TV «t, .i.., _.::.Z 1 . . . I.: H DQSPLAY‘ 0N 'EVALUATNEMEA A lNG-‘w MULT eof P11,- D “SCREEN, g. E EFFECT OF LARG T; s for the Degre ' SSE VI.H .fi. .3 R E W .N ‘ U, AT T S ”N... ‘A m H; m M INN CHARLES G.- BOLLMAN 1970 ' # m , .: i. .‘. ,. . .,. . . . , . lithifiu. , . 421.....6. ..: , . . ‘ .A , If . . ., . .. . . , . ‘ . . . LL3$271r . . ‘ y . , o G . I r . .J v A _ 7‘ a... ... a: . ., . .3... 1: . . a. u. ...,.,_,. «,._._..§.m...._n.q.~.c ‘ .35? b5. qsgaiad t“ - LIBRARY 3 Michigan State ‘ I! 1 University (4 I!!!I”!!!lflzllflllalfllllflfljflflllflfllllllsllllfllflfl SWW, This is to certify that the thesis entitled THE EFFECT OF LARGE-SCREEN, MULTI-IMAGE DISPLAY ON EVALUATIVE MEANING presented by Charle s G. Bollmann has been accepted towards fulfillment of the requirements for Ph. D. degree in Education 0-169 l I f ‘P' KF‘ ‘ I .r' .m’ f ' l ( 1.x -- t ‘ .«v' Gum-fig? me? ,I If. ABSTRACT THE EFFECT OF LARGE-SCREEN, MULTI-IMAGE DISPLAY ON EVALUATIVE MEANING BY Charles G. Bollmann Since the early 1960's large-screen, multi-image displays and presentations have become increasingly pop- ular. Numerous applications of the techniques have been made for educational and instructional purposes but little is known about the non-cognitive impact of such presenta- tions. This study explored the gross, affective impact of a multi-image presentation upon human subjects. The ex- perimental data were intended to shed light on two general questions: 1. Will a multi-image and audio presentation cause greater positive shift in evaluative meaning than a parallel single-image and audio presen- tation? 2. Is the magnitude of shift in evaluative meaning related to the amount of the viewer's visual field which is covered by the projected flmage Charles G. Bollmann area as determined by the viewer's distance from the screen? Using a posttest-only design, the experiment pit a multi-image presentation against a parallel single- image version and a control presentation. All treatments were 10 minutes in length and used ten-foot images from 35mm slides and music-only audio components. The over- all horizontal image area for the multi-image presenta- tion was thirty feet. Random assignment of the seventy-one graduate stu— dent subjects to the groups, simultaneous presentation of the treatments in identical rooms, and the use of the same twenty-scale semantic differential instrument made it pos- sible to ascribe between-group variance on the dependent variable of evaluative meaning to the independent variable of presentation treatment. A prerequisite factor analysis of the SD response data showed that the subjects rated the five concepts on three main "evaluative" dimensions. Accordingly, a separate analysis of variance was performed with respect to each research question for each concept on all the factors found in the prerequisite analysis--a total of thirty-two analyses of variance. The findings were summarized as follows: 1. There is considerable evidence that a systematic main effect was operating on three of the five Charles G. Bollmann concepts but the effect cannot be ascribed statistically to the multi-image presentation. While not conclusive, there was some evidence that more positive shift in evaluative meaning was elicited from those viewers of the multi- image presentation who were situated the closest and farthest from the screen. This reversal of the expected finding seems to warrant further investigation since the viewers were situated from one-third to two screen widths from the projected images. THE EFFECT OF LARGE-SCREEN, MULTI-IMAGE DISPLAY ON EVALUATIVE MEANING By a 0 Charles GE Bollmann A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1970 ACKNOWLEDGMENTS As with most human endeavors, a research project of this kind could not be begun nor seen through to completion without the assistance, cooperation, and encouragement of many people. The researcher owes a heavy debt to many but wishes to express his sincere gratitude to those who es- pecially helped the work along. To the researcher's doctoral committee: Dr. Charles F. Schuller, Chairman: Dr. John Barson; Dr. Charles Black- man; and Dr. Hideya Kumata. Their guidance was simply invaluable in mounting and completing the project. To Miss Ruth Allen who willingly provided the experi- mental presentation and needed equipment. To the consultants who freely contributed their special skills and knowledge: Robert Wilson, James Mullin, and William Allard. Additional thanks are extended to the researcher's fellow graduate students and particularly to Dr. Elwood Miller and Dr. Curt McCarty--both friends and colleagues without peer who patiently cajoled, listened, and en- couraged until the work was done. Lastly, the researcher acknowledges his undying appreciation and love to his wife, Gail, and to Mike, ii Shelly, and Mitch for the valiant and cheerful forbearance which they all showed during these years of study. iii TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . LIST OF TABLES . . . . . . LIST OF FIGURES . . . . . . LIST OF APPENDICES . . . . . Chapter I. II. III. INTRODUCTION AND THEORY OF Introduction . . . Purpose and Scope of the Need for the Study . . Definitions . . . Theory and Rationale for Multi-image Theory . Evaluative Meaning . Assumptions of the Study Limitations of the Study REVIEW OF THE LITERATURE Affective Impact Studies Image Size Studies . . Multi-Image Studies . DESIGN OF THE STUDY . . Introduction . . Overall Design and Methodology Phase One . . . . . Phase mo 0 O O O 0 Phase Three: The Experiment Design . . . . . Treatment Presentations Population . . . . Instrumentation . . Rooms and Equipment . iv THE STUDY Study . the Study Page ii vi xi accomth-I H Chapter ' Procedures . . . . . . . . . Hypotheses . . . . . . . Scoring and Analysis . . . . . IV O FINDINGS O O O O O O O O O O O Prerequisite Analysis . . . . . Rationale for the Factor Analysis Computational Procedures . Results of the Factor Analysis Summary of the Factors . . . Experimental Effects . . . Computational Procedures . Main Treatment Effect . . Effect of Viewer Location . V. SUMMARY AND CONCLUSIONS . . . . . Summary . . Purpose of the Study . Design and Procedures . Analysis of Results . . Findings . . . . . . Discussion and Recommendations . Size of Experimental Population Selection of SD Concepts . . . Selection of Experimental Populatio Viewer Location . . . . . . Experimentation with Individuals Some Larger Issues . . . . . Conclusion . . . . . . . . O O O O a O O O O O O O O BIBLImR-APHY O O O O O O O O O O O O APPENDICES . . . . . . . . . . . . Page LIST OF TABLES Table Page 1. Percentage of VT and VC Accounted for . . . . 65 2. Group Means and Results of ANOVA's on Factor I . . . . . . . . . . . . . 68 3. Group Means and Results of ANOVA's on Factor II . . . . . . . . . . . . . 68 4. Group Means and Results of ANOVA's on FaCtor III O O O O O O O O O O O O O 69 5. Group Means and Results of ANOVA's on Factors A and B . . . . . . . . . . . 69 6. Post-hoe Comparisons of Group Means . . . . 71 7. Subgroup Means and Results of ANOVA's on Fae tor I O O O O O O O O O O O O O 7 6 8. Subgroup Means and Results of ANOVA's on Factor II . . . . . . . . . . . . . 76 9. Subgroup Means and Results of ANOVA's on FaCtor III O O O O O O O O O O O O O 77 10. Subgroup Means and Results of ANOVA's on Factors A and B . . . . . . . . . . . 77 ll. Post-hoc Comparisons of Subgroup Means . . . 78 12. Factor Loading Matrix for Concept SCIENTIST . . 119 13. Factor Loading Matrix for Concept MBORATORY O O O O O O O O O O O O O 1 2 l 14. Factor Loading Matrix for Concept MICHIGAN STATE UNIVERSITY . . . . . . . . . . . 123 15. Factor Loading Matrix for Concept BIOCHEMISTRY O O O O O O O O O O O O 12 5 vi Table Page 16. Factor Loading Matrix for Concept EXPERIMENT . . . . . . . . . . . . . 127 l7. Varimax Rotation for Concept SCIENTIST . . . 129 18. Varimax Rotation for Concept LABORATORY . . . 130 19. Varimax Rotation for Concept MICHIGAN STATE UNIVERSITY . . . . . . . . . . . . . 131 20. Varimax Rotation for Concept BIOCHEMISTRY . . 132 21. Varimax Rotation for Concept EXPERIMENT . . . 133 22. ANOVA for SCIENTIST Across Groups on Factor I . . . . . . . . . . . . . 135 23. ANOVA for LABORATORY Across Groups on FaCtor I O O O. O O O O O O O O O O 136 24. ANOVA for MICHIGAN STATE UNIVERSITY Across Groups on Factor I . . . . . . . . 137 25. ANOVA for BIOCHEMISTRY Across Groups on Factor I O O O O O O O O O O O O O 138 26. ANOVA for EXPERIMENT Across Groups on Factor I O O O O O O O O O O O O O 139 27. ANOVA for LABORATORY Across Groups on Factor II O O O O O O O O O O O O O 140 28. ANOVA for MICHIGAN STATE UNIVERSITY Across Groups on Factor II . . . . . . . 141 29. ANOVA for BIOCHEMISTRY Across Groups on Factor II O O O O O O O O O O O O O 142 30. ANOVA for EXPERIMENT Across Groups on FaCtor II o o o o o o o o o o o o o 143 31. ANOVA for SCIENTIST Across Groups on Factor III O O O O O O O O O O O O O 144 32. ANOVA for LABORATORY Across Groups-on Factor III . . . . . . . . . . . , , 145 33. ANOVA for MICHIGAN STATE UNIVERSITY Across Groups on Factor III . . . . . . . 145 vii Table Page 34. ANOVA for BIOCHEMISTRY Across Groups on FaCtor III 0 o o o o o o o o o o o o 147 35. ANOVA for EXPERIMENT Across Groups on FaCtor III 0 o o o o o o o o o o o o 148 36. ANOVA for LABORATORY Across Groups 0 FaCtOI A o o o o o o o s o o o o o 149 37. ANOVA for EXPERIMENT Across Groups on FaCtor B o o o o o o o o o o o o o 150 38. ANOVA for SCIENTIST Across Subgroups on Factor I . . . . . . . . . . . . . 155 39. ANOVA for LABORATORY Across Subgroups on Factor I . . . . . . . . . . . . . 157 40. ANOVA for MICHIGAN STATE UNIVERSITY Across Subgroups on Factor I . . . . . . . 153 41. ANOVA for BIOCHEMISTRY Across Subgroups on Factor I . . . . . . . . . . . . 159 42. ANOVA for EXPERIMENT Across Subgroups on FaCtor I O O O O O O O O O O O O O 160 43. ANOVA for LABORATORY Across Subgroups on FaCtor II O O O O O O O O O O O O O 161 44. ANOVA for MICHIGAN STATE UNIVERSITY Across Subgroups on Factor II . . . . . . 162 45. ANOVA for BIOCHEMISTRY Across Subgroups on FaCtOr II o o o o o o o o o o o o o 163 46. ANOVA for EXPERIMENT Across Subgroups on Factor II . . . . . . . . . . . . . 164 47. ANOVA for SCIENTIST Across Subgroups on FaCtOI‘ III 0 o o o o o o o o o o o o 165 48. ANOVA for LABORATORY Across Subgroups on FaCtor III O O O O O O O O O O O O O 166 49. ANOVA for MICHIGAN STATE UNIVERSTIY Across Subgroups on Factor III . . . . . . 167 50. ANOVA for BIOCHEMISTRY Across Subgroups on FaCtor III a o o o o o o o o o o o o 168 viii Table 51. 52. 53. Page ANOVA for EXPERIMENT Across Subgroups on FaCtor III I O O O O O O I O O O O O 169 ANOVA for LABORATORY Across Subgroups on FaCtor A O O I O O O O O O O O O 0 l7 0 ANOVA for EXPERIMENT Across Subgroups on Factor B . . . . . . . . . . . . . 171 ix LIST OF FIGURES Figure Page 1. Amount of Visual Field Covered by a Given Image at Selected Distances. . . . . . . 12 2. Organization of the Study . . . . . . . 33 3. Evaluative Discrimination Capacity and Polarity of 40 Scales for 5 Concepts. . . . 40 4. Diagram of Presentation Rooms . . . . . . 47 . Scales Loading on Factor I . . . . . . . 56 5 6. Scales Loading on Factor II. . . . . . . 57 7. Scales Loading on Factor III . . . . . . 59 8 . Summary of Scale Loadings on All Factors . . 61 LIST OF APPENDICES Appendix Page A. Semantic Differential for Pilot Experiment . 100 B. Derivation of Semantic Differential with "Best-WOrst" Technique. . . . . . . ; 103 C. Randomization Procedures . . . . . . . 112 D. Semantic Differential for Experiment . . . 114 E. Factor Loading Matrices and Varimax Rota- tions for Five Concepts . . . . . . . 118 F. Analyses of Variance Layouts for Main Treat- ment Effect . . . . . . . . . . . 134 G. Derivation of Post-hoe Comparison Formulas . 151 H. Analyses of Variance Layouts for Effect of Viewer Location . . . . . . . . . . 155 xi CHAPTER I INTRODUCTION AND THEORY OF THE STUDY Introduction At the 1962 convention of the Department of Audio— visual Instruction at Kansas City, the late Dr. James Finn startled his audience with a multi-screen, general 1 2 session presentation. Although preceded by multi-image and multi-screen techniques in the military, business, entertainment and exhibition fields and by multi-image classroom experiments at the University of Georgia in 1954 and the University of Wisconsin in 1960,3 Finn‘s 1962 presentation was a benchmark in the educational application of the technique. After 1962 multi-image reports and presentations became fairly commonplace at educational meetings and facilities for handling such presentations were built at V_f l"Professional Sights Soar with Finn-Hall Spectacular," Audiovisual Instruction, Vol. 7 (June, 1962), 366-367. ZMulti—image is now being called "multimage" in some quarters. In this paper the hyphenated term will be used, however. 3Richard D. Hubbard, "Telemation: AV Automatically Controlled," Audiovisual Instruction, Vol. 6 (November, 1961), 437-439. various colleges and universities. The importance attached to the technique by Finn can be inferred from this state- ment by Donald Perrin, one of Finn‘s students and associates. In 1963 Finn initiated a course in "designing Large Group and Multi-Media Presentations" as a part of the graduate curriculum for USC‘s [University of Southern California's] Department of Instructional Technology. This course has served as a workshOp and laboratory for the deve10pment of many excellegt presentations of a cognitive and affective nature. Purpose and Scope of the Study This study seeks first to establish that multi—image presentations have more affective impact than similar single-image presentations and then to explore the relation- ship between that effect and the viewer's location with respect to the screen. More specifically, the study undertakes to answer two general questions: 1. Will a multi-image and audio presentation cause greater positive shift in the evaluative meaning of presentation—related concepts than a parallel single-image and audio presentation? 2. Is the magnitude of shift in evaluative meaning related to the amount of the viewer's visual field which is covered by the projected image area as determined by the viewer's distance from the screen? 4Donald G. Perrin, "A History and Analysis of Simultaneous Projected Images in Educational Communication" (unpublished Ph.D. dissertation, University of Southern California, 1969). p. 71. Need for the Study In his recent historical study, Perrin traces the development of multi-screen and multi-image techniques from the 1890‘s to the late 1960's in all fields from pure entertainment to education. Significantly, he remarks, Even though producers have designed multiple image presentations with remarkable skill, the underlying theory has not been verbalized. In 1963 when Allen made his first research study on the use of simul- taneous images in classroom instruction, there were only three prior studies to be found. .‘. . In the past six years the literature has expa ded enormously. The documentation is largely technical and descriptive, and only one new piece of research has been added.5 It is noteworthy that all of the studies cited by Perrin6 7 and by Allen were concerned with the ability of a multi—image presentation to increase cognitive learning more than a parallel single-image presentation. None of them dealt in a specific way with the possible influence on beliefs, emotions, attitudes or other constructs in the affective domain. The complete lack of research in this area is the more remarkable when one considers that people are willing to stand in line for hours to see a twenty-minute presentation such as Labyrinth at Expo 67, essentially an 5Perrin, pp. 88-89. 6Perrin, pp. 88-94. 7Wi11iam H. Allen and Stuart M. Cooney, A Study of the Non-Linearity Variable in_§ilmic Presentation, Report of Title VII Project Number 422 of the National—Defense Education Act. May, 1963 (Los Angeles: University of Southern California), pp. 5-14. impressionistic and affective type of program.which Kappler observed ". . . certainly drives hardest at sensations and emotions."8 The pOpularity of multi-image Spectaculars such as those shown at Expo 67 and the publicity given to light shows and "happenings" in the youth subculture probably had a direct influence in causing media specialists to try similar techniques in education. A mark of this interest was the addition of a "Multimage Festival" at the 1969 Portland Convention of the Department of Audiovisual Instruction (DAVI) as a followup to the well-received ”Media and the Affective Domain of Learning" presentation given at the 1968 DAVI Convention in Houston.9 wallington, Hale, and Conte state in reference to the presentations at the 1969 "Multimage Festival," There seemed to be a high involvement on the part of the audience in the affective domain, and therioseems to be a strong link Wlth the cognitive domain. While there are clear differences of Opinion on the value of the presentations at the "Multimage Festival" and whether or not a given program fulfilled its purpose, there 8Frank Kappler, "The Mixed Media-—Communication that Puzzles, Excites, and Involves," Life, July 14, 1967, p. 28. 9Department of Audiovisual Instruction, Handbook for the DAVI ConventionLiHouston, Texas, March 24-29, 1968 (Washington: DAVI,I968), p. 40. 10Jim Wellington, Pryor Hale, and Joseph Conte, "Multimage Festival," Audiovisual Instruction, Vol. 14 (June-July, 1969), p. 532 T is general agreement that most of the designers seemed to have Operated on the intuitive hunch that the audience becomes more deeply involved affectively when confronted with large screen areas, multiple images, and high fidelity 11 The importance of verifying the sound amplification. affective impact of multi-image presentations is under- lined by the fact that three "Multimage Festival" rooms presented concurrent programs for three days at the 1970 DAVI Convention in Detroit.12 The burgeoning interest in the multi-image technique would seem to call rather urgently for an investigation of its general effects as a first step in studying the principles of design and presentation which should be taken into consideration when it is used. Definitions Before proceeding with elaboration of the theory and hypotheses of the study, the following operational definitions are offered. Evaluative Meaning The discriminative judgment made by an experimental subject on a series of seven-point semantic differential scales of bi-polar adjectives with respect to a given 11This statement is based on extended conversations with individuals who attended the "Festival." 