in fact, measuring communications, would be their ability to differentiate between these "high" and "low" communication groups. In all, 4 departments and 10 projects with a suf— ficient number of members (i.e., 3 or more) to allow com— parison and for whose members both interview and CPS data were available (see Tables 7 and 8) were identified by means of the Organization Chart. The 4 departments averaged ap- proximately 5 members each while data were available for an average of about 3 engineers per project. Again using the Organization Chart, a control "department" and a control "project" were defined consisting of engineers who, accord- ing to the formal set-up, should not have had any significant amount of communication with each other. Five engineers were randomly chosen to be members of the control department and four engineers were similarly selected to be included as members of the control project. These individuals were chosen with the following criteria in mind: 1. No two engineers in the control group should be members of the same department as defined by the Organization Chart; 2. No two engineers in the control group should be working on the same project as defined by the Organization Chart. Engineers were randomly drawn and then compared in terms of the above limitations until the desired number of men 47 had been chosen for the control department (N = 5) and the control project (N = 4), respectively Within each department, the MI response pattern of each engineer was correlated with that of every other engineer who was a member of that department. Similarly, the PI response pattern of each engineer working on a par- ticular project was correlated with that of every other engineer working on that project. Thus, an agreement measure was obtained for each of the 4 existing departments and the control department and for each of the 10 existing projects and the control project by averaging the correla- tion coefficients obtained between all the pairs of en— gineers in a specific department or project. The same pro— cedure'was carried out for both the CPS and the interview data. Two agreement measures, then, were obtained for each department (i.e., a measure based on the CPS response patterns of the engineers involved and a measure based on the responses to the management interview set of questions). Similarly, two measures were obtained for each project (i.e., a measure based on responses to the CPS and a measure based on re— sponses to the project interview set of questions). Thus, an attempt was made to determine the extent to which either or both of the instruments (specific and/or general) could differentiate between the four departments and the ten proj- ects (identified using the Organization Chart) and their control counterparts. 48 Estimates of Communication In yet another attempt to identify the engineers who should have been communicating with each other, 2 in- dividuals within the company, both serving management func- tions in the division and both in a position to be acquainted with the majority of the engineers who served as subjects in the Wickert (1965) study, independently estimated the amount of communication occurring between each engineer and every other engineer interviewed in the 1965 phase Of the inter- view study (N = 60). Approximately 1 month after all of the surveys had been returned, the raters were each presented with a 60 x 60 matrix to be completed according to the fol— lowing instructions: The following estimates should be made as of the first quarter of 1965; they should be based on the conditions that existed in the organization during that time period. On the chart below, place either a 0, 1, or 2 (whichever best describes the interaction between the two men being considered) in the appropriate cell. 2. Considerable official, administrative and/or technical engineering communication. 1. Some or Occasional Official, administrative and/or technical engineering communication. 0. Little or Na, official, administrative and/or technical engineering communication. While the interval, "the first quarter of 1965," may appear to the reader to be somewhat long in light of the month it took to collect the CPS data, it should be recalled that the interviews were still being conducted at this time. In fact, the completed estimate matrices were returned to this writer on his visit to the company to conduct the final interview 49 of the 1965 phase of Wickert's research. Since it was not reasonable to suggest that the raters undertake to complete two such matrices, and since the estimates were to be used to validate both the CPS and the interview data, the rating interval was made long enough to include the time needed to collect both sets of measures. The raters were allowed one week's time to complete the matrix. Both matrices were re- turned on time. Rater l was a Department Head in the company under study. He was, however, in such a position that he had, for a time, served as a member of the Vice-President of Engineer- ing's staff. Rater 2 was a Division Manager not interviewed in the 1965 phase of Wickert's research. He was, in a manner of speaking, the "right hand man" of the Vice-President of Engineering. Both men, as previously noted, were in a posi- tion to knowledgeably estimate the degree of communication between the engineers interviewed. Rater 1's original 60 x 60 estimate matrix was used to construct a 30 x 30 matrix consisting of Rater 1's estimates of communication for each of the 435 pairings possible among the 30 engineers for whom both CPS and MI data were available (see Appendix K). A similar 30 x 30 matrix was developed from Rater 2's original 60 x 60 matrix. This new matrix consisted of Rater 2's estimates of com- munication for each of the 435 pairs of engineers for whom both CPS and MI data were obtained (see Appendix L). 50 In develOping the matrices for the project communi- cation estimates, CPS and PI data were available for 32 en- gineers. Two engineers, however, had to be omitted from the analysis because Rater 1 was not well enough acquainted with them to provide communications estimates. Thus, Rater 1's original 60 x 60 estimate matrix was used this time to con- struct a new 30 x 30 matrix consisting of Rater 1's estimates of communications for each of the 435 pairings possible among the 30 engineers used in the analysis for whom both CPS and PI data were availab1e(see Appendex M). Similarly, a new 30 x 30 matrix was developed from Rater 2's original 60 x 60 matrix. This new matrix consisted of Rater 2's estimates of communication for each of the 435 pairs of engineers for whom both CPS and PI data were obtained (see Appendix N). Inter—rater Agreement Two contingency tables were developed using these data in an attempt to determine the extent of agreement between the two Raters in assigning estimates of communi— cation to the "management" interviewees (Table 9) and to the "project" interviewees (Table 10). Table 11 contains the results of a chi-square test of association performed on the data to test the hypothesis that Rater 1's estimates were not related to those of Rater 2. A x2 value of 13.28 was needed to reject the "no association" hypothesis for 4 'degrees of freedom at the .01 level of confidence. Table 11 51 Table 9.-—Contingency table. Communication estimates of Rater 1 vs. communication estimates of Rater 2 for manage- ment interviewees. Rater 2 0 l 2 z 2 5 13 18 36 (29.2138) (4.8828) (1.9034) (36.0000) 1 101 34 5 140 Rater 1 (113.6092) (18.9885) (7.4023) (140.0000) 0 247 12 O 259 (210.1770) (35.1287) (13.6943) (259.0000) 2 353 59 , 23 435 (353.0000) (59.0000) (23.0000) (435.0000) Note: Numbers in parentheses denote expected frequencies. Table 10.--Contingency table. Communication estimates of Rater 1 vs. communication estimates of Rater 2 for project interviewees. Rater 2 0 1 2 2 2 13 ' 21 9 43 (35.4874) (6.2276) (1.2851) (43.0001) 1 <6 “8 > < 34 4) ( 36 > < 79 ) 5.1977 11. 