ORGANPZATIONAL ORlENTATlON AND READERSHIP OF AIR FCRCE BASE NEWSPAPERS Thai: for flu We» of M. A. MICHIGAN STATE UNIVERSiTY Donaid Mac Donald 1961 THESIS LIBRARY Michigan State University ‘. \\\\ N\\\\\I\\\\\\I\I\\\\§\§§N\ 1; 3 1293 ‘0437 RETURNING MATERIALS: rV1ESch Place in book drop to LIBRARJES ‘ remove this checkout from ; w your record. FINES wiII be charged if book is returned after the date stamped below. I ' l ORGANIZATIONAL ORIENTATION and READERSHIP OF AIR FORCE BASE NEWSPAPERS by Donald MacDonald AN ABSTRACT Submitted to the College of Communication Arts Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of General Communication Arts 1961 Approved 340 hi“, 3 .lfiuijfvcm‘ TUL Donald MacDonald ABSTRACT The purpose of this study was to investigate the relationships between reading of selected management mes- sages in United States Air Force base newspapers and specified audience factors. The independent variables in the study were: 1) orientation toward the USAF as indexed in the study, 2) age, 5) civil education, and 4) the military rank of the respondents. The dependent variable was reading behavior indexed by what reapondents remembered having "looked at or read" in surveyed newspapers. The hypotheses were that reading of management mes- sages would be greater among respondents who: 1) are strongly USAF-oriented, 2) are older, 5) have more education, and 4) are of higher rank. It was also hypothesized that editorial matter perceived by coders as near-to-reader would be better read than material perceived as remote-from- reader. Respondents drew pencil lines through items they remembered having looked at or read when first they saw the surveyed newspaper issues. Editorial content of the newspapers was coded into thirteen subject-matter and six location-of—interest cate- gories. Mean readership scores were computed and rank ordered for selected respondent groups. These rank orders were compared with the amount displayed within each category. Donald MacDonald The homogeneity of the pOpulation was tested by the chi square statistic, for selected management message groups. These were: 1) health and safety,2§ and 3) commanders' columns and editorials. Questionnaires obtained demographic data on respon- dents as well as items used in constructing a Guttman scale of orientation toward the USAF. Conclusions of the study were: 1. Editorials and commanders' columns are better read by USAF-oriented than by other—oriented reapondents. The same is true for retention and training news. Reader- ship of health and safety news follows this pattern only for officers, with other-oriented airmen reading slightly more than the USAF-oriented group. Local news is less well read than USAF or command news by both orientation groups. 2. Retention and training news is read about equally by all ages. Health news, and commanders' columns or editorials receive higher reader attention as age increases. 5. Aside from unexpected high readership of health and safety news by respondents with grade school education, most subject-matter categories were better read by more highly educated groups. Local base news was also about as well read by the lowest education group as by any others. Again, USAF news, command news, and-—in one case--depart- ment of defense news are better read than local base news. 4. Airmen read more about retention and training than officers do.‘ Readership of such things as sports Donald MacDonald is about the same for all respondents, but officers read more of everything else in the base newspapers. This is true for location-of-interest as well as subject-matter categories. When asked what would make base newspapers more valu- able personally, readers asked for more: 1) base plans and policies, 2) USAF news, and 5) news of personalities at their own base. They also say they read most of base news, followed by USAF news and news of their command, in that order. The scoring used in this study does not sup- port their stated reading habits, but does support their desires. ORGANIZATIONAL ORIENTATION and READERSHIP OF AIR FORCE BASE NEWSPAPERS by Donald MacDonald A THESIS Submitted to the College of Communication Arts Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of General Communication Arts 1961 TABLE OF CONTENTS ACHONVLEDGEII‘IEDITLS o o o o O C . O . . C C v LIST OF TABLES . . . INTRODUCTION . CHI’LPTER I 0 CHAPTER II CHAPTER III. CHAPTER IV . . . BIBLIOGRAPHY . . . . . ALEPHIDIX A o o o o o 0 APPENDIX B . . . . . IIPPEIIID I:( C o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o o 89 APPENDIX D . . . . . . APPENDIX E ii ACKNOWLEDGEMENTS I wish to express deepest appreciation to my adviser, Dr. Malcolm S. MacLean, Jr., who strove to make clear to me the relationships between communication theory and practice, and who provided the planning assistance which made this study feasible; to Dr. David K. Berlo, who goaded me into more careful definition of the problem and my at— tack on it; to Professor John Marston, whose perceptive reading and comments have helped orient my thinking about all management communications; and to Dr. Paul J. Deutschmann, under whose tutelage I began to see the impli- cations of analyzing mass communications behavior, and who provided highly necessary material support for the study. Mr. John Prather gave invaluable assistance and many hours of work in designing and conducting IBM pro- cedures, and in supervising the involved computer program utilized for the study. Lt. Col. Arthur Dreyer, Directorate of Information, Office of the Secretary of the Air Force, provided encourage- ment and critical support in gathering data for this exami- nation of base newspapers and their audiences. I h0pe the results in some measure justify his confidence and contri- butions. . Finally, I am obliged to acknowledge not only the patience, but the weeks of monotonous work done and super- iii vised by my wife. Without her arduous efforts the study would have been seriously delayed. iv Tables l 10 11 LIST OF TABLES Source rank (airmen) for keeping informed on United States Air Force activities . . . . Readership of weekly newspapers by page . . . Readership of weekly newspapers by subject- matter, showing differences between reader- ship ranking by reader sub-groups . . . . . . United States Air Force attitudes toward commander's columns appearing in local USAF base newspapers . . . . . . . . . . . . . . . Emphasis on location-of-interest content categories in a USAF base newspaper, ac- cording to four measurement criteria. . . . . Readers, non-readers, absentees, and unscorable respondents indicated by news- papers. 0 O O 0 O O O O O O O O O O O O O O 0 Comparison of percentage of available editorial space occupied by subject-matter categories with mean readership percentages of respondent orientation subgroups . . . . . Comparison of percentage of available editorial space occupied by location-of- interest categories with mean readership percentages by respondent orientation sub- groups. 0 O 0 O 0 O O O O O O O O O O O 0 O 0 Comparison of percentage of available editorial space occupied by subject-matter categories with mean percentage of readership by five respondent age subgroups. . . . . . . Respondent age and readership of selected management communication subject-matter Categories. 0 O O O O O O O O O O O O O O O 0 Comparison of percentage of available editorial space occupied by location-of- interest categories with mean percentage of readership by five respondent age subgroups . Page 12 15 21 55 4O 51 55 55 56 57 Tables Page 12 Comparison of percentage of available editorial space occupied by subject-matter categories with mean percentage of reader- ship by five respondent education subgroups. . 59 15 Comparison of percentage of available editorial space occupied by location-of- interest categories with mean percentage of readership by five respondent education SUbgroupSooo0.0.0.000000000060 14 Comparison of percentage of available edi— torial space occupied by subject-matter categories with mean readership percentages of airmen and officers . . . . . . . . . . . . 62 15 Comparison of percentage of available edi- torial space occupied by location-of-interest categories with mean percentage of readership by airmen and officers . . . . . . . . . . . . 65 vi INTRODUCTION This study describes a channel of military management communication and the audience of that communication, United States Air Force base newspapers and their readers. The gross message stimulus is divided into its component parts according to subject-matter and location-of—interest cate- gories; a sample of the reader pepulation is divided into categories by age, education, economic status and reference group orientation. Then, the study asks who in the audi- ence attends to specified parts of the gross message stimulus. A measuring device has been constructed to obtain a simple report of what USAF airmen and officers remember having looked at or read in the most recent issue of selected USAF base newspapers. The assumption under which most internal publications are established is that these media offer channels for management to use in diffusing its plans and policies. Newspapers are not the only medium used, the list including bulletin boards, public address systems, employee mass briefings. letters. magazines. and counseling sessions. Such observers as Garth Bentley,1 the Opinion Research lGarth Bentley, Editing the Com an _Pub1ication (New York: Harper and Brothers, 1953;, Chap. . 2 Corporation,1 and the National Industrial Conference Board, Incorporated,2 agree in conceding primacy among these media to the internal organization newspaper. The USAF manual prescribing base newspaper organiza- tion and operation states:5 Readership is more than eighty percent of base personnel and sixty-seven percent of Air Force wives. . . . No other medium within the Air Force consistently has so large an audience. . . . .The commander and the information officer should use the base newspaper as the basic medium of the internal information program. They must make every effort to maintain the base newspaper as the major source of informa- tion on the base. Such a description implies great communicative ef— ficiency. Yet, analysis of response to questions on official USAF tri-annual sample surveys indicates that fewer than sixteen percent of today's USAF personnel regard the base newspaper as either the first or second most im- portant source of their own information about the USAF. Data from one survey show other sources preferred.4 (See Table 1) 1Case Book for Industrial Editors (Princeton, N. J.: Opinion Research Corporation, 19535, p. 5. 2"Employee Magazines and Newspapers," Studies in Personnel Polic , No. 156 (New York: National Industrial Conference Board, Inc.), p. 10. 3USAF Manual 190-4, Information Services, Sect. B, par. 1. 4Analysis of the Air Force Information Pro ram, USAF, Sept. 1 O, p. ISI. 5 It may be noted here that the local commander has rather full control of the commander‘s call, daily bul- 1etin, and the base newspaper, and that the Aig'ggggg_gimg§ is a civilian publication not controlled by the Air Ferce. In each instance, the local media received lower ratings by officers than indicated in Table l for airmen, with an increase in reliance on official pamphlets and on the A;; 22292 Timgg. Thus, we can see that the commander's call far outstrips the newspaper as a preferred first or second source of information about the base and USAF. TABLE 1 SOURCE RANK (AIRMEN) FOR KEEPING INFORMED ON USAF ACTIVITIES Best-1959 2d best Best-1960 2d best Commander's call 19.5% 15.5% 27.7% 14.5% Airman magazine 4.5 10.4 5.1 9.8 USAF news review 5.7 7.9 5.0 7.5 Official pamphlets 2.2 5.4 1.9 2.8 Other official 4.2 9.8 4.2 8.9 publications Base newspapers 5.5 12.2 4.9 11.0 Ag; Egggg Timgg 45.0 17.7 58.0 21.2 Civil newspapers 4.8 8.9 4.6 8.7 and magazines . For many years a newspaper axiom held that local news is read more than non-local news, and editors of internal publications--after which USAF base newspapers are pat- terned-~have assumed this to be true among their "captive" 4 audiences. The present study submits that axiom to a test based on the military readers of USAF base newspapers. To do so, it tests reading differences of selected content categories. One hypothesis of the study derives from examination of similar relationships between messages and readers conducted by Merton1 and by Gouldner.2 In brief, each observer prOposed that the psychologi- cal orientation of a person toward the community affects the person's communications behavior. It should follow that those who are community-oriented will attend more to community-centered messages than will persons who are not so oriented. A tentative hypothesis will be that grganizatiog- mpg; widely Egg; gill persons 9; different ggientation. Newspapers, even of this rather restricted sort, print stories about many things. What is printed is largely a matter of editorial judgment, although editors must work within limitations imposed by civil and military managements. (Knowledge of relative emphasis on subject-matter categories, and an index of the attention or interest given to each by a sample of the readers, should provide management and 1Robert K. Merton and Alice s. Kitt, cited in Con- tinuities in Social Research, ed. R. K. Merton and Paul F. Lazarsfeld_TGlencoe, Ill.: The Free Press, 1950), pp. 42-69. 