12Department of Audiovisual Instruction, Handbook for the DAVI Convention,§Detroit, Michi an A ril 27-May 1, 1970 (Washington: DAVI, 1970), pp. 24, if, and 55. concept or word. Direction (positive or negative) and the intensity with which the meaning is held is indicated by the location of the judgment from the neutral, central position. (A fuller, theoretical description of the term is given in the next section of this chapter.) Semantic Differential A measurement and scaling technique developed by Charles Osgood and his associates by which people indicate valuative judgments of concepts on seven-point scales of bi-polar adjectives.13 Best-WOrst Technique A modification of the semantic differential suggested by Donald Darnell which determines the evaluative discrim- ination capacity and positive/negative polarity of a given bi—polar adjective scale in reference to a given concept.14 Multi-image and Audio Presentation A program of projected transparencies with music in which the visuals are displayed simultaneously on three adjacent screens. The entire sequence using three slide 13Charles E. Osgood, George J. Suci, and Percy H. Tannenbaum, The Measurement of Meanin (Urbana: University of-Illinois Press, I957 , pp. 25-30. 14Donald Keith Darnell, "A Technique for Determining the Evaluative Discrimination Capacity and Polarity of Semantic Differential Scales for Specific Concepts" (un- published Ph.D. dissertation, Michigan State University, 1964), pp. 78-83. projectors is under the automatic control of a multi-unit programer to achieve identical performances of the program each time it is presented. §i£gle-image and Audio Presentation A program of projected transparencies with music in which the visuals (drawn from the parallel multi-image presentation) are displayed sequentially on one screen. The entire sequence using one slide projector is under the precise and automatic control of a synchronizing unit to insure repeatability. Multi-unit Programer A specially constructed device which discriminates three different sound frequencies on one track of a stereo audio tape to advance independently each of three slide projectors. A3pect Ratio The height to width ratio of a given visual, expressed, for example, either as 2:3 or 2x3 meaning 2 units high and 3 units wide. Field of View or Visual Field The total seeable area of 180° to 200° situated before an observer.' Viewer Location The distance expressed in screen widths (W) of an observer from a projection screen in his visual field. Theory and Rationale for the Study This study involves the use of a multi—image and audio presentation to develop and/or alter the evaluative meaning of presentation-related concepts. In the design of the experiment, the media presentation is the indepen— dent variable and the evaluative meaning of concepts is the dependent variable. In this section the theory and rationale underlying each of these variables and their postulated relationship is discussed in turn. Multi—image Theory Reference was made previously to Perrin's observation that although the underlying theory of multi-image presen- tations had not been Verbalized, producers have shown much skill in creating such displays.15 However, Perrin does suggest three elements which should be included in such a theory. From the existing body of knowledge there appear to be three major areas which distinguish multiple image communication from conventional use of media. These are: 1. simultaneous images 2. screen size 3. information density.16 All three of the elements mentioned by Perrin are involved in the present study. 15Supra, p. 3. 16Perrin, p. 89. First, with reSpect to screen size, it can be said that many of the multi-image presentations for educational purposes seem to have used the technique simply to widen the total image area into an aspect ratio resembling commercial/entertainment formats like Cinemascope, Todd A0, and the like, particularly when the purpose of the presen- tation has been to touch the observer's emotions or some other component of the affective domain.17 However, various production constraints and limitations of the media available-- primarily 35mm slides, 16mm and 8mm film footage--have usually forced the use of multiple rather than panoramic images for wide-screen educational presentations. One of these factors is the practical limit to the width of the projected image that can be achieved with a given piece of film (transparency) material. In this regard, Ben Schlanger has reported, "For average viewing distances, experience indicates that about 10% in. of image width can be projected within acceptable limits, for each millimeter of film width."18 This is an important consideration in the present study because the total image width for the experimental treatment is to be thirty feet, slightly more than the allowable maximum (35mm X 10/12 = 29 feet) under Schlanger's rule were a single image to be used. 17Supra, pp. 3-4, 18Ben Schlanger, "Criteria for Motion Picture Viewing and for a New 70mm System: Its Process and Viewing Arrange- ments," Journal of the Societ of Motion Picture and Television Engineers, Vol. 7§y(March, 1966), 165. 10 Thus, a second of Perrin's elements--multiple images-- is closely related to that of screen size. But the use of multiple images involves other considerations also. For one thing, a tempo and rhythm can be imparted by the rate and pattern of image changes. In the present study the pace or tempo of the presentation is rather rapid to suggest excitement and liveliness and the pattern of image changes is intuitively calculated to suggest an artistic kind of rhythm in contradistinction to the effect which would be obtained with purely random changes and image conjunctions. For another, simultaneously presented images may either complement or contrast one another. In the present study, the visual part of the presentation is intentionally designed with image redundancy, i.e., simultaneously presented images complement and "repeat" the same essential information. The characteristic of image redundancy just mentioned is related to Perrin's third factor, that of information density. While most of the visuals used in this experiment are "dense" in an information theory sense, intuitively it is felt that the use of redundancy across simultaneously displayed images keeps the total amount of information at a level which is probably close to that of any of the visuals taken individually. The rationale thus far has dealt with image size in terms of overall scale. Specifically, in the eXperiment 11 a ten-foot screen for each image is used so that the multi- image presentation covers a total width of thirty feet as compared to an image width of ten feet for the parallel single-image presentation. However, within the sCale of any given screen width, image size is also related to the amount of the Observer's visual field which that image covers and this in turn is governed by his distance from the screen. Figure 1 shows that the amount Of the visual field covered by a given screen width can be eXpressed as double the angle subtended by a line from a given point to either outer edge of the screen. Most recently built theaters have been designed so that the rear seats are not more than 2W from the screen so that observers seated there will have at least 30° of the visual field covered by the image. Personal experience and a pilot experiment suggested the second purpose of the study, namely to determine if the amount of shift in evaluative meaning is influenced by the viewer's distance from the screen. Specifically, it is predicted that the greatest effect will occur when 50° to 70° of the viewer's visual field is covered by the image area or, from Figure 1, when the viewer is located from 2/3 W to 1 1/3 W from the screen along the center axis. Evaluative Meaning Earlier in this chapter an Operational definition of evaluative meaning was Offered. However, because it is actually the dependent variable in this study, it seems 12 Left Screen 1 Center Screenl I NOTES: Right Screen l I=one image (screen) Width w=width of total image area over 3 screens Angles on of center subtended of center (% I). Angles on Of center subtended Of entire right side axis are by % width screen only left side axis are by % width image area covered by 3 screens (8W). At a given location the angle for k I is approximate 1y 73 Of the angle for 8 W. Total area of the visual field covered by given image or screen area is twice the size in degrees of acute angles at the center axis. Figure l.--Amount of Visual Field Covered by a Given Screen at Selected Distances. 13 appropriate to provide a separate and more detailed theo- retical explication of the term and its relationship to the study. The construct Of evaluative meaning as it is used in this study is heavily dependent upon the theory of meaning presented by Charles Osgood and his associates in The_ Measurement of Meaning. It is felt apprOpriate to present the major highlights Of this theory for two reasons: (1) it serves as the underpinning for the present study, (2) the measurement technique based on the theory is also used in the present investigation. In order to follow the theoretical argumentation it is necessary to understand three psychological terms.19 Significate: any stimulus which, in a given situation, regularly and reliable produces a predictable pattern Of behavior. Sign: a stimulus other than a significate which evokes in an organism the same reactions evoked by a significate. Assign: a sign whose meaning has been "assigned" via association with other signs rather than via direct association with significates of those signs. The definitions for the first two terms are fairly standard in the psychological literature. However, the notion of an assign and the linkage between a sign and an assign provide the key to Osgood's definition of meaning. In Osgood's view, a sign comes to elicit a response formerly 19All three of these definitions are paraphrases of statements made by Osgood and his associates in The Measure- ment of Meaning (Urbana: University of Illinois-Fress, 1957 ' pp. 5-80 14 elicited by a significate by an internal mediating process within the organism. He says, Whenever some stimulus sign other than the significate is contiguous with the significate, it will acquire an increment Of association with some portion of the total behavior elicited by the significate as a representational mediating process.20 The "representational mediating process" is they organism's internalized association of a sign with a signi- ficate and, thus, is the learned meaning of that sign for the organism. Whenever such learning takes place via association between signs without direct association with significates, meaning is "assigned." It follows, then, that Variation in meaning should be particularly character- istic of assi ns since their representational processes [meanings] depend entirely upon the samples Of other signs with which they occur. 1 This theoretical notion has important implications for the present study which, in fact, seeks to develOp assigns via a media presentation. Specifically, the idea is to produce and/or alter the assigned meaning of several concepts held by human subjects by presenting them with a series Of signs--projected visuals and recorded music. In order to explain how this might be accomplished with a rel- atively short treatment and how such an effect might be measured, it is necessary next tO consider Osgood's "logic of semantic differentiation." 20Osgood, gt gl., p. 6. 21Osgood, 23 31., p. 9. 15 A precise explanation of semantic differentiation and its theoretical components is provided by the following statements from The Measurement of Meaning. We begin by postulating a semantic s ace, a region Of some unknown dimensionality and Euclidian in character. Each semantic scale, defined by a pair of polar (Opposite-in-meaning) adjectives, is assumed to repre- sent a straight line function that passes through the origin of this space, and a smaple Of such scales then represents a multidimensional space. . . . . . . . When a subject judges a concept against a series of scales, each judgment represents a selection among a set of given alternatives and serves to localize the concept as a point in semantic space. . . . By semantic differentiation, then, we mean the succes- sive allocation of a concept to a point in the multi- dimensional semantic space by selection from among a set Of given scaled semantic alternatives. Difference in the meaning between two concepts is then merely a function Of the differences in their respective allo- cations within the same space, i.e., it is a function Of the multidimensional distance between the two points. Thus, a subject differentiates between concepts by making judgments on bi-polar scales which indicate direction and distance from the origin of the postulated semantic space. Osgood then completes the theory by tying his con- ceptualization Of "meaning" (representational mediation . process) to the idea of semantic differentiation as follows: Let us assume that there is some finite number of representational mediation reactions [meanings] available to the organism and let us further assume that the number of these alternative reactions (excitatory or inhibitory) corresponds to the number Of dimensions or factors in the semantic space. Direction Of a point in the semantic space will then correspond to what reactions are elicited by the sign, and distance from the origin will correspond to the intensity of the reactions. 22 23 Osgood, 9.": _a_1_., pp. 25-26. Osgood, gt 31., p. 27. 16 SO if a subject indicates his true judgments of a concept on a set of bi-polar scales and if the set of bi-polar scales are, in fact, associated with the concept as far as the subject is concerned, then it can be said that the subject's judgments provide a quantifiable description Of the concept's meaning for him. Of course the description so obtained is only a part of the total meaning of the concept since the scales are but a sample of the entire pOpulation of possible scales which are relevant to the given concept. Because a subject cannot make ratings on the entire population of scales relevant to a concept, Osgood's measurement model and its dependence on representative sampling must be considered. In this regard he says, The essential Operation Of measurement is the suc- cessive allocation of a concept to a series of de- scriptive scales defined by polar adjectives, these scales selected so as to be representative of the major dimensions along which meaningful processes vary. In order to select a set of scales having these prOperties, it is necessary tO determine whgs the major dimensions of the semantic space are. After a detailed presentation of various studies using factor analytic techniques to explore the dimensionality of semantic space, Osgood and his associates conclude, For one thing, it is clear that it is a multidimen- sional space. In every analysis more than three factors have been contributing to the meaningful judg- ments by subjects. It is also clear that these N factors or dimensions are not equally important Ih mediating judgments, or perhaps better, are not 24Osgood, g£_gl., p. 31. 17 equally used by subjects in differentiating among the things judged. Three factors appear to be domi- nant, appearing in most of the analysgg made and in roughly the same orders of magnitude. The three recurring factors have been named "evaluation," "potency,' and "activity" because those terms provide general descriptions or labels of the scales which usually load upon them. It has become customary to refer to these named factors as the major dimensions of semantic space. Of the three major dimensions, that Of evaluation is the most dominant. Osgood reports, A pervasive evaluative factor in human judgment regularly appears first [in factor analysis] and accounts for approximately half to three-quarters of theextractable variance. This study undertakes to produce and/or alter the meaning of certain concepts with a specific kind of audio- visual treatment. As such it falls squarely within Osgood's learning-theory conceptualization of meaning as a "representational mediation process" because the experi- mental manipulation is designed tO form assigns through new associations of sig2§_(audiovisual elements). Further- more, the resulting assigns are to be indexed with Osgood's measurement technique of semantic differentiation, parti- cularly on the dimension which he has labeled evaluation. Although further reasons and the methodology for measuring the meaning of these assigns on the evaluative dimension are zsongOd' _e_t. 22.-O ’ pp. 71‘720 26Osgood, gt_al., p. 72. 