.V o ' ‘§J_D N 'u . .C."..~o. ‘ . . .. '00. , a;.‘.9—‘IIM-N|L'o'olo'g'. .- f.‘ ' °.... '. . co. . . o . n -*‘f¢!lv— i a; . - ‘ ‘ . 'I 0.3" ‘f'0 O... .60. .‘ '..'H ‘..10! DA.. _ I . . .N‘ O a "‘1" OH I ‘l I O .9 ' ' . 0'0‘1 fifllhmwttéh r” ‘ ' - 0 i3: .A.kA . 'I ! 0 l a ‘ ¢| V'Ui‘vvx'v' w- A STUbY. OF TYPES cr- MAGAzms PICTUREAPPEAL AND EDlTORS" ‘ mm! Yo mama READER? pmne-wxwt magnum Thais f6: flu Degree 0? M. A. MI‘CHMANSIATE umvmsrrf~_ Anne li-An Ka-o ' 1964 _ _.______—‘=v mu; . . ' - . 1.0:: ' - 1.1.. . ‘ :-2.;.'..; "an-:42. .v 'a'. 3‘3 ,V, .K 6 .,, . _ .qin;:0;.. , ‘:;rl-'.§ A on.- I "‘o-?g- - -.- ‘ I Q . . "h',l$‘f" ‘ C . THESIS LIBRARY Michigan State University AS'I EDITORS‘ AB} The major que audience research to predict pictun The study ha: terms of similar 1 editom' ability ‘ in: conditions: part Of editors , diffemnt levels This study c using Q-methodglc Phases. The tecl; the second Phase ABSTRACT A STUDY OF TYPES OF MAGAZINE PICTURE APPEAL AND EDITORS' ABILITY TO PREDICT READERS' PICTURE-VALUE JUDGMENT by Anne Li-An Kao The major question investigated in the study is: What kinds of audience research results provided to an editor can make him best able to predict picture values of his audience? The study has two objectives. First, it segments audiences in terms of similar reaction patterns toward pictures, and then it tests editors' ability to predict the explored reader reactions under vary— ing conditions: a) different degrees of editorial experience on the part of editors, b) different types of predictees (readers), and c) different levels of information about the predictees. This study can be regarded as a small-scale operations research using Q-methodology throughout most of the study. The study has five phases. The techniques, item samples, and subject samples vary from the second phase to the fifth phase. The item samples used in the study are based on different ways of using a group of 120 pictures selected from Elf: and £925 magazines. Subjects are selected differ- ‘ently in each phase. The statistical tools used in the study are factor analyses, correlations and analysis of variance. The first phase of the study is a re-analysis of a previous study of picture preference done by MacLean and Hazard. The second phase is . preliminary ex the 120 magazine Feeling. Complex is a comprehensi‘ is a preliminary reactions toward picture values a: tent. In the first m" appeals whic tum of any eubj "Md and third 1 in their liking, 4 “t “1‘! the one: that ‘50 liking 01 e'pecially ideal I {Md' a “Petitec VidOd the editors 1 I“ the fifth Vith annually c and Menu», most 1'. M ”.311de DiCtu Anne Li-An Kao a preliminary experiment of four college students' reactions toward the 120 magazine pictures on dimensions of Like-Dislike, Intensity-of- Peeling, Complexity-Simplicity, and Clarity-Obscurity. The third phase is a comprehensive study of types of picture appeal. The fourth phase is a preliminary experiment on editora' predictions to two readers' reactions toward some 60 pictures. And the fifth phase studies readers' picture values and leads to a large-scale editorial prediction experi- ment. In the first phase, it was felt that patterns of intangible pic- ture appeals which were fpund functioning cross-sectionally for pic- tures of any subject matter can be put to test specifically. In the second and third phase, it was found that the readers were hedonistic in their liking, and that the pictures the readers liked or disliked most were the ones which aroused strong feelings. It was also found that the liking of pictures had a lot to do with self-identification, especially ideal self-identification. In the fourth phase, it was found, as expected, that the more information about the readers pro- vided the editors, the better were the predictions. In the fifth phase” it was found that women of different age groups, with financially or morally secure backgrounds, valued pictures of art and scenery most highly, pictures with people in a "cute" situation »second and pictures of glamour or fame third. Meanwhile, they avoided and disliked pictures of death, violence, and destruction most. It was found that male readers in general valued pictures of sports, sex and action most highly and that they disliked gruesome pictures of death and violence. But, some male readers with less education and lower incomes valued include var, violenc prediction part, it the readers made si; The predictse's Q-sc delegrephic informs nificantly better t Mogrephic informs the junior editor I statistical signif; It is evident editor; Predict be dmmphic infom Method. for. a “M in this Stud} in “Main l‘eact i: fitted in the com 3 Anne Li-An Kao lower incomes valued serious subject matter most highly. Examples include war, violence, science and social problems. On the editorial prediction part, it was found that the four levels of information about the readers made significant differences in the editors' predictions. The predictee's Q-sort of 60 pictures or the Q-sort plus his detailed demographic information provided to the editors can make them predict sig- nificantly better than the minimal demographic information or detailed demographic information. It was found that the senior editor group, the junior editor group and the naive "editor" group did not make statistical significant differences in their abilities to predict. It is evident in this study that reader reaction patterns help editors predict better than the traditional kind of survey research demographic information. Methods for an editor to check the validity of the techniques used in this study and to find out the percentages of his readers falling in certain reaction patterns discovered in a study of this kind are sug- gested in the conclusions. AST EDITOE ' A81 in] A STUDY OF TYPES OF MAGAZINE PICTURE APPEAL AND EDITORS' ABILITY TO PREDICT READERS' PICTURE-VALUE JUDGMENT 3)! Anne Li-An Kao A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Communication 196“ \>\~ojlcojL~_. ax~ ViEWe StUdy 23 It is important that the operation itself should be a reasonable one, such as a person can perform without feeling that it gravely distorts what he wishes to do. Also in the Q-sorting operation, a person can be made to "give away" many habitual modes of behavior and thus his Q-sort can be said to be reliable at any time. In all, Q-sort is an arrangement of Q-items according to an emphasized condition of instruction. It is a person's understanding, representing empirically-possible states of affairs. Q-factors are classes of similar Q-sorts. Orthogonal factoring brings to a head the maximal differences between classes of Q-sorts. Rotation of factors focuses similarities and differences maximally. Q-methodology is fundamentally comparative; it deals with rela- tions and quantitative concepts. The process of an editor's predictive Q-sort will be discussed separately in the design of the fifth phase. 