2Alvin W. Gouldner, "CosmOpolitans and Locals: Toward an Analysis of Latent Social Roles--I," Administrative Science Quarterly, II (1957), 281-506. 5 editors with an indication of general effectiveness of the medium; After all, a message which is not attended to is not effectiveg Conducting the study by correspondence has advantages, but it has more limitations. Since it is difficult to obtain verification of the reader attention to specific items when many items are being tested, a simple and in- ternally consistent measuring device has to be used. A self-administered "reader interest" test was adapted for this study, based on a commercial survey method known as "Readex."l With the consultation of faculty members of the Michigan State University Communications Research Center, this method was revised for the present study and pretested on a small sample of military readers of the Selfridge Air Force Base, Michigan, base newspaper.2 The study categorizes editorial content of several base newspapers by subject-matter, then relates personal characteristics of readers to the attention given to the categories. In other werds, it asks "who reads what" in these publications. The reference group orientation of the respondents was obtained from questionnaire data. Answers were dichoto- 1Alfred Politz Research, Inc., A Study 9; Printed Advertisigg Ratin Methods (New York: Advertising Research oundation, 1 5 , pp. 1-5. 2The pretest was administered through the Selfridge AFB Information Office, with the c00peration of the base information officer, 2/Lt. Scott Walker, on March 18, 1961. 6 mized, then converted into a Guttmann scale of orientation toward the USAF. These mean scores, when used to indicate behavior of respondent groups, provide another index of "who reads what;" this is related to location-of-interest categories across all newSpapers surveyed.1 Coding items into categories was conducted after reference to various studies of categorization, including 5 4 those by Berelson,2 Deutschmann, and Freyschlag, and con- siderable cut-and-fit testing. It was decided to utilize categories as broad as would fit the study, as the most practical means of obtaining coding reliability. The significance of the study lies in its attempt to provide the management of a military organization with evidence of the success of certain messages, and to provide editors with a more accurate profile of the attention given to different items in their publications. Definitions and explanations are presented in the study where they can be placed in perspective, but the 1;; Machine Method pg; Scali_ng _a_§ rm _a_s_ Twelve Dichotomies. WasEIngton AgricuItural Experiment Stations, Institute of Agricultural Sciences, State College of Washington (Circular 529, August, 1958). 2Bernard Berelson, Content Analysisin Comm ication Research (Glencoe, 111.: The Free Press, 555, Chap. 5. 5Paul J. Deutschmann, News-page Content 2; Twelve MetrOpolitan Dailies (E. Lansing, Mich.: Communicatibns Research Center, Michigan State University, 1959). 4Kedrick G. Freyschlag, "Categories for Analyzing News in Terms of Reader-Interest" (unpublished Master's thesis, Stanford University, 1955). 7 intent of the study can now be seen more clearly: it is to index the subject-matter and location-of—interest, and the readership, of a type of management communications medium by indicators not now in general use among military publications. The study is presented in four sections. Chapter I provides a theoretical background and rationale for the study, and states the hypotheses. Chapter II explains and describes the categories and measuring devices, and discusses the survey design, administration, and analysis procedures. Chapter III reports results of statistical analysis related to the main hypotheses and other matters of interest. Chapter IV briefly reviews the means and ends of the study, presents conclusions, discusses what was or was not supported by the findings, and makes recom- mendations for further study. Additional raw data, toget- her with samples of the instructions and test instruments, are presented in the appendixes to assist in future analysis. CHAPTER I Most humans who knowingly produce messages do so with the intent of influencing someone who receives the messages. Most messages have more or less specific in- tended receivers, or audiences. If messages are produced, transmitted, and made available to the intended receiver, there still is no guarantee that the communication process is complete. For, if the receiver does not have similar meanings for the signs used in encoding the messages, he cannot "make sense" of the messages. Or, if the subject of the messages is not seen as relevant or important, he probably will not bother to examine the messages very long or very thoroughly. (Nhat is important to the receiver should be known to the message-source if messages are to be produced which have a good chance of influencing the receiver} Schramm's model of overlapping fields of experience portrays this problem simply and effectively:1 A's B's Experience Experience Experienc- 0-perienc- No Communication Occurs Communication Occurs lWilbur Schramm, "The Communication Process," The Process and Effects 9; Mass Communications, ed. Wilbur Schramm (Urbana, I11.: Univ. of I11. Press, 1954), Chap. 1. 9 @his is to say that the receiver generally must see the message as pertinent or he will not consciously attend to it; The act of attending to a stimulus is an initial phase in perception, leading to interest, consideration, the assignment of meaning, and response. The mere fact of physical transmission of signs does not constitute meaning. If we can accept the analogy drawn by Berlo in describing how we learn meanings, we may arrive at ways of determining whether we are effective in winning attention to our messages. He writes:1 We learn language, or through language, in the same way that we learn many other things. A person's responses must be rewarded if they are to be retained. Stimuli must be presented in terms of the receiver's own interest and exper- ience. . . .A stimulus is perceived (decoding), some internal response-stimulus pattern is set up (interpretation), which we will call mean- ings, and some overt response (encoding) is made. :Thus, a communication is successful only when the experience of the source and receiver are enough alike that they have similar meanings for the messages) The process by which the source determines whether his receiver has a meaning for a message--feedback--is second nature to all of us. In face-to-face communication the source can see, feel, or hear the reaction to his mes- sage. Usually, he can determine lack of meaning as well; 1David K. Berlo, The Process of Communication (New York: Holt, Rinehart an_d w“—""inston, EeT—‘7—o , p. 18 . 10 if the receiver does not react visibly the most frequent conclusion is that some aspect of the message must be changed. Here, the communicator has several alternatives: 1. To ignore the non-response and change the subject. This implies that the whole thing was rather unimportant. 2. To recode, amplify, repeat, or change the intensity of the message stimulus. 5. To make use of other stimuli, and draw pic- tures, pantomine, or write the message. 4. To call in an authority to retransmit the message. 5. To query the receiver about the meaning of the message. _When one is not in a face-to-face communication situation opportunity for feedback is greatly reduced. The source can seldom see, feel, or hear the conse- quences of his messages after they have been transmitted, at least for a rather long while. Man is ingenious, however, and has devised formal and informal ways of compensating for this restriction of immediate feedback. 2 The Nielsen index,l Videodex, and others all use reports of message receivers to determine who was exposed to 5 specific groups of messages. The Pulse surveys and luBig Nielsen News: Instant TV Ratings," 8 onsor, November 1, 1958, pp. 57-59. 2"Videodex Reports 141 Cities, Adds Score of New Markets in Last Quarter," S onsor, October 11, 1958, pp. 40-41. 5"The Pulse--In Home Interviews," S onsor, September 20, 1958, pp. 50-52. ll 1 ask listeners what radio and television Hooper reports programs they have attended to. And newspapers receive letters to the editor as well as Audience Research Bureau data.2 Newspapers and magazines have spent much time and money in developing useful indicators of reader attention and interest, both for editorial matter and for advertising. In doing this, they have employed interviews, question- naires, telephone surveys, and other techniques of asses- sing receiver responses. In addition, these studies have determined, story by story, advertisement by advertisement, and issue by issue, what kinds of people attend to what kinds of messages. RESEARCH FINDINGS: Of particular interest to the present study are the findings of a series of newspaper analyses. Schramm and Ludwig reported a 1951 study of weekly newspapers:5 1. The reader of the weekly newspaper typi- cally reads about thirty-six percent of all the items in the newspaper, about forty-one percent of all news items. 2. The percentage of items read in a weekly tends to vary inversely with the number 1"Hooper Riding the Big Radio Boom," s onsor, October 18, 1958, pp. 42-44. 2"ARB Expands its Measurements," §EEB§2£$ September 27’ 1958, pp. 56-58. 3wilbur Schramm and Merritt Ludwig, "The Weekly News- paper and its Readers," Journalism Quarterly, XXVIII, No. 5 (1951), 501-514. "“““" 12 of pages or total columns in the news- paper. 3. An item on page one is about twice as like- 1y to be read as an item on any other page of a weekly newspaper. 4. Local editorials are better read, on the average, than half the other content cate— gories. 5. More than any other variable yet isolated, localness and especially local names makes for higher readership of a weekly newspaper. Some of the relationships found by Schramm and Ludwig may be seen easily in simple tabulations:1 TABLE 2 READERSHIP OF WEEKLY NEWSPAPERS BY PAGE Men Women All Readers Page one 48% 52% 50% All other pages 26 51 28 NOTE: Figures cited are mean percentages of all items read. (N-1085) Average number of items on front page was thirty-two. Average number for all pages was eighteen. From studies conducted with other newspapers, Schramm and White made further observations about audience attention:2 A substantial proportion of the news readers in. each community . . .are merely headline readers, Ibid. 2Wilbur Schramm and David M. White, "Age, Education, and Economic Status as Factors in Newspaper Reading: Con- clusions," The Process and Effects g£,Mass Communications, ed. Wilbur Schramm ZUrbana, $11.: University of Illinois Press. 1954). pp- 71-75. 15 TABLE 5 READERSHIP 0F WEEKLY NEWSPAPERS BY SUBJECT-MATTER, SHOWING DIFFERENCES BETWEEN READERSHIP RANKING BY READER SUB-GROUPSl Rank-order of Mean Mean Mean categories by N of Percent Percent Percent male readership Items Men Women All Readers Local pictures 94 74.0% 80.1% 77.1% Local columns 18 59.6 59.1 59.4 Human interest 51 45.8 56.6 50.2 Sports 91 42.5 26.1 54.5 Personals 129 42.5 67.1 54.7 General News 785 41.9 48.6 45.5 Local editorials 52 40.5 56.2 58.5 Society 584 24.0 51.8 57.9 NOTE: This includes only the eight categories most read by all readers. Seven more categories had lower readership. lSchramm and Ludwig, pp 501-314. 14 passing on to the comics, sports, or woman's page where their true interest lies. Allport noted that interest in public affairs news, Which we may equate with some kinds of management com- munications, is rapidly exhausted, with readers quickly turning to news of more personal interest.1 Robert Lane writes, referring to factors predisposing peOple to read about public affairs:2 Surveys have regularly shown that from twenty to twenty-five percent of the pOpulation care only about immediate national issues and evince little interest in even the most important as- pects of-foreign affairs. Who those people are we do not know, but in general an interest in foreign affairs is associated with level of education. 