18 set forth in the section on instrumentation in Chapter Three, it can be mentioned here, (1) that this dimension is most closely related to the purposes of the audiovisual presen- tation/treatment, and (2) that this dimension is likely tO detect differences since it usually accounts for twice as much of the variance as any other factors or dimensions in studies using the semantic differential technique. Assumptions Of the Study ‘ifi This study and its experimental design rest on several assumptions. Some of these prerequisite postulates have been referred to or implied in the previous discussion but they are explicitly restated in this section along with those which have not yet been touched upon. First, the rationale for the entire study rests upon the assumption that mediated presentations can be shown experimentally to convey affective impact. When this basic assumption of the study is particularized to multi-image presentations, it becomes the nuclear idea of the main research hypothesis. Second, it is assumed that evaluative meaning, a com- ponent of the affective domain, can be indexed by a paper and pencil measurement technique. The explanation of the construct evaluative meaning given in the previous section of this chapter is clearly dependent upon the theories developed by Charles Osgood as is the measurement technique 19 of semantic differentiation. The development of the spe- cific instrument, explained in Chapter III, is further predicated upon special procedures for deriving an instru- ment which indexes the meaning of concepts mainly on the evaluative dimension. The third assumption--that the first logical step in investigating the affective impact of multi-image presen- tations is to show the existence of a general effect—- follows from the assumptions already stated. There are many independent variables which are of interest to the message designer and which may either enhance or diminish the affective impact of multi—image presentations. How- ever, it appears necessary and appropriate to demonstrate the general affective impact Of such presentations as they are currently designed before investigating specific inde- pendent variables. Next, it is assumed that the experimental population is sufficiently like the "target" population for whom the multi—image treatment presentation was originally designed to allow the treatment to demonstrate its impact. The investigator had to compromise somewhat to Obtain a popu— 1ation with a sufficient number of experimental subjects at the right time, but it was felt at the outset that the experimental and "target" populations are roughly equal with respect to knowledge of and feeling for the content Of the multivimage presentation prior to seeing it. 20 Also with regard to the experimental population, it is assumed that its size is sufficiently large to perform the experiment. Ideally, each of the three treatment groups required by the experimental design should have thirty or more people, but it is felt that a moderate reduction to twenty-five or so should not seriously impair the results, especially for the main research question. Finally, as stated previously, it is assumed that a parallel, single-image presentation equivalent in informan tion content can be assembled on an intuitive basis from the original multi-image presentation. That is, it is felt that the experimenter can select the visuals for the parallel, single-image versiOn from those of the original multi-image presentation so that both will convey essen- tially the same information. Limitations of the Study There are several limitations of the study which should be mentioned. Two are imposed by the investigator to delimit the study and two arise from general research principles. The investigator has chosen two independent treatment variables--multiple images and coverage Of the visual field by the image area——for study in the experiment. Other independent variables which might be of interest (such as rate of change of images and number of images) are left 21 for later experimentation. Also, while two sense modal- ities (sight and hearing) are involved with the treatment presentation, the study is not concerned with issues of multi-channel or cross-channel communication. Consonant with customary research procedures, the results Of the experiment will not be generalizable beyond the experimental population, except in so far as other pOpulations are not unlike the experimentalpopulation. And the experimental results will not be generalizable from the experimental treatment presentation to other multi-image presentations. In short, the results Of this experiment will provide but one piece of evidence for or against the general effect being investigated. CHAPTER II REVIEW OF THE LITERATURE In this chapter pertinent studies which are related to the present investigation are reviewed. Consistent with the purpose, rationale, and theory presented in Chapter I, this chapter is divided into three main sections: affective impact studies, image size studies, and multi- image studies. Affective Impact Studies Audiovisual media have been involved in attitude change studies and other investigations in the affective domain in two ways. First, various media, such as audio tape recordings and film, have been used as stimulus materials because their characteristic of repeatability allows controlled replication in experiments designed to test theories Of attitude change. Studies of this class are considered beyond the sc0pe of this review since their attention is focused upon message variables such as order of presentation Of argument, source credibility, and other manipulatable 252$ elements. Second, some studies have centered on the efficacy of media or specific elements of a medium to induce attitude change. This second class 22 23 of studies is more closely related to the present investiga— tion and pertinent experiments of this type will be reviewed briefly. In a comparatively recent study, Edling summarizes previous experiments assessing the amount of attitude change associated with specially-designed, mediated messages by saying, The consistent finding in these studies has been that initial attitudes are related to the acceptance and retention of new associations, i.e., information per- ceived as congruent to existing attitude is more readily accepted, and retained longer, than informa— tion perceived to be contradictory to initial atti- tudes. . . . By identifying strongly held attitudinal Objects and associating new concepts with them, it appears possible to modify attitudes toward objects not originally highly regarded, and to do so via media, both quickly and effectively.1 Although Edling was speaking specifically in ref- erence to audiences of school children, presumably through senior high school age, there is sufficient evidence in the literature to suggest that the generalization applies to older peOple as well. Merrill found in an experiment with adult males (military reservists) that an attitude film arousing strong fears produced "defensive avoidance“ and prevented attitude shift.2 1Jack V. Edling, Experiments with Educatignal Media Desi ned to Modify Attitudes (Final Report for Project NO. 2-2354 USOE Dept. Of'H.E.W. and Teaching Research Division of the Oregon State System of Higher Education, 1968), p. 9. 2Irving R. Merrill, "Attitude Films and Attitude Change," AV Communication Review, Vol. 10 (January-February, 1962), p. 13. 24 Kumata and Berlo found that college students who listened to a satirical radio drama about Senator Joseph McCarthy entitled "The Investigator" shifted in a negative direction toward congressional investigations and in a positive direction toward the Senator. The eXperimenters suggest that the first finding resulted from "a change in the context of the perceived Object" and represented a successful attack on attitudes related to that Object. The second finding is described as a boomerang effect due to the one-sidedness of the message against the Senator which called into play "the dominance of the pressure toward impartial, fairly presented analyses" which were even stronger for these higher education students than their initial dislike of Senator McCarthy.3 The negative or boomerang results reported by Kumata and Berlo indicate that communication effects in the affective domain are somewhat difficult to predict and control because Of human variables and previous attitudes, especially when those attitudes are covertly held and are not apparent to the investigator. The caution and pre- cision required in this area is also pointed up by an experiment conducted by Miller on the effect of "motion" 3David K. Berlo and Hideya Kumata, "The Investigator: The Impact Of a Satirical Radio Drama,“ Journalismguarterly, Vol. 33 (Summer, 1956), 287-298. ‘25 in film upon attitudes.4 Because this experiment is one of the few which attempted to measure the attitudinal effect Of projected moving and still images, it is parti- cularly relevant to the present investigation and it will be described in some detail. Miller's experiment was designed to test the formal property of motion in film. He states the purpose of the study as follows: This study was concerned with measuring affective (emotional) response to a formal quality (motion) of a communication medium (film) and its effect on infor- mation recall and attitude formation.5 The basic experiment pitted an experimental treatment group who saw a 16mm film entitled Corral against a second treatment group who saw a filmograph6 version of the 16mm film and two groups which each saw one-half of each version, but in opposite orders. Measures used were: a fifteen-scale semantic differential on four general, 4William C. Miller, "An Experimental Study of the Relationship of Film Movement and Emotional Involvement Response, and Its Effect on Learning and Attitude Forma- tion" (unpublished Ph.D. dissertation, University of Southern California, 1967) and Final Report of NDEA Title VII Project Number 5-1731. 5William C. Miller, "Film Movement and Affective Response and the Effect on Learning and Attitude Formation," AV Communication Review (AVCR), Vol. 17 (Summer, 1969), 172. 6Miller defines a filmograph as follows: "A filmograph is a series of still frames on motion picture film, each printed repeatedly a predetermined number of times so that the time each still scene appears in the film is controlled by the normal speed of projection" (AVCR Article, p. 173). 26 film-related concepts; a 22-item Likert-type attitude scale; and galvanic skin response (GSR) recordings. Miller summarizes the three major hypotheses as follows: It was hypothesized that film motion would, of itself, create audience emotional involvement response as measured by GSR, and that this would produce positive audience response, but would not be a significant factor in information recall.7 With respect to the first hypothesis, Miller's results indicate that the arousal of the motion-only group was significantly greater only for the climax portion Of the film but that GSR is a useful tool in measuring audience involvement,8 in spite of the methodological problems in interpreting GSR profiles reported by Becker,9 Levonian,10 11 himself. It must be remembered, also, that and Miller GSR is an individual measure and requires a measuring and recording device for each subject receiving an experimental treatment. Proceeding to the third hypothesis, Miller's hunch that arousal and recall of information would not be related 7Miller (AVCR), p. 173. 8Miller (AVCR), pp. 177 and 179. 9Samuel L. Becker, The Relationship of Interest and Attention to Retention and Attitude Change (Iowa City: University of Iowa, 1963). 10Edward Levonian, Measurement and Analysis Of Ph sio- logical Responsgto Film (Los Angeles: University of Southern California, 1962), Final Report, Grant NO. 704094, NDEA Title VII project No. 458. 11Miller (dissertation). pp. 94—97. 27 was correct--although one must add that the "null" nature of this theoretical hypothesis requires caution in interpretation Of the finding.12 Miller's second hypothesis, that Of positive audience response as measured by the SD, is of the most interest for the present investigation. Miller reports, "Quite to this investigator's surprise, the film generally affected negative changes on this the SD measure,’ and again, "The data indicate that the film, in both motion and filmograph versions, affected greater negative attitude change to most concepts tested."13 One explanation for this finding according to Miller is that the necessity Of running the experiment in the psychiatric building of a Veterans hospital where the re- quired six GSR units were available unsettled the experi- mental subjects.l4 Miller Offers as a second explanation that the subjects may also have expected something more "shocking" in the way of a presentation because of the somewhat extraordinary transportation arrangements and because of the fee they were paid for participation.15 12Miller (AVCR). p. 179. 13Miller (dissertation), pp. 151 and 154. 14The location Of the experiment in the psychiatric ward is mentioned in the dissertation but not in the journal article (cf. pp. 98-101 Of the former with p. 175 of the latter). Such differences in the two reports explain quoting from both sources. 15Miller (dissertation), p. 103. 28 However, in the judgment of the present reviewer, the "boomerang" results on the SD measure may also have been caused by two other methodological artifacts. First, Miller apparently made no effort to check the relevancy Of the SD scales to the concepts Of interest, even though he acknowledges Darnell's finding of concept-scale inter- action.16 Second, all groups received the SD as a pretest and as a posttest measure, and thus may have been negatively influenced by the repeated exposure in cOmbination with the treatment. With respect to the present study, the Darnell "Best-Worst" technique will be used to eliminate the first problem and a posttest-only design will be used to avoid the second problem. Image Size Studies It seems remarkable that so few experimental studies have manipulated the seemingly Obvious variable of image size. The scarcity of research on this variable may account for the fact that most studies make no mention at all Of image size or the distance Of viewers from the screen--a related variable which accounts for the portion of the field Of view occupied by a stimulus display--even though this variable may conceivably account, at least parti- ally, for the results Obtained when other variables are manipulated. 16See Chapter III of this study for a fuller explana~ tion of concept-scale interaction and the Darnell "Best- Worst" technique for determining the relevance and polarity of given scales to given concepts. 29 In a thorough search of the literature only two studies were found which directly investigated the effect Of the size of a projected image. Using images projected with a rear-screen technique, Ash and Jaspen concluded that most learning occurred with subjects seated in a cone 60° wide (30° on each side of the center line) and 18 feet (12 screen widths) deep and that losses in learning can be minimized outside the Optimum viewing area by expanding the area along the angular sides rather than the base Of the cone, thus keeping viewers as close to the screen as possible.17 As implied by its name, the Telekit screen used by Ash and Jaspen was very similar in size to the screen of a classroom television monitor. In discussing the Optimum 18 indicates that conditions for viewing television, McVey a distance of 12 screen widths is usually acceptable as a maximum distance--a recommendation probably drawn (but not so stated) from the Ash and Jaspen study. With respect to a front-projected image, McVey recommends 6% screen widths as the Optimum viewing distance since the display will then subtend an arc of 9° and, "Studies show that the l7Philip Ash and Nathan Jaspen, "Optimum Physical Viewing Conditions for a Rear Projection Daylight Screen," Technical Report No. SDC 269-7-37 Of the Pennsylvania State University Instructional Film Research Program (Port Washington, N. Y.: U. S. Naval Training Devices Center, Office of Naval Research, 1953): p. 9. 18 Gerald F. McVey, "Where DO We Sit?" Educational Television, Vol. 1 (December, 1969), 25. 30 eye moves in well-dispersed patterns of fixation when watching a visual display that subtends a visual angle of 9°."19 Although McVey does not cite it, a study by Enoch did indeed show a change in eye fixations as the display size dr0ps under 9°. However, Enoch makes clear that this finding really applies to static displays to be used in visual search tasks--in his experiment locating a 20 specific design in a black and white aerial map. In view of these considerations, one might question the application of the finding to most television material, to say nothing of large-screen projections, especially of material designed to produce non-cognitive effects. In a more recent study, Reynolds attempted to deter- mine the effect of viewer distance upon presentation- induced anxiety.21 Using the film Subincision, which shows circumsion rites as practiced by certain African natives, Reynolds found that the level of anxiety was more related to the type of stimuli than to viewer distance from the screen. In applying this finding, however, it should be kept in mind that the stimulus film showed scenes which 19McVey, p. 25. 20Jay M. Enoch, "Effect of the Size of a Complex Display Upon Visual Search," ggurnal Ofithe Optical Sociepy of America, Vol. 49 (March, 1959), pp. 281 and 285. 21James C. Reynolds, "The Effect of Viewer Distance on Film Induced Anxiety," in Dissertation Abstracts, Vol. 29 (Ann Arbor, Michigan: University Microfilms), p. 334l-A. 31 very likely were distasteful to the audience--the converse of the intention in the present study. In the previous chapter, reference was made to Perrin's rudimentary theory of large images. Two Of the elements, it will be recalled, were the visual task factor and the visual impact factor. These elements might be subsumed in the notion of "attention value" of an image investigated much earlier by Adams. Basically, Adams found that atten- tion is approximately proportional to the square root Of the size of an area or stated conversely, quadrupuling the image size doubles its attention value.22 Of course there are practical limits to the absolute maximum image size one can attain because of the physical constraints Of viewing rooms, screens, and film-light-lens sytems. But there are theoretical limits to image size also, as Wagner convincingly argues, because the media should attempt to produce an illusion of reality. He continues in commenting about motion pictures in particular, The motion picture is, after all, a performance--a highly complex symbol system. Its power is not in the fact that it reproduces reality, but rather that it intensifies, abridges, and reorganizes the real world, focussing the attention of an audience on significant details, moving the spectator through an arranged, selective seguence of visual cues in a highly garealistic way. 3 22H. F. Adams, Advertising angIts Mental Laws (New York: Macmillan, 1921), p. 107. 23Robert W. Wagner, "The Spectator and the Spectacle," AV Communication Review, Vol. 3 (Fall, 1955), 298. 32 This contention would seem to be especially appropriate to sequences designed to have impact in the affective domain as does the treatment presentation of the present study. Multi-Image Studies In the multi—image area there is also a paucity of studies and those which are reported in the literature were concerned with the learning of either motor skills or cognitive information. Nevertheless, for the sake of completeness, their findings will be discussed briefly. Reference was made in Chapter I to Perrin's statement that in 1963 Allen found only three prior studies in the literature pertaining to simultaneous images.24 A close 25 failed to reveal either such reading of the Allen report a summary statement or the specific studies to which Perrin referred. But the most likely candidates seem to be Roshka,26 Reed,27 and Malandin. However, since Roshka used actual concrete Objects and Reed used nonsense "concept" 24Supra, p. 3. 25William H. Allen and Stuart M. Cooney, A Stud of the Non-Linearity Variable in Filmic Presentation (Los Angeles: University of Southern California,il963). 26A. Usloviia Roshka, "Conditions Facilitating Abstraction and Generalization." VOP PSIKHOL, 4 (6), 1958, pp. 89-96 as reported by I. D. London,\Psychological Abstracts, Vol. 34 (1960). p. 85. 27H. B. Reed, "The Learning and Retention of Concepts: V, The Influence of Form of Presentation," Journal of Experimental Psychology, Vol. 40 (1950), PP. 504-511. 33 syllables and real words on cards as stimulus materials, they are really outside the parameters of interest in a study focusing upon projected images. Malandin, on the other hand, did use projected materials in his two studies. Because these mimeographed reports in French are unavailable to the present investiga- tor, it is necessary to rely on secondary sources for Malandin's findings. Specifically, Allen reports, Malandin . . . found in his studies that younger students could not relate one image to another if they were isolated in time, that is, presented sequentially. 