21:. Interviews Subjects chosen according to the designs were asked on the phone to arrange an appointment for interviewing in their homes. The inter- viewer called on them as scheduled and asked them to perform the Q-sort. First, the subject was told roughly the scheme and purpose of the study. He was told that this study attempts to find out what kinds of people like what kinds of pictures and that these results were to be fed back to some picture editors to test their ability to predict readers' reactions. The subject was told that his own true personal feelings and opinions toward the pictures were most valuable to us and that his identity would remain confidential. Then, he ‘ shuffled. What or written inst clear about the said "yes". He but was first i ao-sonand lea: into the exact After he front of him i on the top and i“ Whine the my final Char “bioets Chan, the Q-distrib, obtained. In the ; editors . “Ere VhICh include quest ion 8 w 21} Then, he was handed a set of pictures, previously thoroughly shuffled. What he was expected to do was explained by giving him oral or written instructions for the Q-sort. ‘He was then asked if he was clear about the assignment. The explanation was repeated until he said "yes". He was given the scores and frequencies of the Q-distribution, but was first advised to sort the pictures into three piles, i.e., most-- so-so--and least. He was instructed to sort these three piles again into the exact given Q-distribution. After he finished the Q-sort with all the pictures spread in front of him in the required chosen form, he was asked to give comments on the top and bottom three choices and the general principles he used in judging the pictures as a whole. He was told that he could make any final changes in the picture arrangement. However, very few subjects changed. After all this, the picture numbers were recorded on the Q-distribution sheet by the interviewer. Thus, each Q-sort was obtained. In the fifth phase, the subjects, both the readers and the picture editors, were required additionally to fill out a series of questions which included all the needed demographic information about them. The questions were written according to some previous audience studies to cover all the demographic variables typically used for analysis of the audience. The readers or the editors were asked to fill out the question forms when the interviewer recorded the first Q-sort. All the Q-sorts were the raw materials for factor analysis. de 5. re Al as Pa: One Com 25 Computed Factors, Rotated Factors, Z-scores, actor-arrays and Consensus T as -“u Computed factors: .If M Q-sorts are collected, the number of correlations between them amount to ML%E;1. This correlation matrix may yield several factors through statistical computation. These first factors to emerge are called computed factors, and they may not be easy to interpret. Then, a rotation of factors is employed. Rotated factors: When the variates in computed factors have been sharply focused by rotation, the factors are the final ones to be in- terpreted. They are called rotated factors. Factor arrays: When the factors are found, then Spearman's weighting formulas can be applied to the relatively distinguished loadings of each factor selected. In so doing, we expect to get rather distinctive patterns of types represented by the factor arrays being thus weighted. z-scores: Factor arrays are presented in Z-scores - the standard deviation units for ready comparison. In other words, each picture in each factor has a Z-score. Comparing all the z-scores, the highly rejected, and highly‘acceptedr pictures by each factor can be listed. Also factor comparisons can be made: these reveal consensus as well as highly discriminating items. The consensus tables: They are the lists of the pictures that all the factors agree upon in the high range, middle range and low range. In comparing pictures for all factors, we can find which pictures one factor accepts or rejects more than the other factors. All these comparisons are in terms of standard deviation units - the Z-scores. Candi? considered 1 maths conte precision of 'fln inferencn The in] stmhrlere: Like-Di Intensi' colplexi Clarity- ACtual 3. Ideal Se: Value: 26 Conditions 9"; Instruction Conditions of instruction were developed, based on the things considered the most critical variables concerning people's value-judgments on the content of pictures. Conditions of instruction determined the precision of what has been measured and the situation under which the inferences were made. The important variables which have been treated throughout the study were: Like-Dislike: How much do you like or dislike what is depicted in the picture? (treated in 2nd, 3rd and uth phases) Intensity-of-Feeling: How strong is the feeling aroused in you by the picture? (treated in 2nd, 3rd and nth phases) Complexity-Simplicity: How simple or complex is the setting or the content of the picture? (treated in the 2nd phase) Clarity-Obscurity: How easily can you recognize what is depicted in the picture? (treated in the 2nd phase) Actual Self-Identification: How much are you actually like what is depicted in the picture? (treated in the 3rd phase) Ideal Self-Identification: How much would you like to be or to participate in