1 2.3 09 79.0000 (258.3149) (45.3310) (9.3540) (312.9999) 359 63 13 435 M (359.0000) (63.0000) (13.0000) (435.0000) Note: Numbers in parentheses denote expected frequencies. 52 Table 11.—-Agreement between Rater 1 and Rater 2 in esti- mating the degree of communication between engineers admin- istered: (a) MI, and (b) PI. Chi-square test of association, N, degrees of freedom, significance level and coefficient of contingency. N Degrees Significance CAB x2 (Pairs) of Freedom Level (a) Management Interviewees Rater 1 vs. Rater 2 219.37 435 4 p < .01 .59 (b) Project Interviewees Rater 1 vs. Rater 2 214.12 435 4 p < .01 .57 Note: Two engineers were omitted from this analysis because Rater l was not well enough acquainted with them to assign communication estimates. indicates that this value was exceeded in the case of both the "management" estimates and the "project" estimates. Thus, the null hypothesis was rejected in both cases and the alternative hypotheses: (1) that some statistical associ- ation did exist between the management estimates of Rater l and those of Rater 2; and (2) that some statistical associ- ation did exist between the project estimates Of Rater 1 and those of Rater 2, was accepted. Table 11 also indicates that coefficients of contingency computed between the manage- ment and project estimates of Rater l and Rater 2 were .59 53 and .57, respectively. Both values of CAB were significantly different from zero at the .01 level of confidence. Since the maximum value that the coefficient of contingency can take on in a 3 x 3 table is .816 (Guilford, 1965), there would appear to be only a moderate degree of strength of association between the estimates of the two Raters in both situations and, hence, only a moderate degree of agreement between them in assigning their communication estimates. Rater 1 tended to "see" more communication occurring than did Rater 2. For example, in Table 9, Rater 1 consistently assigned a greater number of "2" and "1" estimates to the management interviewee pairs than did Rater 2. Similarly, in Table 10, it can be seen that Rater 1 again assigned a greater number of "2" and "1" ratings to the project inter- viewee pairs than did Rater 2. Whatever the reasons for the differences between them (possibly, the difference in their positions in the company at the time they made the estimates) it is apparent that they did disagree. Table 9 indicates that of the 59 estimates of "2" assigned by both Rater l and Rater 2, they agreed only 18 times (i.e., of the 59 M1 pairs rated "2" by either Rater 1 or Rater 2, they both assigned "2" estimates to the same management pairs in only 18 cases. Similarly, they agreed on only 34 of 199 pairs rated as "1," and only 247 Of the 612 cases rated "0." A similar trend was noted in Table 10 for the "project" communication estimates. 54 The problem, here, of course, was one of deciding which of the Raters was "right," or more precisely, which set of estimates, those of Rater l or Rater 2, most accu- rafifly reflected the degree of communication as it actually existed in the division. Since the statistical Operations performed indicated that some agreement did exist, it was decided to develop a "pooled" communication estimate by adding the matrices of Rater 1 and Rater 2. It was hoped that such a combined estimate would cancel out some of their differences and provide for better prediction. Sub— sequently, the management estimate matrices of Rater l and Rater 2 were combined (see Appendix 0). Estimates of com- munication between MI pairs could now range from 4 ("con- siderable" communication), to 2 ("some or occasional" com- munication), to 0 ("little or no" communication). Thus, there were now 5 possible ratings of communication between any 2 management interviewees (4, 3, 2, 1, 0). The same procedure was performed on the "project" estimate matrices of Raters 1 and 2 (see Appendix P). Comparison of Estimates with CPS and Interview Data Three sets of estimates were now available for both the management and project pairs of engineers: first, a matrix of estimates by Rater 1; second, a matrix of estimates by Rater 2; and third, a combined matrix con- sisting of the summed estimates of both Rater 1 and Rater 2. 55 These communication estimates were now compared with the CPS and interview measures. Rater 1's management estimates were compared with the CPS data for the management interviewees in a 10 x 3 contingency table. Rater 2's management estimates were compared with the CPS data for the management inter- viewees in a similar 10 x3 table. A 10 x 5 contingency table was developed using the CPS (management) data and the combined estimates of Rater l and Rater 2. Chi-square tests of association and coefficients of contingency were computed for each of the 3 tables. The CPS data for the project interviews was similarly compared with the 3 proj- ect communication estimate sets of data (i.e., Rater 1, Rater 2, and Combined estimates) and the apprOpriate statistics computed. Rater 1's management estimates were then com- pared with the MI data by means of a 10 x 3 contingency table. Rater 2's management estimates were similarly com- pared using a 10 x 3 table. Finally, a 10 x 5 congingency table was develOped to test the relationship between the MI data and the combined estimates of Rater 1 and 2. Again, chi-square tests of association were performed on each table to test the "no association" hypothesis and contin— gency coefficients computed to describe the apparent strength of the relationship (if any). Similarly, the PI data were compared with the three project communication sets of data and the apprOpriate statistics computed. RESULTS COMPARISON OF THE RESULTS OBTAINED USING THE GENERAL AND SPECIFIC INSTRUMENTS The main problem with which this research was con- cerned was the extent to which an instrument containing relatively general content could be substituted for an in- strument containing content relatively specific to the organization being studied inevaluating communication among aerOSpace engineers. A comparison of the results obtained with the two instruments was carried out by means of 10 x 10 contingency tables. Chi-square tests of association and coefficients of contingency were also computed. It should be recalled that preliminary interviewing in the Wickert (1965) study identified two types of communi- cation, differing 111 both content and purpose, going on in the company. Thus, two separate sets of interview questions (management [MI] and project [PI]) were develOped and ad- ministered. The results pertaining to these two types of communication will be considered separately in the following two sections of this chapter. It should also be noted that in discussing the CPS data in connection with the "manage- ment" analysis we will be referring only to the CPS data that were available for engineers who had been administered 56 57 the management interview set of questions. Similarly, in the discussion of the "project" communication analysis, we will be concerned only with the CPS data for those engineers who had been administered the project set of interview questions. Management Communication Table 12 represents the contingency table develOped to compare the CPS and MI measuresl. The raw data from which this table was constructed are contained in Appendix F ("Matrix of CPS Correlations between Pairs of Engineers Ad- ministered MI") and Appendix G ("Matrix of MI Correlations between Pairs of Engineers"). An inspection of the matrix in Appendix F yielded a distribution of intercorrelations based on the CPS data ranging from —.51 through +.72. In a similar inspection of the matrix of intercorrelations based on the management interview data it was found that the cor- relations ranged from -.02 through +.51. These respective ranges were divided into 10 equally sized intervals (15 , 1An inspection of the 10 x 10 CPS-MI contingency table (Table 12) indicated that a large number of the cells contained expected frequency values which were less than 5. Since the X2 statistic is often difficult to interpret in such cases, the data were grouped (by splitting the cate- gories at the median of the obtained frequencies) to con- struct a 2 x 2 CPS-MI contingency table (see Appendix J). Analyses performed on these data (N = 435) yielded a x2 value of 2.18 (df = 1). The null hypothesis was not re- jected (p < .25). The computed values of CAB and r were both .07. Neither value was significantly differen from zero. HHOO0.0000 HHOO0.00 OOO O HHOOO.HO HOOO0.0 H O AHooo.Nv Ameo.v N O HHOOO.OO HHNOO.O O O HOOOO.NOO ANOON.O NO H HOOO0.000 ANOO0.0 OO H HHOOO.OOHO ANOO0.0 OOH O HOOOO.OHHO HOOON.O HHH O HOOOO.HOO ANON0.0 HO H HOOOO.HNO HNHOH.O NN O HHOOO.NO HOOO0.0 N O N OO.