28 In their own experiment Allen and Cooney found that simultaneous presentation Of multiple images was more effective than sequential (linear) presentation for sixth but not eighth grade students when the treatment was factual and conceptual in nature. The investigators further concluded, (1) Mode of presentation has less effect on learning as the student grows Older. (2) Ability to comprehend and to apply subject matter Of this type taught in these ways [apparently either linear or multi-image] improves with age.2 After the Allen and Cooney experiment there is a hiatus in multi-image studies until 1969 when two appeared. First, Lombard found only a significant difference for girls 28Claude Malandin, "Grouped and Successive Images," Centre d'Etudes st de Recherches pour la Diffusion duFrancais (C.R.E.D.I.F.), Ecole Normale Superieur de Saint-Cloud, France, (Carbon of original in French, undated) as summarized in Allen and Cooney, p. 109. 29Allen and Cooney, p. 108. 34 in learning synthesis skills in eleventh grade U. S. History in favor of a three-screen instructional presentation and even that difference, Lombard suggests, might be accounted for by a significant F value (pre-experiment) between the sub-groups of girls in the two treatment and one control groups.30 Second, Olsen determined that the addition of film to slides and audio tape enhanced the learning Of motor skills but that multi-sensory (audio and visual com- ponents) and multi-image modalities did not significantly affect cognitive learning.31 3oEmanuel S. Lombard, "Multi-channel, Multi-image Teaching of Synthesis Skills in Eleventh Grade United States History" (unpublished Ph.D. dissertation, University Of Southern California, 1969), abstract. 31John R. Olsen, "The effect Of Multi-Stimuli Presentations on Learning Gain," in Dissertation Abstracts, VOl. 30 (Ann Arbor, Michigan: University MicrOfilmsT. CHAPTER III DESIGN OF THE STUDY Introduction This study investigated the effect of a multi-image presentation upon the evaluative meaning of presentation- related concepts held by human subjects. As shown in Figure 2, the study was divided into three phases. In this chapter, first the overall design and methodology of the experiment are discussed briefly to provide a gen? eral background. Next, the two pre-experiment phases for the selection Of the experimental presentation and the derivation Of the criterion instrument are explained in turn. Finally, the detailed procedures followed in phase three, the experiment itself, are presented. Overall Design and Methodology The experimental design to test the general hypothesis that a multi-image presentation has an effect on evaluative meaning pits two treatment groups against each other and against a control group. Treatment Group 1 received a ten- minute multi-image and audio presentation on Biochemistry at Michigan State University. Treatment Group 2 received a parallel, ten-minute single-image and audio presentation 35 36 .hpsum Osu mo coflumuflcmmuonn.m Ousmwm mmsouu Houusoo H a pass nummus m as mm s as name mmasom was mummocou am Home msOflu ImucmmOHm momfiw IHuHOE 03» spas unmeflummxm poaflm ZOHB IUZDh oan .Hmucfiz U coflumasmom unmadummxm oan .HmvSHB m coflumasmom usmEHHmmxmlmnm mama .Hamm d soflumasmom pcmEHummeImum ZOHB ImHmUmmD m mmwummmcnm>wuwmom xmmsumsouum humsuummao gussnusmsss Hammcsluwmm Omamwumouu Mpuwpncmwao vantages; xosmussussws mmmanuuoslwanmsam> mnemonlmnwummumucw mamsflucolmHOEwu ucmuuomawcslpnmuuome O>Hmmmmlm>wuom cmnlcoom mHAdUm .HH .mH .mH .m .0H .hH .mH .N .m .h .om .mH .va .m .MH .H .NH .m .v 65 hb.mh mm.mm mommnm>¢ vwon. vo.om mm.Hm HMNH. mvNH. coma. mHHN. ucmeHmmxm mess. ms.ms mm.ms I mmma. omma. some. ssumssmsommm mess. ms.mw sm.mm I sass. mass. msom. .mmm mmss. as.am ma.ms . mesa. mass. ANNA. Nmsm. snousuommm swam. mo.os s~.sm . mean. ammo. mmmm. mosummm o> asses o> so A s> so A .sssxm. HHH HH H mmososs OOGMflHm> mo msOfluuomoum .HOH pmvsoooud 0> can B> mo mmmucmonmmul.a mamas 66 single-image presentation upon evaluative meaning as measured by a specially—developed SD instrument. Assum- ing unidimensionality of the SD, it was originally in- tended to sum across all twenty SD scales to derive a total score for each subject on each concept and to per- form a separate analysis of variance on these summed scores for each concept. However, in the preceding sec- tion it was shown with factor analytic techniques that the SD scales loaded on Eh£g§_main factors, eliminating the unidimensionality assumption. Thus, sixteen (rather than five) analyses of var— iance were performed in relation to each of the two re— search questions. A separate one-way analysis of vari- ance (ANOVA) was performed to correspond with each column in Figures 5, 6, and 7 plus an "extra" one for Factors A and B which emerged for the concepts LABORATORY and EXPERIMENT . In each case, each subject's judgments for the scale/concept items designated with an "X" in Figures 5, 6, and 7 were summed to derive a total "factor" score on that concept. These summed scores were then used as transformed data for the ANOVA's by treatment groups. Main Treatment Effect ANOVA Results.--The first question posed in this study asked whether a large-screen, multi-image 67 presentation would cause more positive shift in evaluat- ing meaning of five concepts than a parallel single-image presentation. The data for each of the Sixteen ANOVA's (one for each concept on each factor) are given in com- plete form in Appendix F and are summarized in Tables 2 through 5. As shown in Table 2, the F ratios for the concepts MICHIGAN STATE UNIVERSITY (MSU) and EXPERIMENT on Factor I were significant at the .10 level of confidence or better. In Table 3 it is seen that the F ratios for MSU and BIOCHEMISTRY were highly significant (alpha equals .007 and .003 respectively) and that for EXPERIMENT was signif- icant at better than the .10 level of confidence. (It should be noted that an ANOVA for SCIENTIST on Factor II was not performed because of the low scale loadings on that concept.) Table 4 shows that the F ratios for SCIENTIST and BIOCHEMISTRY were significant at the .056 and .04 levels of confidence respectively. Finally, the F ratio for the "extra" factor shown in Table 5 on LABORATORY was significant at little better than the .20 level of confidence. Post-hoc Comparisons.--It would appear reasonable to determine which of the three treatments, if any, are associated with the F ratios significant at the .10 level of confidence or better--the underlined values in Tables 68 TABLE 2.--Group Means and Results of ANOVA's on Factor.I. Means Concept Group 1 Group 2 Group 3 F Ratio aF (n=26) (n=24) (n=21) 801 36.15 33.29 32.76 1.87 .161 LAB 46.15 43.96 43.33 .726 .487 MSU 53.73 51.58 46.33 3.396 4939_ BIO 23.42 23.04 20.81 1.96 .149 EXP 34.19 30.37 30.14 2.48 5221 TABLE 3.--Group Means and Results of ANOVA's on Factor II. Means Concept Group 1 Group 2 Group 3 F Ratio o‘F. (n=26) (n=24) (n=21) SCIa LAB 22.04 20.50 19.90 1.83 1.69 MSU 16.08 16.00 13.09 §;§2. .007 BIO 38.35 34.42 31.38 §;32_ L293_ EXP 27.46 24.87 24.33 2.65 078 aThe low scale loadings on this concept for Factor II did not warrant an analysis of variance. 69 TABLE 4.--Group Means and Results of ANOVA's on Factor III. Means Concept Group 1 Group 2 Group 3 F Ratio o‘F (n=26) (n=24) (n=21) SCI 38.88 35.12 34.24 3.01 .056 LAB 16.42 15.12 14.95 1.52 .225 MSU 26.96 27.04 24.71 1.17 .316 BIO 22.35 19.83 19.38 3.277 .044 EXP 15.65 13.71 13.95 2.238 .114 TABLE 5.--Group Means and Results of ANOVA's on Factors A and B. Means Concept Group 1 Group 2 Group 3 F Ratio OLI“ (n=26) (n=24) (n=21) LAB (A) 22.54 20.79 20.71 1.68 .193 EXP (B) 15.46 14.75 14.43 .51 .601 —__‘._ 70 2, 3, and 4. The Sheffe method of post-hoc comparisons of the differences between the means was used in this secondary analysis. Details concerning the derivation of the necessary formula and its application are given in Appendix G. Since it was not possible to perform an omnibus test via analysis of variance at a pre-specified alpha level for the main effect and since the computer results of the analyses of variance provided the specific alpha level of each F ratio, it was decided to use the Sheffe method to determine for each comparison the minimum value for signif- icance at the specific alpha level of each F ratio. The minimum value (absolute difference between two means) for each comparison is entered in Table 6 in parentheses below those differences which by initial inspection appeared to be great enough to warrant a post-hoc comparison. As shown in Table 6, none of the post-hoc comparisons was significant at the alpha level for the corresponding F ratio. Interpretation of Results.--When Table 6 is compared with Tables 2 through 5, it is seen that seven of the total of fourteen analyses of variance performed on the three most important factors produced F ratios significant at the .10 level of confidence or better. (Since this study is rather exploratory in nature and since there are 71 TABLE 6.--Post-hoc Comparisons of Group Means. Differences Between Concept F. aF MSwithin Group Means Ratio 1-2 1-3 2-3 Factor I MSU 3.40 .039 96.73 2.1 7.4 a 5.2 (7.5) (7.6) EXP 2.48 .091 51.24 3.8 4.0 .2 (4.5) (4.7) Factor II MSU 5.39 .007 11.91 .08 2.99 2.9 (3.3) (3.4) BIO 6.39 .003 44.89 3.9 6.97 3.0 (6.7) (7.0) (7.1) EXP 2.65 .078 25.64 2.6 3.1 .54 (3.2) (3.4) Factor III SCI 3.01 .056 49.07 3.8 4.6 .9 (4.8) (5.0) BIO 3.28 .044 19.00 2.5 2.97 .4 (3.1) (3.3) significant at a a I I 0 Minimum absolute difference in the same row. for comparison to be 72 no dire consequences to result from making a Type I error, it did not seem unreasonable that alpha levels up to .10 warranted further investigation with post-hoc com- parisons.) It is also seen that of the "significant" F ratios, two were for each of the concepts MICHIGAN STATE UNIVERSITY, BIOCHEMISTRY, and EXPERIMENT and that the most significant F ratios were obtained for the concepts MICHIGAN STATE UNIVERSITY and BIOCHEMISTRY. It may be that F ratios with more significant alpha .1evels did not result for all the concepts because not all the concepts are equally related to the treatment presen- tations. The investigator simply selected those concepts for testing which he felt sure were embodied in the pre- sentation--a customary procedure in studies using the semantic differential technique. It is felt that MICHIGAN STATE UNIVERSITY and BIOCHEMISTRY are the most likely of the five concepts which would haVe been shown to be most strongly related to the presentations if a pre-experiment trial had attempted to select the concepts empirically. Granted the assumption that MICHIGAN STATE UNI- VERSITY and BIOCHEMISTRY are the most centrally related to the presentation, one can say that there is evidence for a systematic treatment effect in the experiment. Further support for this assertion is gained from the fact that the Group means are ordered as predicted with but one minor exception. That is, as shown in Tables 2 73 through 5, the means are arranged across the rows from high to low from Group 1 (multi-image) to Group 3 (con- trol), respectively. But the crucial theoretical prediction was that there would be a significant difference between the means of the two treatment groups which received the Bio- chemistry presentations thus showing conclusively the superior effect of the multi-image version. Unfortunately, the post-hoc comparisons of the Group means failed to show at the necessary levels of confidence that whatever sys- tematic treatment effect was present can be ascribed to the multi-image treatment. Furthermore, the difference between the means of Groups 1 and 3 were consistently greater than either the differences between Groups 1 and 2 or that between Groups 2 and 3. Had the difference between Groups 1 and 3 been significant, it could be concluded only that the multi- image presentation was more effective than the parallel single-image presentation when both are compared to the control presentation. Lacking even that finding, it is clear that there is no statistical support for a positive answer to the first research question: will a multi-image presentation cause more positive shift in evaluative meaning of presentation-related concepts than a parallel single-image presentation? 74 Effect of Viewer Location The reader will recall that the second research question asked if the amount of the observer's visual field covered by the overall image (as determined by his distance from the screen) would influence the amount of shift in evaluative meaning for those subjects who re- ceived the multi-image presentation. It was convenient to deal with this question with respect to Treatment Group 1 only so that its investigation is not dependent upon a positive answer to the first research question about treatment effect per se. ANOVA Results.--In order to get at this question, the subjects in Treatment Group 1 were divided into Sub- groups A, B, and C which were located at presumed Optimum, acceptable, and marginal distances from the screen. Thus, analyses of variance were performed across the three Sub- groups in the same way as for the three Treatment Groups. Since the same responses to the SD instrument were used in this second analysis, the ANOVA's were again performed on scores summed over scales with high loadings on the fac- tors for each concept. Thus, sixteen ANOVA's were per- formed exactly parallel to the treatment-effect analysis but with data only from Treatment Group 1. The complete layout of each ANOVA is given in Appendix H and the results are recapitulated and 75 summarized in Tables 7 through 9. Looking at these tables, it is seen that only four of the sixteen ANOVA's produced F ratios significant at least at the .10 level of con- fidence and only three of the remaining F ratios were significant at the .25 level or better. Three of the F ratios significant at the .10 level were obtained in ANOVA's for Factor III, one for Factor II, and none for Factor I. Post—hoe Comparisons.--The Sheffe post-hoc compari- son method was applied to the Subgroup means for the F ratios significant at or below .10. The results of these comparisons are given in Table 11 where it is seen that none of the differences between the Subgroup means was significant at the same alpha level as the corresponding F ratio. Again in this table the minimum absolute value is entered in parentheses below the obtained arithmetical difference which appeared large enough by inspection to warrant a comparison. Interpretation of Results.--The data bearing on the question of the effect of viewer location contain two Surprising findings. As shown in Tables 7 through 9, none of the four F ratios significant at the .10 level of confidence was for the concept BIOCHEMISTRY. It was anticipated‘that the effect of viewer location would be mOSt pronounced on it, the concept presumed most central 76 TABLE 7.--Subgroup Means and Results of ANOVA's on Factor I. Means Concept Subgroup A Subgroup B Subgroup C F Ratio aF (n=9) (n=9) (n=8) SCI 36.55 35.11 36.87 .212 .810 LAB 46.89 44.55 47.12 .301 .743 MSU 53.89 51.22 56.37 1.024 .375 BIO 23.00 23.33 24.00 .109 .897 EXP 33.44 31.78 37.75 1.827 .183 TABLE 8.--Subgroup Means and Results of ANOVA's on Factor II. Means Concept Subgroup A Subgroup B Subgroup C F Ratio 0LF (n=9) (n=9) (n=8) SCIa LAB 22.22 22.22 21.62 .049 .952 MSU 16.22 14.67 17.50 1.507 .243 BIO 39.44 36.55 39.12 .612 .551 EXP 28.33 24.55 29.75 2.739 .086 aThe low scale loadings on this concept for Factor II did not warrant an analysis of variance. 77 TABLE 9.--Subgroup Means and Results of ANOVA's on Factor III. Means Concept Subgroup A Subgroup B Subgroup C F Ratio aF (n=9) (n=9) (n=8) SCI 38.11 35.67 43.37 2.83 .079 LAB 16.55 15.22 17.62 1.03 .373 MSU 26.33 23.22 31.87 6.50 .006 BIO 23.89 20.33 22.87 1.647 .215 EXP 14.22 14.55 18.50 3.32 .054 TABLE 10.—-Subgroup Means and Results of ANOVA's on Factors A and B. Means Concept Subgroup A Subgroup B Subgroup C F Ratio o‘F (n=9) (n=9) (n=8) LAB (A) 23.00 22.33 22.25 .103 .902 EXP (B) 15.67 14.22 16.62 1.202 .319 78 TABLE ll.--Post-hoc Comparisons of Subgroup Means. Differences Between Subgroup Means Concept Ratio F Within A-B A-C B-C Factor Ib Factor II EXP 2.74 .086 22.77 3.8 a -l.4 -5.2 Factor III SCI 2.83 .079 45.86 2.4 -S.3 -7.7 (7.9) (7.9) MSU 6.50 .006 24.80 3.1 -5.5 -8.6 (8.8) (8.8) EXP 3.32 .054 14.16 -.3 -4.3 -3.9 (4.7) (4.7) aMinimum absolute difference for comparison to be significant at 0F in the same row. bAlpha levels for F ratios in the analyses on Factor I did not justify post-hoe comparisons. 79 to the presentation and the absence of this finding is the first unexpected, and unexplained, result. Inspection of the ordering of the Subgroup means in Tables 7 through 10 provides the second unexpected result. With but one exception the highest means (most positive scale/concept judgments) were obtained from the subjects in the locations presumed to be most marginal. Furthermore, although the post-hoc comparisons were not significant at the alpha level of the corresponding F ratio, they were consistently closest in favor of Sub- group C. And had a tabled F value (instead of the values actually obtained) been used in the Sheffe formula, three of the four comparisons in column B—C would have been significant in favor of Subgroup C. Thus, it is clear that there is no statistical support for a positive answer to the second research ques- tion and that there is some evidence to suggest that the best viewer locations were those closest and farthest away from the screen for the multi—image presentation, the locations of the subjects in Subgroup C. CHAPTER V SUMMARY AND CONCLUSIONS Summary Purpose of the Study This study explored the gross, affective impact of a multi-image presentation upon human subjects. The ex- perimental data were intended to shed light on two general questions: 1. Will a multi-image and audio presentation cause greater positive shift in evaluative meaning than a parallel single-image and audio presen- tation? 