what is depicted in the picture? (treated in the 3rd phase) Value: How much do you need and want the picture to be printed in your ideal magazine? (treated in the 5th phase) 27 Exact conditions of instruction based on these variables were stated formally in each phase as on next page. They were given in the same way to each subject we interviewed, right before each Q-sorting. When doubt came up on the side of the subject, discussion was allowed only to the extent of making clear what the given condition was, until he felt ready to sort with no uncertain feeling about what he should do. The conditions of instruction for Q-sorts in the 2nd phase were: 1. Like-Dislike: How much do you like or dislike who is pictured or what activity is taking place in the picture? In other words, how much do you like or dislike the subject matter? The pictures are to be _ graded from those you like most to those you like least, in such a way as to conform to the prearranged frequency distribution shown in Table 8. TABLE 8 THE FREQUENCY DISTRIBUTION FOR THE Q-SORT OF 120 PICTURES USED IN THE SECOND PHASE £3... Most Score 0 1 2 3 u S 6 7 8 9 10 ll 12 13 1a Frequency 4 6 8 8 9 10 10 10 lo 10 9 8 8 6 u M 2. Intensity-of-Feeling: How strong is the feeling aroused in you by the picture? The feeling may be any kind, but tell the intensity of it. We ask you to grade the pictures from those you feel most in- tense about to those you feel least intense about in the same way as mentioned above. 3. Clarity-Obscurity: How easily do you understand what goes on in the picture or recognize who it is or what it°s about? Again, follow I. a g i 1 ‘ < 1 ti Roi 28 the same picture-sorting instruction given above. u. Simplicity-Complexity: How simple or complex is the setting or the content of the picture? Is the subject matter or the setting elementary, unmixed, plain-spoken, common - or the opposite to you? The pictures are to be sorted in the same distribution again. The conditions of instruction for Q-sorts in the 3rd phase were: 1. Like-Dislike: How much do you like or dislike what is depicted in the picture? Judge only the subject matter. The pictures are to be ‘ graded from those you like most to those you like least in such a way as to conform to a prearranged frequency distribution as shown in Table 9. TABLE 9 THE FREQUENCY DISTRIBUTION FOR THE Q-SORT OF 60 PICTURES USED IN THE THIRD PHASE u ____g A Least ' Host Score 0 l 2 3 u 5 6 7 8 9 10 Frequency 3 u 6 6 7 8 7 6 u u 3 2. Intensity-of-Feeling: How strong is the feeling aroused in you by the picture? Sort the pictures in the same manner mentioned above: the same for the next two dimensions. 3. Actual Self-Identification: How much do you see yourself participating in or identifying with who and what is in the picture? u. Ideal Self-Identification: How much would you like to par- ticipate in or identify with who and what is in the picture? Experiments conducted in the fourth phase are described below: Two picture editors were asked to do three Q-sorts the way they would predict one of the assigned subjects would sort. Different levels of information picture editor Like-Dislike d Q-sorts of his Smelling" was loading on fac as the second PPeVious phas. There w. Predictions. here as an in Bane Variety The wh: The maj can increase tures when m: conditions 0 mainly in th COUdit 1. Ba sort the Pi: Sept. 29 of information about the predictee was provided in each sorting. One picture editor predicted the three Q-sorts of her predictee on the Like-Dislike dimension, and the other picture editor predicted the three Q-sorts of his predictee on the Intensity-of-Feeling dimension. "Mrs. Snelling" was chosen as the first predictee because of her high factor loading on factor A in the previous phase. "Mrs. Millan" was chosen as the second predictee as the highest factor loading of factor B in the previous phase. There were two sets of 60 pictures used in this preliminary editorial predictions. One, the same set used in the previous phase, was used here as an information sort. The other 60 pictures, which were from the same variety of categories, were used for the picture editors' Q-sorting. The whole prediction experiment was stratified as follows: The main issue of the experiment was to test if picture editors can increase their ability to predict readers' reactions toward pic- tures when more information about readers is fed to them. Thus, the conditions of instruction for the editors' prediction Q-sorts differ mainly in the level of information available. Conditions of instructions were given as follows: 1. Based on your image of an average adult in Lansing, please sort the pictures in a Like-Dislike dimension the way you think he would sort. Predict his Like-Dislike for the subject matter depicted in the pictures by sorting them according to the distribution shown in Table 10. THE F Lea‘ Score 0 Frequency 3 2. Consj in kindergarter day, reads a ne belief, subscm‘ high School and 60 Pictures the Q-distribut ion . 3' Besid another Similar Like-Dislike d i 30 TABLE 10 THE FREQUENCY DISTRIBUTION FOR Q-SORT OF 60 PICTURES USED IN THE FOURTH PHASE h—H- __ .__A__‘ M Least Th Most- Score 0 l 2 3 u S 6 7 8 9 10 Frequency 3 u 6 6 7 8 7 6 6 u 3 2. Consider Mrs. Snelling. She is 31, a housewife, has one boy in kindergarten and one girl in 6th grade, watches TV six hours every day, reads a newspaper, doesn't go to movies, has no special religious belief, subscribes to £235 and 2235 magazines, graduated from the local high school and has lived in LanSing all her life; now, sort the same 60 pictures the way you think Mrs. Snelling will sort, using the same Q-distribution. 