\OH. Aoooo.HNv HN HOOO0.0 O HOOO0.0 H HOOO0.0 O HNOON.HO m HOOOH.OO N AOON0.0V n Aowmm.mv m HOOOO.NO o AHNOO.HV H Amnmm.v o ao.\mm. .mOHoawsvoum vmuooaxm ouoawv mommnuaoumm OH OHOAEOZI.OOOZ « HOOO0.000 HOOOO.NOO HOOOO.OOHOHOOOO.NHV OO OO AOOOH.O HOOHN.O O H AONH0.0 HONNO.V O O AOOOO.HO HHONO.HO O N HOOON.OO HOOHO.NO OH O HONON.OHO HONNH.OHO OH ON ANNO0.0HO AOONO.HNO OH NN HNHOO.NHO HOHON.ONO NH NN HOOO0.00 HOHOO.OHO N OH HHOO0.00 AOOON.OO O O HOOOO.HO HOOOO.HO H O OO.\OO. OO.\ON. .OOOO Hz Om ANOOO.HOO HOOO0.0HO OOH OH HO OH HOOON.O HOONH.O AOOO0.0 ANOO0.0 O O O O HOOO0.0 HOOO0.0 HOOOH.O HOHO0.0 O O H O HOOON.NO HOOHO.HO HOOOO.O HHOO0.0 O , H N O ANOOH.OO HOOO0.00 HONO0.00 HHOHO.HO O O O H HOOOH.HNO HOHON.OHO HOHH0.00 AONHN.OO ON OH O O HONNO.ONO HOHOO.NHO HOONO.OO HONOO.OO NN NH N O HOOOO.NNO HOOOO.OHO HHNOO.OHO HOOO0.00 ON ON O O HOOOH.OHO HONOO.OHO AOOON.OO HOOOO.NO NH NH OH N HHNOO.OO AOOO0.00 HOmnO.NO HOOO0.0 O O N N HONON.HO ANONN.HO HOOO0.0 ANOO0.0 H O H O ON.\OH. OO.\OO.- OO.-\ON.- HN.-\OO.- mmo .OHAOB hoamwcfiuaoo .NH OHQOH Aoooo.mv N HOOO0.0 O ANOO0.0 O HOHOO.O O HHONH.O O HOOO0.0 O HOOO0.0 H HOOH0.0 O HOOON.O H HHHOH.O O ANNOO.O O OO.-\OO.- HOOOO.NO N HOOOO.O O HNOO0.0 O HOHO0.0 O HHONH.O O AOOOO.O O HOOOO.O O HOOHO.O H HOOON.O H HHHOH.O O ANNO0.0 O Hm.l\mo.l w OO.\OO. m¢.\OO OO.\OO NO.\NO. Hm.\cm. ON.\ON. OH.\OH. OH.\OO. NO.\NO. Ho.\¢o. VIVG MHIAHHINI INHNHDVNVN 59 points per interval for the CPS data and 6 points per interval for the MI data) to make up the classification categories in Table 12. Section (a) of Table 13 contains the results of a chi-square test of association performed on the data in Table 12. The hypothesis tested was that the results ob- tained by means of the general measure (CPS) were not related to the results obtained by means of the specific instrument (MI). A X2 value of 76.79 was Obtained for 81 degrees of freedom. This value was not sufficiently large to allow rejection of the "no association" hypothesis at a meaningful level of confidence. Since p < .75, the null hypothesis could not be rejected.' Section (a) of Table 13 also indicates that a value of .39 was obtained for the computed coefficient of contingency. This value of CAB was not significantly different from zero. Thus, on the basis of the value obtained, the strength of the associ— ation between the two sets of data would appear to be relatively weak. These findings were further supported by the results of the analyses performed on the 2 x 2 contin- gency table which was developed on the basis of the CPS-MI data (see footnote p. 57). The null hypothesis was again rejected on the basis of the X2 value (2.18) obtained (df = 1). The value of the computed coefficient of con- tingency was .07. A phi coefficient was also computed. 60 The obtained value of r¢ was also .07. Neither of these values was significantly different from zero. (The 2 x 2 CPS-MI contingency table may be seen in Appendix J.) Table 13.--Comparison of results obtained with CPS and (a) MI, and (b) PI. Chi—square test of association, N, degrees of freedom, significance level, and coefficient of contingency. Significance x2 N (Pairs) df ' Level CAB (a) Management Communication CPS vs. MI data 76.79 435 81 p < .75 .39 (b) Project Communication* CPS VS' 67.46 435 81 p .90 .36 PI data a Note: One engineer who had been interviewed on two separate projects was omitted from this analysis. Project Communication The results obtained with the CPS and the project interviews respectively, were compared by means of the 10 x 10 contingency table shown in Table 14.2 The raw data used to construct this table appear in Appendix H ("Matrix of CPS Correlations between Pairs of Engineers Administered PI") 2An inspection of the 10 x 10 CPS-PI contingency table (Table 12) indicated that a large number of the cells contained expected frequency values which were less than 5. Since the X2 statistic is often difficult to interpret in 61 and Appendix I ("Matrix of PI Correlations between Pairs of Engineers"). An inspection of the matrix in Appendix H yielded a distribution of intercorrelations, based on the CPS data, ranging from -.36 through +.77. In a similar inspection of the matrix of intercorrelations based on the PI data (Appendix I) it was found that the correlations ranged from -.11 through +.54. These ranges were divided into 10 equally sized intervals (12 points per interval for the CPS data and 7 points per interval for the PI data) to make up the classification categories in Table 14. Section (b) of Table 13 contains the results of a chi-square test of association performed on the data in Table 14. The hypothesis tested was that the results ob- tained by means of the general instrument (CPS) were not related to the results obtained by means of the more specific measure (PI). A X2 value of 67.46 was obtained for 81 degrees of freedom. This value was not sufficiently large to allow rejection of the "no association" hypothesis at a meaningful level of confidence (p < .90). Section (b) of Table 13 also indicates that a value of .36 was obtained for the coefficient of contingency computed to describe the strength of the association. This value was not significantly such cases, the data were grouped (by splitting the categories at the median of the obtained frequencies) to construct a 2 x 2 CPS-PI contingency table (see Appendix J). Analyses per- formed on these data (N = 435) yielded a x2 value of .61 (df = 1). The null hypothesis could not be rejected (p < .50). The computed values of CAB and r were both .04. Neither value was significantly differen from zero. 62 HOO0.0000 OOO OOO0.00 O OOO0.00 O NOOO.NHO NH HOO0.000 OO OOO0.000 OO HOOO.OOO OO OOO0.000 OO OOOO.NNO NN OOO0.0NO ON OOOO.OHO OH N NOOO0.0HO HOOO0.000 HOOO0.0NO HOOOO.OOHONOOOO.NOO OH NOOO0.0 O HONO0.0 O NOOO0.0 O HOONO.V H NNNHO.HO O NONOH.NO O NONON.NO N HHONN.HO O ANOO0.0 O HOONN.O H mn.\no. OO HOOON.O o Aommm.v c AHONm.Hv o Amoo<.mV o .ANNHm.mV O NONO0.00 OH ANOON.OO O HOOOO.OO O NOOOO.NO H HOOO0.0 N oo.\mm. .mOHOamsooum vouommxo muoamv mmmmnuamumm OH mumps52|.0uoz a. ON NOOOO.O O HOONN.O , O HONOO.OO . O HHHO0.00 O Amomm.HHv mH ANOO0.0HO mH Annam.mHv mH Ammwm.MHv OH ANOON.OO N NHOHO.HO O Om.\m¢. .m «CH ONNHN.V H HOOOO.V H NOOO0.00 O HHOO0.00 O Ammun.mHv HH AOOON.NNO HN Ammmq.mmv nu Ammo<.mHv NN NOOO0.00 O HOOOO.NO N NO.NHO. .Mumn Hm NO HOOO0.0 O HOOO0.0 N HOOO0.00 O NOOO0.00 O Aowmm.mHv 0H Aomoo.mHv NH Aqomn.omv mH HHOON.OHO HN .AmMMH.0v O NOOHH.NO O om.\mH. .m> mmo HOOO0.000 OO NOOO0.0 H HONOO.V . H HOHNO.NO O HOON0.00 O HOHO0.00 O ACHmm.HHv HH Amoem.HHv OH NOHO0.00 N NOOO0.00 O NOOHN.HO H mH.\mo. OH Ho. map pm OHON soap HOOHOMOHO OHHOOOHmHanm who: m Oo mOsHm>m .oocooncoo mo HO>OH Ho. .HOOOO HO OOO NOO OOH u 3.32 . ON .6. O m Onp pm Ohms Eopm HOOLOOOHO OHOQOOHMchHm Ohms Hoppcoo UGO OH am .O .O .m .m .H mpoohOHO mom m mo mmsHm> 2 One OOOO ONO er» tom NN u 2v OOHOHOHLOO egos mmmochno OOH OOHOB Ho>o OEOHH mo Hones: 0:» OOO OHOOnOHO some OH OHOOcHwOO Oo mpHmO no Hones: osp z 2 "opens "Amv OCOprHopuoo mepo>m Omozp OOHELOOOO Op pom: was OHOEHOO wOHZOHHom one H O O O O O OH O O O O O HONHOOO 2 OO. HH. HH. NH. OO. OO. OH. ON. OH. HH. OO. .O.O OOO. NHO. OHO. ONO. OOO. OOO. OHO. OOO. HHO. NHO. NOO. oonHsO> NH. OH. ON. NH. OH. OH. HN. ON. OH. CNN. HN. OHOO see: OOOO HO HOV O O O O O OH O O O O O HONHOOO 2 OO. OO. OO. ON. OH. OH. NO. OH. ON. ON. OO. .O.O OOO. NOO. OOO. OOO. OHO. ONO. OOO. NNO. NOO. OOO. OOO. OccOHaO> ON. OO. NN. HH.