2. Is the magnitude of shift in evaluative meaning related to the amount of the viewer's visual field which is covered by the projected image area as determined by the viewer's distance from the screen? Design and Procedures Seventy-one students from two graduate classes in the College of Education at Michigan State University were randomly assigned to two treatment groups and a 80 81 control group. Treatment Group 1 received a ten-minute multi-image and audio (music) presentation on Biochemistry at Michigan State University. Treatment Group 2 received a parallel ten-minute single-image and audio (music) pre- sentation on Biochemistry at Michigan State University. All of the 35mm slides for the parallel single-image pre- sentation were drawn from the multi-image program and the music track was identical for both presentations. The Control Group received a single-image and audio (music) presentation on instructional technology using 35mm slides and audio component different than the other two presen- tations. Treatment Groups 1 and 2 were also randomly assigned to pre-selected seats in order to investigate the second research question on the effect of viewer location for those viewers experiencing the multi-image presentation. All treatments were administered simultaneously in three identical lecture—auditoriums. All equipment (in- cluding automatic programers and xenon-arc slide pro— jectors) was housed in projection booths. Since all images were ten feet wide, the total horizontal screen area for the multi-image presentation was thirty feet. The same criterion instrument was administered to all groups immediately after the presentation treatment. The criterion semantic differential (SD) instrument was derived in a pre-experiment trial and consisted of 82 twenty bi-polar adjective scales of seven positions. All twenty scales were shown in this trial to be relevant to and constant in polarity on the concepts BIOCHEMISTRY, MICHIGAN STATE UNIVERSITY, SCIENTIST, LABORATORY, and EXPERIMENT. However, unidimensionality of the scales was not supported by factor analyses of the combined SD ratings of Treatment Groups 1 and 2 on each of the five concepts. The analysis of the data with respect to the two research questions was performed in accordance with the three main and two "auxiliary" factors which emerged in the preprequisite factor analyses. Analysis of Results Seven of the sixteen one-way analyses of variance performed with respect to the first research question produced F ratios significant at the .10 level of confi- dence or better. However, none of the post-hoc compari- sons of the Group means for those analyses were signifi- cant at the same alpha level as the corresponding F ratio. Sixteen one-way analyses of variance were performed similarly on the SD ratings of the three Subgroups in Treatment Group 1 to deal with the second research ques- tion. Only four of these analyses produced F ratios 83 significant at the .10 level of confidence or better and none of the associated post-hoe comparisons were signifi- cant at the alpha level of the corresponding F ratios. Findings The findings based on the experimental data can be summarized as follows: 1. There is considerable evidence that a systematic main effect was operating on at least some of the concepts but the effect cannot be ascribed specifically to the multi-image presentation. 2. There is some evidence of a systematic influence of viewer location for those subjects who re- ceived the multi-image presentation but the in- fluence cannot be ascribed specifically to those locations initially presumed to be optimum. If anything, those locations presumed to be mar- ginal were the best. Discussion and Recommendations The findings summarized above indicate that this experiment did not produce compelling statistical evidence for positive answers to the two research questions. Never- theless, there are two reasons for concluding the report with a section on implications for further research. First, even had the results been more conclusive, the issues would be far from settled. Without further 84 testing, the results would not be generalizable to other multi-image presentations nor to other populations. Each experiment, of course, must stand or fall on its own merits and each finding supported by experimental evidence is just that--a separate finding. Only after many mutually sup- portive findings have appeared can a principle of design be enunciated and perhaps integrated into existing theory. The second justification for this concluding section follows from the first. If further research is warranted (and the present investigator feels strongly that it is), it seems salutary, using the hindsight of experience, to suggest some ways for modifying the assumptions and improv- ing the methodology so that a replication of the experiment might obtain more conclusive results. Size of Experimental Population First and most routinely, the experiment should be replicated with a larger experimental population. Measures of central tendency perform best statistically when samples contain at least thirty subjects, more than were available for the groups in this experiment and nearly three times the size of the subgroups. Selection of SD Concepts The second improvement concerns the SD instrument. Great care was taken to select the adjective scales which were relevant to the concepts of interest. Had similar 85 precautions been taken in selecting the concepts which were relevant to the treatment presentations, the results might have been much more conclusive. As pointed out previously, the concepts were arbitrarily chosen by the investigator—-a customary procedure in studies using the SD technique. With hindsight it is suggested that all but the concept MICHIGAN STATE UNIVERSITY (MSU) are very simi- lar and that they may have represented, in fact, essentially one larger concept--perhaps an overarching idea like SCIENCE--to the experimental subjects. If such were the case, the results obtained would assume greater impor- tance, for significant F ratios were obtained for MSU on two of the three main factors and for at least one of the four "science-related" concepts on all of the main factors. In a replication, then, much more care should be used to select mutually exclusive concepts shown to be related to the treatment presentation. Selection of Experimental Population The relatedness of the concepts just discussed sug- gests yet another way to improve a replication of the experiment. It was initially assumed that the experi- mental population of graduate education students would not be substantially unlike the sophomore, beginning Bio- chemistry students for whom the presentation was originally designed with respect to knowledge of and feeling for 86 Biochemistry. However, if as now suggested, four of the concepts represented substantially the same idea for the experimental subjects, it can be argued that the graduate students had quite hardened meanings for the SD concepts because of their additional experience with and exposure to science courses. Previous research has shown repeat- edly that firmly held beliefs and attitudes are very presistent and the between-group variances obtained in this experiment would then appear quite remarkable. Accord- ingly, it is suggested in future research that the experi- mental population and the intended audience for a given presentation/treatment should be matched as carefully as possible. Viewer Location The second research question concerned only one of many independent variables which may be involved with the effect of multi-image presentations. Although the small N and the weakness of the finding bode extreme caution in interpretation, the experimental evidence found is quite surprising. How might it be explained? Inundation and Normal Viewing.--To begin with, there were no apparent differences between the subjects who sat the closest to and those who sat the farthest from the multi-image screens.1 If, then, the differences on the 1That is, there are no discernible differences statistically nor demorgaphically as revealed by data on 87 dependent variable between those subjects taken together as a Subgroup and the other two Subgroups was due to their respective locations, it may be that those who sat close to the screen were "inundated" by the presentation and those farthest from the screen simply received more impact because they experienced the presentation at the "normal" two-screen-widths distance. Audience Dynamic.--There are other possibilities as well. For example, there may be an "audience dynamic" of the nature that those who sat more in the middle of the seating area perceived themselves as part of a cohesive group experiencing just another class presentation while those seated at the extremes reacted more individually, feeling more like guinea pigs than part of a group having a common experience. Viewer Preference.-—Or perhaps the impact of a pre- sentation is related to whether or not a viewer is able to sit where he prefers to be seated. Such a notion might be checked with an exPeriment designed around matched groups, one which is allowed to choose seats and the other for which seats are assigned. Visual Acuity.--There also may be interactions among the visual acuity of the subjects, seat location, and size the questionnaire completed by the subjects. Neither were there noteworthy differences in kind nor degree on the free response items on the questionnaire. 88 of image. In this experiment the variable of visual acuity was assumed to be randomly distributed throughout the experimental population but it might be sell to control for it with a pre-experiment, individually-administered test. Experimentation with Individuals Passing on to another matter, it should also be men- tioned that experimentation using individual subjects as the unit of analysis may be very useful for determining the most fruitful independent variables for further inves- tigation. Some of the variables which would seem to be of interest are: image size; viewing angle; placing multiple images on one screen versus individual screens for each image; pacing and rhythm imparted by independently-changing, simultaneously-presented images; and previous knowledge and attitudes of the viewer. Also, individual experimentation offers the possi- bility of using autonomic nervous system responses, such as galvanic skin response and pulse rate, as affective measures both for separate analysis and for correlation with indirect measures such as the SD technique. Again, a caution about generalizing may be in order. Any variables found to be of special interest with experi- ments using individuals as the unit of analysis should be checked in experiments with groups. One should no more 89 generalize to groups the findings from experimentation with individuals than one should generalize from an isolated presentation/treatment to all such presentations or from a given population to all other populations. Some Larger Issues The suggestions for additional research given thus far have dealt with quite ordinary and obvious variables. There yet remain some larger issues which deserve dis- cussion even though it is not easy to suggest ways of dealing with those issues experimentally. Suppose for the moment that multi-image presentations are actually a new medium rather than an extension of a simpler and older medium. Such a state of affairs would elevate the level of discourse for it would seem necessary then to devise completely new presentation and design principles. Existing principles, crude and intuitively based as they are, would be inadequate. For example, if Marshall McLuhan is right that, ". . . in the electronic age, data classification yields to pattern recognition . . ."2 perhaps multi-image presen- tations should be organized (or disorganized according to conventional standards) to train the viewer to recognize patterns rather than to arrange the images (as was done 2Marshall McLuhan, Understanding Media: The Exten- sions of Man (2nd ed.; New York: The New American Library, Inc., 1966), p. viii. 90 in this experiment) in ready-made groups like plants, animals, laboratories, equipment and so forth. McLuhan's distinction between 222 and 9991 media may have even more relevance for multi-image displays. He contends that media which present high—definition images are "hot" and non-involving whereas media with' low-definition images are "cool" and force the viewer to 3 Could it be become involved by completing the display. that "hot" high-definition images (in this case projected transparencies) in effect become "cool" and involving when they are changed rapidly or when many of them are presented simultaneously? McLuhan has predicted that we have seen only the beginning of the school dropout problem because many of today's youth simply find no correspondence between their school experiences and the "electrically configured" world in which they live.4 Although it would be foolish to surmise that multi-image presentations may be a panacea for returning relevance to the school program, it does not seem too much to say that some of the possibilities should at least be investigated. 3For a fuller discussion of this distinction see Understanding Media and Marshall McLuhan and Quentin Fiore, ThepMedium is the Massage (New York: Bantam Books, 1967). 4McLuhan, p. ix. 91 Conclusion In conclusion, it can be stated that this study has provided sufficient experimental evidence of the affective impact of multi-image presentations to warrant additional experimentation. Also, the rather surprising results with respect to viewer location should certainly be verified with new experiments. However, to prevent dissipation, fragmentation and duplication of effort in additional multi-image research, it would be well for future investigations to be coordinated in some way. To that end, this investigator stands ready to share information, ideas and results with all others who also hold a belief in the importance of exploring the effects of multi-image presentations. With such frequent and frank exchanges the day may be hastened when multi-image presentations might make a substantial contribution in improving affective as well as cognitive learning which is, after all, the ultimate goal. BIBLIOGRAPHY 92 BIBLIOGRAPHY Adams, H. F. Advertising and Its Mental Laws. New York: Macmillan, 1921. Allen, William H. "Audio-visual Communication." Enc clo- pedia of Educational Research. 3rd ed. Edited By Chester W. Harris. New York: The Macmillan Company, 1960. Allen, William H. Effects of Audiovisual Materials on Changing the Attitudes of Culturally Disadvantaged Youth. USOE Sponsored ProjECt Final Report No. BR25-8378, ERIC No. ED020298. Los Angeles: Univer- sity of Southern California, 1968. Allen, William H., and Cooney, Stuart M. A Study of the Non-Linearity Variable in Filmic Presentatigp. Final Report, NDEA Title VII Project No. 422, ERIC No. ED003563. Los Angeles: University of Southern California, 1963. Ash, Philip, and Jaspen, Nathan. Optimum Physical Viewing Conditions for a Rear Prgjectibn Daylight Screen. (Pennsylvania State University Instructional Film Research Program) Port Washington, N. Y.: U. S. Naval Training Device Center, Office of Naval Research, Technical Report No. SDC 269-7-37, 1953. Becker, Samuel L. "Interest, Tension, and Retention." AV Communication Review, 12 (Fall, 1964), 277-291. Becker, Samuel L. The Relationships of Interest and Attention to Retention and Attitude Change. Final Report, NDEATftle VII Project No. 739, ERIC No. ED003600. Iowa City: University of Iowa, 1963. Berlo, David K., and Kumata, Hideya. "The Investigator: The Impact of a Satirical Radio Drama." Journalism Quarterly, 33 (Summer, 1956), 287-98. Bodger, Lowell A. "Ultra-Wide Screen Systems." American Cinematographer, 43 (July, 1962), 424-26, 440-44I. 93 94 Burris-Meyer, Harold, and Cole, Edward C. Theatres and Auditoriums. New York: Reinhold Publishing Corp, 19640 ' Campbell, Donald T., and Stanley, Julian C. "Experimental and Quasi-experimental Designs for Research on Teaching." Handbook of Researchgon Teaching. Edited by N. L. Gage. Chicago: Rand McNally & Co., 1963. Croft, Roger G.; Stimpson, David V.; Ross, Walter L.; Bray, Robert M.; and Breglio, Vincent J. "Comparison of Attitude Changes Elicited by Live and Videotape Classroom Presentations." AV Communication Review, 17 (Fall, 1969), 315-21. Darnell, Donald K. "A Technique for Determining the Evaluative Discrimination Capacity and Polarity of Semantic Differential Scales for Specific Concepts." Unpublished Ph.D. dissertation, Michigan State Univer- sity, 1964. Department of Audiovisual Instruction. Handbook for the DAVI Conventiong‘Houston, Texas, Mardh 24—29, 1968. Washington: DAVI, 1968. . Handbook for the DAVI_Convention, Detroit, MiChigan, April 27-May 1, 1970. Washington: DAVI, 1970. Edling, Jack V. A Study of the Effegtiveness of Audio- Visual Teaching Materials When Prepared Accgrding tgthe Principles gf_Motivational Research. Final Report, NDEA Title VII Project No. 22 , ERIC No. ED003532. Monmouth: Teaching Research Division, Oregon State System of Higher Education, 1963. Edling, Jack V. Experiments with Educational Media to Modify Attitudes. USOE Sponsored Project No. 6-2454, ERIC Number not available. Monmouth: Teaching Research Division, Oregon State System of Higher Education, 1968. Enoch, Jay M. "Effect of the Size of a Complex Display Upon Visual Search." Journal of Optical Society of America, 49 (March, 1959), 280-286. Fondiller, Harvey V. "Magnamultiscreen At Expo '70." Popular Photography, May, 1970, pp. 102-103, 152, 156. 95 Furbay, Albert L. "Influence of Scattered Versus Compact Seating on Audience Response." Speech Monographs, 32 (June, 1965), 144-48. Gibson, James J. "A Theory of Pictorial Perception." Audio-Visual Communication Review, 2 (Winter, 1954), 3-23. Hall, Edward T. .The Hidden Dimension. Garden City, New York: Doubleday, 1966. Hanson, Marvin L. Effectiveness of Film in Changing Parental Attitudes. USOE Sponsored Project Final Report No. BR-6-l638, ERIC No. ED020659. Salt Lake City: Utah University, 1968. Hays, William L. Statistics for Psychologists. New York: Holt, Rinehart and Winston, 1963. Heise, David R. "Some Methodological Issues in Semantic Differential Research." Psychological Bulletin, 72 (December, 1969), 406-422. Hochberg, Julian. "The Psychophysics of Pictorial Perception. Audio-Visual Communication Review, Vol. 10 (Sept.