3. Besides what you have been told about Mrs. Snelling, here is another similar set of 60 pictures in the order sorted by her in the Like-Dislike dimension not long ago. Her top and bottom choice comment were that beauty, flowers, Sports and scenery interest her and that she hates children suffering. Now, please study her sort of pictures, and then sort this previous set of pictures again the way you think she will sort, in the Like-Dislike dimension. In the second picture editor's prediction Q-sort on the second predictee, the same level. of information was given for the first, second and third Qwsorts, except this time on the Intensity-of-Feeling dimensiono In the second sort, the demographic information about Mrs. Millen in- cluded: a woman aged 31, doctoral candidate in communication, has no children, doesn't watch TV, reads a newspaper every day, goes to a movie every gious belie traveling i has strong She doesn't individual .' graphic inf< and then the The ma audience of two readers test 0f the Picture edit ChOSen made There definition 0 Cf tWo Signi ing the piCt CEdUpeS for A. Va 31 movie every two weeks, subscribes to 2225 and EEEE magazines, no reli- gious beliefs, knows photography and reads photography books, has been traveling in Montreal and Cornell, likes auto racing and fencing. War has strong impact on her, and she likes unusual character depiction. She doesn't have interest in masses where no identification of the individual is present. In the third sorting, besides this same demo- graphic information, her Q-sort was demonstrated in front of the editor, and then the editor was asked to sort the same set of pictures once more. “2321;225:1222 The main idea of this phase is to study values of a well-rounded audience of different backgrounds for magazine pictures and then choose two readers to represent two major patterns or types of reactions. The test of the editor's ability to predict involves different kinds of picture editors using different kinds of information about these two chosen readers. There are several major issues in this design. They include the definition of one certain type of picture appeal for Q-sorts, the choice of two significant readers to be predicted, the variables used in test- ing the picture editor's ability to predict and the method and pro- cedures for testing. A. Value-dimension as the chosen type of picture appeal to be tested. In the previous phases, variates like Like-Dislike, Intensity-of- Feeling, Complexity-Simplicity, Clarity-Obscurity, Actual Self-Identification, Ideal Self-Identification were studied. Here in the final phase, it is hadly possxbl same time, bec large amount c additional var Q-sorts and ed But, wha specifically s a well-rounded in the picture In this Consid YOU; by id ~ Very much ' PiCtur‘e he' way as to . tribut i0n 2 Score Frequency Please Considerin ment nOt W 32 hardly possible to test many variates (sorting instructions) at the same time, because even a few readers and editors require a fairly large amount of time and effort for testing on one variate. One additional variate means twice or three times the required number of Q-sorts and editors in testing. But, what dimension should be chosen? Since we could not study specifically several dimensions of picture appeal distinctively, we chose a well-rounded term to test reader reactions; that is, to test the values in the picture in the eyes of the readers. In this value dimension, the condition of instruction for reader's Q-sort is specified as follows: Consider very carefully a magazine that would be ideal to you; by ideal,.I mean one that has pictures you want and need very muc t at would have high values for you. Pick out the picture here that represents chTfIEE. of picture this maga- zine would be most likely'to have. Please sort the pictures according to their values for your ideal magazine in such a way as to conform to the following prearranged frequency dis- tribution: least valuable most valuable Score 0 l 2 3 u 5 6 7 8 9 10 Frequency 3’ .3 6 8 u Please sort the 60 pictures as accurately as possible, considering only your very own individual feelings and judg- ment not what other people or friends would think about them. The reason for suggesting an ideal magazine in the mind of readers is because the term "value" is hard to define. The ideal magazine in readers' minds puts readers in a concrete hypothetical situation and is believed to make readers think about values in pictures for various kinds of reasons. Finally, finding what readers want and need can be practically valuable. B. The Eighteen malenfemale; y backgrounds. with instructi were correlate factor-loading and a middle-a chosen as the We sets of pi Showed the ind among the high Patterns, In on two factors differences in choice I SiDCe for different c. The was '7“ Of e. W0 or more pepiment. The files; it had Variables T. 33 B. The choice of two predictees. Eighteen readers were chosen representing different combinations: male—female; young-middle-aged-old; and college-high school-grade school backgrounds. They were asked to do two Q-sorts on a value dimension with instructions as above. Their Q-sorts of both sets of 60 pictures were correlated, and factor-analyzed separately. Studying the two factor-loadings, a young male reader, who was a grade school graduate, and a middle-aged female reader, with a college level education, were chosen as the two predictees. Correlations between their Q-sorts on two sets of pictures were -0.185 and -0.07. The low correlations showed the independence of their picture-values. Their Q-sorts were among the highest factor-loadings for two different types of reaction patterns. In other words, they were the two most representative readers on two factors and the most independent from each other. Also their differences in sex, age and education happened to make them the ideal choice, since it is interesting to see if ability to predict differs for different readers. C. The design for testing the picture editor's ability to predict was analysis of variance. The analysis of variance tested the effects or the combined effect of two or more experimental variables. It was a 3xux2 factorial ex- periment. The term factor denotes the treatment and experimental vari- ables; it had three variables and there were 3,u,2 levels in the three variables. The three variables and their levels (see Table 11) were: Kinds of Editors - the Pre- dictors “- Senior Editors (Professional Editors 8 P h°t°8Pap he rs ) 3H Qasort TABLE 11 DESIGN OF THE FIFTH PHASE: THE