380 we may see that attention is quite broadly given to or withheld from news according to its salience; As the news is seen close to us, geographically, politically, or psychologically, we tend to become interested in it. Schramm and White found that age, education, and economic status are associated with the attention or lack of attention given to certain kinds of news in general newspapers. They noted that higher education, age, and economic status tend to correlate positively with reading of delayed reward items. "Delayed reward" may be defined 1Gordon W. Allport and Janet McFadden, "The Psychology of Newspapers: Five Tentative Laws," Public Opinion Qu uterly, IV (1940), 702-705. 2Robert Lane, Political Life (Glencoe, 111.. The Free Press, 1959), p. 158. 15 simply as that news which is not seen as having direct and visible effect on the reader, his purse, family, or activities. Its effect is either so hidden from view or so remote in the future that many readers do not perceive its relation to them.1 The relationships found by Schramm and White may be explained as a function of developing personality, in which youth has not yet internalized the concept of civic duty, being more interested in the here-and-now, while older and more educated persons, who would tend to occupy higher economic status, have internalized this concept so that "their duty is their pleasure." INTERNAL PUBLICATIONS: That readers of company publications do attend to news seen as personally and immediately rewarding is borne out by a report made by Walter Barlow of the Opinion Research Corporation:2 Where the news was of universal or local interest, readership and retention figures were high. But if the news gimmick did not apply to the local area, the figures were likely to be disappointing. Not all observers have arrived at the same conclu- sions; one editor engaged in research for industrial communications, James McCloskey, has noted other impressions, lSchramm and White, pp. 71—75. 2"In-Plant Communication: A Seminar Report," Cornell Conference Reports (Ithaca, N.Y.: Cornell University SchooI of Business Administration, 1955), pp. 68-69. 16 although he is not specific as to the source:1 Insofar as questionnaires can be meaningful in so intangible a prOposition as what employees want to read in their company magazines, the surveys indicate that most readers get as much if not more than they like of such staples as departmental notes, sports, woman's page, and gossip items. What many 92 want more of is news about the company's operations, and future plans and prospects. These are things that affect their livelihood. Some previous surveys have indicated that "news about the company's operations, and future plans and pro- spects" has been thought by management to be in the "delayed reward" category. At least, Whyte indicates as much in writing of what he considers a colossal failure by American industry in telling employees about business.2 Summarizing studies of employee communications programs, the National Industrial Conference Board, Inc., reported these modal primary objectives:3 1. To promote unity in the organization. 2. To promote understanding, help employees become better acquainted. 5. To promote two-way communication--the inter- change of news and ideas among employees at all levels in the organization. 4. To recognize employees as individuals and as groups. - lJames McCloskey, Industrial Journalism Today (New York: Harper and Brothers, 1959), p. . 2William H. Whyte, Jr., Is Anybody Listening? (New York: Simon and Schuster, 19527. ~ 3"Employee Magazines and Newspapers," Studies in Per- sonnel Poligy (New York: National Industrial Conference Board, Inc., 1955), p. 11. l7 5. To give employees information about the company 6. To provide a vehicle for management phil- OSOphy. [The USAF and its commands have regarded base news- papers as industry regards its "house organs," allotting thirty percent of the total newspaper evaluation score to their use as a "management tool," and another thirty per- cent to the newspapers' support of "USAF information objectives." It fully supports the dicta relating to the power of "local" news in gaining attention to the papers as a channel of management communication. Typical instructions are! Strategic AirCommand:1 If it's worth page one, it certainly should carry the local angle. Air Training Command:2 Everything appearing on the front page should be locally written copy or copy that has been localized. Air Defense Command:3 Great emphasis is placed on localization of news service materials. Other official directives concern various standard practices used to increase the attention-getting qualities of news items, pictures, and feature stories. These are generally taken from journalism texts or from civil news— paper style-books. 1 2"Spot Checking ATC House Organs," Air Training Command, 1960- SAC Editor's File, SAC Press Service, July, 1960. 3ADC Regulation 190—4, August 22, 1960. 18 Perhaps the best-known description of the uses of the company publication is found in Bentley's primer for company editors: 1 It is the sole channel of communication between management and employee which has won established acceptance and, as a result, is the most power- ful and influential that can be used. Such a publication should be planned with a fourfold purpose: 1. As the company's mouthpiece, it explains and interprets policies simply, fully, reasonably, thus making them understood and creating a desire to cooperate with them. ‘It explains why certain steps are taken, enabling workers to know exactly what is expected of them and why. - The employee paper builds morale, preventing friction among departments and individuals. . . .It welds diverse employee groups into one large 'family' by focusing attention on common aims and purposes. . . .By humanizing the employer, it creates better feeling between officers and employees. Since every comprehensive employee relations program has social and recreational angles, the employee paper is a spark plug. It an- nounces coming events. It maintains enthus- iasm while the events are in progress. It gives credit publicly to winners and other outstanding performers. It encourages personal ambition by recog- nizing and publicizing promotions and good performance. It promotes Obedience to rules and regulations by pointing out the reasons why conformity is essential. It is a spur to safety programs, economy drives, sales drives, courtesy campaigns. Bentley then notes that the pages must include: 1) what the company wants printed and, 2) what the employee lBentley, Editing the Company Publication, pp. 5-4. l9 wants to read. He then concludes that:1 Unless readers read it regularly and thoroughly, all effort put into company promotion is lost. . . .Interestingly enough, the dividing line be- tween what employees want and what the company wants is not a sharp one. . . .The majority of employees-~believe it or not-~always Show in— terest in good stories explaining company poli- cies, how products are made, and other promotional material. AUDIENCE RESEARCH METHODS: In all the cited studies, two principal methods were used to obtain employee feedback, or reader feedback, about what messages were attended to: interviews with persons presumed to have been exposed to the medium, and questionnaires to similar groups of freaders." The mes- sages in the publications were not categorized by subject- matter in most cases, nor were they ranked by space or frequency emphasis and correlated with readership. Ludeke reported on the editorial research program of the Curtis Publishing Company, which may be used to illustrate:2 The backbone of our editorial research program is is a continuous series of surveys designed to provide us with specific information concerning each editorial item in the particular issue of the magazine being surveyed. This specific in- formation shows us the proportion of the persons exposed to the issue who: . 1. Recalled seeing the item. lIbid., pp. 23-24. 2Herbert C. Ludeke, Ma azine Editorial Research (Philadelphia: The Curtis Publishing 06mpany, n.d.), pp. 5-4. 2O 2. Read at least some of the item's main body of text. 5. Read all of the item's main body of text. 4. Expressed enthusiasm for the item. Finally, we get a rather complete record of the characteristics of the person providing the in- terview, ranging from age and sex through his, or her, occupation, education, and income. . . We begin to see the outlines of who reads what with what effect. This is an accurate, if brief, summary of the con- ventional method of gathering news media feedback. But not all organizations have the time, money, or manpower to utilize this expensive method, so questionnaires are used. The USAF uses questionnaires for very large samples of commissioned and enlisted personnel, obtaining feedback on attitudes toward the internal communications channels, and obtains an indication of retention for a few specific items. The sources of information about the USAF in general, weapons systems, and highly important personnel policies comprise a significant portion of these questions. The October, 1960, sample survey, for example, asked more than 10,000 persons, "What is your opinion of the Com- mander's Column in your base newspaper?" The breakdown of answers is contained in the table on the following page:1 Publishers of USAF base newspapers thus obtain lUSAF Tri-annual Sample Survey, October, 1960. 21 TABLE 4 USAF ATTITUDES TOWARD COMMANDER'S COLUMNS APPEARING IN LOCAL USAF BASE NEWSPAPERS Do not have a base newSpaper . . . . . . . . . . . 10.4% Have a base newspaper, but: It does not have a Commander's Column . . . . 9.2 I do not recall a Commander's Column, as I do not read the base newspaper regularly. .28.5 My base runs a Commander's Column, and I find it: Over-all Airmen reading of the and evaluating Sample Commander's Column Very good. . . . . . . . . 14.0% 27.0% Good . . . . . . . . . . . 24.4 _47.0 Undecided. . . . . . . . . 9.0 17.0 Not good at all. . . . . . 1.2 2.0 22 useful feedback, but do ppp know whether their reader pro— files correlate highly with those indicated by research on other publications. The question remains, "Who reads what?" ' The questionnaire method has been criticized, as have all methods of obtaining response without Specific retention evidence, as being grossly inaccurate. The, method has, accordingly, been put to test. Frieberg as- 'sessed the relative validities of unaided, partially-aided, and aided recall questionnaires used to determine whether subscribers to a group of semi-technical magazines actually "looked into" the issues surveyed. The aided recall mail questionnaire sample of 1296 subscribers was interviewed, with an average report of having read ten percent less than had been indicated by the mail questionnaire.l Ludeke placed magazines in rooms with one—way win- dows, then observed the actual page-by-page "reading" 'behavior of subjects. Queried on their reading, most re? ported having read or looked at somewhat less than was actually true.2 Both of these examples may be considered as types of validity assessment for reader reports upon their own reading behavior. Albert D. Frieberg, A Comparative Spudy of Mail 1 Questionnaire Techniques for Measuring Readin p:_Business Pub ications (Chicago: Putman Publishing ompany, 19595.. 2Ludeke, Magazine Editorial Research, p. 4. I '\ 25 Politz conducted a comparative study of three methods for rating the interest given to printed advertisements,’ including two interview methods and one aided recall mailed rating system. While the patterns of attention were simi- lar as indicated by reader response, the tests were not completed on the same magazine issues and cannot be care- fully validated. The conclusions of the study, however, report confidence in the reliability of the mail method used.1 APPLICATIONS TO THE PRESENT STUDY: The mail method tested by Politz is a reader pp: terest survey which reports the number of pe0ple whose previous reading of an advertisement left at least enough impression for them to remember--when they see it again-- that it was of interest.2 This is the method which-has been adapted for the present study, in combination with a questionnaire not relating to readership of a SpeCific newspaper issue. The reading measurement device is self-administered, requiring the respondent to draw a vertical line through the middle of each editorial item which he remembers having looked at , or read, pp far down pp hp remembers having read pp when lPolitz, A Study pf Printed.Advertising Rating Methods, p. 5. 