- OO. NO. HO. HH. NO. OO. OO. NHmO see: OOOO OOO HOV HONOOOO OH O O N O O O O N H OOOnOHm . mpmo Hm NOV ocw ammo HOV "pom OHOOnOHQ Hoppcoo 02m HOSpoO How OOOHO H>OO OHOOOOOO use .m.z .moosmHHm> .OOOHHOHOHHOO ommho> .. Emmrplsa - .11-731: 2 CE? 2. New rate those you by. we 3%, + l. .317 T231. iii-I . ’Wfi‘°3"¥-fl. 13‘1“?” irked plus («a-3.1256: the plus £1) + 2.. ;_Km€-m dict-mile of the procizime 11mm, mark she one which is 11....3t «- 3." Can type and fem aicim‘oian crumb by a. (1); the one which 13 o 11.. 2.3 willing to my}: lung ten-um mt mat Want by a ('2), e: c. - 5. fibreaeee prmmtnese in arrival n the exgmleo :; 0 6. - 38 neat 1.11 hi!) appearance: 1 + 7. A lees not play fsmvcxdtee _ 0 8. £33 an accounting unclaimed 5 + 9. _ Ber-38 not. criticime others. - 1.0. Asks for extra duties 2 + 1.1.. iii-ghee his deeieimzr-s’ camfzaly 0 .2. ifs-me to melee 8 {44M ism-z'ozsssion - , + - Emwwle ~ E??? 3 F? 3.. ‘ 11m mite those you have ""14 .3. ,1... .1: *5 11:1: meg: efi'ZFITeTifmfi sized minus (-). Using: the mixes 3 + 2. Karma details of the prcwfiuree . cell's-m, mark the one which is 1 - 3. 1 Can type and. ted-re dictation 30+ 1Werrt by a ('1) 5 the one 0 11». Ia Milling; to war-k 12mg ham lab. is :1 mt lemt irn'rieit by a 2 - 5. Sarcasm prematmufi in arrive]. .3, etc. See the exmmleo O 6. ; is meat in his gemwamnce l 4- 7 . 12398 not play favorites . O 8. E33 an accounting backgrmrml not. rate the items you mark-223511 5 + 9. Dame not criticism others tb zero (0). Do not spend too much 3 - 10. Asks for mm duties we on any one item or set of items, 2 + 1.1. Makes his decisions camfiilm ' 0 12. Haunt-s to mks 3. gm impression Riordan-1y on EEK? l. v I 1 «I' :4 - 5"» - , ~ 1. s l ( , $ 7 ‘ , ’ ~ 1 ‘ ~ 6 I ‘ I» . ‘1 ’ 1 1,. ‘ ‘ ‘ l‘ "l '2' LC; ‘3 ‘ I) i “I ‘ ‘3 a -,4 . A 3 . a ‘\' . .. s ’ 5 y ,. ’H | _ v A 11': g z ‘ '. 3 . .— v ',A‘ " AA,‘ !- I a “ ', J ’ u- 4. 1r i: d 1. 1:41. 6; .. ' J4 J ‘- 1 I in": : ... _. V 1 . I ' 1.. p i ,J‘ v‘- . l ‘ ‘3" .‘ '9 s i ’ ' ”.44 ' .. x :1 ' ,t' I ' : 0 ." " “1‘ 'x #5. ‘ . ~. ” ' £4 ' ?' \l h 9 "‘2 J x ... r' 'J’ 4 w I , l, .3 ’ :r', - 6 f '. ._ I 6 -.-. >6 “1: f ,r. , I I"; i 9“ a ' 4‘" &( L ' '8- "3‘9 k x L ’ 3 " I '- 411 q'o‘ " o. Aw" I . n ‘ f,‘ $. .7 r“ \ fl) .9 “i f' - L "_r . 'f - -. n. . -. '~‘ . , ._. :‘D . ‘. - 4 . . ~6'. 3 ,. . ’ . »- -- . .1- .- -‘ . .4", T 3 — - § 8 ‘ ‘ . - 1': ‘i -: 02.. 1' ‘ ' .‘1'. 19 L.’ - ,.‘ '2 ~ _ 4 ":6 ~ K. a x : , . - :4. 6,: l ‘ \ g- . 4‘. “no.4? . v 2."; ..'.- 1‘ '- ‘13} '-- .. . ~‘. 5'. 0!: 11* a; O 3. f”. L. 31': t. 5651:": in 3 9 ‘1": - I: ‘6' w .. »‘-.v {...-16:»; 5242'-» .'= .; 541‘“ 34 3‘; lug. Cd. ': 5’51»! 41‘! £3 . 1’? ' r- . . " ' . . I ‘ ‘ 4 " . "6.. . ”7 -'. '»' ' 4" Q‘. v’1l'f‘».' ’ "‘ K“ '3 “. 1L 9:“ _1"' ".3 tr 1‘: '6 ‘ 'Lf‘p 1.; .. ly.w_'-».:,..-.':x1'_ 1 4. .~ ,5. [- -"\,!4.'4-I.v_l 4.“. 94 ‘ ., .4 -% ,‘. '\ ’4": reagg I “3": ~~ ‘,'r_ -‘l 4 2‘: 1 * "3. T C r-:&-"-'r ‘ I: ""1" 2‘ -- u“! 2:! :{) A 3’ (fly—I L1) tn ‘_ J _(1‘ :3. ‘ err-1 "j- :2?- 23-h“ ‘9‘ «f ... '6'; _ hypo? “(IQ-'- ’ {’1 fluts lt-‘TA 1’1; . .1 " -.- ._ 2 J. . .".'. . m... ...: l . .' fin: .. 1-} .‘ . ~ 1' . u :- ) _ .‘ . 1 ., _Q, . F-. .' .t‘ . . _, e yua 2:16. -6-1 :53: are 1:: 111,11: .6 1.6:: fair :6 r 16'1“; .6 :2 6...}: 45:1} :61“: c 1 1.2;: 3.3 . ' “I“ l ‘ ~ .u.o- n.- .mv-‘r .«auvu \ ,1- .v| .-. .... 4.- , . .... _ ~4.~ .. ‘3.' .44 . . ~ " .u-.... I L‘ ,. . _ . Iniv v - ’ fl -. . ‘ . 1‘!" \Vfiz‘i“ 32” t hf: {l ’w' ".45).?“ t... t' :1 1 31:: 3'" 1.7411,." ”7.“? ‘5 ‘ 7" -: J '1'}? C i? l C 7 Ed 1'. l . ' : .4 5:"! is i‘r I j! 1 kn o ' “in ... “ _ ‘f‘: r - , J . _ ,4. .... ,*—‘-,_.V,_" ‘3. :4. g c. -‘ “a :a‘w- - ... ... -: 1.5\| :- ..- 4t. '_ 1" .‘ 5 i , I“ q ’ ‘. y + a v m "f , », .. t» - 4 ' fl . -' .: z 1 " 4: J v” ' ‘ ,-_‘.~ n‘;‘ 1‘. I 2' ..‘f" ., I ~ ’ _ i L. (i ":- ‘ I :9, .‘ ‘ .l .3,’ FIE '1 : ’1 \1 L 4 t)" t'. '0' % |;‘x' ‘1 9 1:.— 1.’ .':~‘ .f: t‘ "n- a %‘ fl' 7, E 1 A1. 1."): 31;, '_ (K E; ’ 4.123;»! [4'4 9:62:4‘...‘ 1. 3.. ~. ~- . . 1 . (V ,. .h 1 1 ~~ .3 0.. 91.35) 1 .6 on; e is: :20 ”8.516 4 :: 3 5:21.: 1:0 . v1; 9'- m‘~n --MP‘W“ “"ww'tr “way, an s 9”;':--- '1'- I I.._---’- . ‘A 11-21." ‘1‘Ofizfv-U‘ V11»- :11- »2. «4 1 I -,. .' : :0.- lfv‘ 1‘ "I“ Typfir- "An‘fi'VI-it—mlr' 6- 1.": scam-u. .L. -.' .m' .- u‘. A I am "-: ) b- 3"»3‘. ‘— .%..."w‘a *6 ':_ r \..v $- 1:. ..L: I: _ ”“733. {as} ix'kra t) nor—run; a») w 6‘ 5" 6 ~ - -. . 'f '5 ’2 z ...-6. . 2:» .‘~ «fig-...... 4‘ .. .6 i "'1 a "A, 1::- it 13‘" ‘6'“ «63/ 51142.3: 4.; .‘ ’13:: 6 £332, 4‘. .2. 1.6.23: 16:: .~ "7)“ ‘ ...) '1' J? . u. ‘i. ... . A _ ‘.'_ 9. 4,‘ _ 'fi .',__ g' ‘6 :- “15";538 5:13: [VJ-2562 34c: + -5) {QWSK11’ 6:? .4733 “111,13 16:16.21 5 1 : ':~ , . 1. . -. 6. '1 . 4 13..- " . ,.-'. fl “ 3 V .- ....‘ i '4.- :» LL?K27<,';. .23 a5, {)1 554.34... ”"2133 t; VJ": ’11 31* .510 6566.61. ., LB 6.? ‘s 12116-3516?- S I 0‘ «.KJ '1 fl . f." - ‘ a“ 1; 5' ‘-d '4 ~ I? '. ._Ku -" a W". _-- ~ ‘- f“ ,t, .x ‘ '. I ZEN; i" LL16 1:3 11) a”: 2296-.» at 61* 25.5.. J15 1.1.? 8631.3. ~-.;;"T . :. :1»). .: 22.5.1'6153161‘3 . -.. ,a' . .. .‘ ‘ 5 l L166; 1 ~“ '2: 11:66 0 WW: K‘ w f‘rr a..." " 1.4: wn “”6 W66- .6. .. l 62:. k. m 1.: his 1': w : x $81., .1; 6», 4.41.! J a) 391: '3. J. U 3.1.”: 4 v? "g p ~ 2 I ‘ \, . i 'f ..6‘ 1 -.'~' 1 4‘ . -1” {I} . ., 0 5.35.4”: :1 “F" ‘3‘). .r 65' 5; 138 a '32.?“ l" :7“! 86‘? U; at! ‘14,: '3 63. My 1 .'" v » - . v.- ‘- ' k? -' n , ‘. ; 4 *1.“ , .".'~. '._ . .“V, .- .g . r_ - n? f». v . '4 3 . ... r. ’-I. ’1' t.. do uuli‘fi» ._ £666.13". 1.} i': ' :39. 3 $.33 E..:. r1731}? ‘3‘) . . -' '2;.:..;:,‘:K’c I: T; 8,111.3 3" ’13 161m” 16: .6. tszr‘r‘ 16.63:, in £—:.:;1'11¢6.‘a: g? ...:m. 529 T8 5.6:: 015 2.6666528: 5:36.; '8‘??? 13:61.“ 2.6.123 5m 516:9; it '1; 332$ ’11: 6141.3” Q,"L"'t "I. an fiFA"," 4 "935: LG Lu"? 2211:.54‘1’K3133 ii 11?; 661336361" .1 ‘3 tr; t3 .36»;- r 35,111.. “3123155. 6‘: 8316:1181: {TV-.234“- d} a ““‘“"' - ,, , . 3‘: .u... f. my»- 1 ._ . a» 5. T1- ' ’.. n '6 1» -— -. . Wm 4'4 - .. - ..4'. - ' ~ '3‘ 1. ‘ t 4 ‘ A 3 . J". '_ [I -. 5 6 4 u N; .31 C ~6 I f - 3 g .y‘ . v. _ .‘v-h . '1 ‘J BMU~J§F~ 7 ‘ L; ..L a “Li";- E‘} :' Livia.-- J. ":12: 6 Di 2--‘~ t" cfi- : 2T1 «‘6 :r‘ {1‘94 6.45.1 1:58 o :4. WM" 21;) 8 {311564655 vigmfeua 61635571“ '2" 31"!»5‘51 . (55.6661: 3. 62-6666 1:71 'bi a I!'E:£§L‘"”fft" c5? 5:382:63 ‘6 66-66661. ~21 KG-(fl?‘é! Ritz-56:6: 11,16- is 1231‘ f, 8621:6116“, '6»: . :11; Think; re? a: prob 36:16 :3 15:1..251c661. “r 11'8“ ‘2' “Ms 84666665129: mm get-65; 56.62.986.681 by t:t?~.6:;K.-w-6‘6 13.12 26632:. ‘18 X8 46‘223166 m 16866625.: 66 ($361.32: in. m 2:: 8 ‘Wa'l Tfiiinéfi 86368668138 7.66:: i.ei-Hw {-..-3265. mi '35.: .151:3!;‘;h§§!-a Gama” 6mg! a-me'y 66mm“ (3:5. am 5826:: 56:61:65: prfll .. BEERQWEQ r1173" 4 V" 55-“?! {56"},6: " 8633,51! ‘Z’ff’a 679.1%:3 (76:16:25,633 68.63:. a. " t: :K 16; 66:63?“ 663.213; a: :65. 'c:st~..r‘r€:;f‘:fi 113' 1.11.0 386‘35 673165.661)? Ms? ‘1‘th 616:6»: 56 s ”a 'v for 55:- - '1" — _‘ c. O 7"???» .9} .14 26:“ ;; Q 0 0 A J as (.27-«‘1 my: 5'6»: N i-‘ 4 U g: H O -u u- Han M :a—wrumt . F4 ' C: O baa-r-..» ..— —‘ _I 7"“ —--.-.e-- .nrvaur -; run" an“. cum-I- wind.“ ‘5 3:636 3.2.3- 764463J76'4'61352’QTffi'! '11:" psuzA-ns. L‘s‘ ‘ 1w 51.11 26 5:586:26 06633061661sz fwmww“ y FEW: 63:”.9111..d J: 3:; 18 £666.21: :6} 1:; 015mm”: 62. is 6263663183 p?! '-. Lama": 160 84:66:83 6. 65:. 726-: 38.12.1662; 0f 1.3.5.8 'Seremazzha ‘66 “6,606.: 5:. {3625113 5:11 :3 L186 6' ' 81:25.3 12:: 81:66 0:1; 3 ‘20 Thinks 1:: 5:. 1655.: 6:6’ 66:31:63.6. 3:3 6:3. thm d.).fti'ic15.31t;! To Tfiafii to 1*;"23-5536 a fi..:.hr"£ ‘63:: :61:- 6‘66:- 5:68 1‘66»): 64:66:63 801511.751 1616363 :4 it :3 “66.1218. 63:16:11 .22; 6:76:68 1’}; 661623.611 2‘0 1:165162 HH.): in maul/1:1 5:11.118 6: .K‘ :K (16‘5966LL£.“:¢ 11:0 5:566:66: 66:65: 36 663666.: 1:30.111 9. 71:12:62 5:66:66 35*..‘1'5‘313'8112’3- of 666'}; 56.31:.» I131. tau: 66:1":61'f‘61? 6m uY;*-"=!£-i'-L§?£el 6:}? 0.314;. 1}... ~13: 5. 