EDITORS' PART Kinds of Levels of Information Number The Male and Editors - Provided the Editors of Female readers the Pre- about the Readers Editors to be Predicted dictors fig 1 Minimal Demographic Editor 1 Information Editor 2 Senior Editors 2 Detailed Demographic Editor 3 (Professional Information Editor u Editors 8 ' Photographers) 3 Predictee's actual Editor 5 Q-sort Editor 6 u Detailed Demographic Editor 7 Information and a Editor 8 Q-sort 1 Minimal Demographic Editor 9 Junior Information Editor 10 Editors (Journalism 2 Detailed Demographic Editor 11 Majors Who Information Editor 12 Have Some .l Experience in 3 Predictee's actual Editor 13 Picture- Qnsort Editor In Editing) ’ u Detailed Demographic Editor 15 Information and a Editor 16 Qosort * "Naive 1 Minimal Demographic Editor 17 Editors" Information Editor 18 (Junior or ” Senior Educ 2 Detailed Demographic Editor 19 cation Majors Information Editor 20 Who do not Know Photo- 3 Predictee's actual Editor 21 graphy) Q-sort Editor 22 u Detailed Demographic Editor 23 Information and a Editor 2n 1. Di Th di 58 Th 3T 8T h-l l. 35 Different kinds of predictors--the editors: Three different kinds of picture editors were chosen accor- ding to their experience in picture-editing. We chose eight senior editors, eight junior editors and eight naive "editors." The senior editors were professional editors, picture editors, and photographers of certain publications around the Lansing and East Lansing areas. They were full-time employees and most of them were in their uO's and 50's. All of them had a college degree or above. Most of them were married and had children. Five of them had annual incomes from $10,000 to $15,000 and three of them had annual incomes from $7,000 to $8,000. The junior editors were junior or senior journalism majors with some experience in picture selection. The eight chosen for our study were from Michigan State University, and most of them had worked or were working for the Michigan State News, the week—day newspaper of the university. Some of them were photographers for Michigan §£2££_N2:§3 Their ages were in the neighborhood of 20; they were all single. They came from families with an average annual income around $5,000 to $6,000. They liked sports and cultural activities. The naive "editors" were hypothetical editors. All of them were junior or senior education majors who did not know photography. Their names were found on card files of the education department. They were asked on the phone about their knowledge of photos and all the ones who didn't know photos were asked to do the Qasorts. 2. of PE T} 2. 3. 36 There were four different levels of information about the two predictees assigned differently to different kinds of editors. Four pairs of editors of each kind were given one of the four different levels of information about the two readers before they predicted them. The levels of information given to the editors were: 1. Minimal demographic information: age, sex, and level of education (Condition of instruction A). 2. Detailed demographic information: (Condition of in- struction B). 3. A Q-sort of the predictee: (Condition of instruction C). u. A Q-sort of the predictee plus detailed demographic information: (Condition of instruction D). Two different readers to be predicted. The chosen male and female readers mentioned before were assigned for each of the 2n editors in the random and controlled order no matter what information about them was given. That is, each editor was randomly assigned a number within the group he belonged, and each number indicated one certain kind of condition of instruction. The male and female readers were randomly assigned to each editor. If editor A was randomly assigned to predict male reader first on the basis of minimal demographic information, then, the next editor who was to predict on the same basis was asked to predict female reader first. Thus, the effect of order in predicting the two readers was randomized and controlled. (see Table 11 again) Al. twice, 81 in the ea The were corr pictures. terms of c The readers' Q The 1 Pictures an fourth and The e: 37 All the 2“ editors used the second set of 60 pictures to sort twice, and each time predicted one reader. The list of pictures is given in the early part of this chapter. The editors' predictive Q-sorts of the second set of pictures were correlated with the readers' actual Q-sorts of the second set of pictures. Individual editor's ability to predict was assessed in terms of degree of correlation. The sum of the D-squares between editors' Q-sorts and the readers' Q-sorts were used as raw scores for analysis of variance. The results of 18 readers' value judgments on two sets of pictures and 2a editors' abilities to predict are given in the fourth and fifth chapters. The exact written conditions of instruction for editors' pre- dictive Q-sorting are given in the appendices. RESL First 2! The Maci The Badger Vi] Variables conc They listed sc picture. Such Publication , BEE intereSt CHAPTER III RESULTS AND INTERPRETATION OF THE FIRST FOUR PHASES First Eh ase: Re-anal sis of "Women' 5 Interest in Pictures: T e Ba ger Village ”Study" The MacLean and Hazard study: "Women's Interest in Pictures: The Badger Village Study" has thrown some light on the important variables concerned with the selection of publishable pictures.1 They listed some possible elements which may make an effective picture, such as news value, size, suitable subject matters for the publication, etc. In their study, they studied the variable ic- m interest . Before going into a re-analysis of their study, a summary of the findings is cited below: Six major appeals proved to account for nearly all measured variation in interest for 31 out of 51 pictures selected from Time and 0.8. News E £2£$2.R82°rt for 19u9 and 1950. Briefly, those appeals were: I. Idolatr : Successful, glamorous, wealthy women, es- pecially in a "Happy family" situation. Adolescent girl's dream. II. Social roblems: People causing "trouble" like strikes and riots, peopIe "on the wrong side of the tracks," people who do not "fit in" to popular ideals. III. Picturesque: Salon pictures, moody pictures, dynamic pictures, pictures portraying lonliness and escape from other people, doing things with hands. lMalcolm S. MacLean, Jr. and William R. Hazard, "Women's Interest in Pictures: The Badger Village Study," Journalism Quarterly (Spring, 1953). ‘ 38 accident . VI . E; big audie: The defi merely meanin, ETeat variety The Six matter. Withz' same elements Six groups of In the “man "as alrr fulfillment. Familiarity a Picture may ‘ "as ginn as the subject pictul‘es . 