2The commercial practitioner is Readex, ng., St. Paul, Minnesota. The method was deVeloped to serve pub- lishers of smaller magazines and newspapers, and is used on advertising and editorial items. 24 he first went through the issue. In addition, each respondent is asked to provide two sets of data about himself: 1) demographic information, and 2) answers to questions determining the intensity of his commitment to his unit and the USAF. What may we expect to find from the resulting data? Our findings maybe divided into three areas: 1. The emphasis given to various subject- matter and location-of-interest categories, indexed by percentages of available ed- itorial space displayed. 2. The degree to which a e, education, and rank (economic status? among USAF military readers results in the same kinds of read- ing behavior as found by Schramm and White. 5. The relative readership of specific command messages by persons who are highly oriented toward the organization and by those who are less committed. Since Merton described the "cosm0politan-loca1" continuum among community influentials, and Gouldner applied a similar description to a college faculty, communications researchers have shown increasing interest in the implica- tions for communication-generated behavior. From a purely practical point of view, it seems fruitful to attempt to determine a relationship between reference—group orienta-- tion and attending to certain kinds of messages. What the USAF describes as "management" items in its t0pical analysis of base newspapers encompasses a vast range of subject-matter, including: 1) editorials and columns by base and other commanders, 2) retention and training in- formation, and 5) safety programs. Are these three attended 25 to differentially by those with a strong USAF orientation and by those with other orientations? The same questions may be asked about readers along the age, education, and rank scales, both among enlisted and commissioned readers. Indeed, the conclusions of the USAF tri-annual information surveys indicate there are differences; as rank increases there seems to be a greater conviction that internal information media provide enough of the right kinds of news. There is, then, a possibility that military management is operating like the religious evangelist who preaches hell's fire and brimstone to the least needful of his congregation-~those already in church. Schramm's findings noted above, that age, education, and economic status are associated with reading of "delayed reward" news items such as editorials, government activities, and public service appeals, are of interest to military and industrial managers. In the USAF, whether first-term 'enlistees and junior officers read retention and training information or not is of paramount importance. These considerations lead to two hypotheses: first, that attention pp organization-centered messages gill pg greater ppppg ppganipation-committed pppsqns than among ppppp,ppp ppyp lgpp strong, pp ppp-gpppp, commitment, and second, that attention pf pppp groups 31;; pg greater pp content categories perceived py coders as near-to-reader, and less for categories perceived pp remote-from-reader. 26 THE MEANS USED-—AN OVERVIEW: The present study does not attempt to do more than establish in one situation a procedure for recurrent de- scription and analysis of message presentation. ITo evaluate newspapers we need to know what messages they contain and who reads the messages; Appraisal and evaluation have been largely based on subjective judgments of fellow message- producers, and on assumptions such as: pl) a house organ "captive audience" reacts to the publication just as it does to the city newspaper or country weekly, and 2) avail- ability guarantees readership. Neither proposition seems supportable until tested. The group administration of questionnaire and news- paper marking survey was chosen for use in this study in consideration of its relative similarity to previous feed— back methods utilized by the USAF. There are other methods of obtaining reader reports about communication behavior, and some of those methods could provide more exact ans— wers to the questions asked in this study; they all involve, however, much more cost in terms of time, manpower, and money. On the other hand, a simple questionnaire cannot garner the same amount and range of information for tabula- tion and analysis. This study seems to tread a middle way. The reading measurement used here does not discrimi- nate by percent read of individual items, nor does it demand evidence of retention of the subject-matter tested. It 27 does not, however, seem subject to systematic bias. It is intended to provide an index of "readership" roughly.as follows: 1. Eyposed: read at least one item in the newspaper, but not this item. 2. Glanced pp: read any of this item. 5. Interested: read past the headline or lead of this item. ‘ When one has measured "readership" of the content categories in a periodical he is faced with the necessity of deciding just what he measured. The usefulness of measurement is contingent here upon the applicability of the measure to editorial management; thus, the content of messages must be categorized according to the purposes of the investigation. As Berelson says, content analysis stands or falls by its categories.1 The matter of devising categories has been studied by journalists, psychologists, and others for at least sixty years, with somewhat less than total agreement on the criteria for categorization. Many of the attempts are catalogued by Freyschlag, who arrives at a specification of content categories based on reader interest.2 Virtually all investigators during the past decade have concluded, as Berelson did, that a rigid set of categories cannot lBerelson, p. 88. 2Freyschlag, pp. l9—57 '28 be found, and that content analysis should employ the categories most meaningful for the study at hand.1 This study deals with two types of categories: subject-matter and location-of-interest. Berelson describes content categories as comprising "what is said." He Be— lieves that categories Should be as broad as is consistent with the purpose of the investigation, to increase the reliability of the coding Operation.2 Freyschlag tried to steer clear of "general" categories, feeling that single-coded categories may have several sub-elements in them. I prefer the former view for two reasons. First, a study of this kind requires high coding reliability; while double coding may sometimes provide subtler data, broader categories may be equally useful. Second, as Markham and Stempel have found, such criteria as headline display serve quite adequately as indexes of category emphasis. Their study of twenty-four Pennsylvania daily newspapers resulted in the conclusions that:3 It appears that the performance of the three different measures--space, issue frequency, and headline display-~was nearly identical. . . .Space measurement is the best—known and most standardized content analysis technique. The findings of this study suggest that some l 2 Berelson, Chap. 5. Ibid. 3James W. Markham and Guido H. Stempel, III, "Analy- sis of Techniques in Measurin Press Performance," Journalism Quarterly,.XXXIV, No. 2 (1957 , 187-190. 29 criticisms of space measurement are erroneous. . . .Space represents not just the volume alone, but symbols, issues and themes expressed. Nafziger, MacLean, and Engstrom comment on division of content into subject-matter, source, and type categories. While they agree that newspaper reading is a voluntary act and that reader interest must be aroused, they enumerate significant cautions for interpretation of readership scores. They posit that the reader's choices may depend on a transient mood, available time, or the competition of other stimuli. They further caution against concluding that a single test is indicative of continuing behavior. This, of course, proscribes confident ascription of reader "interest" from results of a one-wave survey. As they note:1 The fact is that we have little information by which to judge how closely related are readership and interest. Application of theory to the present study, with discussion of the choices made, are discussed in greater detail in the following chapter. lRalph O. Nafziger, Malcolm MacLean, Jr., and Warren Engstrom, "Useful Tools for Interpreting Newspaper Readership Data," Journalism Quarterly, XXVIII, No. 4 (1951). p. 456. CHAPTER II This chapter includes the survey design, description of the categories and measuring insturments, the sampling procedure, and a report on the survey administration. VARIABLES: The present study was designed to determine whether the readership of specified USAF management messages varies significantly with the reference-group orientation, age, civil education, and military rank of a sample of military readers of the newspapers. The independent variables in this study are the respondent characteristics of age, education, rank, and reference-group orientation. The dependent variable is reader attention given to the management messages. CATEGORIES: The editorial content of eight USAF base newspapers was divided into thirteen subject-matter and six location— of-interest categories. Each editorial item in the news- papers was coded into one subject-matter and one location- of—interest category. The procedure chosen for categorizing subject-matter was evolved from a t0pica1 analysis form used for periodic evaluation of USAF base newspapers, together with repeated applications of the categories to actual base newspapers. The major categories were broken down into sub-categories 51 for easier location and assignment in the course of content coding. The list is included as Appendix C to the study; however, the over-all categories are: 1. Health and safety. 2. New facilities, units, and equipment at this base. 5. Training and retention at this base. 4. Base management (a revised and limited list). 5. Military service and social clubs. 6. Education, culture, and religion. 7. Dependents' affairs and housing. 8. Military recreation, amusement, sports, and hobbies. 9. Promotion and recognition of military and civil employees. 10. Commanders‘ and others' messages, and editorials. 11. Community relations. 12. The "big picture" of command, USAF, and Department of Defense. 15. Miscellany, including military personals not otherwise codable, humor, and fillers not otherwise codable. The location-of-interest categories were based on certain characteristics of the items aside from subject- matter. Virtually all studies heretofore cited suggest that "localness" is a prime determinant of readership. Starting with the most local area and proceeding to the most general area, the categories used in the present study are: 52 1. This base. 2. The adjacent civil community. 5. Air command or numbered air force in line above this base. 4. USAF in general, with no local application noted in the item. 5. Department of Defense, including NORAD, Army, and Navy. 6. Other. An operational definition of the location—of-interest categories presented some difficulties, although Deutschmann's description of them as "where-to-reader" categories appears to be a happy solution in many respects.1 As with subject-matter categories, the number of items, column inches, or percentage of editorial space could con- ceivably result in widely divergent pictures of content structure. To test this, a fourteen-page newspaper was analyzed, using a variant of the abovelocation-of-interest categories. Items were coded in four ways: 1) by frequency, 2) by column inches, 5) by percentage of available editorial space according to first theme expressed, and 4) by propor- tion of individual items containing discrete location category content. All except the last were coded with the criterion of having the desired characteristic present in the headline, lead paragraph, outline, pictorial matter, lDeutschmann, p. 44. 