- OCto’ 1962) p 22-54. Hovland, Carl 1.; Janis, Irving L.; and Kelly, Harold H. Communication and Persuasion. New Haven and London, ConnetiCut: Yale University Press, 1953. Hubbard, Richard D. "Telemation: AV Automatically Con- trolled." Audiovisual Instruction, 6 (November, 1961), 437-39. Justin, J. Karl. "Lecture Hall and Learning Space Design." Journal of the Society of Motion Picture and Tele- Vision Engineers, 75 (March, 1966), 183-190. Kappler, Frank. "The Mixed Media--Communication That Puzzles, Excites, and Involves." Life, July 14, 1967, p. 28. Kraus, Sidney. "Modifying Prejudice: Attitude Change as a Function of the Race of the Communicator." A! Communication Review, Vol. 10 (January-February, 1962), 14-22. Le Baron, Walt. "In Defense of Multi-Media." Revised reprint of a paper presented at the joint Project ARISTOTLE-American Association for the Advancement of Science Symposium, "Communications and Multie Media in Education," AAAS Convention, Dallas, Texas, 96 December 27, 1968. Santa Monica, California: System Development Corporation, 1969. Levonian, Edward. Measurement and Analysis of Physiological Response to Film. Final Report, NDEA Title VII ' Project No. 458, ERIC No. ED003572. Los Angeles: University of Southern California, 1962. Lombard, Emanuel S. "Multi-Channel, Multi—Image Teaching of Synthesis Skills in Eleventh Grade United States History." Unpublished Ph.D. dissertation, Univer- sity of Southern California, 1969. McLuhan, Marshall. Understanding Media: The Extensions of Man. 2nd ed. New York: The New American Library, Inc., 1966. McLuhan, Marshall and Fiore, Quantin. The Medium is the Massage. New York: Bantam Books, 1967. McVey, G. F. "Environment for Learning," University of Wisconsin, Madison, n.d. (Mimeographed.) McVey, G. F. "Where Do We Sit?" Educational Television, Meisengeier, Richard W., and Stevens, Raymond P. "The History of America--A Multi-Media Approach." Paper presented at the Symposium on Research and Utiliza- tion of Educational Media for Teaching the Deaf, February 5-7, 1968. American Annals of the Deaf, 113 (November, 1968), 1015-19. Merrill, Irving R. "Attitude Films and Attitude Change." AV Communication Review, 10 (January-February, 1962), 3-13 0 Millard, William L. "Multimedia Instructional Techniques, Facilities, and Services for College Teaching." Journal of the Society of Motion Picture and Tele- vision Engineers, 75 (September, 1966), 825-27. Miller, William C., III. An Experimental Study of the Relationships of Film Movement and Emotihnal Response, and Its Effect on Learning and Attitude_§grmation. Eihal Report, USOE Bureau of Research BR-5-0734i} ERIC No. ED017172. Los Angeles: University of Southern California, 1967. 97 Miller, William C., III. "Film Movement and Affective Response and the Effect on Learning and Attitude Formation." AV Communication Review, 17 (Summer, 1969), 172-181. Olsen, John R. "The Effect of Multi-Stimuli Presentations on Learning Gain." Dissertation Abstracts, 30:2235-A (1969). Osgood, Charles E.; Suci, George J.; and Tannenbaum, Percy H. The Measurement of Meaning. Urbana: University of Illinois Press, 1957. Perrin, Donald G. "A History and Analysis of Simultaneous -Projected Images in Educational Communication." Unpublished Ph.D. dissertation, University of Southern California, 1969. "Professional Sights Soar with Finn-Hall Spectacular." Audiovisual Instruction, 7 (June, 1962), 366-67. Reed, H. B. "The Learning and Retention of Concepts: V, The Influence of Form of Presentation." Journal of Experimental Psychology, 40 (1950), 504-11. Reynolds, James C. "The Effect of Viewer Distance on Film Induced Anxiety." Dissertation Abstracts, 29:3341-A (1968). Roberts, Alvin B. and Crawford, Don L. "Multiscreen Presentations: Promise for Instructional Improve- ment; An Approach to the Study of the Civil War." Audiovisual Instruction, Vol. 9 (October, 1964) 528—530. Roshka, A. Usloviia. "Conditions Facilitating Abstraction and Generalization." VOP PSIKHOL, 4 (6), 1958, pp. 89-96, as reported by I. D. London in Psychological Abstracts, 34 (1960), 85. Schlanger, Ben. "Criteria for Motion Picture Viewing and for a New 70mm System: Its Process and Viewing Arrangements." Journal of the Society of Motion Picture and Television Engineers, 75 (March, 1966), 161-67. Sells, Saul B., and Trites, David K. "Attitudes." Enc clo- pedia of Educational Research. 3rd ed. Edited By Chester W. Harris. New York: The Macmillan Company, 1960. 98 Silber, Kenneth H. "An Experimental Study of the Degree of Affective Response Elicited by Several Mediated and Non-mediated Instructional Methods." Dissertation Abstracts, 30:1879—A (1969). Smith, Arthur L., and Schlanger, Ben. "The Colonial Williams- burg Theatres for a Wide Screen Participation Film." Journal of the Society of Motion Picture and Tele- vision Engineers, 70 (September, 1961), 681-84. Snider, James G., and Osgood, Charles E., eds. Semantic Differential Techniqpe. Chicago: Aldine Publishing Company, 1969. Snowden, Terrence J. "Comparison of Three Types of Multi- media Presentations." Dissertation Abstracts, *25:6322-A (1964). Turabian, Kate L. A Manual for Writers of Term Papers, Theses, and Dissertations. 3rd ed. rev. Chicago: The University of Chicago Press, 1967. Twyford, Loran C. Jr. "Educational Communications Media." Encyclppedia of Educational Research. 4th ed. Edited by Robert L. Ebel. London: Collier-Macmillan, 1969. Wagner, Robert W. "The Spectator and the Spectacle." Audio-Visual Communication Review, 3 (Fall, 1955), 294-300. Wallington, Jim; Hale, Pryor; and Conte, Joseph. "Multimage Festival." Audiovisual Instruction, 14 (June-July, 1969), 53. Wickline, Lee E. The Use of Motivational Films to Favor- abiy_Change the Attitudes of High_School Students Toward Science and Scientists. NDEA Title VII Project Final Report No. 729, ERIC No. ED003598. Charleston: West Virginia State Department of Education, 1962. Williams, Anthony V.; Peterson, James; and Paul, Robert. "FACTOR AA." Technical Report No. 34.1 of the Computer Institute for Social Science Research, Michigan State University, May, 1969. (Mimeographed.) APPENDICES 99 APPENDIX A SEMANTIC DIFFERENTIAL FOR PILOT EXPERIMENT 100 INSTRUCTIONS The purpose of this instrument is to measure the meanings of certain things to various people by having them judge them against a series of descriptive scales. (It has absolutely nothing to do with the grading in this course.) Each page contains a concept at the top followed by eighteen adjective pairs which are separated by a scale containing seven positions. You are to check at one of the positions along each scale to indicate where you rate the concept at the top. Here is an illustration - 3 : 2 : l : 0 NICE : : : u u H u u N u 0. DJ AWFUL This rating scale is bounded by the words "NICE" and "AWFUL". The more "NICE" you feel the concept is, the farther to the left of the scale you would place your check: the more "AWFUL" you feel the concept is, the farther to the right you would place your check. If it is hard to decide if it is "NICE" or "AWFUL", or you feel the adjective pair is not relevant to the particular concept. place a check in the central space, under the zero (0). This means "undecided" or "irrelevant". There are no right or wrong answers. The best response is what ygg feel is appropriate RIGHT NOW. Do not spend too much time on any one item. PUT DOWN YOUR FIRST IMPRESSION. Please be sure to place a check in one of the seven positions on each scale. Do not go back to pages you have already completed. When you finish a page, please continue to the next. Remember, even though some of the items may not seem to make much sense, this measurement technique has proven to be very valuable when the respondents do the best they can in honestly indicating their first impressions. 101 KIND BAD STRONG TRUE DIRTY SERIOUS UGLY PASSIVE HARMONIOUS SIMPLE POSITIVE MEANINGFUL OLD ATTRACTING WISE SOFT UNSUCCESSFUL USUAL (Concept) H 102 AWFUL GOOD WEAK FALSE CLEAN HUMOROUS BEAUTIFUL ACTIVE DISSONANT COMPLEX NEGATIVE MEANINGLESS NEW REPELLING FOOLISH HARD SUCCESSFUL UNUSUAL APPENDIX B DERIVATION OF SEMANTIC DIFFERENTIAL WITH "BEST-WORST" TECHNIQUE 103 DEVELOPMENT OF SD INSTRUMENT INTRODUCTION As noted in the text of the proposal, the criterion instrument to be used in measuring the dependent variable, connotative meaning, will be a specially prepared form of the Semantic Differential. In order to minimize the possible interaction between scales and concepts and in order to maximize the discriminability of the SD, the scales to be used 'will be determined with the Best-Worst technique with POpulat- tion B prior to the actual experiment. Most of the ideas and material given below is drawn from Darnell's 1964 dissertation. PREPARATION OF SD MATERIALS The steps to be followed in preparing the SD booklets to determine the discriminability and polarity of scales on the concepts BIOCHEMISTRY, MICHIGAN STATE UNIVERSITY, SCIENTIST, EXPERIMENT, and LABORATORY are: 1. Choose forty scales (which intuitively appear to be appropriate) from Osgood's lists. Write scales in polar position as given by Osgood. 2. Number scales for purposes of random drawing for order. 3. Prepare individual slips for the numbers 1-40. 4. Draw for order of presentation according to follow- ing decision rules: a. Flip coin to see if lst draw should be "+" or "-" in polarity where "+" polarity is order given by Osgood and "-" is the reverse. b. Draw slips at random, one at a time and record the numbers in order drawn. c. Write scales, alternating each in polarity after the first as given in a. above. 5. Prepare sheets of scales in order determined as in 4. with 20 scales per page. There will be 2 sheets of scales for each concept. 6. Assemble booklets of sheets so that each booklet contains sheets for 3 concepts. The first booklet, then, would have sheets for concepts 1,2, and 3; the second booklet sheets for concepts 2,3, and 4: and so on.-—5 ”kinds" of booklets for the 5 concepts: 1,2,3; 2,3,4; 3,4,5; 4,5,1; 5,1,2. 104 105 ADMINISTRATION The prepared SD booklets will be distributed randomly (but in the sequential order shown above in step 6) to students in two regular classes in the College of Education (MSU) with a total N of 50. Thus there will be approximately thirty people making ratings on each scale-concept item. (50 subjects)? (5 "kinds" of booklets )x appearances 30 with 3 concepts each per concept subjects across kinds per of booklets concept The instructions for the exercise are given in Exhibit 1. Essentially, S's are told to mark a "B" for "Best imaginable" example of the concept apg_"W" for the "Worst imaginable" example of the concept on each scale. ANALYSIS OF THE DATA Scales for which the null hypothesis given below can be rejected will be judged to be evaluatively discriminative for a given concept. Ho: The number of subjects who placed their response to the "best example of the concept” to the left of their response to the "worst example of the concept" is equal to the number of subjects who indicated the opposite direction of preference. Scoring Looking at a set of responses to a scale- concept item, each subject who places "B" on the left of "W" but not in the same scale interval scores a "plus". Each subject who places both marks in the same cell scores zero and drops out of the sample. A subject who places a "W" on the left of "B" scores a minus. The ratings (raw scores) will be entered on scoring tally sheets set up as shown in Exhibit 2 so that the ratings of all subjects on a given scale-concept item will appear in the same row. Analysis The sign test will be used to determine the significance of the ratings across subjects on each scale- concept item. Actually, a null hypothesis of no difference between the number of "plus" and "minus" ratings will be tested for each scale-concept item or row on the scoring tally sheets. A scale will be said to have an evaluative discrimination capacity for a concept if, and only if, the null hypothesis of the sign test is rejected at the 95% level of confidence. The statistic to be used in each case is the exact binominal pro- bability of the entered data. 106 Since there will be about 50 subjects involved in this rating exercise and each subject will rate 3 of the 5 concepts on all scales, there will be a maximum of 30 ratings per scale- concept item. The effect of a "B" and a "W" in the same scale position is scored 0 and reduces the N for a given item. There— fore, in preforming the sign test, the number of "pluses" or "minuses" necessary for significance given the total "N" of "+" and "-" scores for a given item is as shown in Table i, TABLE 1 To be significant at .95 N of +'s ANQ_ level of confidence -'s on an item EITHER +'s QR_-'s must é, * 5-8 0 9-11 1 12-14 2 15-16 3 17-19 4 20-22 5 23-24 6 25 7 *Based on two-tailed binomial probabilities given by Helen W. Walker and Joseph Lev, Statistical Inference (New York: Holt, Rinehart and Winston, 1953), p. 458. Rejection of the null hypothesis will be shown on the scoring tally sheets by entering a "+" if there are signifi- cantly more "pluses" tallied or by entering a "-" if there are significantly more "minuses" tallied. No entry in the last column will mean that the null hypothesis could not be rejected and that therefore the given scale does not exhibit evaluative discrimination for that concept. The results of all the sign tests will then be entered in the matrix shown in Exhibit 3, A "+" will indicate that the subjects showed a preference for the adjective on the left and a "-" will indicate subjects' preference for the adjective on the right. Thus, when this matrix is completed, it will show which scales evaluate on a given concept and the polarity .of each such scale. The scales for the criterion instrument for use in the experiment can then be selected from this matrix on the basis of their evaluative discriminability. The decision rules which are contemplated are: l. A scale must discriminate on at least 4 concepts to be considered for use. 2. Every concept must be discriminated by at least 4 scales. 107 A Survey of Judgment Criteria This study is part of a larger experiment concerned with a specific media design and presentation technique. This exercise is designed to find out what criteria people use for making judgments about a specific set of things--what kinds of questions would they want to ask about one of those things before they could decide whether it was a "better" or "worse" thing of its kind. For example, you probably don't care whether your friends are large or small, but that's the first question you would ask about a pay check. You may not care whether your automobile is red or green, but it makes a difference in apples. If we had a hundred years to spare, we might be able to answer this question by discussion, but this study is an attempt to get an answer more quickly than that. On the following pages are scales with adjectives at each end that look like this: left : : : : : : right The intervals on these scales may be interpreted as extremely left, quite left, slightly left, neither or both, slightly right, quite right, and extremely right. Of course you are to substitute whatever words occur at the left and right ends of the scales. At the top of each page is a concept, such as DOG. What you are to do is to think of the best imaginable and the worst imaginable examples of the class of things named by that concept (in this case, the best imaginable DOG and the worst imaginable DOG) and indicate where you think the best and the worst examples fall on each of the scales on that page. For example, if you happen to like large DOGS, and you don't care much for small DOGS, you might indicate that the best imaginable DOG is extremely large and the worst imaginable DOG is quite small. Of course, your best DOG may be "gentle" and your worst DOG "mean," but you will have an opportunity to indicate that on another scale. 108 Indicate your feeling for the "best" example by marking a "B" on the scale in the appropriate place. Indicate "worst" by marking a "W" in the appropriate place. Your responses might look like this: DOG large B : : : : : "': small mean W : : : : : f' gentle green : : : BW : : __ : red The latter mark indicates that you don't really care whether a DOG is green or Eeq_or that this scale just doesn't apply to DOGS. With concepts such as ELEPHANT, MONSTER, or RUBY you might feel that one of the extreme positions on the scale describes aii the members of the class, in which case you should mark "BW" in the extreme position. Just make sure that you have two marks on every scale. You will find two different sheets of scales for each concept, or stated the other way, you will find each concept on two different sheets of scales. subjective bright worthless healthy poor positive light fair sad reputable old hard ugly clear simple interesting intuitive sharp untimely sacred (Concept) 109 O. O. O. objective dark valuable sick rich negative heavy unfair happy disreputable new soft beautiful hazy complex boring rational blunt timely profane 110 (Concept) awful : : : : : nice loud : : : : : : soft delicate : : : : : ° rugged good : : : : : : bad cold : : : : ° : hot constrained : : ' : : - free shallow : : : : : deep fast : : : : : : slow usual : : : : : : unusual active : : : : : : passive defensive : : : : - - aggressive important : : : : - : unimportant changeable : : : : : : stable pleasant : : : : - - unpleasant 10W : : : : : : high serious : : : : . {7 4g humorous dirty : i: : : : : clean true : : : : : : false small : : : : : : large strong : ° : : : weak 111 Questionnaire The information requested below will be helpful in analyzing and using the results of this survey. Thank you for supplying the information and for your cooperation with the survey. Please indicate your Sex: Male Female and your Age: years. Please indicate with an X on the appropriate line whether or not you are currently enrolled (Winter term, 1970) in the following courses: ED 431 Educ. Media in Instruction Yes (Mr. Bruce Miles, Instructor) Ed 83l-A Educ. Media in Instruction Yes (Dr. James Page, Instructor) ED 83l-B Educ. Graphics in Instruc. Yes (Mr. Don Wilkening, Instructor) ED 821-A Curriculum Construction Yes (Dr. Charles Blackman, Instructor) NO NO No NO APPENDIX C RANDOMIZATION PROCEDURES 112 RAND OMI 3.7.7? I ON INS TRUCT I ON S 1. Count number of men present. Count out that number of slips coded Blue and place in container marked MEN. Write that number here . 2. Count number of women present. Count out that number of slips coded Black and place in container marked WOMEN. Write that number here . 3. Keep each stack of slips in order, both before and after depositing appropriate number in container. 4. Shake each container thoroughly before drawing starts and between draws. Let men draw only from "MEN" and women only from "WOMEN". 5. Read the following to the class: This evening, thanks to the cooperation of , you are asked to participate in a short experiment. Your assistance will be invaluable and in return it is hoped that you will find the experience interesting. For the experiment this class and ‘_ taught by __must be divided randomly into three equal groups. In a moment you will be asked to draw a slip to accomplish the randomization—- the men drawing from one container and the women from another. 6. After each class member has drawn (each container should be empty) read the following: All men should have a slip with a Blue diagonal mark and all women should have a slip with a Black diagonal mark. Everyone's slip is also marked with the group he is to join and the room for that group. To avoid confusion, we will move to the appropriate rooms in three separate groups, beginning now with group 1. 