39 IV. War: War, fear of involvement in war, horror at the gruesome results of war, glory of American armed power and sympathy for the victims of war. V. Blood and Violence: People suddenly dead through crime or accident. - VI. Spectatog_Sports: Action pictures of Sports which command big audiences. ' The definition of "interest" used in their study was very flexible, merely meaning how interested the interviewees were in the pictures. A great variety of reactions emerged under the term "interest." The six major appeals seemed to be grouped by category of subject matter. Within each appeal, elements of interest varied. Also the same elements of interest could be found cross-sectionally throughout the six groups of interest. In the first group, "Idolatry," a relatively wealthy and famous woman was almost the perfect object for identification, for dream-wish fulfillment, by some of the young married women in Badger Village. Familiarity and pleasant associations with what is depicted in the picture may determine the liking of the picture. Lack of recognition was given as reason for lack of interest. Adverse moral judgment of the subject matter played some part in the negative reactions to the pictures. A In the second group of interest, "Social Problems," social con- science reaction was given as one important interest factor. Striking feelings, we; social res onsibilit , sad Eli-identification, and curiosity were the reactions aroused by the pictures in the people. Failure £2 recognize the subjects again was the main reason for low interest pictures of this category. Avoidance of social problems was revealed in some of the comments made on low interest pictures, too. In the tensity 23: fe as qualities Negative reas: that particul In the 22:, horror, 22 and 232 the t°l>-I‘ated for 1“Winter In the W (3 seemed functj reactions age In the were aSSOCIa1 O No". as that despite no In the third group, "Picturesque," likes.subject'matter5‘22: tensity 2£.fee1ing--power, strength, pictorial composition-~and contrasts, photographic qualities, were mentioned in the comments, as qualities which contributed to the attraction of the pictures. Negative reasons were "didn't seem," "92 interest i325; liking'm that particular activity." In the fourth group, "War," reactions seemed to involve £222, horror, curiosity, patriotic feelings and sympathy. £22: gszgland identification appeared to be highly important in some of the top-rated pictures. Dullness and dislike 23 Er- were the reasons for low-interest pictures. In the fifth group, "Blood and Violence," a great deal of fear-thrill due to intense feeling toward others or self-identification seemed functioning as an element. Avoidance was found in the negative reactions again. In the sixth group, "Spectator Sports," favorable reactions were associated with general liking for the pictured Sports, and vice versa. Now, as a whole, we can see from the analysis cited above, that deSpite the different nature of the six distinctive groups, some elements of picture appeal included in "interest" can be found cross-sectionally. Elements include: Like-Dislike (Personal interest-~subject matters and value judgment) Self-Identification (Feelings concerned with being or liking to be the depicted charac- ters and situations or the reverse) Intensity of Feeling (Impact: from the visible or in- tangible forces) H1 Clarity-Obscurity (Visible settings of angle, light, action, contrast, position, etc., plus realism and familiarity) Simplicity-Complexity (Degree of understanding of what is depicted) In fact, the chief essentials of content are appeal and im- pact. Visible forces such as a closeup, action, pattern, or ex» treme contrast of whites or shadows will arrest the eyes of the reader, and cause reader reSponse. Intangible factors-~the interest, drive, and feeling in the picture--such as adventure, ambition, combat, daring, escape, love, mystery, romance, self-preservation, sex, survival, suSpense, anger, familiarity, hate, sympathy and so on, are the forces to arouse a response in the reader. Impact is the emotional reaction of the reader which produces responses running from casual interest to violent action. A picture's im- pact can result in reSponses very mild and audible.l Another indication of these cross-sectionally functioning elements of appeal is found in the factor-to-factor correlations. For example: 1. Correlation between "Idolatry" and "Social Problems" is .23. This may be due to self-identification either ideally or sadly which draws interest. 2. Correlation between "Social Problems" and "Blood and Violence" is .45. The nature of the subject matters in these two groups is somewhat similar. It seems that intensity of feeling is the same element of interest for both groups. lKalish and Edom, 23. cit., pp. 79-80. 