53 or in the body of the item if there was no other indicator. When all items were placed in discrete categories by the above methods, the categories were ranked and put into a table: TABLE 5 EMPHASIS ON LOCATION—OF—INTEREST CONTENT CATEGORIES IN A USAF BASE NEWSPAPER, ACCORDING TO FOUR MEASURING CRITERIA Rank-order of items in the newspaper Criterion Base Town Command USAF Other Number of items 1 4 5 2 5 (N871) (n=59) — (n=4) (n=ll) (n=l4) (n=5) Column inches 1 4 2 5 5 (N=504) (n=519) (n=29) (n=77) (n=65) (n=16) Percent available 1 4 5 2 5 editorial space (65.5%) (5.8%) (15.5%) (14.4%) (5.2%) Percent available 1 5 2 5 4 editorial space within items This table lends credence to the use of item count and percent of available space as adequate indicators of category emphasis, as noted by Markham and Stempel.l To insure coding reliability, a pretest was con— ducted. Four USAF base newspapers, ranging in size from fifty-eight to 105 editorial items, were coded into the nineteen content categories. With no verbal instructions to the coders, but using a list of representative t0pics lMarkham and Stempel, p. 94. 54 subsumed under each content Category, the following results were obtained by four coders: 82.8 per cent coding agree- ment for subject-matter categories, and 80.4 percent agree- ment for location-of-interest categories. When coders were briefed, and the words "at this base" affixed to two subject-matter categories, the mean percentages of agreement increased to 89.2 for subject- matter and 91.1 for location-of-interest. Arbitrary judgment of the author was used in four cases, but discus- sion with two coders resolved all other differences. Coding of the eight newspapers used in the survey was accomplished by three coders, all familiar with USAF command structure; this point had caused the majority of non-agreements in the coding pretest. The mean percentages of coding agreement were 87.9 for subject-matter categories and 94.5 for location-of-interest categories. READING MEASUREMENT AND SCORING: The reading measurement used is simple to monitor, easy to complete. Reading scores for individuals and groups can be obtained quickly and objectively. The measure was pretested in two ways. Four persons were asked to read a base newspaper, and were administered the-survey the following day, with excellent results. Then, a small sample of military readers was surveyed for their readership of the front page of their base newspaper; there was no brief- ing of the respondents, and only a single exchange of 55 letters with the base information officer, who supervised the airbase pretest. Seventeen of the nineteen persons to whom the survey was administered completed the test adequately, although there were some minor variations in "readership" marking. Two respondents returned blank newspapers, but did not indicate whether they had seen any of the survey issue of the newspaper. Reading measurement was designed to indicate, for those items a respondent remembered having seen when he first examined the newspaper, approximately how much had been seen or read. This measure has not been proven sus- ceptible to interval scaling, so correlations were computed using the Spearman rank order correlation coefficient. This statistic has an approximate power index of .91 as compared with the most powerful correlational measure, the Pearson product-moment coefficient.l Reader "attention" is defined as reading the headline and/or the lead paragraph of an editorial item, or "seeing" a picture. Reader "interest" is defined as reading beyond the headline and/or lead paragraph of an item, or reading a picture outline. To obtain an index of reader "attention" the fre- quency of both attention and interest scores was summed for every reader within each category, and then expressed as a lSiegel, Sidney F., Nonparametric Statistics for the Sehavioral Sciences (New York: McGraw-Hill, 1956): pp. 15-23 0 56 series of percentages of the total number of items dis— played in each category. To obtain a reading interest index the frequency of interest scores was summed within each category and then expressed as a series of percentages of items in each category. This approximates an index of depth of reading. These percentages figures, summed over all readers of a category, and divided by the number of readers, pro- vide mean reading attention or interest scores for all readers of each category. REFERENCE GROUPS: There remained the problem of indexing reader orientation. A questionnaire was administered to determine: 1) the respondent's attitude toward his supervisors, 2) his attitude toward the USAF as a career, 5) his feeling of "belonging" to his unit, and 4) whether he felt he is given a fair chance to express his feelings about base policies which affect his welfare. Frequency of question- naire responses provided the data for a scale of orientation toward the Air Force, away from it, or "neutral." Ques- tionnaires were administered to all respondents present for the survey, whether they had read the surveyed issue of their base newspaper or not. HYPOTHESES: Hypotheses considered were: 1) more attention will be given to certain management messages by USAF-oriented 57 respondents than by other-oriented respondents, 2) more attention will be given to content perceived by coders as near-to-reader by USAF-oriented respondents than by other- oriented respondents, and 5) reading attention given to specified management messages will increase with the age, civil education, and military rank of respondents. The message-groups chosen were: 1) health and safety, 2) retention and training, and 5) editorials and commander's columns. As with all content coding for the present study, placement criteria were that the desired characteristic be present in the headline or lead paragraph of a news item, or prominent in the outline of a picture. THE SAMPLE: The sample of this study was devised and administered under the supervision of the USAF Directorate of Data Systems and Statistics, upon request of the Office of Information analysis branch. The statistical services officer at each of the eight bases selected for the survey was instructed to choose about 150 airmen and 150 officers for the sample, with a possible total of about 2400 respon- dents. The method of selection required choosing every app card from base personnel files, varying the choice interval according to the airman or officer base population. For each base the initial card was chosen by lot. A com- plete list of these instructions may be found in Appendix A. _ 58 Respondents for the survey totalled 1576; of that number, 1059 reported having read the surveyed issue of their base newspaper. The 852 persons reported as "not available for duty" at the time of the survey could not be tested for similarity to the group present for duty. There is, however, some indication of possible bias presented by the nature of the absentees. At Otis AFB, for instance, the statistical services office determined that while flying personnel comprised only 57.8 percent of the base p0pulation, they comprised 47.0 percent of the total absentees. This sample bias remained fairly constant among all the newspapers for which there were many absentees. The officer-airman per- centages at bases surveyed were close to the USAF popula- tion: the USAF distribution as of November, 1960, was 15.9 percent officers and warrant officers, while the bases surveyed were slightly under-represented by having an officer-warrant officer p0pulation of about 14.8 percent. The absence of flying personnel from the sample at three bases introduced further bias. Members of flying units have a higher civil education level than members of the USAF as a whole. In addition, the fact that officers of flying and medical units were least represented in the sample surveyed provided a bias in reading about certain units and topics. It was assumed that flying and medical .personnel will have more than the average reading interest .1n.hea1th and safety items, and will have more than the 59 average reading interest in news of the base units to which they are assigned. : At Otis AFB, again, flying officers accounted for 72.4 percent of the total officer absentees. Flying squadron airmen, however, accounted for only 55.5 percent of the total airman absences, perhaps reflecting the some- what lower percentage of enlisted personnel assigned to flying units. ADMINISTRATION OF THE SURVEY: Respondents at each base were requested to meet at central testing points under the supervision of statistical services personnel. McGuire AFB apparently administered the survey at each squadron, under the supervision of squadron personnel. Other bases apparently complied with the USAF instructions for survey administration. Each reapondent was provided with a survey kit: 1) a copy of the survey instructions for respondents, 2) a two-part questionnaire, 5) a c0py of the base news- paper issue being surveyed. Survey instructions requested completion of the data-gathering phase between eighteen and forty-eight hours after initial publication of the newspapers. The time limits were set to allow enough time for normal reading by base personnel, but to preclude inordinate amounts of forgetting. It is probable that some bases could not comply with the time requirements; most USAF base newspapers are 40 TABLE 6 READERS, NON-READERS, ABSENTEES AND UNSCORABLE RESPONDENTS INDICATED BY NEWSPAPERS Newspaper Total Absent or Non Surveyed Sample Unscorable‘ Readers Readers Mather AFB 291 074 075 156 McGuire AFB 209 066 O82 O61 Eglin AFB 512 085 074 155 Dover AFB 500 182 025 095 Otis AFB 505 129 057 159 Shaw AFB 502 024 108 170 Lockbourne AFB 500 O60 052 189 Barksdale AFB“ 504 142 066 096 ‘Fourteen questionnaires were so incomplete they were removed. ‘*Differs from number stated in letter of transmittal. published on Friday mornings, which necessitates a Saturday morning survey. Since many military organizations try to use only skeleton staffs on weekends, this may have caused further sample,bias. I have no record of exact times of survey administration, but received notice from one base that Saturday morning administration caused some of its rather large number of "non availables." Persons not reading the survey issue of their base newspaper were requested to complete the questionnaires, and turn in a newspaper marked "did not read this issue." The respondent instructions requested each reader to draw a line vertically, top toward bottom, of each 41 editorial item he remembered seeing upon first reading the newspaper. If none of an item was seen, there would be no mark drawn; if the headline and first paragraph were read, the line would be drawn that far; if more was read, the respondent would draw a line down as far as he remembered having read. When all newspaper-marking and questionnaire work was completed, the parts were stapled together, turned in, and mailed to the author. TABULATION AND CODING: As in most similar studies, the readership and questionnaire responses were coded onto International Business Machine (IBM) punch cards. The reading measurement device used here did not allow precoding, so a transcription method was devised. For the questionnaire answers, number values equivalent to the possible answers were assigned and then transferred to tabular sheets which corresponded to the eighty columns found on the IBM punch cards. Newspaper coding sheets were dispensed with for this study. Instead, each newspaper editorial item was serially numbered on master—c0pies of each paper; then the reading score for each item in each newspaper was marked on the face of the item with a felt-tipped pen. A "l" was entered if the respondent had marked the headline and/or lead paragraph, or pictorial matter alone. A "2" was entered if the respondent had marked more than the headline and/or 42 lead paragraph, or the outline for a picture. -In the case of humorous cartoons and the like, any line was automatically interpreted as a "2." Following this, tabulators transcribed the reading scores onto tabular sheets for later use in key-punching Operations. All questionnaire responses were also trans- cribed to tabular sheets after being coded with the felt- tipped pen for high visibility. Individual scores for reading attention or interest in each content category were obtained by use of a specially develOped program in the "Mistic" electronic computer at Michigan State University. The percentage reading scores, at attention and interest levels, were then punched onto each respondent's IBM master data cards. Since all editorial items in a newspaper were serially numbered, and were tabulated in like order, the items for a given content category were widely scattered over one or two IBM cards. This necessitated the prepara- tion of special "input" decks of cards, gathering all item scores for a given category into one "field" which could be used in computing percent read of all items displayed in a category. ANALYSIS OF CORRELATION AND DIFFERENCE: To obtain correlations from the data punched onto IBM cards, mean reading scores for re3pondent groups within single newspapers were combined. Mean rank-orders were 45 computed, and the rankings tested for correlation with the amounts displayed, Similarly, the mean reading scores served as cutting points for construction of chi-square tables to test for significant differences in reading behavior by subgroups of the sample p0pulation. Cards were sorted into groups based on reference-group orientation, age, civil education, and military rank; the number of respondents above and below the mean was determined, and the chi square tests applied. GUTTMANN SCALING FOR REFERENCE-GROUP ORIENTATION: This scaling technique demands that a large propor- tion of the respondents answer scale items consistently. Essentially it consists of finding a set of questionnaire responses in which:1 Endorsement of a given item is accompanied by endorsement of all items that are less extreme and rejection of all items that are more ex- treme. When the set of items is consistent enough in this way the scale can be assumed to be unidimensional. It then should provide a reliable measure of response to the relationship it describes. The relative consistency is indicated by a computed coefficient of reproducibility. For the present study a set of scale items was chosen from the questionnaire administered to all respon- l"A Machine Method for Scaling as Many as Twelve Dichotomies." 