5‘9: 11:: 063.831 :3 , 13?.6 Cfl’fifl‘fs ‘66:: “$515.65? 1213‘; ca :13; war ‘L’Ti F7643 “Cita- hi”‘~¢1-:M.flv -—-— Influ‘m— . .4 .1- A , . o .. _ , .. .. .: . -.- : an» . v 2». 2, :a...-‘H: ...a 4 .;.... . -' 5 V I - \ ,. ‘ ‘ I w.- 5. x \- u —. I: . t ( _l : M V 3 g . .. . 3 . . , . . . . V .-, I . , q -.-. 4 . < a y ‘5 _2 .. . . -.' . - ‘ :7. I :- . _. 2 : . 2 :d I. ’ ,5. . - . V, 3 ‘ ' . 2 . . I - 5 . ..' ; ,1 f ; ‘.~ ‘ 5 ~. '.~ '_.“ ’ II I) .1‘ '."~ '. . . 3. . . , .-i_, .~ 55 ..' 1 .' I ._ ‘ : «: .. i 'l | a "I “‘ 1 Y" ' ‘5 ’ "’2‘: ' (,IO. pv‘,'.."1" ‘A 1 "’ ’ r”""" ' " . : .2 52‘ ..‘5‘ . 2 3 . ‘ I .i : .‘ 5..4 ..‘l K. J ,i.’- ' ‘J :1~ ‘_.' ' .5‘ . ..‘5 ‘2"... .‘2 : ,v .. I . . ’ ' . ' 5 . :1 . f 21 5. \- v l e ' -£ -v 4. - . l W G "1" u ~ ' 2' o . . V 5.“! . a . .5‘33 1"“ ".‘,'2‘ -.-! ‘ f! ,‘ ,I_./ 5‘ ~‘7- an ~ V..’ I 2‘ I) + I) :‘At:. .».>‘.(. t 1‘ "- lw .. . a.) I'm" K’L‘ u 5m-‘ .L l. 5' '-. ‘,.;. ‘a I...- ) " 2 .I ..' v 4‘ 4*‘v I; 4’3- '2 2' n V r- .9 -. _r V -’ _ . 'e , g. . _ ‘ . . . . . 9“. Ug;\:.‘: . -; \9 . r: t..- .9 f) T f‘, 25.". ,II ‘,_ «‘ ' a". .. ,_ ’1 5‘~.. 5‘: 1. r .‘r 7» -p h.’ ¢ .;., ..;‘; h‘gJ L5?» 0.2 .4 t 5.; ,5: .46 ‘- 9 . u .. -- '. . I5 I“. . III In I! ~ L:- ~ -‘m-fll— yum :— l I 9 .‘ F .._. :‘ . . . . w: .3 -..." . -..22 .. ,. . 4.; . . 31......” i 5'. [5.1.5.2723 (“.51-3:1. 1.33) Cu: 4.53; - 225* ”LL? 5.: g 1 L312 34.6.4.2"; 2,.I I: 3.15.; ..2. -.';‘I£.;.'.—, 3 w .5 .-~ I.» :. .2 ‘3. - : :7 .2 1:: 5511553.. ..: 1 «5 :II. 5e: 132' I3. 5.55:2: 52.3-32.5. .2....V.-5-: 2-5: I. .- :5 52.2.2.3 1': I 3" O T” w. 2;I: 5'3 a ' 'I ._-: 21- ~53 '2 2: ' e C: , '3 -1. 'vk) L‘O \ ) L’:£: (it 0‘3. “.;-_'I. 3‘ ~l {:1 (5.5 .- 5‘ 5 ' . a) ..‘m 9;“. hr.” ‘I-I‘fi . -,,' \f- ‘ ‘ "L 5.. .._.. .3.,. . 1.... . ..:-VI ..I'I . .5: -..3 .2 :5: . 2. i 9 LA ’I: “i tiJi? '-§'.-.’2(. ““1353 : 1.135.516! V5.5: £4221 “.1 133.3. 3:14.; ..:‘ 'l . ,3 ‘2‘. 2‘ "_:.; ‘_,_ .1", f. '2 4‘; A __ 'v ; ,' g 1 «"LH; ”'5‘." - .‘_. 4 ,..,O:2 :5‘.:511I:.2.'.,; .3: 3:32.. ...: «3.53.25; 3.9 3.23-:- 32 75 i 13 mm 1 ~ 3252.. .2“ n .I-; ..V ' ‘ .. 1r“ :_ ;_V5 F.,: J. .5 _. 2.‘ ; .3- (2 a _. 2‘--.‘ .... ’,_ ”,1 ' 1 1.2.1.5: L533.) 3.15 ..:. 32.1221 “n 11'535-52321' 3.22.) :55... 33,252:- .22:.-: a... 1 25:5: 55.1-4.3 8‘ U .. l : . .. I. .' . .- .1 2 ‘ .-2 2. .. 2...“- ‘ ‘ ‘. 7352 C593 II-‘ij‘a: 3.23; ? ”’52:: ":75 3. 2.1-525731 792.37; [I 2 53.7.0 1155.31?-I‘3..r2"5."i.7é“22'2 2‘ 3'3 V2“ (an. 5"" filmy-KM L‘wstlmu “MJL- J—— I ”~“u-auumcwIWonU-‘ho- mix»: -.:-u1-r.¢—ui-u tau-..au'u-Mfi‘. 3 n . 5 e ’.~‘.‘-u'a‘__v-..- ‘M'w _ e "..:—..: -——- . . v-u—vsou -_._ ..:-s ' ':._ - tn :5 v u; 1"» -"._ ,:-‘I 2 e' v-I) ~' .;. . - 2 ;= s - 3 .a z . T C i t-n~ig 1H3.) Ul‘bt); ;- ‘ L'-.—s‘é0 (’34.; '54 (an 5 Q’i"~-1 f:&.l. :4 ‘us-Iog- ..AI‘J m N ‘- r' a I! '\ ‘3 u ’l W ‘ 5. .‘V’é. .“V ‘ 1. "i ' ‘l o £3533 at . 3:: 13:33.15: 5:3':;..I....:-~.~ ...:3 5.1.18 1sz .10; $5.335355.555.352.221; 335552.255 techs-55.7...zgzarf-53 ‘12:) {..: .15.: -I33 :3 g: .. 3.25.223 new" 51“. 1:5 ':’.‘.' 5155“”! y 15'}? ,I'..5‘V53 3:531:22 1531.5 3.03 I233 32.}: tzfi..'--.I..':I§. 201’". "3.5519332 511.135. 5V1". 1.! 2‘53? 31 1:532 E: : IQ. 3V1.--."3‘I2:-;:z to try to 3.3313533 C23: $3.315 53.1.]. of the: 3.5.5:: 3.1.3.1551; in 5:; "$22,313:..3eafia P793533: 55555555 1113:" when run: .1” 53.51;: g :13 35:52-15:53 33;:5:.33:L.L.3:~5~53_ ”3.2: 1:: 1333 3:534:20 32:33.25: U332"; act 5151:! 3352353.; .52.:2, 15- V: {:53 315.1: .23.: ..::ca 531‘ d. :VIT 3"”3'531’41: 81‘0: 15.53:: .;.}... 151.53.571.55 55.52“: 5525:5333)! to 55515;: W» 5 hunt ”5.535337251535233: 3.3.2. 55225525553132: 5 ’o D5352": 113-553.", 642“? up $5651.13“ 3*. in 155:1?12f315; {:73 555315-531; £52253: 0 5.. _L.’ O “ 3 a. C121 :5'. 25’. 3.3.9.37 35:35:..351 ' O (“I .. 4 O 3 3:) (Zn-2‘: 533m ‘3» 53.3 f-V m --.‘§an . —=‘ M “WW” 4;“- '7 A ”30 Kym“, L315 Luna UK) ..:-‘3 13%.:UJ" 3.253;!2; it: 35.1132. :55 “55333: 3:72;;113 when ‘23 the c5373}: .352. 1:52.11 r - irJO 1"? w...‘ ” I ,; - 73 : 1.2-.32: 1.2.: 53:52:? mm... IV (:1... i‘ 5.3:: 55 .11: £225,355; £5213;ng 522.5 .533 -. 342.51»:- 1531.11}, 5:: 355?}; he 13355355513 his 2.23.553 33:35.51; 3:552.“ :5" Pin 6.175;: .55 :11: 11:53:23 1:11:53 even-mu 3:55:13 535535,, 3 15.244133953353331; V3.5 6.525211%.LJ 2‘5-51 VI 3.231%“: L21? 5:52.535: n J. C251“ 333“}...1“, ' ‘Y {fit/‘33 ‘(Efl'i‘i‘ii 51’22‘VTTif‘-DW:I:...:-IE t') B. T“: ..: 5.13.7043 «h: i:- t w 523 5:) a; ' a... ”I . 3 u (fa if! C‘ a ‘N o In}, 5'32: 3;}.323 ‘13:: t": 1:335: 01’ $33: naval 5.4-4 3751'“ .53: mi 11".: "7:: P”VV.:.....¥3 v:3:.1'..=.211?.-le asf'r‘l 3:333:34 :~ 1:135. 1:: 1.53 3 (.;qn‘ - 2' v - . '5 5‘ I A,‘ ,.- '.2¢-.rv win- #8375 35::2‘2 flit tau-”.53 tukf'b Um? . {'11 54.7% all}. Jarfigf 55 . 5257221.. .2; V *‘\5'V""’ '5"! ' ..r. - .‘1 ‘. . T 22, ,3. . . . 3.1.: 3*“ :23": 353212-553 has 52.43.13... any 3:0 5.33555. ,2..2:V ._._ a5, 33: 2 75:33.5 .231, “In . . 3,5,- , ..., 5 . 213:.“ 9 9'3 5'31. 9"- 3 31.? 1.2.5.33 6372' 3.573%“: 03.9% "£0 54:1" Fr.- 5 .. LA- .. 14')? u j _‘ -"‘ " " 1.”..- u- ' m-~~ an '0 N .- aw ‘fl-mfl-V-Q .w: . —-..~-—.w- .. "luv“ .p.. . .Lw.x.~..t4- I 112 APPENDIX D Management Code Sheet Psychology'867 JMNHSsDbfifiaflwriCDDE:flhET Ednmany1966 Interview No. Coder Q. 1.1 Q. 2.11 1. 1. 1.1 ' 1.1 __ 1.11 __ 1.12 __ 1.111 1.13 __ 1.112 1-14 ...... 1.113 . 1.15 __ 1.110, __ 1 3 ...... 1.12 __ 1.31 __ 1.2 __ 1.32 __ 1.1+ 1-5 ...... 1.412 1-51 —..—— 1.L+13 1.6 ....— 1.#2 1.7 ____. 1.421 __ 1.8 __ 1.L+3 1.9 ._ 1.44 *1.10 1.115 __ 1.101 __ 1.5 *2. 1.51 1.52 3- 3 2 1. ‘ ' _-— 53 3.21 2. 3.22 __ 2.1 __ *3.3 .... 2.11 2012 *4. -_— 2.13 2.2 Q. 2.12 2921 1 ;2.3 1.2 2-4 _— *1.25 ‘3- 1.26 13.2 __ 1.3 3° —‘_— 2. 2.1 Q. 1.2 2.5 -——- 1. an: 2.7 i: -———— *2.8 ‘ ———- 2. 1.21 3 9 ‘-'—' 1.22 '-———- 1.23 _ *3.5 __ 1.2h *4. 1.203 4.1 1.204 n. *4.2 #14,.3 Interviov'flo. Q. 2013 0 WW NH P'P’ \nxn bah: ADFJ 1.4 r??“ £555 IHI N K1») -2- Revised Management Code Sheet February 1966 Q. 3.12 ‘I’ UWWUUUWUUWUUUUUUWUW I C O Coder . o o OOOCDMtU O. HHH P‘PJP‘P‘P'P‘P‘ O O 0 00000000000 OOOOOOOOOHWOQmM-tWNI-d WGQ¢M¢UNHO *6.1 113 -3- Revised Management Code Sheet February 1966 Interview No. __~_ Coder Q. 3.21 Q. 3.23 1. 1.1 1' 1‘1— 1.ll ' 'i—i-B- 1.111 1 15 1.1112 ———— 1.111u 1'2 ““21 1.112 i'zu-———— 1.113 ' -———— 1 12 --- 1.26 _____ 1.121 1 261 1.122 1.1:; -- 1 262 1.126 1.263 ____ 1013 "*" 1.26h-_____ 1 131 1 3-—-—— 1.2 1'31 1.3'-—- 1.32 _‘_‘ 1-33 . 2 *1.3n .3, *2 __ 3. 4.4.1 3-1 ———— 4.2 3°2-———— 4.3 *3.3 ____ *4.4 ____ 4. *4.3 Q 3.22 *4.“ 1. 1 1 __ Q- 3 3 1 101 1.102 1 1.2 ____ 1.1 1.21 _____ 1,2 1.23 1.3 u... 1.4 *2 1.5 1.6 3~ 1.7 3-1 ____ 1.8 3-11 ____ 1.9 3.13.____ *1.01 *3.3 ____ 2. 4 *2.1 4.3.____ *2.2 *4.4 ____ *2,3 *4.5 *2.4 -4— Revised Management Code Sheet February 1966 Interview No. ..-mm-- H __.7_“_”__ Coder Q- 3-3 (Cont'd.) Q. 5.1 (cont'd) *2.5 1.15 *2-6 .... 1.151 *2-7 ...... 1.152 *2-8 ...... 1.16 *2-9 _— 1.161 *2-01 _— 1.162 1.18 3- 1.181 *3-1 1.182 *3.2 _____ *1.19 *3.3 ____, 2. 4. 2.1 4.1 __ *2.11 4.2_____ *2.12 4.3 __ 2.2 4.4 __ 2.21 4.5 __ 2.211 4.6 ...... 2.212 14-7 ____ 2.213 4.8_____ 2.214 4.9 ...... *2.215 __ ai‘le.01 __ 2.22 2.221 5 2.222 *6.5 2.224 *6.6 ' $2.23 7. 3. *7.1 __ 3.1 __~ *7-3 ____, 3,2 ____ *7.4 ____. $3.3 _____ *7-5 ____. *7.6 ____ 4. *7.7 ____ *4.1 *4.2 Q. 5.1 *4.3 1. 1.11 Q. 5.21 1.111 1.112 1. 1.12 1.1 1.121 1.2 1.122 *1.3 1.13 1.131 2 1-132 m____ 2.1 _____ 1.14 2 1 1.141 2. 1.142 114 -5- Revised Management Code Sheet February 1966 Interview No. Coder Q. 5.21 (cont'd). 2.14 2.15 2.16 2.18 [..: O H H O\OCD'\10\{:'NH H ** NNNNNNNN 1H *3. * u) @- Note: Any number so marked * is not to be used in agreement analysis. 2-23-66 rhb APPENDIX Eiwrfl Project Code Sheet 115 PSYCHfiLOGY 867 Revised Project Code Sheet March 1966 Interview No. Coder Q. 1'1 Q0 2.0 (cont'd) 1°11 .... 1.1133 ~12 ...... 1.11331 i-13 ...... 1.11332 ""— 1-1" ._ 1.11333 1.1% ...... 1.11331. 1%? .... 1.11335 ° _ 1.11337 "‘— *1°18 ...... *1.1135 *1.12 Q. 1.2 ---- 1.1 Q. 3.1 (Leave out) 1.11 1.12 """' Q. 3.1. §_I_1_c_l_ 3.5 11a": ' -——-— 1 11 1.15 ' ...... *1.17 w “'13 —— 1.2 1.18 1 21 “‘19 —— 1‘22 "'_"' H.101 __ 1'2 _— 1.102 __ 1°23 _— *1.25 Q. 2.0 1%.?!) ._— 1.31 1‘i 11 1.32 ' 1.33 1.111 1 3h _— 1.1111 *1'35 ...