7. Lead the group to room 106-B. 113 APPENDIX D SEMANTIC DIFFERENTIAL FOR EXPERIMENT 114 Group Seat iNSTRUCTIONS The purpose of this instrument is to measure the meanings of certain things to various people by having them judge them against a series of descriptive scales. (It has absolutely nothing to do with the grading in this course.) Each page contains a concept at the tOp followed by twenty adjective pairs which are separated by a scale containing seven positions. You are to check at one of the positions along each scale to indicate where you rate the concept at the top. Here is an illustration - NICE : : : : : : AWFUL This rating scale is bounded by the words "NICE" and "AWFUL". The more "NICE“ you feel the concept is, the farther to the left of the scale you would place your check: the more "AWFUL" you feel the concept is, the farther to the right you would place your check. If it is hard to decide if it is "NICE" or "AWFUL", or you feel the adjective pair is not relevant to the particular concept, place a check in the central space. This means "undecided" or "irrelevant". You are to rate the concept at the top of the page on all the scales on that page. There are no right or wrong answers. The best response is what ng_feel is apprOpriate RIGHT NOW. Do not spend too much time on any one item. PUT DOWN YOUR FIRST IMPRESSION. Please be sure to place a check in one of the seven positions on each scale. Do not go back to pages you have already completed. When you finish a page, please continue to the next. If you have any questions, please ask them of the monitor now. He will tell you when to begin making your ratings on the first page. 115 timely dark fair bad beautiful worthless true passive nice negative reputable unimportant interesting sick clear unpleasant strong sad deep dirty (Concept) 116 O. O. I. untimely bright unfair good ugly valuable false active awful positive disreputable important boring healthy hazy pleasant weak happy shallow clean 117 QUESTIONNAIRE The analysis of your responses will be much more meaningful if you will provide us with a little additional information. Therefore, please answer the following questions. 10. What is your age? What is your sex? What was your undergraduate major? What, if any, is your graduate major? Please indicate the number of courses (College level and above) which you have had in the following: Chemistry Biochemistry Other Sciences Are you a teacher? If yes, what grade or level? In a few words please state what you think the presen- tation you have seen this evening was originally created for. That is, what might its purpose have been outside of this experiment? Before this evening, how many multi-image presentations had you seen? Please write below any other comments about the experi— ment or presentation which you would care to make. Please check below any of the entertainment films which you have seen: The Thomas Crown Affair The Boston Strangler Charley NOW--please turn your set of papers over. Do NOT go back and change any responses. Please remain seated until you receive further instructions. THANK YOU FOR YOUR COOPERATION 1:1 APPENDIX E FACTOR LOADING MATRICES AND VARIMAX ROTATIONS FOR FIVE CONCEPTS 118 1119 ov-.s oavo.o. cann.o. .m««.o. oama.= svo°.o. osmo.au «n.d.: o~u«.o. nouo.o unau.eu oo.«.o sena.o omm«.o ~ono.=. acna.o. emm«.o. «66”.. vma«.o. nonu.° on .emsezmHom pamocoo soc Masses schemes sopomesu.ma vmc«.ou oon~.o omno.ou onmn.o non~.o «moo.e onn~.ou «ow«.a csoa.on osuu.o nn~«.on nmoo.o von«.ou menu.on omsu.cu soc~.on nooo.on caco.ou vvo«.e nmne.a auoo.o mvcu.o .nnmn.o. n~mo.o ooam.oc «05°.o smoo.o omnn.°. nkm°.o. coco.o sono.o n°"~.o afloo.o. onmo.o smco.ou nomd.o mouo.o mamu.o- undo.ou ~0mn.a ommo.o. macs.a soma.o. ok-.s coda.o «cko.o. amen.o ~nmo.o unmo.o smmc.a. soma.o. sooo.°. oono.c nmko.o. onma.c «soa.o. omms.o ocss.o. oowo.oc mano.o 0.03.: -ma.o mnk~.o cms=.o. ooc~.ou oo~a.o do~=.s. oovo.c nos=.° mamo.o. amon.o. .~v~.°. ono«.o. «do=.o. coo=.o cond.o «mos.o a.na.o .o¢=.s osn~.o ono~.o Hoao.o. oooc.o. Hmsa.o onco.o coa~.o kamn.o. va~«.e nnoo.o omen.o. oeda.ou ono«.° c~o~.o soa~.o. sac”.s svea.o moms.° ns-.°. cmo~.o. Nefiu.o. kn«¢.o ocoo.a. onna.a onOH.o. onma.a oooo.o. soon.o vn«a.= usefl.o. conn.s mwvo.o Nnflo.s. mono.o. «can.o. envo.oo ncna.ou s~no.o moon.a. n°¢H.o. vooa.o Anon.o. ocfis.=. kakn.s. msoa.s. nmva.o ndvn.o cavc.o anon.o ennaco. mono.o Hov«.or nfloo.s. mkwo.o. oomo.o. oemo.o. ~no«.o mooo.o naaa.au o~c«.o nnmm.o mqmw<9_ mnn¢.a anew.ou nnmn.o coco.o. cmoa.o noon.o noma.o Hwno.o. ankv.o. mkk¢.s. cano.o vamv.o cmcv.o. Amma.o moc¢.o. moan.o o=¢«.oa seno.o nmmv.o rmon.oc mna¢.o coon.o fl. 3 J . r...ro C eman.o seem.a mamk.o unco.o onck.s mokk.o 4564.0 mafik.o nacu.o «mm~.o nucn.o amok.o mcc~.o «coo.o mamh.o «eco.o noeo.c ‘0 Pl N '4 2120 vomo.su vmao.s «owo.o enos.o coca.o. sous.° anno.s some.a caoo.o o-0.o smno.oo osmo.s moss.s onoo.a ~on«.o. cans.o. ouma.o ovno.o. ones.s. onuo.o. ~noo.a «ons.an .mas.su 0559.9 woos.ou coas.su kmss.eu anes.s ovfle.s» oome.o ~o.s.ou onwo.s .aos.su nosa.o «nos.o «ons.ou skwo.on ssos.ou nvao.s nono.ou on oouo.a o-o.o oon°.o. nuao.o. mooa.ou ammo.o. onco.o. onca.o neaa.o oeso.o. m~oo.° ovmo.o mosa.ou as~«.oo oono.o mmmo.ou anus.c unso.o nswo.o oouo.on on cvco.o “moo.o ”masts. onn°.o. kmaa.a oedo.o. oo~o.o .moo.a. voc~.° sona.e caoo.o. ovwo.o. oomo.o. nomo.° co~°.e. ams«.° vaoa.o. oomo.n. vo~o.o sona.a. kn A.c—9200VNH mqmst meeem zeeHmon scmosoo soc saunas wssemos sopommuu.ea mamas fimnm.o moma.o. nemc.o $059.0 enm~.=u monfl.o nomm.= wweo.c. owc¢.ou nunn.ou nmun.o mmo=.ou «woc.on cuvv.o eoso.ou aecn.on cnmn.on avo¢.o maca.o occa.c. oomv.o xmco.o hvmv.o mmv~.o onem.n camn.o amen.o casm.o cmmk.o oesv.= weno.o “men.o mcvn.o oflfio.c «aao.o omvo.o «men.o aflkn.o ocuo.o o:na.° 0H cw NH «a ca IL24 9509.9 0999.9. 0999.9. 0999.9 5099.9 0909.9. 9099.9. 9959.9. 9999.9. 9099.9. 9999.9. 9099.9. 0959.9 «009.9 0599.9 9999.9. 0099.9 -09.9. 9909.9. 9599.9 0099.9»- 0099.9 0999.9 0909.9 9999.9 9999.99 9909.99 9909.9 0909.99 9959.99 9909.9 5099.9 9509.99 9909.99 5599.9 9099.99 0559.99 0099.99 9909.99 «vco.o on 0099.9. 0099.9 9999.9 9909.9. 0999.9 5509.9 9959.9. 0999.9 ~o~9.9 00~9.9. 9959.9 0599.9. 0099.9. 0009.9. 0909.9. vooa.o. 9009.9 9mn9.9 n9~9.9. cano.o on 9999.9 0999.9. 9599.9 0009.9 0909.9. 5099.9. 0909.9. 0999.9 0099.9 0559.9. 9999.9 59~9.9 9009.9. 0909.9 9099.9. 0009.9 5099.9 9009.9. 0999.9. 0999.: 59 9009.9. 9999.9 9059.9 0~99.9. 9599.9 9999.9. 0909.9 9909.9 0009.9. 9959.9 5999.9. 599~.9. ~009.9 9509.9 9999.9 9909.9 0059.9. 0599.9. 9999.9. nooo.ou on A.U.pnoov¢H mqm¢8 0999.9. 5999.9 0909.9 9909.9 0~09.9. 0959.9. 9099.9. 9009.9 0009.9. 9999.9. 9009.9. 0099.9 9099.9. 9099.9 090999 5599.9. 5599.9 0099.9 0999.9. 0-9.9 0d 9509.9. 9999.9 0599.9 9009.9. 0999.9. 9909.9 9009.9. 9099.9. 5509.9. 0999.9. 5059.9 9959.9 9099.9. 0999.9 5509.9 000~.9 5509.9 9009.9. 5099.9 0009.9. Yd 5099.9 0599.9. ~909.9 9559.9. 0099.9 990~.9 9009.9 N599.9. 500959 9999.9 0999.9. 0~09.9 5999.9 0099.9. 5009.9 5059.9. 0099.9. 9959.9. 090~.9. ~059.9. nd 0909.9 9959.9 0059.94 moon.9. 0999.9. 5999.9 5909.9. 9059.9. omco.9. «~99.9. 0~09.9. m5mo.9 9909.9 c5m9.9 0059.9. 9909.9 9999.9. c~o9.9 5059.9. 0~m9.9 NH 9999.9- 0995.9. 0999.9 995~.9 «m09.9 5009.9- 9099.9 5059.9- 0999.9. 5005.9. 9099.9. 9599.9 mmn9.9 5v59.o 9999.9. nvn9.o 5909.9 n009.c «H V9 n9 N9 99 99 2125 999999999999 9990999 999 9059.9. 0999.9 9005.9 5099.9. 9909.9 9009.9. 5509.9 0999.9 5959.9. 0099.9. 5959.9 9999.9. 9509.9 9099.9 5999.9. 5009.9. 5509.9 0999.9 9995.9. 0955.9 99 9055.9 0959.9» 0599.9" 9959.9 0959.9 0999.9» 99cn.9n 9959.9 0999.95 9559.9 9599.9 0995.9» 5009.9 5999.9» 0909.9» 5599.9 5999.9 9099.9 9999.9" 9009.95 5909.9. nomo.o. 9099.9 9509.9 0905.9 0995.9 0099.9 0999.9. 5099.9 9959.9 ~o~9.o 0099.9 omwn.o. 9099.9 0909.9. 9999.9. 9959.9 5995.9. Noon.o. 9599.9. 9599.9 5999.9. 9999.9. 9059.9. 0509.9 9099.9 9999.9 9509.9 5999.9 0999.9. 0599.9 9005.9. 0599.9 9559.9 9999.9 9909.9. 5599.9. 5999.9 9095.9 9509.9. 5095.9. 9099.9 5009.9. 9090.9 9099.9 9990.9 5099.9. 0959.9 9999.9. 9999.9. 5959.9. 9599.9. 0595.9. 0599.9 0090.9. 9999.9. 9099.9. 9559.9 0009.9 5555.9 9909.9 5909.9 5059.9. 5995.9 9959.9 0999.9. 9999.9 9990.9 9099.9 9009.9. 9099.9. 0599.9. 9099.9. 0999.9. 9509.9. 9559.9 0099.9. 0005.9. 5999.9. 9099.9 9999.9. 9990.9 9059.9 0509.9 0505.9. 9999.9. 9509.9. 9999.9. 0059.9 0099.9. 9555.9 0999.9. 5999.9 9090.9 0905.9. 0909.9 5059.9. 5009.9. 9509.9. 0990.9. 999992 wz9cmoq nopommuu.m9 m9m¢9 0999.9. 9995.9. 0999.9 0909.9. 0090.9 9995.9 5909.9. 5909.9 9959.9 9090.9. 959959. 5099.9 0909.9 9905.9. 9999.9 0999.9. 0999.9 9959.9. 5099.9. 0599.9. 9999.9. 9509.9 0999.9. 9959.9 9950.9 9905.9 0599.9 0999.9. 0055.9 9999.9 0099.9 0909.9. 9999.9 nn99.9. 9995.9. c05m.o an0n.o 9900.9 90 0.9 I .L 9955.9 9905.9 nn99.o 0090.9 5005.9 nmm0.o 99 N9 99 09 L126 0000.0. 0000.0 00~0.0. 0000.0. 0000.0 0000.0. 0.00.0. 0000.0 0000.0. 0000.0 ~0m0.0. 0000.0 0000.0 0000.0 o-0.0u n~m0.0n «000.0- 0000.0- 0000.0 0000.0. 0N ~n~0.0n ~n0«.0u 0~m0.0 0m00.0 0.00.0» 0000.0 0000.0" 0000.0 0000.0» 0000.0" 00'0.0u vm00.0n ~vno.cu 0000.0 0000.0 0000.0 nnm0.0 ocmc.ou 0000.0» 0000.0» ow n00".0 0000.0 m~00.0. 0000.0 ~moo.0o 0000.0 0000.0 0000.0 ~m00.0: 0000.0- 0000.0 0000.0 0000.0. sano.o 0000.0 0000.0. 0000.0 0000.0 0000.0- ~000.0 on 0000.0. 0000.0. 0000.0 0000.0 0000.0. 0000.0. n000.0u 0000.0 0000.0 0000.0 0000.0 0000.0 0000.0. 0000.0. 0000.0. 0000.0. -n0.0. 0~00.0 0000.0 0000.0. 54 0.0.00000 m0~u.ou 0000.00. 0~00.0. 0000.0- 0000.0 0000.0. 0000.0 0000.0 0000.0- 0000.0. 0000.0 0000.0 0000.0. 0000.0 0000.0 00-.0 0000.0. 0000.0. 0000.0 oon=.o .0 a .ma mqmus.ha mqm¢e I. 5.1 III - . .. , .: mama. - mm.o m~.o mo.aa an.oH ~o.- ~¢.Hm o> x omom. mo.“ He.” .mn.m oo.m mo.a~ -.m~ e> x - an HHH . H moaocu 1 - x xuuwclcmmao . pomm. . omn«.°. eomo.o. uddmd4dl. moko.°. «na~.o . no«=.o om 30Hamsmlmmmc mp~mnp .Nfid.°. o~so.o. n~o~.o flown.a. eflmo.o. nflm¢.o o~ cmmlhmmms . . pom». -n=.o. uov~.o; umaadiq monv.o. finmn.o nmwo.o ma A Xmm3umcouum . ¢omo. -~a.a. maco.o. ¢ooa.° «amn.o. mmmn.a awmmn c NH ucmmmoncnuucmmmem . mmoo. «m~¢.o. oocn.o. cou~.o osfio.o :uumfl1q aflm¢.° o“ . hum numm U hm¢m amn=.ou enno.o. ooon.o onn«.o 0Nr~.m .finc¢.o ma xofim-»guanmn K Mhmo. soa~.o. vaco.au :vm~.o a~m«.ou floov.o mnm«.o - v“ mcfiuonlm:aumwumuca wmmh. “so“.oc . maw~.oc coma.o mova.o aoom.o uvno.o n“ . ucmuuomeflcsuucMuHOQEH mama. , mnua.o. onma.o. anoo.ou mno~.o. cuna.o admded Nd manmusmwumacnm Amusmwu mwmn. nmmu.oc Hono.u. no«9.cs ooco.os unco.o. cflwm.o Ha o>aummwccm>wuflmom . ooom. aNOu.a. dd.~.o. ommd.= momfl.o‘ fimh~.= vvv~.o OH Asu3MImUflc cwmn. oocc.°. oomo.o econ.o Hovo.o. .«Mdm4d. UONH.O o m>wmmmaum>fluum anew. cowo.o. oofio.o. osmo.o coca.°. coma.o (Manda a . 7 omamwuosuu mama. mama o. mmmaao. Novo.o. «Dov.o. =mmno.d. morn.o N mmmfi fiquBIwHDMDH m> amhm. «moa.ou coma.oc «Hoo.o Hon~.oo ~no«.o ouwo.o o >Hmzl : gunman mums. ~oo=.o. vmmo.o amm~.o 0559.9. (abN.a nomw.o m ong-coom memm. odv«.ou nnmd.o. so-.a. ~¢nfl.oo noa¢.o nmr~.u o . uwmucslufimu ammh. s.»n.ao oonH.o- coco.o oomn.po (new.a xvaw.o n . . xuauuugmfiun «can. omna.o. .-~.o. owua.° omoa.o. INquqq. qmfic.o N uawswucauxaveau e~pn. . “anu.o nnm..o. anno.o oNno.o. ”~m~.o «moo.o a t N: , a n v n w w .moz~ouu.mfi mqm¢e mmmm. mm¢ho moch. mmmm. mvom. mhhh. Hoam. mmHh. ¢o¢®. mom». «mam. hemm. mmnh. cham. Homm. mo¢m. Nmmh. Ammv. omam. mmom. mvmh. hows. 30aamnm HHH 4 HH H moau x v~.m ma.va m¢.ma H~.~H mm.om 8> x >uuwvlcmmao flavn.m ~v¢a.o. H~m~.o hofin.c . mmwa.c ON I wmo omm~.o o~c~.c oooa.° vmwcwwmnn rnq¢.o 0H onnn.o cu n.= odon.o comm.o Nero.o- ma. xmm3: couum mamn.o vmmm.g onofl.o neon.o vccv.o “fl ucmmmmamcslucmmmmam . afloo.o onco.o envm.o n¢cfl.o an“~.m 0H xwmmlumeo ~mn~.ou wvo~.o «non.c van.o moan.o ma xoflms nuamm: mnmo.o Ncnm.o mxmo.o cocm.c caa3.o- ¢H mcwuonnmcflumwaucH moma.° mnflm.o m\na.o (90:.c «c5M1e nH ucmuuomEaCSIucmuuanfl nNHH.o “noa.o anom.o mone.s- nebw+qn NH manmusawumfifilmanmuzmwu mafia.o. omofl.= onm~.a rqmm.= _ Ha m>wummmCIm>HuHmom onoo.o “can.= mano.a nama.o momn.o OH H9w3mvmofic ¢~v«.o moca.o oumn.o uncn.o aoan.o o . m>wmmmmlm>fiuom o¢c~.o unno.= c¢-.o oanu.n rrmouc n wmamMImsuu Hunfiio «~c~.o oomn.o Hafio.o. havn.o N mmmazuMOBImanmsHm> oflco.o “066.9 cnm~.o fleas.o- merm.c a xamSIaswfiusmon caco.o moon.= omfim.o n»:a.o cnr~.o m m nmnunoo naofl.o onm~.= H¢c~.= flama.n “505.9 ¢ uwmwcsluwmu ammo.o. «aflm.= 0355.0 (mo«.m Hflmn.c n . xumnnu: “up cog“.o. cema.= H¢h«.o wc¢~.= a-~.o N nanfl.o onm~.= mom~.o. n v n aHoEaucslxaoEflu moom.o nw.m.o N H l3]. NBHmmm>HZD mefiem EdonOHZ ammonoo How.zoapmpom NQEHHQ>II.mH mqm<9 HHH HH H moeo x Nose. mh.b vH.NH mm.HH mn.0m 9> x ova. mnNN.o. ozon.o «ncm.o . NNmN.o 3N BOHHmLmIQomo mNHm. Naon.= anNa.o. oNn¢.c [QMan.q 0N cmmtxammn _ mmoh. omvo.ou nsvo.o «nau.= nmmw.c ma xmw3tmcouum mmmm. Noe“.o NNoN.o o=m¢.o 4432141 NN ucmmmeQCSI cmmmmam mmvn. nmmN.o caofl.o m¢HN.o cqrn c «H >NmnlumeU cmmo. onmo.=; cMca.a moca.a .m¢cn.a nN xuNmnstNmm; nmmm. NoHN.o nnmn.= =fimmq1q. nvoo.o. «N mafiuOQImC+umououcfi Amen. «Hoo.o mnov.o oono.o _..N c «a ucmuuomEHcstucmquQEfl mvam. mkon.a vomo.o Nozo.o mv:z.a NH .umwclmanmusmmu mme. NCNN.° NNCN.° N¢oo.g. ooxN.o NH . w>fiummmCIm>Hufimom mush. Nvam.o Noum.a enca.; mnad.o 0N H3w3mlwwflc ammo. NmNo.o nofin.o mNNH.o INNN.C N m>wmmmmuo>fluom mmmo. Nnvo.o. comm.o wmoa.e on.N_o a 0m“ mmlmsuu mmmm. cano.oc ocfio.o vmcn.o vowN.o N mmoflcuuo3nmanmsam> cons. NNo¢.= .mmmqu. Nflmo.a- NQNn.o o xamSIHJWHusmofl mmmm. Hmvfl.o noofl.o mafifi.o. havsqd. m manooom ovvm. mNNn.o. Nonn.o oan=.o. mwaNN.a v uwmucznuwmu m¢¢o. cuvo.o. camn.a cch.= onmN.N m ‘ xumc-uzmflun Nvmo. ochn.°. NnNN.o oomn.o .umqu4dn N aduEHucsuuamEfiu ovcw. ovs~.a no'H.o vnmm.o mnac.o H N: ‘ n N v 2132 NmBmHEmmUOHm pmmocoo Rom SOHDMpom Nmaahm>ll.om mqm<8 mmmm. mmbb. «qu. oomsw oNNm. omam. Noam. NNom. Nose. «mom. NmNm. whom. vam. mops. moch. mmmn. mflmp. ANN“. Hams. «man. amen. NNNN. HHH , HH H moeu x Nm.m mm.mH mm.m No.NN om,mN mo.mH 9> x xuuflfiucmmao om¢=.o. oNN°.o. Nono.o vmvo.o. wiqmmm1dp . unco.o. CN 30 mnmlmwwc coco.o . mafia a‘ mno~.cc «950.0. vmma.os «mam.o 0H Ummtmmbnfi «NNN.o NmNN.o voo°.o NNvo.o NNNN.N. omoo.o NH xmm3lmcouum NmNN.o oamm.o Oomv.o. Nomn.a. NNNN.O. Nemo.o NN ucmmmmamczlucmmmoam ovN¢.o ncho.o. 0¢vv.o. NcNo.o cgmm.n. NoNn.o ON xnmcuummao I¢Nou.a NoNN.o oomN.o. ocoo.o. “Nao.o. .oucn.o m“ anmI>memws mm.=.°. «NON.o caoa.o. Nnm¢.o. «mno.o. NNNo.o. Va mewuonnmcfiumwumucfl aNo=.c Hch.o neoN.o. «Nmo.o. quo.n. NNc¢.o ma ucmuuomswcsnuchuanw cNN°.o. moN¢.o NNNc.o. ONNN.O. NmNN.N Naco.o NH ‘ wanmusmwumflwumanmusmmu amm=.o. NNNN.o NNvo.o oHNN3N. cano.o. ONNN.O NN m>wummwcum>awamom HaoN.o Namo.o emacio NoNN.c. NNmN.o. oNNv.o NH H5w3mnmuflc ovmo.c c0oo.o NNNH.o. Nevo.g. (ch_n- Nmm¢.n o m>fimmmano>flu0m NNON.° nNNN.o ONoN.o. NNvo.o onNN.N. VNNN.0 a 0mammtmsuu noo~.° . omcmwo «Noo.o. msoa.o. mmnN.N. NNNN.N N mmmacuuo3loanm3Hm> ONNN.O NNNo.o NNmN.o. nofiv.o. NNNo.o. .INMWN.O o >Hm9IH3Mawsmmn mNoa.o NNnv.o emoN.o. mNNo.o. rnco a. anN.o m nmnuwoom NemN.o NNNn.o o¢NN.=. ¢¢0N.o. NnmN.n. .Jmnmuan v uNNMCSIuwmw nnn¢.o onnn.o Ncoo.o NwNw4du Nomn.a. Namfl.o n . meouuszun oNN..o Nonn.o NVeN.=. quN.o. .Imamw4du NONN.° N >am€aw¢5I>HmEau .vua.o o.N°.o. mmoq.o. came a. rNN°.°. oNnN.o N o . n v n N N 2133 ezmsHmmmxm pmmonoo Ho mmmm. vowh. hvhh. HhOb.. “NHm. Hemp. Hmmm. mmNn. ohms. mnom. Noom. moon. mmHN. Nmmm. mHmw. ovHN. memo. thh. momm. mNmN. Hmvp. moms. N: H onpmpom NmEHHm>II.am mqm x «m.m NH.HN mv.NH mo.m Hm.NH oN.mH 9> x huHHUImeHU «No;.o. HNO°.D v0¢N.o momH.o. Naoo.o- JIMNMM4N. 0N 30HHmnmummm© oNVH.o ooHH.o onN.a ooHH.o (onn.o mnvN.o a“ . tmmlwopmz nm°=.o. Nenv.o Nono.o NNNo.o emcn.o ununMH4d. . NH wmwmwmcpuum cho.o NHmN.o moon.o mnoH.o. comq.o ovem1q NH ucmmmm mcslucmmmmam mNnN.o NHHN.q. mnNN.a NNNo.o. («no.a. Gurn.o 0H xumnlumwa _ Hoo=.o nmmH.o .qumw4dn vnHH.o cnmn.o NNCH.N m“ M wUHmI%1-Hmm£ Nmm~.o omNN.o ocno.o oNHn.o. cho.cr ¢NNM4¢H NH mcHuonlmcHummumHCH omc=.c .IwMMMJdI vnHo.a- oovo.o- m»mq.o H¢NN.O NH unmuHOQEHcsuucmuHomEH cvoa.a NHNN.° NNVH.o Namn.o- a mNuN.a_ NNmo.o- NH wanmuzmemelenmusmmu oHNH.o oNNo.o Nsmo.o CNHN.O- uHmH.o HQNH.N HH 0>Hummwcum>HuHmom NNVH.° Homo.o mann.o nNeN.o- HOGN.: NNHH.Q 3H Haw3mlmuHc .m.N.o cho.o nom¢.o 0N¢H.N NNma.= oHNv.N o m>HmmeIm>Huum mmc=.o. .Idudedn noHo.=- caOH.o- Nva.o NHHH.° m . ‘ wmamwlmsuu vNoN.o. mmmN.o ogmm.a Nva.o. cmVH.: mowa.o N mmwanuuo3loanmsfim> onoo.o Nam¢.o «Hmo.o cNOH.o. ..maaw.q Nan:.o- o xamSIHSMHusmmn «ov=.o Nomv.o o~H~.o unmo.o mean.o av,a o m omnuooom ONNN.a nNcN.o HNN~.3 NnNH.a. pmmu.o Hocm.o v Hanuczuuamu mmo=.o. Nncm o moon.o moom.o. ano.=. «NNH.Q m . xumnuuanua . mona.a OHNH.° .Iawdu4dl vaNN.o. NNND o chH.o N adoeaucSIXHmeau .Iflauuqdu «NoH.o HHo°.° Homo.o. NNNH.° ooHH.o H o m c n N H APPENDIX F ANALYSES OF VARIANCE LAYOUTS FOR MAIN TREATMENT EFFECT 134 tlllIll l|| 'III'..-’I|I.. I T .1 ,0 I» n .- .H_H0pomm no mascao mmomo¢ BmHBzmHom pom ¢>ozH_J~mzoomh >xoumH Hzmozmmw: m 4 m < H m u 2 N _ a q > u a m N u 7.x (NI-.431 Joz¢ll.mm mqm¢e oaN°.c oNUH.o «H» m .N:. omzH_4NNNNONN u.Hm_Hroowh¢u zo HNCJNHNQ .x m_ ugm Hzmazmmuu w J a 4 h m U z < m w < > u U m a w > U! U I tl‘iiliu’lttssn I! 'll -0, l'l'olo pl 1|." ‘ .H Honomm so mQSOHU mm090¢ NBHmmm>HZD mB¢Em z¢0HmOHE ooo°.o «Hon.o «H» m .Nm. owm<:on H2N~u_uumoo zo_H.4mmmou MJNIHJDz 3N o¢ONNaNo.vaN 3.1.: I I .. - .vOmnooNN.oo _ no NNonmnHo.\Nn¢ .‘ on°.o oooon.n nmeoHNon.NNn N ONNonqnc.Nno .HNMN N-No >H_4.N NNNUNHNu an: .NNH vx m. m4m4_m«> Hzmozmmm: .>roomp¢u zu u a Mom ¢>OzOHH¢II.mN Ema/WE mvno.o mnnN.o (Hm m .Nm. ommH_J.m - muz«u_g.zu.m .xoaamn . an .numa . uc Nuaacm macaw A." .x m“ m4m<~m<> >moomH szozmmmn A>roowb u 0 m H m > . q .... -’. .‘l-’ I..- l. O. . ....... ’I: .‘u;‘ l-Jv.¢\l \cl.’ .(lrlu I} - I. .III Iv‘,.v i- lull..o VV, N .. III.II|. ’Iltl fl - . c - - - - - N o . - : --s; ,l - - N . q, - 3 1; I .H Honomm :0 masono.mmomo¢ Bzmszmmxm Hem ¢>oz ,xouqua . H zmaxw .va .x m. m4m<~a<> quozmmma .>roow»¢u Incw com z u n m ~ m > J. a z. a mozoumm.mwum-. an”.-»zwmuam WNHN “Nam. NNMNNHN . .. -- - -. k. . ,i - IL40' v I.IIIII.III IIIIIIII II II .HH Honomm so mmsoHu wuoH N Hmoeoz».4~m >moomHNo -.. - muzoumn m«.o osxuuxoo mzuk hzwccao rw<4 wuzum ammu<4m «racewpcu 104m «on z10 an I ovwo mtmh hzmcmau «Hun u.4 m<4 .IIIII III I III ..II.II.II.II II II .x m» o.« » w4m<~m<> kzmozwmwn m.u 2:. h ¢.< > u o 22 k0 wmm>uflq7 m h m y 4 4 z . .III JNHOH wwaxoomp _No muusom I, ' ’ II II.IIIIIIIIIIIII.IIIIII.I IIIII‘I IIIIIIIIIIIIIIIII 1L - - 4 :1I. - - - .HH Houomm :0 mmsoao mmopo< weHmmm>HZD me¢9m z¢0Hmon mom <>oz --= -z: .-, zmoz.u.u.zc~m wxomnam: 130mm 3m: .rzouwhx - Is., no .wowo .a I.x m..m4m<~x«> >¢oom»«u;-.=..: ..I .uud .x mu m4m<~m<> pymozmnwo I II- III IIII. I I no wuuanm u o I m H m > 4 < 2 ~ I...IIIIIIIIIIIIII. .I III zz no mmm>4-u.om mqm¢e HHH HH H moeucm mama. o¢.HH mm.mH ¢B.OH oa.¢a mm.