44 dents. The items used were: 1. Do you think most supervisors are really interested in your ideas and suggestions? 2. Do you feel that you are a real part of your squadron or wing? 5. Generally Speaking, how do you feel about the Air Force as a career? 4. Do you feel that you are given a fair chance to express your feelings on plans - and policies which affect you here on the base? These items formed a scale with a coefficient of reproducibility of .8862, slightly lower than expected, but serviceable. Respondents were trichotomized into groups labelled as high, neutral, or low in USAF orientation, based on their scores in this scale. The high USAF orienta- tion group numbered 287 and consisted of those persons who made the desired responses to at least three of the four questions. The neutral group numbered 810 and consisted of persons who made one or two of the desired responses. The group assumed to be low in USAF orientation numbered 584, and consisted of persons who answered none of the questions in the desired manner. The three groups number less than the sample (1576) because any respondent not providing pppp response to all these questions was eliminated from the scale construction. It was felt that persons scoring high on this scale would be more strongly USAF oriented than those who scored low. Thus the central hypothesis, that USAF-oriented persons will tend to read more of certain management 45 messages, could be tested. Rank order correlations and the chi square were the statistics used to test the sample population along the orientation continuum. The USAF-oriented respondent, generally, said he: 1) will remain in the USAF as long as possible, 2) believes that most supervisors are sincerely interested in his ideas and suggestions, 5) feels he is a real part of his USAF unit, and 4) thinks he receives full opportunity to express his feelings about plans and policies which affect him on the base. The other-oriented respondents are rather skeptical about their supervisors' sincereity, tend to feel that "I just work here," say they will retire as soon as pos- sible--if they stay until retirement at all, and do not feel that management listens to, or answers, their sugges- tions and complaints about base policies. The middle or "neutral" group of respondents is likely to fall anywhere between the two more extreme views. For purposes of this study it was desirable to separate definitely, the polar orientations, hence their relatively small Np as compared with the middle group. CHAPTER III Air Force base newspapers are read about as well as the average civilian daily newspaper, and somewhat less than civilian weeklies. Military readers of base newspapers who remember seeing an item at all are more likely to remember having read a substantial portion of the item. And the fact that an item is "local" does not guarantee it high readership by the military audience. The first difference is so pronounced that all readership percentages in this study were based on persons who "looked at or read" the items scored: there were almost no respondents who admitted to reading just the headlines. The second difference, the non-local material was read more than local, has implications for military management. Management-centered communications fared very well in the Air Force base newspapers surveyed, bearing out the contentions of researchers in business and industrial papers and magazines.1 The variables of age, education, and military rank affect reading behavior about as expected--with one notable exception--but reference group orientation appears to have much less effect upon the reading of management communica- tions than any other of the variables examined. 1McCloskey, Industrial Journalism Todgy. 47 SUBJECT-MATTER: Information concerning new units, facilities, and equipment on a base was the best-read of all subject-matter, although it ranked eleventh in percent (5.9) of available space occupied. Least-read categories concerned: 1) education, culture, and religion, 2) dependents' affairs, and 5) sports, recreation and hobbies. As the latter two of these may be said to fall in "immediate reward" categories, the Schramm-White findings about newspaper reading and the vari- able of personal impact of news upon the reader are not entirely supported by the present study.1 It is possible that inclusion of religious material with news of educa- tional programs and Opportunities and the like may have served to depress the reported readership of educational material. Likewise, the readership of miscellany, which in this study includes jokes and cartoons, may be biased by the inclusion of filler items--the odd facts and historical references used to fill small bits of editorial space. This study showed such material ranking no higher than seventh for any major respondent group. Health and safety items were read more than items in five of the other thirteen subject-matter categories. Such news was most highly read by reSpondents who have only a grade school education (52.7 percent); only community 1Schramm and White, pp. 71-75. 48 relations news was read more (54.4 percent) by this group. Retention and training news was read as much or more than eight other subject-matter categories. The exceptions to this ranking included: 1) grade-school- educated respondents, who ranked this category as 9.5 among thirteen, 2) college-educated respondents, who read seven categories more than retention and training. 5) officers as a group, and 4) officers considered by reference-group orientation. Both of these last two groups gave eighth ranking to this news category. The mean readership for this subject-matter category was thirty-five percent. Reading of retention and training information varied with education more than with other variables, but the variance was not significant at the p.05 level (x2=8.55). Editorials and commanders' columns ranked fifth in over-all readership, varying from first ranking by respon- dents in the "28-55" age group to eighth ranking by those in the "less than 22" age group. Respondents who are less than twenty-two years old read 25.4 percent of the editorials category, while those over forty read 45.5 percent, sup- porting one of the Schramm-White findings.1 Reading of editorials varied most with reader age, but the variance was not significant at the p.05 level (x2=8.11). The category "The Big Picture of USAF and Department 49 of Defense," including policy and interpretive news, was ranked third or fourth in readership by most respondent groups. This interest in the broader aspects of the mili- tary organization, paralleling Barlow's findings,1 may be seen again in considering the content of location-of-interest categories. LOCATION-OF-INTEREST: News of the command to which a base is assigned was highly read by all respondent groups, although it ranked fourth in percent of space occupied. Information about the Air Force in general, without reference to a reader's own base, was the next most highly read, varying from second rank to fourth.‘ Department of Defense information and news of the home base alternated in holding third and fourth readership rankings. The location-of-interest cate- gory termed "other"--often an undefinable center of empha- sis--was accorded fifth ranking by virtually all reader groups. There were so few items about adjacent communities that the bottom rank of this location-of-interest category may not be justified. The only groups, however, who reported reading more than twenty percent of this category were: 1) respondents with a grade school education, and 2) those at least forty years old. Virtually all respondent groups read at least 55.0 percent of the command news category, with a range from 1"In Plant Communication; A Seminar Report," p. 68. 50 51.0 percent to 45.9 percent. The most local category, "This Base," was read a maximum of 55.5 percent by "neutral" oriented officers, with an average of 28.9 percent. ORIENTATION: Sixteen of the twenty-six possible response combi- nations to subject-matter categories indicated a higher readership percentage for "neutral" oriented respondents than for either extreme of orientation. Among the airmen, none of the management message categories focused on by the present study was included, but "neutral" officers read more of all three than did the officers rated as USAF-oriented or other-oriented: 1) health and safety, 2) retention and training, and 5) editorials and com- manders' columns. The coefficient of reproducibility obtained in pro- ducing a Guttmann scale from questions used to index orientation (CR = .8862) is the highest obtainable with the questions asked, and appears to encompass the logically correct questions. Predictability was not increased by reference to single questions about respondent attitudes toward the Air Force, their supervisors, their units, and their Opportunities to speak and receive answers about base-policies. Considering just the two extreme orientation groups USAF-oriented persons read more than other-oriented persons in most subject-matter categories. Relative readership of 51 omH.+ as ..u :,c.. m ng .um oem.+ue mmo.and. 800. n.“ asses meHHmHe one mxdmn maflcmoh Mo msoHpsamHHoc s m.l. l.w. s H.He p o.um HH w.rH o m.sm a 0.0m s s.oH_mop5pmom HpsounoA w hqmaamomwm m o.Hs m n.ns s H.nm m m.mm a o.mm s o.mm s w.sm m o.sH oncpoHH me one mH «.mH r h.mc m m.mm oH m.mm oH m.om m m.mm m m.mm NH m.m msoHpchm mpwqdaSOr m o.Hm H n.ns m o.Sm a «.SN m s.mm m m.mm m H.sm o o.m unoproH s .mddg nHoo .mHanopHem o m.sm o m.oe o m.Hm n m.mm o m.mm m m.nm m w.mm a o.m soHanmooom a noapoaonm oH m.sm 0H s.om oH H.sm a m.sm m H.mm HH m.mm m.0H o.mm m m.sH moHnnom s soH umonoom .mpnod mH o.MH MH c.0m mH r.Hm mH m.mH mH o.mH MH m.mH mH o.nH n H.oH msHsHH manoesmmo HH c.sH . m.md mH 5.0m mH o.wH mH o.MH NH H.0H NH m.mH w r.m sonHHom s .ows Iaso .soflpmose m n.3m HA I... H. «.mm HH 0.0m a o.mm OH m.mm m.oH o.mm mH ¢.H mnsHo mespHHHz m n.s A o.ms m m.mm H m.mm m n.Hm m m.mm m m.om H o.mH asosomsss: one m.n m.« , s.mm a c.0m s s.nm H s.ms m H.mm m o.mm m m.s wstHmw a GoHpGOpo H g. . u o.ms H s.sm m m.mm m m.mm H m.mm H m.Hs HH o.m psosmHssm a noH hHHHosm .mpHs: so m.e m.um s.m_ w n.0e a m.sm m m.:m a H.mm m m.mm OH s.s mpoHsm.s spHso scam .. Exam : scan a seem 3 Adam a seem .a seem a msHman thmmmm nonomopso mmsnzu Homanv HHnmuzu dsHmuzu HHmuzv dowNAZV HHmmuzq an Adam psoonom woppoZIHoonnsm sonpo Hsgppoz ssh: nospo HanSom ma.p whoesom WWHduzu mHOUfiHMO Ammunmw umahwd Had mmbomwmbm ZOHE< mo mo< mo mo¢ezmomma mo zomedmzoo m mflmda 54 AGE: Readership of retention and training news, base management, education, culture and religion, sport and recreation, and miscellany was not significantly affected by respondent age-group. Readership of such subject-matter categories as: 1) health and safety, 2) new units, facili- ties and equipment, 5) dependents' affairs, 4) promotion and recognition, 5) editorials and columns, 6) community relations, and 7) the Big Picture did increase with respon- dent age, as seen in Tables 9 and 10. When the age of respondents was considered, reader- ship of management messages was no less than 25.0 percent in any case, nor greater than 47.0 percent. It will be seen in Table 10 that, while readership of health and safety news increased gradually as age increased, readership of retention and training news apparently was not affected by age, hovering close to 55.0 percent for all age groups. Editorials and columns, however, Showed a constant and fairly large increase from each age group to the next older. Readership of location-of-interest categories was ranked similarly when equated with respondent age. Gener- ally, readership increased with respondent age, although by as little as 2.1 percent in the case of command news. The percentages and ranks may be seen in Table 11. EDUCATION: Previous research has indicated that attention to the mass media tends to increase with greater education. -55.. mmo.