—_— 1.1112 3136 _- 1.1113 - _— *1.1115 "'— 1.112 "— ' ‘1.11211 Q' h'1 1.11212 ' "'" 1-1 1.11213 __—"' 1.11 __ 1.1122 1-12 1.1123 _____"" 1.13 __ 1.11211 *1.17 __ 1.1125 — 1.18 __ 1.1126 *1.1127 1.113 1.1132 1.11321 1.11323 116 -2- Revised Project Code Sheet March 1966 Interview No. __ Coder 0. 1.2 Q. 5.2 1.1 1.11 _____ 1.1 1.12 _____ 1.11 1.13 _____, 1,12 "' " 1.11. 1.13 1.15 1.1h 1-16 1.15 1.19 1.16 —'— *1.102 1.17 ""‘ 1.103 *1,101" 1.2 “"' *1.23 Q. 6.1 Q. ’403 1.1 1.11 1.1 1.12 1.11 1.13 —" 1.111 1.16 —""' 1.112 *1.19 "‘_' 1.113 ""—— 1.11h 1.115 Q, 6.2 *1.116 1.12 1.1 1.121 1.11 *1.125 1.12 *1.126 1.13 "'”' 1.1h 1.153 Q. h.h 1.1h5 1.1h9 1.1 "*1,17 1.11 1.111 1.112 _____. Q. 7.1 *1.113 *1.114 1.1 1.12 1.11 1.121 1,111 *1.122 1.112 ' ' ' *1.123 1.113 ""’ 1.1132 1.11321 _____ Q. 5.1 1.11322 1.12 1-1 1.121 1.11 *1.123 1.12 1.13 1.19 *1.20 117 -3- Revised Project Code Sheet March 1966 Interview No. Coder Q. 7.2 Q. 8.2 (cont'd) 1.1 1.116 1.11 31.117 1.111 1.12 1.112 ”' 1.121 1.1121' 1.122 1.1122 "‘“' 1.123 1.113 1.121 1.1lh 1.125 1.115 ' 1.126 ""' 1.116 *1.127 1.118""" *1.119 *1.1101 "” 1.12 1.121 1.122 """ 1.12u 1.125 1.127 1.128 ““" 1.129 1.1201“"’ *1.1202 *1.1203 .— g _ 1.1 1.11 1.111 1.112 1.113 1.12 *1.121 - I Q. 8.2 1.1 1.11 1.111 1.1111' 1.1112 """ 1.1113 1.112 1.113 1.11h 1.115 1.1131‘ 1.1152"‘”"' .meome Hun—Hoot 03... cu wovgou down 96: u we mmSHm>. ... mN. HN. we. mm. No. NH. HN. NNVHN. «n. ««. mH. mo. NH. mm. N«. mo. am. NN....mm. mN. Nm. w«. cm. 118 Nm. um o«. mH. Nm. H«. o«. No“- mm. Hm. m«. Hm. m«. oo. o«. cm. H«. em. 00. m«. co. oH. N«. no? mN. Om. mN. mm. w«. «HP On. mm. N«. mo. mN. Hor o«. m«. mm. Nm. CH. mm. HN. MN. mm. Hm. mN. mo. HN. om. 0H? NH. m«. on. nN. mo». OHP m«. mm. oH. o«. «mu. ««. mm. m«. mm. mH. mN. a0. NH. OH. mm. m«. mm. mo. 00. «H. H«. Nm. Hm. mo. woP mm. mm. mm. no. on. Non-0o. NN. mo. om. mormorNN. Hm. NN. «Hp-om. No. N«. «N. moPmHn... no. mo. NN. Hor. mH. oH. mm. NH. no. Hor Nm. om. mH. o«. oHPN«. Ho. NN. 0N. HH. NHVON. cm. H«. Hormm. er N«. o«. 0N. m«. mor««. No. MH. Hm. wH. OH. moPONVoo. mH. mH. mNP NN.! mH. NmP mH. HH....HHP HNrONP «Hr mN. 0N. mH. mo. 0H. mN. mm. Hm. MOVOHP oN. Nm. mN. «H. «o. mN. NH. NN. NH. Nm. «m. Hm. HN. oH. mm. NN. 0H. Nm. co. wN. «oPNNP m«. we. mH. mm. mNP m«. NN. Nm. o«. «m. «o. No. 00. HN. ON. NNP«o. Ho. MN. NH. mo. mH. nN. mH. oHPmm. NH. mm. on. mHPNmP ON. mH. Ho. NH..I wN. NHPHN. No. HH. NN.! mHP HHP N«. NH. HN. «or no. «N. mN. m«. 0N. wN. Hm. on... m«. wH. on. o«. HN. Nor mH. om. Nm. No. mN. HN. NNP N«. OHP wH. N«. NH. mo. OHP Nnr NH..I HH.. Nm. «N. Nm. on. NN. mo. mH. NN. 0N. 0H. m 3.: 3m mmmflmwmflmfl NS 8 S 3 2 2 fl mmmmszzm .3oHPHmuGH uaoammmnmz wmumudeHfié 939595 «0 muHmm nomauom mGOHumHmuHoo mmu mo NHuumz NN.. Non-mo. wN. HN. em. mH. mN. «m. mo. «NV 00. NH. HN. «m. mN. m«. on. HN. wH. 0N. 0H. no. co. co. om. m«. oN. mm. N«. Nm. NH. NN. NH. cor no. moPON. mH. NN. o«. c«. MN. o«. N«. «m. «o. «O. «H. nHP mo. NN. mm. mm. NN. «HP Ho... Ho..- Nm. Hm. mm. «H. mN. no. «G. HH. a0. mNP 0H". «H. NoPHH. NH. 8. 0H. 0N. «N. «o. HNM .m NHvdmmq< an. no. mm we. 8...? we. oo. me oh. no. we mm. as. me «oraoree mm. 3... S 3. Sr am 8r 2. mm mm. mm. Nm as. ea. on an. no. mm 3. 8... mm 8.: SF mm 8r 2. on 3. 3» mm me. no. em NN. ON. mm mm. mmvau Bree... om mm. 3r 3 NN. Hm. ea an. «N. ma mm. «2.2 2. mo“. 2 ma. NH. ca 2» S. N 3. 8. ... 8. e I I H e H 811331119113 OO .moUMHO HmaHomO can on vacancy coon m>m£ H mo ON. ON. OO. «H. HO. OO. NN. NN. «N. ON. «N. OH. OH. OH. NH. OH. ON. NH. NO. OO. OO. OH. ON. OO. «N. ON. ON. O«. ON. ON. NH. ON. OH. 119 O« N« O« NH O« k . HN. ON. ON. ON. ON. OO. ON. NH. OH. «H. OO. HN. NH. OH. NH. OO. OO. NH. OH. ON. ON. NH. OO. ON. HN. ON. «N. ON. ON. NN. NN. NN. NO. OH. ON. ON. HN. OO. NO. ON. HH. ON. ON. NN. «« O« OO OO N O O N NN. «N. ON. OO. OO. OH. OH. ON. OO. HN. «N. HO. NH. N ON. NN. OH. ON. OO. «O. ON. HN. OH. «O. OO. OH. NH. HO”: OH. OO. OH. OH. ON. ON. ON. OH. OO. NO. OH. NN. OH. ON. OH. OH. OH. ON. ON. «H. ON. «O. HN. OO. ON. HN. HO. HN. OH. NO. . NO. .OINNOIHOM Ommquozm OH. OO. OH. OH. NN. ON. NH. ON. HN. OH. NO.I OH. ON. OO. OO. OH. HH. OH. NH. ON. OH. «O. OH. N ON. NN. ON. «O. NO. ON. NN. NN. NO. OO. NN. OO. N moSHm> « OH. HH. OO. OH. OH. OH. ON. OO. OH. mo 0. «H. ON. OH. OH. NH. OH. OH. OH. NN. N HN. OO. OH. HN. OH. ON. ON. ON. HH. ON. OH. NH. OO. ON. ON. OH. OH. ON. OH. «H. OH. OO. OH. «N. OO. «H. OH. OO. OH. ON. HOr.ON. ON. OO. «N. OO. NN. ON. ON. ON. ««. NH. NN. ON. OH. «O. ON. OH. OH. ON. OO. OO. OH. OH. OH. O«. OO. ON. ON. OH. HO. OH. OH. OH. OH. NH. OH. OH. OO. OH. OH. ON. ON. HO. HN. ON. OH. «O. HN. ON. OH. «O. ON. OH. «O. ON. OH. NN. OO. OH. ON. «N. NH. ON. NH. ON. ON. NO. OH. OO. «N. OO. HN. OH. HN. OH. OH. H«. OH. OH. OO. NH. HH. OO. HN. «H. OH. OO. OO. OH. OH. OO. OH. NH. OH. NH. ON. «H. NN. OO. ON. OO. OO. HH. OH. NH. OH. HN. OH. OH. OH. OO. HN. OH. ON. OH. NN. NN. ON. OH. OH. ON. ON. NH. ON. OH. OO. ON. OO. OH. OH. OH OH NH HH OH N .mummGHOam mo muHmm newsman maOHumHouuoo 3mH>nmuaH uaoamwmnmz mo xHhumz .O xHanma< OO. OO. OO. OO. OO. OO. OO. NO. ON. OO. OO. OO. HO. HH. NO. OO. OO. OO. HO. OO. OH. OO. OO. OO. HO. OO. OH. OO. OO. HO. OO. «O. OO. HO. OO. OO. OO. OH. OH. OO. OO. OO. ON. ON. HO. OO. OO. HO. OH. NO. OH. OH. OO. HO. OO. OO. O«. OH. OH. OH. OO. HO. OO. OO. OO. OH. OO. OO. OH. NO. OH. OO. ON. ON. OH. NO. OH. OO. ON. OO. OH. HO. OH. NO. w” «1 PH SHHHNIDNE .mmomHm HmaHoov 9.5 9. Oougon down 9.6: H mo moHHHm> Om OH. OO. H«. ONP OO. OO. NO. OO. HH. OO. ON. HO. «O. OH.I NN. OO. OO. «N. NO. HO. OO. OO. O«. OO. OO. OO. ON. OOn.OO.....«O OO. «O. OO. O«. OO. HN. O«. OO. ««. O«. «O. «O. NH. O«. OO. OO. O«. OO. OH. H«. «O. H«. OO. «N. OO. «O. ON. NH.I NO «O. ON. OO. NO. OO. NN. O«. ON. OO. «O. OO. HO. OO. NO. H«. OO. NH. NO. HO. OO. OO. ON. O«. NO. NH. NO. OO. HO NO. OO. OO. OH.I NO. NO. «O. ON. OO. O«. OH.I OH. NN. OO. O«. NH. OO. OO. OO. NN. HO. OH. ON. ON. ON. HH. OO ON. OO. ON. OH. «Or OO. ON. O«. HO.I OO. OO.! ON. OO. OO. NO. NH.| HO. OH. ON. OH... «N. «H. OO. ««. OH. O« ON. ON. OO. OO. OO. ««. ON. N«. N«. «O. O«. OO. N«. O«. N«. OO. «O. ««. ON. «O. O«. NN. ON. OO. N« OOPHO. NH. O«. ON. ON. OO. NO. «N. N«. OO. NO. OO. «N. HO. OO. NO. OO. NO. ON. NN. OH. HO. O« «H. OO. OO. O«. ON. OO... O«. ON. OO. OO. NOPOO. OOO.. OO. OH. OO. HO.I OO. «OPOO. ON. OOr «« OO. O«. OO. OO. HO. OO. ON. NO. O«. «O. H«. O«. «O. HO. OO. OO. H«. OO. HO. «H. «NPO« ON. NO. «O. ON. NN. HN. OO. N«. ON..ON. OH. HN. «N. NN. «N. NO. OO. ON. ««. ON. NO OO. HH. ON. NO. ON. «O. O«. OO. NO. OO. OO. NO. O«. OO. OO. O«. «H. OH. OH. OO OO. ON. ON. O«. OH. O«. N«. ««. NO. «O. O«. OH. NH. NO. «O. OO. NN. NNPOO ON. OO.! ON. NN. OO. N«. OH. NO. OO. O«. OO. ON. ON. OO. O«. OO. NH.I «O OOP ON. OO. H«. OO. ON. H«. O«. OO. «O. NO. ON. NO. NO. OO. OO. O OH. NO. OO. NO. N«. OO. ON. ON. NO. NOV OO. HO. OO. OO. «H. H OH. N«. ON. NO. OO.! ON. O«. NH. NO. ON. NN. NN. NN. OHPO H«. O«. «O. «N. ««. N«. O«. HO. ««. NO. OO. ON. «H. O N«. O«. ON. N«. OO. N«. «O. OO. NO. OO. NO. HOrN O«. OO. OO. «N. N«. OO. OO. OO. OO. HN. OOP ON NO. «O. N«. OH. H«. ««. OO. H«. OH. NNP OH OO. ON. OO. OO. ON. OO. NH. OHP OH.I NH «N. OO. O«. N«. NN. ON. OH. OO. OH O«. O«. OO. HN. N«. NN. OHP «H O«. OO. OO. «N. ON. OO. O NN. HO. NH. OHP NH.