¢m sm.mH o> x mash. ~m.m mm.mH om.o No.HH oormfl mo.mH 9> x xuuaclcmwau ¢¢mm. om¢=.o. oNHo.o. “ono.o vmvo.o. 14mma1au .nnco.o. cw 3o mnmlmomn oombu 0000.0 . munm_a mwom.0c vn00.0o vmmfl.0u vmam.0 0a @chmmmmx ommm. v-~.o amd~.o coco.o nkvo.o “fivm.m. omoo.o ca xmmBImcouum omam. mmm~.o onmm.o oom¢.oe Homn.o. mnmm.o. nemo.o NH ucmmmmHQCSIucmmmmHQ moam. ovmc.o ncho.ou o¢vv.o. ac~o.o agnm.n. floun.o 0H >Nmnuumwao mmom. .Idmau.a «oc~.o oomfl.o. ovoD.o. fiwmo.o. .05cn.o a“ xoamuxmemwz mmsm. mm.=.o. «ao~.o eaofl.o. hqmc.o. «mmo.n. mhmo.o. ¢H mewuonumcfiummuquH «mom. odo=.o amv~.o noo~.o. vmmo.o. mfiqo.n. Nficv.o ma ‘ ucmuuomEHCSIuchuOQEH mmwm. osmo.o. nn~¢.o nmmv.o. o~h~.o. cmfifl.n “aco.o NH wanmusmmumfivlmanmusmmu whom. amm=.oc nxnm.o Nflvo.o oahmsa. cano.o. omnm.0 a“ . 0>Hummwcnm>awamom vmmm. aac~.o uamo.o cn001o moflm.o. cmmm.o‘ oakv.o a“ H5w3mumuwc noun. 0¢mo.° «coo.o Amma.o. eavo.9. .Idmmmimh homv.o o m>flmmmalo>flu0m moon. ama~.o no-.o omofl.o. ~\¢o.o onfia.n- v,nn_o a omHmMImsuu mmmn. n¢o~.o omcsjc «Hoo.o. mac“.o. wmn~.n. mhn~.o n mmwanuuo3loanmsam> mflmn. oa~a.o okco.o numfl.c. nofiv.o- mmfio.o- .Immm~.o o >HmansMaw5mmn Hmmh. m~o=.o «anv.o ono~.o. mxmo.o. rhea o. ¢~n~.c m Umnucoom Hams. u.m~.° mnmn.o o¢fifl.=. ¢¢o~.o. cmmm.n. .IANdm4d. v uflmmcsluwmu «mfln. nnnc.o omnn.o acco.o mcnm4du flown.o. «Hm~.o » _ xumnuunmflun swan. o~mc.o monn.o s¢c~.=. vnv~.°. nflnm.a. mo~«.o m aHmEa Gala meu nnmn.. vvau.o ovso.ou cmoq.o. came o. rsfio.o. oHnH.o a m; o a v n w « .muzhaqu m0»bII.aN mqm x vamp. vm.o 0H.H~ m¢.~H mm.m Hm.~H on.ma 9> x huuflflucmmao thh. v~oa.ou «noo.o vovm.0 moma.0. 0000.0- . nmn0.a om BoHamLmImmmU Hpon.. omVfl.o ooHH.o o~m~.o ooHH.o «can.o mnv~.o 9" _ tcml>CLcL hNHm. 0000.0; uvnv.o mono.o mmwo.0 emcn.o :.n. . mfl NWmfitmcDuum aqmn. oaco.o mamm.o moon.o mnofl.o. comv.o mmemfiq NH ucmmmo mcsnucmmmwam Hmmm. m~n~.0 namn.a mnam.0 mxmo.ou «150.0. cnrn.o ofi awmzlumwao mmmn. Hoc=.o nmmfi.o .IddflN4du. vnHH.o cnmn.a mocfl.o m” H .Uam1%1 mm. oven. nmm~.o ocm~.0 ocno.o oxfin.oo camo.c: LPNMMNJL H r v” mafiuonImCfiummumucw whom. omc=.o .IwmmM4d: vnflo.=. oovo.o. c»mv.o w¢n~.o wfl ucmuuoaeflcsuucmuuome meow. 0.03.; mflu~.o nmva.o mamn.o- .1nwmw1q- ammo.o. m“ manmusmmumflnlmanmusmmu moon. onmfl.o ammo.o mnmo.o cxmm.o- “Hm“.o morfl.o HM . o>aummmcuw>fluwmom mman. mmva.o Homo.o mmnn.o n~¢~.o. moom.; mnfid.o 3H H5w3mumufic mmmm. env~.o .Iamum4m. nomv.o omcfl.o unma.a oaxv.o . o m>flmmmmum>quum mamm. mmc=.os nqumqqda moHo.=. oaoa.o- nknq.3 nama.o m . wmamwimsuu m¢Hn. emo~.o. mmn~.o .Imqmw.q= mflv¢.o- cm¢fl.: mo¢~.o m mmwanquBImHQmDHm> memo. onco.0 nmmv.0 «H00.0 vxoa.o. >u¢nc.0. o HszHDMflusmmn much. voc=.0 ~o~¢.o omflm.o homo.o mean.o acre 0 m vmnucoom mmmm. cmax.o mucu.° fl~m~.a mnud.3. rmmu.o “oqx.o v adamcsuuamu mmmn. mmo=.o. «new a moon.g maom.o. snflo.=. vmqfl.o m . xumnuunmfiua amen. wona.o oama.0 .IANdedI v00~.oc nnm0.o cuvfl.o m adoEaucslaamEAU mwmh. . .IdeuddI «~o«.o Haco.a non°.oc «mna.o oonfi.o a w: o m c n m g .31? ~3an .40»qu 3mg <50? APPENDIX F ANALYSES OF VARIANCE LAYOUTS FOR MAIN TREATMENT EFFECT 134 .‘l. {1.11 I t 31.1 :1!!! vlul'. I . \. I ‘ 1 II- :‘ililtal. till-9|. 4 Li. :0 o (51" 'Illlll‘iliul , .l Il.y t ..H.nopomm so masono mmomod emHBZMHom Mom <>oz»_J_m >uoumh bzmozmmma {I ~>roowb u o m H > J a r « mazOUmm Hm.o- mam» hzwmmau hm‘J muzpm ammu44m . 7?<)JJOI uuzq~7q> u: wry» I :I 343! Lfn. .EUIIDD E ‘r‘ 1., .. .1.Iu{.‘.\.«.‘ II‘EI‘q- 1136 .H 9090mm so mascao.mmoho< wmoek.4_mOZ yxcomyqu a‘J .fiwfl .x m. u4m<~m<> pzmozwmua .>room»¢u zu u o mazaumm «a.a m .p,»zmm¢3u ~m<4 muzum owma<4m onxonxoo mb<3 an ; ndon mtap » wa130 'r H J: ). 7 ¢L hr. ”UCJCW '9..'..'i 1“., t" ‘5“-..II (I. O4 . 'Iv ..‘ 11 1-!1- qp‘i my .......1r‘l= O . .I ‘Y I1l1"lltul. ’IOII I , I- .H Hopowm so mQSOHO.mmoHo< WBHmmm>HZD medem z<0HmUHE Rom ¢>oz._4.mroom».u zu 0m: .mm« .x m. m4m4~q<> pzmozmmm: w J m < » memymw an owomxaflo.wm~\ J«»:~ no mxonwnao.\nnr umhapmmpnu vmxhwc m o\nnnvnc.\no u5umonuh no .mum: >m00mp u mum(fifim no 23m 3 .C n > .J < I <1 r." .‘ ‘....’..v.l...!h ‘1‘ 0“!“ fr..-"lll.lllllll I. 0"- . nu - - 3 1; , - -a ‘ - ¢H-nopomm :0 wmsouu.mmopo<-mmamHzmmUOHm pom ¢>oz¢un.m~ Mgm¢e V mena.o mnn~.o «pm m .mz. amm»_4~m >moomp pzmazwmmu .»roomh u o a _ U) Joz¢nu.om mam¢e «moo.o , oco~.o , . . chm m .mm. omm>~J~¢ pxnombqu « zwaxm .vma .x m. wqm<~a<> pzmozmmm: .>room».o :u(m mo“ z‘m: mh u o mazouum «m.= mx“» »zwmm:u »m«4 muzpm omma<4u eh\nd\oo wb<3 on I ndOH L[~. _Zuzrnu v..(.JJ)._ .L).(~I1> L0 Li. J Hy. . tzflié‘c: . . . , » e - - - l o . - :. -g,-- , - : z . o, .5 1; - .H nepomm :0 mmsoHomeono< Bzmszmm m Mom ¢>oz».J_m goz yxoowhqu . « xmaxw .vmfl .x m. m4m<~a<> hzwozmmm: .>roowp¢u ru u o m _ m » 4 « a q oozoumm «M.o mtg» pzmmmau hm<4 muznm omma<4m Oh\nd\ou Wh‘i on I Mao." mluh b2w¢¢bu 23(LJJDB b)..-(~t.€> g... huI‘Jiwa‘ W.FEHE§ES_. ‘.A i— . -- - -- ----i-ii---ii-i-ii i..!-i|lif;: ------:.-i-..i!....-i,.313---: 2.- - - 4 1 -. .HH Hopomm so,mmsono wmoao<,amoeoakngum¢mo¢¢ o_»m.+<»m g m¢ moz‘u_ufizo_mluxo¢aa< to nmomo‘. .uo wuzuom . u=o¢u .a ..x m" w4m<~¢<> >moomp¢u 7-: .- o‘g .flua .x m_ w4m<.¢.> kzmozwmmo armoowwcu xu u o m a m > J < 2 c ll- '1-.lll. ,nllA II . c I n! n I. . . .‘a 8"! .. ’Y'isv'lalll-..l ilv|'.ltl' .. II?! |.|| illfill mnzoown mdoo mtub Pzwcflbo rad-a muz—m me1(4w ah\NN\oo mh k0 mmm>4<7(- ll. .01Ari ..- .II ‘ C‘lc ’. ’z‘ht'b) v... llulcllv I? 14? . 1L .- - - ‘ AH -.1+ - - - .HH Honomm no mmsopo mmoao< waHmmm>HZD meoz¢uu.mm mqm --1 :.- AAAAA :muz.u~u.ze.m nxomnu‘2.‘. -- --i;:;aa;i- .- ;‘ no .mowo .--;i-.;iq..:‘ no muaaom 3m: .rxaump‘u xuz I- i..- I l l'1iill'll'lllllvlll'rvil":ll|.lll'. gnome .« ..x m—.m4m<_x‘> >¢oom».u;. .: ..- . .-d .x m” w4m<.a‘> szazmmmo . u o.. m S m r 4 < 2 4 22(1440m m02<_&¢> to nun>J12¢ IIIIIII.4'|IIII ulI.‘ .le. oooflmooo... .- ‘1 ,|oo oovoc«o«.~o~ u omoofio Noxoomoooo;;: .- ,,- zmzoooo .oou .x no mooooao> ozmozmamo .oooowooo xu I, .‘ ; no wuuaom m _ n > J 4 7 o Uthorli L) 0U3.JK.CK ..I z .1 rt . . .._ I. .. , Ir: .».v. (I. ..:...I..... ,I I III IIIIIIIII II III I II IIII I I I III I I III- I II I II I I II I II III I I II I I I II I I I O I I I I I I I I, I III I I I IIIII I I II I II III IIIII I I I I III I. I II I I I I I III I s .I.I II I I II III I I I I I I I II. I... I II III III II I I I .II I II II II II I I. 1.... I I» II,II III I. II I, I I of I I I IIIII‘IIIIIII I.II.IIII IIII II ~ III III I I‘ll III.» II ,- I I II .I I I .I III III I I III I I II I I I II I I III II II IIIIIIIIII II II I. I I I o I II I. I I I I I I I I I III I I I I I I I I I II. I II III I IIIIIIIIIIIIII III'I. III I III I. III I I I III. II IIII I III ,II I I, I I I. , . I II I I II I I III I I III I IIIIII II I I II I. II I I II III I II I I I. I. I III I I I I I I I I I II I I I I I n o I I IIIII I. I I II I- I II I I I III. II .I .I. I -II I I I II III I I III I II. II I I I I 5 III IIII.I.I II II II . III IIIIII IIIIIIIIII I III III IIIIII III -- : 3-- “wwwosflmmm- mm mmmomwlw $34 Hamzwmwwwwigmm.,.¢mmz.4uun on -Hmfi -- coho.o woomno «om m .N:. omzooom u ozw.uuuuwou zo.»<4mmaom mgauooo: onvonooo.ooofl . J‘Foo azaI-:- o.o~.n.o.m~: :1. g... . : oo ammomnoo.nvofi omoooow».o zozooo oso.o onvoonm . 1.- acomoooo.oo . . . noodfionfi.ooo .o.»m o no >o.o.o.oooo o_»oo»‘om u wo.oom z.my zoomwou mmo moz.u.u.zo~m wxoaa a :.I . - .:Ia:;-- . - - I - no .momo .; . . . . . no moaoom aoooo .« .x no moooomo> >ooomooo. « omoxm ..~« .x ma moo<_o.> ozmozmowo - armoamp-..u o mnzouwn dd. 0 mw~moumh 4 < 2 . m;~» ozmmaau omaq muzom nmma44m ON\NN\OO wkco on I ovno mt.» pzmcmau .- . - , .,,;. .. , ZZ‘IJJDD LLJI'anl) s) 7’).I.III “WHw:Houomm no.mmsopo mmonoq emHezmHom Rom.¢>oz uuz ymouwhao wuzm~um «cud .x mu w4m<~m<> rzmozwmm: ~>moom».o ru . h o m * m > 4 . 2 « .wazoomm o».o wx: Ipzmmmau pm<4 muznm omma<4m 3\\’d.\°: HI- .I '\\I‘ J.... .I.Iar...l\: #145 o~.o.e .HHH.gouomm am mmsowu.wmogo< wmoaoz¢nu.mm mqm¢e . -a~.o (pm m .«m. omm uuzmoom.«u zu¢m «on z >¢ouw».u .Hma .x m" w4m<.¢<> pzmozmmma .w 4 m ‘ p m u z < a «‘< > m o m _ m > 4 q 7 « 22(EJJ3D M03<~31> k0 mauxdirfi is... 4‘4- .‘OOQ'!. Iv.‘ III» . .‘c. :1146 ..HHH youomm no mmsohw mmoao< weHmmm>H22 meoz<:u.mm . nnno.¢ owma.o <>u m .um. omm._4.m.mo.¢ u.»m.»‘»m U muz‘u.u_za_m .xozma. .mazoomm u«.o wt.» wzmmaau pm 4 uuz~m omma<4u Dh\.d\0° Db43 HI I 'h d In.» bl mambo nmaaonso.~n mo.moaso.on wm 3 no N zoowwz* no .moma n¢mo~o¢n.odnm mxmmnmo~.~v~m °~o°~n¢n.- muw¢oomh hzmozwawa ~>moomh.u xu«w gag a(mz mh<¢ go wua;cm w u z « _ a < > u o m _ m > 4 q z a ttt$ ', I. v'. I. Il'.-|l‘l‘.| [I‘ll Il).vl| !. I: . PO“... . ‘ II I l i 2147 .HHH popomm no mgsoau-mmoao¢ MmemHzmmUOHm pom «>oz muz‘u_u_zu.m-.xomumn - , no .mcme . kc wuancm macaw .« .x mu w4m<~¢<> >xnowh pzwozmmwa .»roow».u ru¢w can z u o m fl m > J 4 7 4 .nazouum ua.o wx~»‘»zm¢¢3u .m‘J muzgm omma<4m oh\&d\oo mbp.4.m muz.u_u.zu.m..xozuan no .momo kc wuwgcm macaw a zwmxw ~>moow».u zuqm «on z >¢oomb hzmozmmwa w u z a _ m 4 > a o m _ m > J < 2 a D.O.‘..|ala): In): {oz-l) \JIJ‘J . IJ‘I.‘ r... 1 f! E 3 i u.— I I'll III“ .. . II .tl‘tyt- rl'qolb--1111 v 9 4 1 \ I 1«| 6. .¢ Hopowm :0 masons mmopo< wmmw¢mammmmmmm1¢>am4uu.angmqm«e «5.9.. nsd~no (rm m aux. awcp.4.o .izrzu:::ti-lni9¢:1u|i!-Ii:.::i:xuo*ummwa-,;sk,;:- ‘--. :. . no wucnom , ;.;aaomo;..a 3.x m. m4m m>moowhqu c<4 .«m« .x m” m4m<~¢.> bzmozmmua .pxaomp‘o‘:o‘ u o m _ m > 4 q 7 . «av ‘ 1|: 1 n-i‘ I ! 4 I4“Ot 0 ulnllui'lini".n"l¢ll|lll' III. I'll" IerCD I! n".t.l'|vil?, II.| wazouun ua.a mx.»,»zm¢¢:o >m.4 muz.m ammu<4m as\ou\oo wrcn on I onvu mtuh >2wuzao ,‘- ,.‘. : ‘. i zz kc umqucrc nu - - :4 l , . ‘ - .m Honomm Co masonu mmOHo< BszHmmmNm Mom <>ozN~4~o moz >¢oowh¢oow»¢o zu .k 0 m _ m > 4 q 7 N I I'll .In I llliii ill Illllii {iii} ‘I.I I I . lU7(»&Q> k0 VWW>J<..(1 . z m .ma<4m mozoomm dd 0 mummihzwmmam ”MM“ mWEW» mnulxau 22oz~sa_m«mo¢m u_am_»<»m L ax.nom zaul Inoumxi muqamam us gum uuqu14, muzao.u~zo_m .xozam‘ no .maau . uc Luxncm msozumam .N .x m. w4m«_m<> .noomaao wuzm_um .cma .x mu mam<~z<> pzmuzmawg .>roum».o zu a o m a m . a q a q mozomun «a.a mt.» pzuzmao _n44 suzun aun1ozauu.mm mamaa omN°.a mona.o «by m .N:. ama muz.u_._zu_m,.xoz¢¢¢ uo .mwss . ea mzoxomom am vx mu w4¢<_m<> >qnomp hzmozmdma .>roow»-o ro u e m . m . . q a q DDZODMM "4.0 HX—b PZwICDU hm(J muzum. 3wmu<.._m \a‘\ul“3’ ' ‘1‘)- , all I l.v‘05c'. ’IIO: 15E3 H.90poma so masonmpam mmoaoa weHmmmeHz: maaam zameon pom a>oz»_4.m muz«u_a_zo.m .xomam. no .mowo do womacm maozomam .N .x ma w4m<_m<> .eosmeau am: .NNa .x m. m4m<~m<> pzwozmawo .>rouw»«u zo u a n _ m r a q _ ; )s t’- b)- mazooam ou.o wrap pzwmmau pman mw¢_m ammuxuw 1159 O I mozoomm ua.a .!O|-| .u‘l!""nu 1".‘1 :1 I. caco.o m m “we. omm2 zmnuukumcu zonh<4mm¢ou w4m~wqat Noo.° .pamm u-uo »._4.mamomm uuzao~a.zo.m .xoxaa< .>xoamp.u zu .nNa .x m. m4m<~x¢> hpmozmmma .moma m u z < _ z ¢ > u o vxnmfloen.ccv ooomoaoo.nmv nmnnao'n.¢ mua J ,.Y_(.9.JJ)VL muhwfim. HJPQNTGD .1. mwumrnmwxu kaipmx mmawoeuezm.u.uumou zo.~.4m¢¢ou MJN_»J=z mN ooaoeonc.naan - .Heomeo...nv A nN «NNKNNNN.aoa noa.o ~e~Nu.a acacnoon.oN N canoecoa.nma .» >¢oumh pzmozmuma .»rouw».o zu‘m can z u o m _ m > 1 .oazoomm « .o m:.» :szmaau hm<4 muzam om ma<4m ch\ed\oa m: on s nuca ctr» meDU Z? an "_uyarm .UUVQ~Y«\> an», Uflr.>J<...( l’le‘UuUb‘u {Ni(.b .OAJLI 2‘“! JP‘K‘IIWI' ‘ EIIIII‘I I 1 | .u.’ Lr‘II o - -14 .l . .. .. : -, K-,. 1 .I C oftisali; - I U. i: c tillllloco-..‘ lll‘h- - l i - . II Ill!\l.| I'll ll'l'. I! III. till“. IIII. III' I I it-.- I i u. 1 t '1'» 1r 2 2:, .:.,.;_2 .. . ---;-..---;i:;-,s .:-i _.;3. ,-;-. - .2. g- 1 . - -- - I , - 6 - -- - . - -1- -l- -3- I- 1. --...u -1 I - -0 ; 1a. , w - . - .HH Honomm So masonwnsm mwono¢ NmoeOz¢ll. mv mqm¢e N..o.. .noo. a «pm m .Nc. am¢.aom g _ . C- - : . , .zm.u.uuuou zo.».m~«.oe u4.~»4=: . . . ., - . - nu. -z- «asao.~«..~.!. a:::s::-- ”N Nno.o i s.Nwocono _.-- .NonaNNoo.o . , ..g 1,N 1.: .h.»» a no >p.4.maco¢n ubpm_»<»m u-.-; amazon zamx. mozau.u.zo.a .xoaga. zoomwmu . - . L «nonnaoo..o. J.ho» mm.¢oom»¢u z_:»_: I: In". ‘ll‘ll 1 . I.‘ Isoaowdoum.uov .nnNN.nNo.« ; :-;-T-- mm.moawa -: --: ;:uz:::::.:; --I|:lusa.-:.---z -I:A-:;-..uoramomci,--azulrl -:r:!:::;-- ;. .;:no.muc3ou.i «saunas» .N .x as m4m<_¢<> >¢oompao :i;. -, -:.: o<4 .uua .x m. m4m<.¢<> pzuazmama . .scoomaau xo«m cog z--:u;o a _ n s 4 . z . nnzoowa mdou ml-r. blur-.30 rn‘J wuzun Dmmn(4w 9:1- - I I i: ,tar - -III ch\NN\OO lh(a B! I Q'fio Utub FIUICDD . -. . y‘rlttJJDD. - ub?¢~lq> LO nuhfiJctl I’Nloa_w.bv(u 10‘” '01.. 2‘“! night‘s‘l" -I; - v- .- .... u or --. II t ... . i-.. .l. .3191. I: 0- -2 l.----l I: ult- If -21... ,.-|iu|-ll. Incl-ta. xiii-21:: 53.11.13.1- - . - . I.¢.I--:llla.l.bnll--l»n -I|u:l:-!n--.-. III .lIiIIeue!.Iil.n.l.I!-..;n|-. - : - . ,-.--:- - 2. - .. -- -- - , -.---- ........f.-. :---------.t-- ,0 1--. -- -- - - - .- -- -2 -- -. -.-.-._...-.--i-l-i.. .HH 33$ no mason-wag $83. Magma->2: mass-a 2.8%on .Hom <>mz¢....-U$w..-ama§--..---... -. onaa.. _ <~w.m .Nx. am¢ - - - -.wm¢.u~._zo~m ”mamagm.;;:1- -.- :;i-t-::.e - z =:-.---:-.-. no .mumn:-:...-:zz.:::-:-;:u.;: --- - -- -ko.mu¢=om- .maoaamam .N _..x m. m4m<~mq> >¢oame«u -:- . ; : am; .NNa .. a" N4m<.¢.> pzmazmama .>¢oomh¢u xo«u «on z¢wx mp<¢ no nuh>4¢>< G’zocuh‘u 1“! It I. ha‘l1'l‘ .‘o! .1-111;l:l!1:s1112111111111 1I111-1- .111 11-- - I; .1 H 1 1 3 6.- 1 - 1 51.11.11 -- 1 - -- .1 1 11! 1111-1 1|..- 1-.- 11 - 1,- 11 . 1 .HH nopomm no masonmnsm mmono< amemHammoon ham a>oz<11.mv.mqm»_4.uamo¢N :mnz«u_u~zoam .xo¢.N< 0". «gauche. - 1 ompnu+<»m.u . 91-..: 01,.Ilnl1l 1| maoaooam :mzuoum .>¢oom»«o xo1 I 1..I V I o lbll lat.“ I!!! ’0’} liY-"v 1.1 | 1 .l llvl’l ‘1 .|l ‘ On Ivlnl .11 I. - - u . I-I. II lllll.ll.'||l\lll1'l|.ll.l‘l nu _somaovcn.aoo.. an.mo«n.sno N auspawom.cc.- -1- maaaom z.m: . xoammzu - mmm<=om.uo ram 1 11--11---:_.- no .mamn..-111.111111-1111.1-111.1 .m .x mu mao<~c<> >¢oam»«u .nmu .x mu m4m<~¢<> hzwazmuwa .1m.u.nm<:h- .m u z ¢.. msm->1-1u o- m .-» > u . I'll". I '0 thtJJDD 1: .I ‘l'l'.|u.|.‘-‘.5 'Il l I Nutty...) Lb . thc» -»m.¢oow»«o 2.xp_g -nm_¢aomhao 7mmzewa Noz._¢.> .go muaaom- 2 c. bhh¥dtt‘- R’Kcomir(u 10‘..- .Pb-UII 2": {In-’1lh1llrlsl 1 '. I “II l l 111 l I I III I I I I II II III III III! I fII.1IIIIIII III I I o I I n I II. I IIIIIIIIIIO. II II. I I ' ‘ I ’ IIIIIII - I 1‘ - I l I I... I II- ‘I I I I ‘ - -‘l‘- - I -‘1 'III. 'I‘ ' I I 'II I I I It -‘1 - -1 I I I I I. I II III I II I I II II II III I III I III IIII II II II-II I III I IIII IIIII IIOIIIIII IIIIII I I a IIIII II I -I II I III II I. I I I I 1 II III I I I I, I I I I I. I I, I II III I III IIIIII IIIIIII III .II I- I .III-1‘ I II II III I . It I I- I I I I II II III O.I III IIIIIIIIII .Il I.IIIIIIIIII,I.III II I I III II III I. I1 0 I . Q I II II IIIII I I IF I III IIIIIII I IIIdI. I III III I II I n I... I I I I .IIIII II. I .II .‘II III." III OIIII,‘ III III I. II I II- I I I I I I . c O O I II I I III! I I I. III IIIIIIIIIIIIIII I II III I I I II III IIIII III III IIIIIII III II I III II . .I. II I I I I I IIII III II I III II. I II I II- I 1011 I 1! I.I II... I II- I II 1' I II I II I I | I I I I II.I'IIII I III I I IIIIII III II .1H1H Nopomm :0 masoampsm mmofié azmfimmmxm .Hom $6311.: WmflaI-II, A: III III III IIIIII IIIIIIIIIIII I I III IIIII III. IIII NNNN.N cone. 9. , . «Nm N .NN. NNNNNNN N 1 . - . . -..- -. - Nzu.u.NuNoo zo.»N_4_NNNNNN uppnsNNNm.N- muz.u...ze.u ..onN1.. III. III-III. IIIIIIIIIIIIII w¢m<4 muan ommI szazmnwa NN; NN, ..IIIIIIII I Naonooon.No- .1-;-..,. . , 1 N .Nr NooNNNN- No-WNNNN. II .III ,I II.Iv1l «nonmqo¢.nvo NNNNNNNN. NNN IIIII IIIIII I .x m— mam<~¢<> >¢oomh .-.1 -. 11No.mNNaom o u n > 4 < z N ”9"ht.’ t D'EJ'.‘ Joz¢ny§ mam: 2.«. . ovn¢.o - «pm m .uc. amc‘3o. pzm_o.uumou 20.».4wxmoo m4:.»4:x ow enovonmo.¢HnH J uuz >¢oomh szozwmmo ~>moompau zu u o m . m > 4 ¢ 2 I .aazooum «w :2. apzmzzwu pm<4 moznm omml<4m 05\ «\09 D» “I t (and wtpb hzwttno . Z7 k0 mum>412< 166i vnnzonmm «a.o ah\&u\oo «Hun ~>moom~.u xuoz¢uu.m¢ mam¢e I .;r .’ 4.41: .rbbll Ilzsw..‘ .1~ soo~.° om¢<=ou uzm.u~uuuoo zo_H.Jw¢.ou “4m_p4:z nun.o .H‘Hu u-uo »._J_m.mo¢m u_pm.»‘»m H muz ymoowh ~2mozmmwa 72(IJJDT Jqpo» mmnmouwh no wuaaom w J m < h w o z < u m < > a o m H m > 4 « z < m03<~t<> no mumxd<2( ['10‘11t ' . I YJIIIi I .HHH- Houowm no masonwnsm mm0ho¢ HHHmmm>HZD MH¢Hm z<0HmUHE Mom <>oz¢ll. av mqde ~Hon. a . oHoo.o (Ham . .mm. am¢«aou HzmHuHHHmoo zo.»n4m¢aau m4u.»4:z aw HomnnHoo.~on quoh - onmmnaoe.¢~ om onmmncn..CHm mummcumh».4_m‘mo¢m o._m.H muz‘u.m.zo.o .xcmugu no .mcwo . mo wueacm asexumam .m .x mu w4m<.¢<> >¢oumH Hzmozmmma ~>m°omp.u :u‘m gnu z‘mz mH‘¢.¢mm « wroag<. m 4 a . H m u z < H a < > a o m _ m I J . z « a» zmmm: no »m«4 mqu» amma<4u >1: 0. 0 con“ wrhh bzmmIDU .mozoumm H4.o w: onxpuxoo a (2(14403 wuz no wum>J<7I 1-- I ‘1 168 .HHH Honomm so mmsonp m mmoHoH HmHmstmoon pom ¢>oz muz »moamH Hzmozmmwa .m 4 a 4 H m.u 2 q H a < > u o m H m I 4 < z 4 22(tJJDD muquz<> In mum>J¢r< “'uv.|' Ib|-|| II...!'J I‘lua.‘ nlA' ,o'ltl,ll xlr 130-.., 169 [HHH mmpoww :0 mammnwnsw;mmono¢ ezmszmm m nom «>oz moz«u.u_za_u..xozmm. ., --. . A to .momo . ho muaocm anoxumam «N .x m. w4m<~m<> >¢ouwh kzmozmuua ~>moom~.u :u.w co. z¢mx m»<¢«mmm ¢ mxoaa<. m 4 c « p m u z ‘ g a < > u o m u m > J « z . nazooum o«.o mt.» pzmauau pm<4 mozum omma<4u onxmaxoa up k0 mum>Joz¢uwummfim§ once. a .cua. a «pm u 3.: am¢<=om ; - pzm_u_aumou 2a.»‘4u¢¢om “4;.paax n~r ‘ . «nmnmaov.~nn Jako ... a.:auli=‘2-s::;xz-;.;;.s:-:;:i.azansgnaoooo~n9¢ .:I:|lai-v---n~_-a;.i.;nooooomc.o~n,‘.-i wwwmoompqu z~1»”1 ~o¢.m ouncanu . - ”Noouoov.u- :. ., 1 ;~ . ocmnmqoo.~ mwnmoowh-4_o muz ,xoom» hzmozmmwa .>¢oom»¢o.xu¢u-¢ou,z¢ux m~<¢<¢mn:¢.nxo4J¢. iu;4.a.«:fi::;mmu.z { .:¢-<.>-. u o m a m y 4 < z . nozoomn cu.c w...» #201830 thJ wuznm ammnzam :33: m»: on- r on: 1m...» $953.. - . -- ; -. 223343 371:; L.; $91.14 C _ a .IIII .II1II.|£I II II IIIIII III I IIIIIIIIIIIII III II II I I III III I II I I I III II I (I II I ,. I yIIII III I! II III II I.I,I‘IIIII I I I III. II I.IIII..I IIIIII,IIIII II..I..IIII III II .I‘ III-II !.I, III. II.IIIII I I I I . .... I.IIII.II ‘II‘ I III _ II IIII II I I1 . III II I IIIII I , .IvIIlnII III . II.I.I .IIII III IIIIIIIIIIII IIII III.I I I III IIII II: I I v I III I III I I II III I..IIIIII I .I I‘II I III IIII I I II III-II (III III I III IIIIII IIII I III III I I I I I. I II IIII III I I o I II I ,,,,, II I III I I I I II II I‘ IIII II. I I II. ..IIIIIII II I IIIII .3. I IIIII I II I I I I I II I. I. .r . I : I .III . II I l I IIIII I IIII IIIII III III I I I I II I I I I III I II I . I I II I I II I I III III)! I! I.I III‘I III I. III IIII I I I ,I I II I, I I I I III III II I I I I, I I II IIIIIII I I I I IIII II II. I II I I [I II III . : III .I. I I I I .I I I I I I I I In I I I I I II I I I 1 . I I I III I l I I III III-II III I I IIII II I IIII I III III. I - 3.-.! .m Monomm so masonmnsm among EmsHmmmxm mom §oz<.....mm.mq.m¢a o¢oaru magnupj-- -g,i; «rm m aux. am¢ moz«o_‘_zo_u “mamas. i-u;.ne-! I;-::§i:::a:;;:.;lua;izuo umomag. g..;-: ..:i 1,... ac wuanom -,.: 1.. t ., I .;.,i:;a=oaomam- .w :.x m~,w4m >¢oom»«u « amaxw ..~a .x m. m4m<_¢«> hzmozmamo .rcoumhau xu-: u o - m u m > 4 < 2 . I III-I :IIIIIIIIIIII‘II III.IIII‘I I II mnzoown «u.o ox.» pzuzcao um.4 moz.m nmmg<4m Oh‘Dd\°D ”ECO 1.0” I UHJflTiutfih FTWKCDUI; -. ., . . . ZZ kc mmm>4<7< HICHIGnN STATE UNIV. LIBRRRIES m\IW‘IWIIH‘I”WNWI’llWWWINIWIHIIHI 31293101995359