+u.M mm sue mso.luu Hmm.+uu smH.+uu manta sanmHe ens mxdwn mqwemon Ho mQOprHounoo m m.wm m s.mm oH m.sm o H.mm m. n.mm s s.oH meanness quommom w mumHHoomflL s H.mm m s.nm m s.¢m m H.mm s m.Hm m o.sH onspon mHm one m m.mm a m.Hm m m.mm oH s.mm m s.mm mH m.m msoHpsHom spHssssoo m m.ms m o.os H o.om m p.0m m s.mm o o.m unoppoH e .msssHoo .mHusopHem n m.sm m m.om m s.mm a m.mm m m.um n o.m aoHoHsmooom s soHpoSowm HH m.nm HH m.mm HH o.mm m o.mm HH H.mm m m.sH moHnnom w .HOHpmopoom .mpnomm mH o.mm NH m.om mH m.mH WH o.eH mH H.mH m H.oH ansHHH .npsoosonon MH s.mm mH H.mH mH s.nH HH H.mH mH s.mH m m.m donHHom a .OHSpaso .Qofipmoscm oH 0.6m oH m.mm m m.mm NH m.mH oH m.mm MH s.H mnnHo anspHHH: m s.mn m s.om m.m m.mm m m.sm H 0.5m H o.mH psosomsss: omsm o o.mm s w.om m.m m.mm s m.mm m o.mm m c.s mcHsHsna s soHrsopom H m.ms H o.ns m n.mm H m.ms m m.sn HH o.m psosHHssm a , .moHpHHHosH .mpHsa 3oz m m.Hm o m.Hm a m.mmw m. mm m. mm oH s.s Summon s ansom Harm Hasm . seem a. seem Adam .HsHmnHH msHmnHm monomopso oHuz o Huz «womnzv H owwqu HsmHu z.W an scam psoonom noprszupoonnsm no>o the os on u am mm 1 mm mm n mm mm can» umaH mmbomwmbm mw< BZMonmmmm Mme Hm mHmmmmndmm mo mu< mo mwdszmommm mo zomHmdmzoo m mqm¢a 56 H.0n c.3e s.sm H.mm m.Hm . onspoHH mHm one m.ms o.o¢ o.mm 6.0m s.nm unopsoH s .nsssHoo .mHanopHsm m.sm m.om s.mm m.mm m.mm soHpHdmooom s soHnoSonm 4.0m . s.om m.mm m.sa . o.sm psoSomsssz omsm o.mm m.mm m.mm m.mm o.nm msHsHsna a soHpsopom m.ms o.as s.mm m.ms m.sn .psoansem a moHpHHHosH .nann 3oz we.HH sm.Hm am.mm am.sm A am.mm Spousm s ansom W . A _ Hmmanv omHqu Hmomuzv Hommuzv HsoHuzv moHnomopso nose eds or mmnsm mmumm nmnmm mm sane muoH nonsenspoomnsm mmHmommado mmaaon< monom HHH no @cmaaoo HHd m o.Hm w m.¢H o m.HH @ m.mH w m.nH m m.m mqudaaoo psoomncd w ®.mm s m.Hm m o.®m a ¢.®m m m.mm H m.m® mmmm mHna Adam a scam m seem .a Hate a Adam a msHanH ssHsmHn uoHnowopso INmmHuzq o Huz dmomuzv mom «2 dmeuzq hp Harm psoonom pmohouqHImoIaOHpmooH Hm>o can or on I #m mm I mm um I mm mm nag» mmmq mmbomdmsm mw¢ Bzmnzommmm m>Hm Hm mHmwme«mm ho modszmommm z< mo muHow humpHHflL N N.Os m s.Om m O.sm N O.mm m s.Hm H 0.0H Osceomsss: onsm O m.Om a m.Om N 0.0m H O.Om m.m m.mN m O.s msHsHsn% w QOHpquo H O.Hm H N.Ns H H.am O H.Om HH N.NN HH O.m psosNHaem a .moHpHH IHomm .mpHQD 3o 5 O.Hm a H.Om O O.NN u H.mN N O.Nm OH s.s -mpousm a ansom seem a seem a seem a scam a scam a asHHuHH ssHHmHO moHsomopsO HNNNuzO OHNNuzv HmNsuz OMHHuzO HONAZH as Home scooped Hepnsznsoomnsn amcHHoO HHH omsHHoO osom Hoonom n Hm mosses NHIO Hoosom oesnO wmbomwmbm ZOHBHm Hm mHmmmmndmm mo mc< mo modazmommm MO zomHmdmzoo NH mqmon< monom HHN no OquSoo HHH O 0.0H O m.OH O m.mH O m.o N m.Om O m.N SpHdns Iaoo amoomwOH m N.Nm m . H.Om m s.ON m 0.0N m O.nN H m.NO mmmm mHna qum N qum L& Harm & Harm & stm .& SOHmmHQ hmHmqu OOHHomopwo ANNNuzw AHmanM AmNSAZ AmHHuzu NONuzv hp qum pnoonom pmonopsH omoHHoO HHS oOoHHoO seem Hoonom n Hm mousse NHIO Hoonom oost IHOInoHpmooH mmbomommm ZOHBHm Mm MHmmmmn< ho m¢< mo madazmommm ho zomHmone monom HH< no oqmaaoo HH< O 0.0H O O.NH O 0.0H O O.N OpHsSSSoO pseosnea O 0.00 O 0.0N s 0.0N H O.NO seem ana assm a scam a scam a santh ssHamHn moHnoOepsO OOHsuzv HOHOnzO «ONOHuZV an scam espouse pmsnopsHueoIdoHpsooH OHOOHHHO nman¢ mnmcwom HHN mmMOHhho 924 zmzmH< Hm mHmmmmndmm mo mo< mo m¢ot Checking ATC House Organs. Air Training Command, 0. . SAC Editor's File. Strategic Air Command Press Service, July, I965. . ADC Regulation 190-4. Air Defense Command, August, I960. . Anal sis pf the Air Force Information Program. September, I960. UNPUBLISHED MATERIAL Freyschlag, Kedrick G. "Categories for Analyzing News in Terms of Reader-Interest." Unpublished Master's thesis, Stanford University, 1955. Reply to Attn. of: Subject: To: APPENDIX A DEPARTMENT OF THE AIR FORCE Headquarters United State Air Force Washington 25, D.C. AFASC—5P8 25 Mar 1961 Survey of Base Newspaper Readership TAC SAC MATS ADC ATC ARDC (Statistical Services) 1. The Office of Information, Office, Secretary of the Air Force is interested in determining items read by officers and airmen in base news- papers. This is part of the Air Force's program to increase the effectiveness of base newspapers as a tool of command management. To obtain the readership data, the Office of Information re- quests support for a research study being accom- plished by an Air Force graduate student at Michigan State University. 2. Statistical Services Officers at the bases noted below will survey approximately 150 officers and 150 airmen. a. Barksdale AFB 0. Shaw AFB b. Dover AFB- f. Mather AFB c. Eglin AFB g. McGuire AFB d. Lockbourne AFB h. Otis AFB A random sample will be obtained by selecting from officer and airman files as follows: Base Total Officers Selection 800-899 or less Every 6th card starting with 4th (4, 10, 16, etc.) 900-1099 Every 6th card starting with 5th (59 ll, 17, etc.) 1100-1199 Every 8th card starting with 7th (7, 15, 25, etc.) 1200-1299 Every 8th card starting with 2nd (2, 10, 18, etc.) 1500—1599 Every 9th card starting with 5th (5, 14, 25, etc.) 78 Base Total Officers Selection 1600 or more Every 11th card starting with lst (1, 12, 25, etc.) Airmen Selection 5500-4999 or less Every 24tH card starting with 16th (16, 40, 64, etc.) 5000-5499 Every 54th card starting with 17th (17, 51, 85, etc.) 5500-5999 Every 57th card startin with 28th (28, 65, 102, etc.§ 6000-6499 Every 40th card starting with 55th (55. 75. 115, etc.) 6500 or more Every 44th card startin with 24th (24, 68, 112, etc.§ 5. The survey will be accomplished within the period 7-10 April 1961. Respondents will be provided with a survey package within a period 18-24 hours after the most recent issue has been published. The survey package for each respon- dent-will include a COpy of the issue of the base newspaper being studied, a copy of the test in- structions, and the two part questionnaire. 4. A zerox copy of the test instructions and questionnaire are attached for reproduction purposes. 5. Base Statistical Services Project Officers will notify the Base Information Officer immedi- ately on receiving this instruction of the survey requirement for 525 copies of the base newspaper published on or about 7 April 1961. 6. Respondents will be called together in groups under the supervision of the Base Statistical Services Project Officer. 7. Respondents who report they have not read the issue being surveyed should write "not read" on the newspaper BUT still fill out the question- naire. 8. Respondents who are absent from duty within the 18—48 hour survey period will have "not available for duty” written across their copy of the base newspaper by the Statistical Services Project Officer. The blank questionnaire will also be attached to the base newspaper. 9. The survey package-~150 for airmen and 150 for officers-~will be mailed with five unmarked 79 copies of the base newspaper to the following address no later than 12 April 1961: Major Donald MacDonald 514 Lexington Street East Lansing, Michigan 10. Acknowledge receipt of this material by mail. Upon completion of project, an information- al copy of transmittal should be forwarded to Directorate of Data Systems and Statistics. AFASC-5P8, Attn: Mr. Mills, Washington, 25, D.C. 11. This requirement is exempt from the assign- ment of an RCS (Reports Control Symbol) per para- graph 7c(5), AFR 174-1. FOR THE CHIEF OF STAFF /s/ Vasco E. Gulino VASCO E. GULINO, LT. COL, USAF 2 Atch Chief, Personnel Branch 1. Offset master & Directorate of Data Systems Printed cy and Stat (Base only) 2. Printed cy (l) (Com'd only) APPE JVDIX B WHAT DID YOU LOOK AT IN THIS WEEK‘S BASE NEWSPAPER Your base newspaper aims to provide you and your family with information about the Air Force and your Air Force neighbors which you don't receive any other way. To meet your needs and interests--the editors must know what you read: You are one of a small group of representative read- ers I am asking to help me, by showing what items interested you when you first looked at this issue of your base news- paper. By showing this interest now, you may help to provide all editors a better way to serve you in the future. I assume you have by now looked through your regular 00py of this issue about the way you usually do. If so, please follow the instructions below without readingt ’He paper a ain, pp without reading more pf_pp' 1. Look at every item on every page (except adver- tisements). Include news stories, editorials, columns, feature articles and pictures. Start with the front page and work through to the back, item by item. 2. If you don't remember having looked at or read the item at all, leave it unmarked and proceed to another item. 5. If you do remember having looked at or read the item, start at the pppiT and draw a line down the middle,_ to thet 01nt where_ on fiooke do or read previousI . is wiW asoft penc1lL or 5a II-p01nt_ pen. 4. Don't read a story to find out if you have al- ready seen it: Mark only the items which you ulooked at or read pp first seeing this issue p: the a er. 5. If you don't remember having looked at or read anything on a page, mark "DID NOT READ" where that applies, please complete the two short questionnaires attached to this page. This will not identi_y p23 personally, but is needed if I am to tell what kinds 0 eo 1e are interestEd in wEatT‘ Then T313 tfieh newspaper once, inse5t_the question- naires into it, and hand both to the person monitoring this readership test. Thanks for your help. You are cooperating in an im- portant effort to improve your base newspaper so you will have the most interesting reading. SURVEY PACKET--PAGE 1 OF 5 ' 81 THIS IS ONE OF THE TWO QUESTIONNAIRES. YOUR ANSWERS WILL NOT IDENTIFY YOU BY SQUADRON.OR NAME. THEY WILL BE USED TO HELP US DETERMINE "WHO READS WHAT.” IF YOU HAVE TROUBLE UNDERSTANDING ANY QUESTION, ASK THE MONITOR FOR HELP. THERE ARE NO TRICK QUESTIONS--AND NO RIGHT ANSWERS EXCEPT THE FACTS. All questions require ONLY ONE reply: *****#*********************¥**# 1. Age at last birthday: 2. Years of active duty completed: 3. Marital status: Married ; Single ; Other 4. Residence: a. Barracks or BOQ: b. Base apartment or house: c. Off-base apartment or house: 5. Highest civilian education completed: (mark only one) a. Grade school (8th grade or less): b. Some high school: c. High school graduate: d. Some college: e. College graduate (or more): 6. How many on—base clubs and organizations do you belong to: a. None: b. Less than three: 0. More than three: (including three) 7. How many civilian clubs and organizations do you belong o: a. None: b. Less than three: 0. More than three: (including three) 8. Your present rank: THERE IS ONE MORE QUESTIONNAIRE: BE SURE YOU HAVE MARKED ANSWERS FOR EACH OF THE QUESTIONS ABOVE. INCOMPLETE RETURNS MAKE IT VERY HARD FOR US TO ANALYZE THE FIGURES. gOW TURN TO THE LAST PART OF THE TEST, WHICH HAS THREE AGES. SURVEY PACKET - PAGE 2 CF 5 82 THIS PART OF THE TEST ASKS A FEW QUESTIONS ABOUT YOUR FEEL- INGS AND HABITS. YOUR ANSWERS WILL IN NO WAY IDENTIFY YOU TO ANYONE IN THE AIR FORCE-~AND THERE ARE NO_RIGHT ANSWERS! PLEASE ANSWER FRANKLY, SO WE CAN "DRAW A BETTER PICTURE" OF THOSE WHO READ BASE NEWSPAPERS. FOR EACH qUESTION, PLACE AN "x" BEFORE ONE CHOICE-93 FILL THE BTANK'TRF—VID'ED" "— ***#*##****t*****#************ 1. Generally speaking, where do you really get the low- down on what is going on at your base? Fellow workers. My supervisors. Base Newspaper. Ill Commander's Call. Other (what?) 2. Do you think most supervisors are really interested in your ideas and suggestions? Yes, definitely. They have some interest. They have little or no interest. I have not thought about this, either way. 5. Do you feel you are a real part of your squadron or wing? I feel I really belong. I feel that I just work here. Sometimes I feel one way, and sometimes the other. 