|O O«. OO. ON. OOrO «O. ON. OO. « HO. NNP O OH. N H SHHHNI‘DNH 120 OOHOHOOOSSOOOOOOOOOOOOO OHOONMMONOHHHOHOHNONOMMH OOOOzHOzO .BoHPHounH nomnoum Omuwumafiafi... muumaHOcm mo muHmm amo3umm OGOHuMHmHHOO Omo mo Knuwz .m xHOGoOHF. .O NO. NN. NO. NH. NH. OH. HH. ON. OH. NN. OO. OO. OH. HO. OO. OH. «O. NH. OO. OO. NN. NH. .meMHO HmEHooO 25 cu @0930“ down 96: .H «o mosHm> .5 OO. OH. NN. NH. NO. NN. OO. OO. OO. OH. OO.IHO. HH. NH. NO. OO. NO. HO. HH. OO. «N. ON. OO. HOP NN. OO. NO. NH. «O. NH. OO. ON. NH. OO. HN. OO. NH. OO. OH. HO... NH. HN. NO“. ON. NO. OH. HN. Oor OO. ON. ON. ON. OH. «O. OO. HO. OO. NN. O«. OO. NH. OH. NO.I OO. OH. NH. NN. «H. OOP NOPOO. OH. NO. OH. OO.IOO. OH. OO. «H. «OP«O... OH. «H. OO. HOPOH. NH. OH. HH. OH. ON. OO. OH. OH. OO. ON. OH. OH. ON. OH. ON. OO. OH. OO. ON. NN. «O. HH. NH. HOP OO. «H. OO. HN. HO. OOr «H. NO. HN. OO. H«. OO. ON. OO. OH. OO. ON. ON. HN. NH. «H. ON. 121 ON. OO. ON. OO. OH. «H. HO. OO. Nov OO. OO. ON. OH. «O. OH. OO. «O. OH. OH. ON. HH. NH. ON. OO. «H. HO. NH. OO. OO. OO. NN. HOVNO. OH. HN. NN. HH. NH. OH. N«. NOPOH. NN. NN. OO. OO. ON. HH. OH. ON. OO. ON. ON. NH. HN. OH. OO. NO. HN. OH. ««. NO. ON. OH. «N. «H. «H. OO. OO.I «H. HN. OO. OH. «O. ««. NN. OO. OOPOH. NN. NN. OH. HOPOH. OO. OO. OO. ON. OH. NH. «H. «N. OH. ON. «O. OO. NH. ON. OH. NN. ON. OOPON. OO. NO. HO. OO. H«. OO. OH. HN. ON. OO. «O. NH. OH. OO..I HO. NH. ON. NH. «O. NO..I OO. ON. «O. «H. ON. OO. ON. ON. OO. OH. NO. NO. NO. OH.I«O. «N. OO. «O. NO. OO. HO. ON. «O. NO. OH. ON. NO. OO. OO. ON. NO. OH. ON. NN. OO. HH.". OH . OO . NOP OH. ON. HO. OH. «O NO HO OO O« N« O« «« O« NO OO OO «O OO HO OO ON NN ON OH NH OH «HO O O “O mmmmzHUzm .mumoGHOam no can awm3uom.,mn0HumHmuuoo 30H>uoucH uoomoum mo xHuumz .H xHOnomm< OH. OO. ON. OO. «N. HH. OO. OH. OO. «O. OO. OH. NH. OO. NH. OH. «N. ON. OO. HOP OO. OO. HH. «O. OH. OO. OH. OO. HH. OO. OH. NOuIOH. OO. ON. «H. OH. HH. OH. NN. OO....OO. ON. OH. «O. HN. HN. OH. NO. NN. OH. OO. «H. OO. ON. OO. OH. OO. OO. OH. I « OO. um. ON. NO. OH. OO. «H. OH. OH. ON. HH. OH. OH. OH. O«. NH. HO“. ON. ON. O«. «H. OOP OH. NOP NO. NO. OH. M HH. «O OH. NO OO. HO OH. OO NOPO« OOPN« ON. O« ON. «« HH. O« NO». NO OH. OO ON. OO HO. «O NO. OO NH. HO OO. OO OO. ON HN. NN NOP. ON NO. OH HOP NH NO... OH «H. «H NN. O OO. O OH. O OH. « OO. O OO. N I H H SHHHNIDNEI Contingency Tables for Regrouped Data 122 APPENDIX J (a) Contingency Table. CPS vs. MI CPS Data -.65/.24 .25/.84 z .123 78 201 —.0u/.19 (115.52) (85.u8) (201.00) Cd 4.) O 127 107 234 o -20/-55 (13u.48) (99.52) (234.00) H m 2 250 185 M35 (250.00) (185.00) (“35.00) (b) Contingency Table. CPS vs. PI CPS Data -.u1/.30 .31/.78 2 10a 110 21H m -.1u/.13 (99.87) (114.13) (21u.00) 4..) m 99 122 221 S 'l“/°55 (103.13) (117.87) (221.00) E: z 203 232 835 (203.00) (232.00) (H35.00) APPENDIX K Matrix of Management Communication Estimates-Rater 1 Engineers 6 7 10 ll 12 l3 l6 19 2O 21 22 24 25 26 29 32 33 36 37 38 39 4O 44 45 46 47 48 53 u OOH r—lr—lr—IO .OJOr4r4O r-{J‘KOb-Or-l r-it—i 123 OOH OHH r—{Or-i O 2 O O O 0 O O 0 0 O 0 0 O O 0 0 O 1 1 1 1 1 O O O O l 1 O O l l l 0 O O 0 O O O O 0 O O O O O O 0 1 1 l 2 O l l 2 2 2 l l 0 1 O 1 1 1 O O l l l l l O 1 O 2 53 APPENDIX L Matrix of Management Communication Estimates-Rater 2 Engineers 6 7 10 11 12 l3 16 19 2O 21 22 24 25 26 29 32 33 36 37 38 39 4O 44 45 46 47 48 53 4 OOHO r-iOr-INO rfiJWOt~CDH Hr4 OOO HNO 000 H O 124 21 OOOOOOOO'OOOOOOOOO HOOOOOOOOOHOOOOOO OOOOOOOOOOOr—IOOOOO OHONHHOOOOOOOOOOO szxooxmm mmmmmmm saeeutfiug KONCIDQOZ'LOKONGD 0000000033333: 53 APPENDIX M Matrix of Project Communication Estimates-—Rater 1 Engineers 5 8 9 14 15 17 18 23 27 28 30 31 33 34 35 36 37 43 44 46 47 49 50 51 52 54 4 3 r—lr—INO r-ir—iNOO Hmmzmoo 125 HONr—IO Ot-it-IOOOt-l I—IOOOOOO OOOONONO ONOI—lr—IOOO OHOa—lr-IOI—IOt—l ONOOHOOO r-th-INOt-lr-{O oc3c>Oc3c>Hr4 0) OH OO HO O O JeeuISug CO CO CO OO OO H0 00 CO CO OOO OOO ONO 000 000 OCH OCH OCH OCH OOr-i OOO Ot-IO OOO OOO H00] 000 O0 r-lO OO HO H0 00 00 H0 00 NO OOO NOt-l HQ 00 NO r—IO NO NOON OOOr-l Or-IOO NOON ONOO OOOr—l OI—IOO 1 O 0 1 0 0 O O O 0 O O Rater 1 did not know them well Two engineers have been omitted from this analysis. Note enough to provide estimates. APPENDIX N Matrix of Project Communication Estimates——Rater 2 Engineers 5 8 9 14 15 17 18 23 27 28 3O 31 33 34 35 36 37 43 44 46 47 49 SO 51 52 54 4 3 O OO OOr—I OOOO HHOOt—l HNHOOr-l Ot-lr—lt—iOr—{r—i Hmmzmoooxz H O O O O O H O O OO 00 OO OOO OOO 00 126 ON 00 OOO 00 001—! OOO OOO OOO OI-IO OOO I—IOO O-r-{O OOO OOI-l OOO OOO NOO OOO NOO OOO OON OOO OOO OOO OOO OOO OOO OOH OOO r-lr-lO Or-ir-l CO CO 00 CO OH 00 OO O OO OOH OOO OO ON OOO OOO OOO OOO OOO OOO OOr-i OOO OOO CDO OCDri (DCDO 000:: riOCU OCDCS OCDF! oc3c> oc:c> OCDri or4O (DCDO O4Or4 OCDC> OCDC) oc3c> .HCDO OCDO oc3c> oc3c> or4c> OCDF' OCDC) OCDC) HCDO <3c>o Or-IO Rater 1 did not know them well Two engineers have been omitted from this analysis. Note enough to provide estimates. APPENDIX 0 Matrix of Combined Management Communication Estimates Engineers 6 7 10 11 12 13 16 19 2O 21 22 24 25 26 29 32 33 36 37 38 39 4O 44 45 46 47 48 53 4 HHNO m O OOO HHO HHO mzo O O H 127 OHIO OHOO OOOO OOH NOH HHOO HHH-:1” OHO HOH MJNO OHOO HH:1’N OOOO OHHO OOMH OOON OOOO HHOO HHOO OOOO NMHO OOO: OHOO HH-S'O MHHH OOH-11' HHS O 00 OH 00 0000 0000 MOMO NOOH OO-er MONMO HOO HNOONHN N:U\\OO\N NNNNNM saeeuISua H«Z'KONOH M\O[\ Ch UNKDNQ) HH mmm m :f-ZI'ZJ'Z' 53 00 O: m :1': APPENDIX P Matrix of Combined Project Communication Estimates Engineers 5 8 9 14 15 17 18 23 27 28 3O 31 33 34 35 36 37 43 44 36 47 49 50 51 52 54 4 H ON HOO OOOO OOMON OOOr-‘lOt—l O: HO OO NO NO HQ 128 OMOu—‘l HOO u-h-l NOMOr-i OOO r-'|OO OOr—l OOO OOO HOH ON Or-l OO MO 00 O: OOO MOO OOO OOO (”OOO OI—IO Ot—IO OOO OOO MOO OMO OOOO OOr-l O-tl‘O OCH OCH OOO ONO OOO Or-iO MOO OOO NOO-=rOO MOO MOr—l r-IOO MOO I—IOMNOO OOH ONO ONO ONO OOO Ot-IO OON ONt—I NMOOMOONOOOOOOOHOOOOMOHHOOr—l NMMOI—IMH-fl'I—lv-lr-lOOr-{zrr—{NOOHNZ'OOHNr-IN r-INMt—iOr-it-{Ov—lr-IONNNOOOOONONNONr—lr-INO saaautfiua Contingency Table. 129 APPENDIX Q (a) CPS (MGMT) Data vs. Rater 1 Rater 1 Estimates 2 1 0 z 1 0 1 2 “'65/'°51 (.1655) (.6437). (1.1908) (2.0000) 0 2 0 2 “~50/‘°36 (.1655) (.6437) (1.1908) (2-0000> 2 10 7 19 “'35/"21 (1.5724) (6.1149) (11.3126) (18.9999) 2 15 24 41 3 "20/"06 (3.3931) (13.1954) (24.4115) (41.0000) 9 8 20 50 78 f:-'OB/'09 (6.4552) (25.1034) (46.4414) (78.0000) ° 6 38 64 108 g '1O/°2” (8.9379) (34.7586) (64.3034) (107.9999) 5:. m '25/°39 (7.6966) (29.9310) (55.3724) (93.0000) 9.: o 9 14 45 68 .40/.54 (5.6276) (21.8851) (40.4874) (68.0001) 5 5 11 21 '55/'69 (1.7379) (6.7586) (12.5034) (20.9999) 1 2 0 3 .70/.84 (.2483) (.9655) (1.7862) (3.0000) 2 36 140 259 435 (36.0000) (139.9999) (258.9999) (434.9999) Note: Numbers in parentheses denote expected frequencies. 130 APPENDIX Q (b) Contingency Table. CPS (MGMT) Data Vs. Rater 2. Rater 2 Estimates 2 1 0 2 -.65/-.51 (.lé57) (.2313) (1.6230) (2.0800) -.50/-.36 (.1857) (.2713) (1.6230) (2.0800) -'35/--21 (1.0846) (2.5870) (15.4884) (19.8300) g '-20/‘-05 (2.1678) (5.5809) (33.3713) (41.8600) : "°05/09 (4.1341) (10.5393) (63.3366) (78.86800) :5: °1O/°2” (5.71103) (14.6483) (87.2814) (10888800) g -2“/ 39 (4.9?72) (12.6338) (75.4690) (93.3300) '“O/'5“ (3.5354) (9.2330) (55.8816) (68.8800) '55/'69 (1.1%03 (2.8883) (17.8414) (21.8800) ”0‘8“ (.1886) (.4869) (2.4845) (3.0800) 2 (22.3397) (59.8801) (35388802) (43588800) Note: Numbers in parentheses denote expected frequencies. 131 .mmHocmsvopm oopomoxm ouocoo mmmonuconmo CH mamnesz "opoz Azooo.mmzv Ammmm.mzmv Hoooo.mHHV HHooo.mmv Amoco.mHV Amoco.wHV H mm: Ham MHH mm mH mH Hmmmm.mv Aamos.Hv HmmHH.V HommeV HHHNH.V HHHmH.V . H. m o H H o H :w \o AHooo.HmV “Hzmm.HHv Ammm:.mv Ammmm.Hv Homom.v Hommm.v mo.\mm. Hm HH 3 m H m AHooo.mmV AmHHm.wmv szoo.sHv Ammmo.mv Amem.mv AwMHm.mv :m.\o:. 0 mm m: 0H m m a & Hoooo.mmv Ammom.mmv AmmmH.:mv Amsmm.mv Hmwzw.mv Hmmzm.mv mm.\mm. 3 mm mm :m w m m \) HHooo.mOHv AHzmm.Hmv Hmmmo.mmv memm.mv Homm:.:v Homoz.zv :m.\OH. Wm wOH om mm m m m mm “Hooo.msv Hammm.zav HHmmm.omv HHmmm.mv “mamm.mv Hoamm.mv mo.\mo.u . ms om 0H m H m (. Amoco.H:v Amowm.mmv “momm.OHV Ammsm.mv Hommm.Hv Ammmo.HV mo.u\om... m H: Hm :H a o m m Ammmm.wHV Amwms.OHv Ammmm.qv Hamms.Hv Amoms.v Ammws.v Hm.u\mm.u H m H Hoooo.mv AmmmH.Hv AmmHm.V AmmsH.v Hmmm0.v Hmmwo.v mm.:\om.