3. Cor aga sue a, Int and MacLeax ing sensible detected pap they can mak Calfunctior Mich? and ject matter Based Study. fouz appeal for The cally Whig each Of f: Obscurity H2 3. Correlation between "Picturesque" and "Blood and Violence" is .51. The dramatic scenes in the "Blood and Violence" group again give strong feelings as the Picturesque group does, such as power, strength, the visible appeal. a. Interest in "War" was related slightly to interest in "Blood and Violence" and "Spectator Sports." MacLean and Hazard felt that a study like this would allow see- ing sensible subject matter controls on future experimental work. They detected particular kinds of elements in each group and suggested that they can make studies within the group appeal study, or that elements can function between the groups. They suspected that clarity and aims plicity and dramatic pictorial qualities may prove helpful when sub- ject matter is held constant. Second 2222:: Preliminary m of Type-s- .o-f'. Picture Appeal Based on the re-analysis of MacLean and Hazard's picture interest study, four definite variables were chosen to test types of picture appeal for this phase. The purpose of this second phase experiment is to find out specifi~ cally which pictures are higher than the others in reader's ranking along each of four dimensions: Like-Dislike, Intensity-of-Feeling, Clarity- Obscurity and Simplicity-Complexity. After sixteen Q-sorts by four subjects were collected, the data were intercorrelated and factored. The process used here was William Stephenson's QoMethodology centroid factor analysis as discussed in Chapter I I o Three fa by rotating th factors were 0 Computed shown as below ‘d h. fl -— COWPUted Variable Number no? 1 e 3 2 e C 3 e 7 ‘4 " 9 7 s I 2 6 . '2 7 '1 8 ‘I ': 9 ° 4| '10 O I ll 0 4 l2 . 13 . ll.- .' 15 u e 16 a /. ”3 Three factors emerged as significant and clearecut results, by rotating the computed factors four times. factors were obtained through a weighting process. Factor arrays of three Computed factors, rotated factors and the weighting schedule are shown as below: CENTROID AND ROTATBD FACTORS OF THE SECOND PHASE TABLE 12 Computed Centroid Factors Rotated Factors Variable Factors Variable Factors Number F1 F2 3 u Number F1" F2" F3"° F” l .3“ .07 .00 -.10 1 .25 .53 -.06 -.08 2 .ou -.72 .12 -.u9 2 .15 -.68 s.20 -.u7 3 .70 -.07 .22 -.45 3 .73 .06 -.16 -.us 4 -.70 .09 «.51 .07 u -.86 .ou .10 .01 5 .23 .35 -.31 -.31 5 .01 .U7 -.01 ~.39 6 .28 ".2” ‘o30 -015 6 012 '008 626 =.#l 7 .38 .0” .26 .21 7 .95 .08 .10 .19 8 “027 “'01” -9'41 ‘031 8 ‘cu2 -.09 -001 -339 9 052 o3g -.10 'el2 9 :37 053 e12 -022 10 .27 -.21 =.05 -.2O 10 .23 -.11 .05 -.29 11 .45 “.30 -.96 037 11 019 “006 e77 ‘516 12 .42 -.25 -.38 .u9 12 .20 -.05 .77 -.Ol 13 .uo .68 ~.26 -.22 13 .16 .82 .00 -.28 14 ~.10 -.55 -.1u -.16 1“ -.10 -.50 .09 -.27 15 .35 -.63 -.03 -.01 15 .35 -.50 .29 «.26 16 -.30 .03 ~.l2 -.15 16 -.32 -.01 -.1u -.08 WEIGHTING SCHEDULE USED IN nu TABLE 13 THE SECOND PHASE Factor Representing Loading Weight Variables ' F1 3 .73 .l.563 1" -086 “30303 7 .HS .560 F2 1 .53 .737 2 -068 -le265 5 .97 .603 9 .53 .737 13 .82 2.503 1” -050 -0066? F3 11 077 10891 12 .77 1.891 45 Three factors Factor A m- Factor A shows that for these subjects, portraits or dynamic actions of familiar people in social activities tend to be clearer and simpler for understanding. TABLE 1n PICTURES THAT FACTOR A ACCEPTS AND REJECTS: SIMPLICITY AND CLARITY Accepts (Standardized Q-scores lu-12) Marilyn Monroe Beach beauty in bikini Brigitte Bardot Riat Hayworth Baseball pitcher b a close-up Misses America of 1959, 1960 and 1961 A football player's muddy face - a close-up Olympic track star - at her final dash Nixon family on the beach Kennedy family sitting on the lawn lOlst birthday - Grandma Moses and her cake Telephone booth cram - a close-up Nixon campaign - shaking hands Championship boxing - Johansen 8 Patterson Rejects (Standardized 0- scores 0-2) Cathedral dome painting Metal net Machine hands Light ball Glass equipment Plastic models of human body organism Big'machine .Firemen and the dead BishOp performing a mass Flood scene - long range Radar Space lady Farm scene Boys and big waves This factor is strongly represented by subject I and II's (husband and wife A) Complexity dimensions. Clarity-Obscurity and subject I's Simplicity- Pictures considered easy to understand tend to be the same to both subject I and II and tend to have simpler subject matter of pictorial compositions. Sex, sports, and politics are the can They include Clarity and Brigitte Ba: known to the distracting track star, Moses, champ Scenes, clog On the other pictures are Pictures hav. with not mud to be Unders1 aI‘OUSed ConfL clear- and (201': the SPace lac 1‘ some acti‘, an Unusual a: scene and 31?. painting take '46 are the categories of the highly accepted pictures in this factor. They include portraits of sexy movie stars on the extreme end of Clarity and Simplicity, e.g., Marilyn Monroe, Rita Hayworth, Brigitte Bardot. These are undoubtedly sex symbols universally known to the people. Their portraits do not have any attention- distracting settings. Other pictures, such as the Misses America, the track star, the telephone booth cram, the lOlst birthday of Grandma Moses, championship boxing, and a political campaign, are dramatic scenes, close-ups or action shots from the current events spotlight. On the other hand, science, destruction, off-beat and sudden death pictures are placed on the rejection side of the seale. The science pictures have few people in them; most of them show static objects with not much background and require considerable verbal explanation to be understood. Pictures in the off-beat and sudden death categories aroused confusion and curiosity in the subjects but are considered un- clear and complicated. Pictures like the Pope, the French vendor, the Space lady and the orchestra conductor show people in the middle of some activity yet without additional ground. Also, ahots taken from an unusual angle puzzled the subjects most. Examples include the flood scene and slum bedroom taken from high above and the cathedral dome painting taken as a closeepptfrnm directlybbelow. In pictures ranked higher on this factor than on the other two factors (see Table 15),Clarity and Simplicity seem highly related. Pictures of sex, sports, politics and performance are more of Factor A. Further, they are either portraits or dynamic actions of something fa- miliar. Pictures of social parties, destruction, sudden death, designs, off-beat material, science, and arts belong more to the other two factors than to this factor. #7 TABLE 15 PICTURES WHICH FACTOR A ACCEPTS MORE OR REJECTS MORE THAN FACTORS B AND C M..— A A.