4. Generally, are you interested in base newspaper editor- ials and the commander's column? Yes, I am very interested. I usually read them, but they don't help with my problems. I seldom read them, unless my friends are talking about them. I am not interested in such items. SURVEY PACKET — PAGE B'OF 5 85 Generally Speaking, how do you feel about the Air Force as a career? I plan to stay as long as I can. I will probably retire as soon as possible. I will stay - providing I am kept in the right job/training area. I probably won't stay until retirement. I plan to get out as soon as possible. Fellow workers. My supervisors. Air Force people assigned to counsel about these things. Civilian friends. Other (Who?) Do you feel you are given a fair chance to express your feelings on plans and policies which affect you here on the base? No, I get no chance to do this. I can speak, but they don't pay much attention. I am not qualified to speak about such things. I get some chances, but not enough. I get full opportunity to be heard--and answered. Do you think your base newspaper does a good job in keeping you informed about base plans and policies? No, they never tell us anything. They give us some information, but don't make it very clear. The newspaper does a good job on this. I'm not interested in this kind of information. Are you a supervisor? Yes No SURVEY PACKET - PAGE 4 OF 5 10. 11. THIS 84 Do you read your base newspaper? Which Seldom or not at all. Yes, fairly regularly, and the type of article I find most interesting concerns: AF wide and command policies and activities. Base policies and activities. News about AF personnel on the base. News about the civilian community. Squadron items. Entertainment, sports, and hobbies. Information of interest to dependents (Social, schools, chapel, scouts) Advertisements. Want Ads. of the following comes closest to telling you how to make your present base newspaper more useful to you personally? Have a base newspaper, but I don't read it regularly enough to give you an opinion. More international news. More USAF news (e.g., Minutemen, Counterforce, personnel). More major command news (Policies, activities). More stories on base personnel, human interest. More Sports and recreation news. Less stories about the brass. Less social news. More articles of concern to dependents (schools, chapels, social, scouts). Less advertising Less sports and recreation. IS THE END OF THE TEST. THANK YOU FOR YOUR HELP. TURN IN ALL PAPERS. SURVEY PACKET - PAGE 5 OF 5 1. APPENDIX 9 CONTENT CATEGORIES ARRANGED BY SUBJECT—MATTER OF ITEMS Health and Safety: a. b. C. d. e. f. Air safety, safety discipline, traffic control, reporting. Ground safety, including auto, home, shop, fire, electrical, taxiing,-weather. Accident reports, on ground, in air, driving, home, shop, athletic, and freak. Physical fitness, fatigue, weather cautions, use of facilities such as steam room, weight require- ments, fitness tests. Immunization, pet shots. Medical cautions, self-care, drugs for fliers, fatigue, food cautions, weather care. Facilities, units, and equipment at this base: a. b. c. d. e. f. g. h. 1. New or remodelled buildings in progress or planned. Aerodrome expansion or modernization. Meteorological equipment installation or moderni- zation. Alert facility development. New aircraft, major overhaul, or modification. New construction or maintenance machinery or equipment. Mobile training detachment arrivals, departures, or enlargements. Flying squads of mechanical, inspection, or clerical specialists. Activation, deactivation, or reactivation of units. Retention and Training, applied to this base: a. b. c. d. e. f. g. h. i. J. k. Enlistment appeals. Enlistment campaign or progress reports. Requirements for enlisting or re—enlisting. On—the-Job-training (OJT)- Upgrading requirements. Frozen specialist categories. Retirement rules. Promotion rules. Officer retention news and appeals. Formal USAF schools, officer training school, academics and preparatory schools, including base level, clerical and management schools. Proficiency pay. Base Management, at this base: a. b. Facilities use. Economy programs. 5. c. d. e. f. h. i. k. 1. U1. no 86 Uniform requirements and availability. Courtesy campaigns Military discipline. Vehicle use and care. Courts martial, AWOL, man-days lost from work, discrepancy reports. Pay and personnel policies at the base, new administrative procedures. Commanders' call. Internal open houses. New or shifted key personnel. Base personnel assignment policies. Processing on or off base. Base unit activities, to include word or picture- stories on shops, offices, sections, and visiting VIP's. Schedules of: Base exchange. Commissary. Cafeteria and snack bars. Working hours. Uniform wear. Service, NCO, and officers' clubs. Sick call. Theaters. 9) Chapels and religious organizations. 10) Busses. CDQOWW-PW [UH VVWWW Military Service Clubs: 3. b. C. d. e. 'fo 8. ho Features on coming events Elections. Personnel. Membership and eligibility. Formation or dissolution. Purchases or decoration. Reports of social events already held. Reports of official or quasi-official events held. Education, culture, and religion: a. b. c. d. e. f. g. h. i. j. k. Midnight oil. Bootstrap. Institute of Technology. Extension Course Institute. USAFI Scholarships. Qualification examinations. Off-duty night classes. Chapel affairs relating to military personnel, aside from schedules and meeting announcements. Story or essay contests. Library and bookshelf news. 87 Dependents' Affairs: 8. b. c. d. e. f. g. h. i. J. k. l. m. n. o. p. School rules, eligibility, curricula, costs, tran- sportation, lunches, honors. Wives' clubs. Youth clubs. WiVes' athletics and games. Youth athletics and games. Boy and Girl scouts, CAP, etc. Vacation plans and reports. Dependents' Assistance PrOgram news. ' Red Cross activities and awards: includes Grey Ladies. Food news. Teas, coffees, cocktail parties, luncheons. Personals. Columns by or about dependents. Dependents' activities in church groups. Tours. I Cultural and/or recreational classes. Recreation, Sports, Amusements, and Hobbies: 3. b. c. d. e. f. g. h. i. J. k. 1. Crafts and craft shOps. Athletics; including all peeple except dependents. Special interest clubs. Variety and dramatic shows. Shooting programs. Columns about athletics, and about field sports. Travel news. Hobbies in the news. Model airplane contests and displays. Private flying, in and out of Aero clubs. USO activities for base service personnel. Auto repair, collecting, racing. Promotion and Recognition at this base: a. b. PeOple promoted. Monthly, quarterly, or annual awards for: 1 Air crew. 2 Ground crew. 5 Crew chief. 4 Airman. 6 Squadron. Local safety awards. Local "heads up" awards. Cash, bond, or pass awards to military or civilians. Educational achievement awards to military. Civilian awards to military, and vice versa. 10. ll. 12. 150 88 Editorials, Commanders' columns, Chaplains' messages, and a. b. C. d. guest editorials. Commanders' column. Chaplains' messages. Letters from senior commanders. Editorials, local and guest. Community relations: a. b. C. do 90 f. g. ho 1. Jo k. 10 Parades and festivals. Charity and community fund drives. Open houses. Kids . days 0 Expositions. Police and fire cooperation. Messages from civic officials. Guest tours. Business-industry-education days. Fire fighting assistance. Civil defense. Participation in such organizations as: 1) Service clubs. 2) Lodges and orders. 5) Toastmasters (not base clubs). 4) SPBSQSA, etc. Leaders in youth activities (not on base). Special displays. Church activities. Members of civic committees. Noise—abatement council. Big Picture: Army or Navy news. Weapons systems. Satellites and space systems. Registration and voting procedures. Bonuses. Retirement, and retirement employment. Department of Defense and USAF personnel policies in general terms, not applied to this base. h. Communism explanations and background material. Miscellany: a. Cartoons and pin-ups. b. 00 do e. f. Bond slogans. Fillers about odd facts, including military history. Features not codable above. Personals about military peOple, not codable above. Local retirement ceremonies. APPENDIX 2 RESPONDENT REPLIES TO USAF-SPONSORED QUESTION INCLUDED IN SURVEY PACKET: DO YOU READ YOUR BASE NEWSPAPER? 222 Seldom or not al all. Yes, fairly regularly, and the type of article I find most interesting concerns: 600 AF wide and command policies and activities. 422 Base policies and activities. 122 News about AF personnel on the base. 021 News about the civilian community. 146 Squadron items. 182 Entertainment, sports and hobbies. O6§ Information of interest to dependents (Social, schools, chapel, scouts, etc.) OQQ Advertisements. 010 Want Ads. The 1576 respondents made 2016 selections in their answers to this question. IBM master cards indicate only the initial choice of each respondent; totals above include all other choices as well. The modal response by those who report reading their base newspaper is "AF" wide and command policies and activities," while news of base policies and activities is chosen next most often. The question asking respondent choices for material making base newspapers more personally useful (see Appendix E) reverses this ranking. The two questions do not elicit choices in a parallel 9O manner, so direct comparison is difficult. The findings of the study tend to support the choice of USAF policies and activities, with lesser but significant interest in items of local application. APPENDIX 9; RESPONDENT REPLIES TO USAF-SPONSORED QUESTION INCLUDED IN SURVEY PACKET: WHICH OF THE FOLLOWING COMES CLOSEST TO TELLING HOW TO MAKE YOUR PRESENT BASE NEWSPAPER MORE USE- FUL TO YOU PERSONALLY? 222 Have a base newspaper, but I don't read it regularly enough to give you an opinion. 126 More international news. 282 More USAF news (e.g., Minuteman, Counterforce, personnel). 126 More major command news (policies, activities). 222 More base news (policies, activities). 202 More stories on base personnel, human interest. 088 More sports and recreation news. 022 Less stories about the brass. 022 Less social news. 026 More articles of concern to dependents (schools, chapel, social, scouts). 022 Less advertising. 022 Less sports and recreation. The 1576 respondents made 1812 selections in their answers to this question. Again, IBM master cares indicate only the initial choice of each respondent, while the totals here include all choices made. Here the modal response by those who report reading their base newspaper is "more base news," with "more USAF news" in second position, and "more major command news" in fourth position by number of choices. 92 The difficulty in analyzing by comparison can be seen in the different wordings used in the two USAF-sponsored questions. In the first, a respondent must choose command news if he desires to indicate USAF news, while these are made separate choices in the second question. In view of this difficulty, and of the relatively large number of multiple choices made to both questions, the following suggestion is made. Respondents apparently do not feel one choice suffices; they might be allowed a three-way choice for 2229 category in such question as these. The choices would be to indicate reading behavior, or desire for reading material, at low, medium, or high levels for each category.l' In such a case, the latter question considered here might be presented in somewhat the following way: FOR EACH ITEM LISTED BELOW, WHICH COLUMN COMES CLOSEST TO TELLING HOW TO MAKE YOUR BASE NEWSPAPER MORE USEFUL TO YOU PERSONALLY? I don't read it regularly, and have no choice Want All Right Want More Now Less International news. USAF news (minuteman, Counterforce, etc.). ' Major command news (policies, aCtiVitieS)o 1National Industrial Conference Board, Report No. 156. 93 Base news (policies, activities). Stories on base personnel, human interest. Sports and recreation news. Stories about the brass. Social news. Articles of concern-to dependents (school, chapel, social, scouts). Want More All Right Now Want Less Further, since the present study has noted that many known non—readers report believing base newspapers do a good job and editorials interesting and useful, there is support for the hypothesis that anonymity must be assured even more than is guaranteed here, if completely frank replies are expected. In other words, there is a pronounced "halo" effect associated with answering questions seen as socially acceptable. U :‘i L1 R001“ U LIL ROOM U ‘5 ONLY lMAM“