- m o H H o o Hoooo.mv HomMH.HV AmmHm.v HmmsH.v Hmmmo.v Hammo.v Hm.u\mm.u m H o o o H H o H m m .H mopmsHpmm ooCHnEoo mmmeHpmm ooCHnEoo .m> damn ABzqu mmo on @ xHozmmmH .oaome mocmwcfipcoo APPENDIX E (a) 132 Contingency Table. Management Interview Data vs. Rater 1 Rater 1 Estimates 2 1 0 z 0 2 5 7 "°”/°Ol (.5793) (2.2529) (4.1678) (7.0000) 3 9 10 ‘ ‘ 22 ‘ '02/°07 (1.8207) (7.0805) (13.0989) (22.0001) m 2 19 4O 61 g '08/'13 (5.0483) (19.6322) (36.3195) (61.0000) 9 14/ 19 4 33 74 111 3 ° ° (9.1862) (35.7241) (66.0897) (111.0000) H > 11 31 58 100 E ‘20/°25 (8.2759) (32.1839) (59.5402) (100.0000) 2 8 34 43 85 j '26/'31 (7.0345) (27.3563) (50.6092) (85.0000) c 6 11 20 37 g '32/'37 (3.0621) (11.9080) (22.0299) (37.0000) 3 38/ 43 0 0 9 9 g ° ' (.7448) (2.8966) (5.3586) (9.0000) 2 1 1 0 2 °““/°“9 (.1655) (.6437) (1.1908) (2.0000) 1 0 0 1 '50/'55 (.0828) (.3218) (.5954) (1.0000) 2 36 140 259 435 (36.0001) (140.0000) (259.0000) (435.0001) Note: Numbers in parentheses denote expected frequencies. APPENDIX R (b) 133 Contingency Table. Management Interview Data vs. Rater 2 Rater 2 Estimates 2 1 0 z 0 3 4 7 "'O“/°Ol (.3701) (.9494) (5.6805) (7.0000) 4 5 13 22 “OZ/'07 (1.1632) (2.9839) (17.8529) (22.0000) 2 8 51 61 g '08/'13 (3.2253) (8.2736) (49.5011) (61.0000) “ 1 15 95 111 : .14/.19 (5.8690) (15.0552) (90.0759) (111.0001) 3 20/ 25 8 13 79 100 Z ° ' (5.2874) (13.5632) (81.1494) (100.0000) (1) 4s 2 ll 72 85 5 '26/°31 (4.4943) (11.5287) (68.9770) (85.0000) 4.) 4 4 29 37 g '32/'37 (1.9563) (5.0184) (30.0253) (37.0000) 0 O O 9 9 §’ '38/'”3 (.4759) (1.2207) (7.3034) (9.0000) 9 1 0 1 2 E .44/.49 (.1057) (.2713) (1.6230) (2.0000) 1 0 0 1 '50/°55 (.0529) (.1356) (.8115) (1.0000) 2 23 59 353 435 (23.0001) (59.0000) (353.0001) (435.0001) Note: Numbers in parentheses denote expected frequencies. 134 .moaocosompm ompoooxm mpocmv mmmonpcmnmo CH whonEsz "opoz Hoooo.mmv Ammmm.:m:v Hoooo.Hamv Hoooo.mHHv Ammmm.HHv Hmmmm.sHv H mm: HHH MHH mm HH HH HHooo.HV Awpom.v Awmmm.v Apmwo.v Aszo.v AHHHQ.V mm.\om. H o o o o H N Hoooo.mv HommH.Hv HmmHm.V AmmHH.V Ammmo.v AHHHO.V m:.\::. 2 m o H o o H w Ammmm.mv AHOHH.mV Hmhmm.mv Ammom.v HHHHm.v HzmHm.v m .\ m. s m m o o o o n H m Hmmmm.mmv Ammoo.Hmv AmHHm.mv AmHHm.mv AOHmm.HV HOHmm.Hv Hm.\mm. w Hm ow s m m m a Hoooo.mmv Azzom.m:v Amomo.mmv AHomm.Hv AHHHm.mV AHHHm.mv Hm.\mm. .1 mm 2 om m m N w Ammmm.mmv AmHmb.wmv Hosam.mmv Hmmmm.mv HmHmH.:V AmHmH.:v mm.\om. m OOH Hm mm m m m A Hoooo.HHHv Hmsmo.mmv Hmzmm.mmv AHHmm.mv AHmmm.:V AHmmm.:v mH.\HH. m. HHH HH mm MH H H M Hoooo.Hmv “Homm.:mv Hoozm.mHv Homo:.mv AHHHm.mV AHamm.mv mH.\mo. 0 Ho mm Hm m o . m . m Ammmm.Hmv Homm:.mHv AmzHH.mv AHHHH.HV AMOHm.V AmOHm.v Ho.\mo. 0 mm m m H H m AHooo.Hv AHHHH.MV AHmHH.HV AmHmm.v AHmmm.v Awmmm.v Ho.\:o.: H H H m o o H o H m m : mmumEHpmm omcHnEoo moumefiumm oocHnEoo How m HHHzmmHH .m> mama 30H>pmch pamEowmcmz .mHnt mocmwsfiucoo APPENDIX S (a) 135 Contingency Table. CPS (Project) Data vs. Rater 1 Rater 1 Estimates 2 1 0 z 0 0 2 2 ‘°1”/’°08 (.1977) (.3632) (1.4391) (2.0000) 0 2 4 6 ‘°O7/‘/Ol (.5931) (1.0897) (4.3172) (6.0000) 4 2 11 17 'OO/°O6 (1.6805) (3.0874) (12.2322) (17.0001) m 3 8 22 33 g '07/°13 (3.2621) (5.9931) (23.7448) (33.0000) 0 6 7 40 53 :; 'l“/°2O (5.2391) (9.6253) (38.1356) (53.0000) 8 21/ 27 9 18 65 - 92 'g ‘ ' (9.0943) (16.7080) (66.1977) (92.0000) 51 28/ 34 12 19 73 10" m ' " * (10.2805) (18.8874) (74.8322) (104.0001) m 5 15 60 80 O '35/'“1 (7.9080) (14.5287) (57.5632) (79.9999) 4 5 29 38 '42/'”8 (3.7563) (6.9011) (27.3425) (37.9999) 0 3 7 10 '“9/'55 (.9885) (1.8161) (7.1954) (10.0000) 2 43 79 313 435 (43.0001) (79.0000) (312.9999) (435.0000) Note: Numbers in parentheses denote expected frequencies. Three engineers were omitted from this analysis. APPENDIX S (b) 136 Contingency Table. CPS (Project) Data vs. Rater 2. Rater 2 Estimates 2 1 0 z 0 0 2 2 _,14/—.08 (.0598) (.2897) (1.6506) (2.0001) _ _ 0 3 3 6 ~07/ ~01 (.1793) (.8690) (4.9517) (6.0000) 2 5 10 17 °°O/°O6 (.5080) (2.4621) (14.0299) (17.0000) 1 8 24 33 g '07/'13 (.9862) (4.7793) (27.2345) (33.0000) Q 3 6 44 53 I: '1“/°2° (1.5839) (7.6759) (43.7402) (53.0000) 0 a) 3 12 77 92 -3 ~21/-27 (2.7494) (13.3241) (75.9264) (91.9999) E 28/ 34 1 16 87 104 ~/ - - (3.1080) (15.0621) (85.8299) (104.0000) U) m 1 7 72 8O 6’ '35/°“1 (2.3908) (11.5862) (66.0230) (80.0000) 2 4 32 38 .42/.48 (1.1356) (5.503“) (31.3609) (3709999) 0 2 8 10 .49/.55 (.2989) (1,4483) (8.2529) (10.0001) 13 63 359 435 (12.9999) (63.0001) (359.0000) (435.0000) Note: Numbers in parentheses denote expected frequencies. Three engineers were omitted from this analysis. 137 .mHmmHmcw man Bong popuHEo who: mpomchco mmpne .moHocosvmpm ompoooxm opococ mmmocpcogma EH mpmoesz "opoz HOOO0.0MOO HOOOO.OOHO AOOOO.NOV HOOOO.HOO HOOOO.mHV HOOOO.OV H OOO OOH HO HO OH O HOOOO.OHV HOOO0.0V AmOmO.HV HOOOO.V HHHOO.V HOOOH.O OO.\OO. OH H H N O O HHOOO.Omv Hmmm0.0HV HHOHO.OV HOOOO.mv HOOOO.HV HHOOO.O OO.\HO. mm mm m m m N O HOOO0.000 AOOO0.000 AmmOO.HHO HHOHH.OV HOmHO.OV AHOOO.HO HO.\Om. m OO OO NH O O H \) HOOOO.OOHO AOOOH.HHV AOMH0.0HO AOHOO.OHO AOOMH.OO AHHOH.HV m.\OH. a OOH OO OH OH O O O m HOOOO.HOV AmOHO.mOV HOHHH.MHV AOHO0.0V AOOH0.00 HOMOO.HV OH.\HH. ”w mm mm ma m N H o HOOOO.HOV HOOOm.OmV HOOO0.0V HHHHH.OO AHOHO.HO HOOOO.HO OH.\ H. MO mm HO O O H O O 0 AOOOO.mmO AOOOO.NHO AOmOH.OO AHOOH.mO HHONO.HV HOHO0.0 .mH.\HO. 0 mm OH H O O O m HOOOO.HHO HOOOO.HHO HOmmO.mv AOHOO.HO AOHmm.V HOHOm.O O.\OO. OH O m m H H O AOOOO.OO HmOHH.OO AHOOO.V HMOOO.V HOHmm.V AHONH.O HO.-\HO.- O m H H O O HOOOO.HV HHOHm.HO HHOON.V AHmOH.O HmOHH.O AOHOO.V OO.-\OH.- m H O O O O H O H m m O mmpmermm OmcHOEOO .m> OHOO ApomOopmv OOO mmmeHpmm oocHoEoo HOV m HHOszHO .mHan zocmwchcoo 138 APPENDIX T (a) Contingency Table. Project Interview Data vs. Rater l Rater 1 Estimates 2 1 0 2 2 0 8 10 "lu/“08 (.9885) (1.8161) (7.1954) (10.0000) 2 6 21 29 ‘°O7/"Ol (2.8667) (5.2667) (20.8667) (29.0001) 4 8 65 77 'OO/'06 (7.6115) (13.9839) (55.4046) (77.0000) 3 07/ 13 6 18 74 98 8 ' ' (9.6874) (17.7977) (70.5149) (98.0000) 8 14/ 20 10 17 66 93 H ' ° (9.1931) (16.8897) (66.9172) (93.0000) 5 7 18 41 66 g '21/'27 (6.5241) (11.9862) (47.4897) (66.0000) H 8 9 21 38 g '28/'3“ (3.7563) (6.9011) (27.3425) (37.9999) .2 2 3 12 17 g '35/°”1 (1.6805) (3.0874) (12.2322) (17.0001) Q4 42/ 48 O 0 4 4 ' ° (.3954) (.7264) (2.8782) (4.0000) 2 0 1 3 '“9/'55 (.2966) (.5448) (2.1586) (3.0000) 2 43 79 313 435 (43.0001) (79.0000) (313.0000) (435.0001) Note: Numbers in parentheses denote expected frequencies. Three engineers were omitted from this analysis. APPENDIX T (b) 139 Contingency Table. Project Interview Data vs. Rater 2 Rater 2 Estimates 2 1 0 z 0 1 9 10 '°1”/‘°08 (.2989) (1.4483) (8.2529) (10.0001) 0 6 23 29 ‘°°7/"°1 (.8667) (4.2000) (23.9333) (29.0000) 1 5 71 77 'OO/°O6 (2.3011) (11.1517) (63.5471) (76.9999) 9 2 17 79 98 § '07/'l3 (2.9287) (14.1931) (80.8782) (98.0000) 3 3 12 78 93 g '1“/'20 (2.7793) (13.4690) (76.7517) (93.0000) :> H 2 11 53 66 3 '21/-27 (1.9724) (9.5586) (54.4690) (66.0000) c: H 2 8 28 38 g ‘28/'3u (1.1356) (5.5034) (31.3609) (37.9999) g .35/.41 1 3 13 17 E (.5080) (2.4621) (14.0299) (17.0000) 0 0 4 4 '“2/'”8 (.1195) (.5793) (3.3011) (3.9999) 2 ' 0 1 3 .49/.55 (.0897) (.4345) (2.4759) (3.0001) 2 13 63 59 435 (12.9999) (63.0000) (359.0000) (434.9999) Note: Numbers in parentheses denote expected frequencies. Three engineers were omitted from this analysis. 140 .mfimzamcm wasp Eonm voppHEo who: wpomcfimcm moose .mmfiocoswmpm Umpomdxm opOCmv ammonucmmmd CH mpmoesz “mpoz HHooo.mmzv Hoooo.mmmv Hoooo.mmv HHooo.mHV Hoooo.:mv Hoooo.mv H mm: mam mm m: :m m HHooo.mv Hmmmo.mv Hmsmz.v Asmmm.v AmmoH.v HHmmo.v mm.\m . m H o o o m a a Hoooo.zv Hmozs.mv HHosm.V Amomm.v Asomm.v mewo.v .\m . 1 a H o o o o w: 3 MW Hoooo.HHv Homzm.HHv Homm:.mv Hszm.Hv Hmsmm.v HsHmm.v H:.\mm. m NH OH H N o H 1 HHooo.mmV Hmmmo.omv HHmH:.mV Hommo.mv Hmomo.mv Hmmms.v :m.\mm. 1 mm Hm : a m H w Hoooo.mov HmmHm.m:v Hmooz.mv Hamem.mv Hszo.mV Hmmmm.Hv sm.\Hm. a mo em 0H m m H M Hmmmm.mmv Hm0H5.mmv Hmmmm.MHv Hmmsm.wv HOHMH.mV HHHNm.Hv om.\:H. a mm mm mH o m m m Hoooo.mmv HmmmH.sov Hmswm.mHv HHmoz.mv Hmooz.mv Haemo.mv mH.\eo. 0 mm o» mH m . m H w Hoooo.esv H:Q:F.va Assam.OHv Hmzmz.av Hmmzm.:v HHmmm.Hv wo.\oo. e as so 6 a m H Hoooo.mmv Asomm.mHv HmmmH.:V Hooom.mv Hooom.Hv Hooom.v .1 .1 mm om m m H 0 Ho \50 Hoooo.OHV Hoomm.mv Hmmm:.Hv Hmmmm.v HaHmm.V Hmoom.v wo.u\:H.u OH w o H H o u o H m m : mmpmefipmm Umcfinfioo ‘ on a xHozmmma mopwEHpmm oocHnEoo .m> mama 3oa>mmucH poonOQm .oHnme aocowcfipcoo MICHIGAN STATE UNIVERSITY LIBRARIE L|| Llll LlllllllLlllll HS