__L FA) FBFC #fl .1 Misses America of 1959, 1960 and 1961 Scottish dance Symphony orchestra Brigitte Bardot Marilyn Monroe Beach beauty in Bikini Rita Hayworth Kennedy campaign - autographing Nixon campaign - shaking hands and keeping track of his watch Nixon giving a Speech Republican campaign - public gathering with slogans Baseball game - fourth base Football player's muddy face - close-up Tennis player Olympic track star - at final dash Championship boxing - Johansen and Patterson Baseball pitcher - close-up Chinese kid in Communist Commune working Preaching - Billy Graham Black and White intermarriage - May Britt and Sammy Davia.]ro Peeking eye - close-up Freedom riders in the bus Crowds and policemen - long-range Burned school house in ashes Sudden death Peeking German soldier Soldier crawling Victory comes to soldiers Military weapons Prisoners in Cuba Skiing Gravestones Telephone booth cram - close-up Telephone booth cram - long-range Head operation Dentist and patient 11(3ch Shah's bride and peace doves Ford family in daughter Anne. 8 debut party Prince's wedding Queen of.Iran at opera theater King's wedding Football game entertaining - "tiger" marching - 80th wedding anniversary Senator Sam Rayburn and baby grandson Bish0p performing a mass Little girl and dog Hemingway sitting and reading under a tree with his pipe and dog Old orchestra conductor Little girl and fish Republican big shots Slum bedroom Freedom riders singing on the campus Negro college students Kids and flood ' Firemen and the dead Beach widow crying Boat tragedy and frightened kid Soldiers at sea to be saved Mother and 12 pairs of shoes Cello concert near the river bank Castle - long range French woman selling fish French vendor Boat man Space lady Light ball Radar Metal net Little boy and doctor Machine hands lst prize roses Cathedral dome painting Farm Sunset fishing ‘ —-——-——- Factor B “- Factor B has shown a hedonistic principle in preference and intensity of feelings. TABLE’16 PICTURES THAT FACTOR B HIGHLY ACCEPTS OR REJECTS: LIKING u WA A ‘ A AAA .1.. Accepts (Q-scores 19-12) Kennedy family sitting on the lawn lOlst birthday - old lady and her cake Debbie Reynolds and daughter Nixon family on the beach Castle - long range Sunset fishing Farm scene French woman selling fish Hemingway sitting and reading under a tree Brigitte Bardot lst prize winning roses Senator Sam Rayburn and baby grandson Little girl and fish Beach, fisherman, tree and fish Beach beauty in bikini Misses America of 1959, 1960 and 1961 Khrushchev and Kennedy 80th wedding anniversary Rejects (Q-scores 0-2) Hanging of Mussolini and his mistress Bloody man on the back of a truck Soldier crawling Death in the jungle Juvenile delinquency - boys leaning against'the wall Man being caught Prisoners in Cuba Searching the dead's pockets Dead soldiers on the beach Airplane crash in New York Prisoners Hungry boy - a close—up Peeking eye - a close-up Chinese student in the Commune Flood scene - a long range Gun soldier parade Threat in the gang Hedonistic is defined as the striving for pleasure and the avoid- ance of unpleasant feelings. People are basically very much alike in their emotions. On the very basic level - starting at infancy - are such persuasive drives as hunger, the need for sleep, the instinct for sex, the need for companionship, the search for approval, the drive toward activity, and at the same time, a desire for comfort. We develop our ability to love; we learn ways to avoid trouble, deal with people, 49 watch our health and take on responsibility.1 Based on the above principle, the pleasantness of the pictures seems a major element for people's like-dislike. And the result of this factor is an evidence for it. The heaviest factor loadings to represent factor B are four subjects} Likerpislike dimensions and subject I and IV's Intensity-of-Feeling dimensions. All subjects tend to like and dislike the same subject matters, and their likes and dis- likes both give them intense feelings. Pictures on the acceptance side include young marriage, design, art, performance, glamour, peace, and fame. They are pleasant, feminine, relaxing, appreciative, comforting and relieving. On the other end are death, misery, social problems, destruction, and tools of war; they are pathetic, regretful, tense and unbearable. In the subjects' comments, hedonism was revealed: "I like democracy, pleasant way of life and politics. I don't like crime, war, povertyé" "Beautiful, vivid, impressive, and bright life gives warm, peaceful and excited feelings, also a truth-seeking feeling; life full of crime and misery seems unbearable for me to take." "Unspoken human love and warm relationship between humans gives me a feeling of relaxation and easiness." "Sports, women, and a little bit of politics are the things I like just like most of the other men do. The older you are, the more you dislike war and destruction." In the list of pictures that Factor B accepts or rejects more than the other two factors (see table 17), we will see that the principle of hedonism again follows. it lStephen Baker, Visual Persuasion (New York: McGraw-Hill Book Company, 1961). 50 TABLE 17 PICTURES WHICH FACTOR B ACCEPTS OR REJECTS MORE THAN FACTORS A AND C ~A_-L A. FB) FAFC “A Shah's bride and peace doves Ford family at daughter Anne's debut party Prince's wedding Queen of Iran at opera theatre Nobel party in Sweden Skating on the stage Football game entertaining - band marching lOlst birthday - old lady and her cake Golden wedding party - several hundred couples 80th wedding anniversary Bishop performing a mass Little girl and fish Judo - kids in judo uniforms Debbie Reynolds and daughter Nixon family on beach Kennedy family setting on the lawn Republican big shots Khruschchev and Kennedy both smiling Chinese kid in Communist Commune operating machine White Democratic girls campaigning Girl criminal Castle - long range Big bridge - close-up Great building and reflection in water Firemen and the dead Mother and 12 pairs of shoes French woman selling fish French vendor Boat man Big machine Metal net Glass equipment Plastic models of human body organism Cathedral dome painting Farm Sunset fishing Clouds Beach, fisherman, tree and fish FB