READING PERFORMANCE OF VISUALLY IMPAIRED PRINT READERS USING STANDARD PRINT, LARGE PRINT AND MAGNIFICATIDN 'Ihesis for the Degree of PR“ D. MICHIGAN STARE. UNIVERSIW IOHPI DOCK 19H This is to certify that the thesis entitled Reading Performance of Visually Impaired Print Readers Using Standard Print, Large Print and Magnification. presented by JOHN BOCK has been accepted towards fulfillment of the requirements for Ph-D- degree in mm .7. J flQZ' / 25‘ iwk. / {‘\ Major professor (3 Date 52’ 53 / , 1/917 9. 0-7 639 LIBRARY Michigan State University ABSTRACT READING PERFORMANCE OF VISUALLY IMPAIRED PRINT READERS USING STANDARD PRINT, LARGE PRINT AND MAGNIFICATION BY John Bock Sixty-four visually impaired elementary age print readers from eight locations in Ontario and Michigan were individually tested under four reading conditions. The purpose of this study was to seek objective information regarding the relative effectiveness of standard and large print, used with and without magnifica- tion in facilitating the reading skills of visually impaired elementary age print readers. Specifically the study sought to determine whether the reading speed, accuracy and comprehension scores of visually impaired elementary age print readers would differ when reading twelve and eighteen point print with and without magnification. The children were individually tested with four parallel forms of the Gray Oral Reading Test. Reading speed, accuracy and comprehension scores were tabulated. The prediction that there would be no difference between performance on large and standard print as measured John Bock by reading speed, accuracy and comprehension, proved to be accurate. It was concluded that visually impaired elementary age print readers, as a total group, (when no differentiation is made for acuity level or eye defect) perform equally well on standard and large print. The hypotheses which predicted that there would be no difference in performance of visually impaired elementary age print readers as measured by reading speed, accuracy and comprehension, when reading either standard or large print as compared with reading standard print magnified, were all rejected. It was concluded that magnification of standard print was less effective in facilitating the read- ing skills of visually impaired elementary age print readers than either large or standard print without magnification. This conclusion was also based on comparisons within the entire group. The prediction that there would be no difference between performance under conditions preferred by the subjects and alternate test conditions was found to be correct. It was concluded that the subjective judgement of visually impaired elementary age print readers was not to be relied on as an indicator for selecting the most appropriate material or reading condition. The supposition that age, intelligence and acuity level would significantly affect performance could not be proven. Some supporting evidence was found, to suggest that acuity level and eye defect may be factors which do influence performance. John Bock Evidence in this study suggests that large print may be marginally better than standard print but that it is not significantly so. From the information gathered in this study it would seem that there may be subgroups within the larger population of visually impaired ele— mentary age print readers for whom large print is more important. This study also produced information which indicates that there are some children for whom simple magnifica- tion of standard print is beneficial. READING PERFORMANCE OF VISUALLY IMPAIRED PRINT READERS USING STANDARD PRINT, LARGE PRINT AND MAGNIFICATION BY John Bock A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1971 ACKNOWLEDGEMENTS The completion of each phase in an individual's professional development provides an opportunity for taking stock. At such time one is reminded of the debt which he has accumulated. A thesis represents the involvement and efforts of many people, and although the researcher accepts full responsibility for any weaknesses which may appear, the credit for strengths must be shared. The writer has been the beneficiary of considerable assistance from so many sources that it is not possible to mention by name everyone who has played a role in this study. This does not diminish his appreciation of their individual and collective efforts. Grateful acknowledgement is extended to the committee members Dr. Richard L. Featherstone, Dr. Charles E. Henley and Dr. Larry E. Sarbough for their assistance and encouragement. Special thanks must go to Dr. Charles V. Mange, not only for his interest and guidance as chairman, but especially for his friendship and inspiration. The interest and assistance of the Ednalite Research Corporation, which supplied the Ednalite Masterlens for use in this study, was greatly appreciated. ii The Bobbs—Merrill Company is thanked for permission to reprint the Gray Oral Reading Test. To Ruth, his wife, the writer wishes to express affectionate thanks for continuing encouragement and support. iii TABLE OF CONTENTS ACKNOWLEDGEMENTS . . . . . . LIST OF TABLES. . . . . . . LIST OF FIGURES . . . . . . Chapter I. INTRODUCTION. . . . . Problem. . . . . . The Purpose of the Study Major Questions . . EXploratory Questions II. REVIEW OF RELATED LITERATURE The Visually Impaired Reader. . Characteristics of Print Magnification. . . . Illumination . . . . The Reading Process. . Experimental Design. . Evaluation Criteria. . Research on Print Size. Research on Magnification Summary Statement . . Hypotheses. . . . . III. METHODOLOGY . . . . . Sample . . . . . . Independent Variables . Print Size . . . . Magnification . . Intelligence Score and the Materials . . . . . Equipment . . . . . Procedure . . . . . The Interview . . . The Scoring. . . . iv Page ii vi ix Chapter IV. FINDINGS OF THE STUDY. Sample . . . Analysis of Data. Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis 10 Hypothesis 11 Hypothesis 12 \DmflmkfikwNI-J Exploratory Analysis V. INTERPRETATIONS OF FINDINGS AND RECOMMENDATIONS Sample . . . Analyses of Hypotheses. Print Size . Magnification Reading Preference Summary of Hypotheses Testing . . . EXploratory Analyses Age . . . Intelligence Acuity Level Summary. . . Problem . . Purpose . . Methodology. Analysis of Data Conclusions. Recommendations REFERENCES . . . . . APPENDIX A . . . . . APPENDIX B . . . . . APPENDIX C . . . . . Page 42 42 50 53 55 57 58 6O 61 62 64 66 68 68 68 7O 75 75 76 76 82 86 87 87 87 88 88 89 89 9O 91 91 92 93 96 105 107 109 Table 10. 11. LIST OF TABLES Distribution of the sample of age and Sex 0 O O O O O I O O O O O O 0 Visual acuity . . . . . . . . . . . Major eye defects of children in the sample . Children eliminated during final sample selection . . . . . . . . . . . . Overview of hypotheses to be tested . . . . Mean raw scores on reading speed, accuracy and comprehension for total sample of sixty-four subjects and partial sample of forty—four subjects . . . . . . . . . . . . Analysis of variance of reading speed scores for total sample of sixty-four subjects and partial sample of forty-four subjects. . . Analysis of the reading speed (words per minute) difference scores between large print without magnification and standard print without magnification (means, standard deviations and Tukey comparisons) . Analysis of the reading speed (words per minute) difference scores between standard print without magnification and standard print with magnification (Means standard deviations and Tukey comparisons) . . . . Analysis of the reading speed (words per minute) difference scores between large print without magnification and standard print with magnification (Means, standard deviations and Tukey comparisons) . . . . Analysis of variance of reading accuracy scores for total sample of sixty—four sub- jects and partial sample of forty—four subjects . . . . . . . . . . . . vi Page 43 46 47 49 51 52 54 55 56 57 59 Table Page 12. Analysis of the reading accuracy difference scores between large print without magnifi- cation (Means, standard deviations and Tukey comparisons) . . . . . . . . . 6O 13. Analysis of the reading accuracy difference scores between standard print without magni- fication and standard print with magnifica- tion (Means, standard deviations and Tukey comparisons) . . . . . . . . . . . 61 14. Analysis of the reading accuracy difference scores between large print without magnifi- cation and standard print with magnification (Means, standard deviations and Tukey comparisons) . . . . . . . . . . . 62 15. Analysis of variance of reading comprehension scores for total sample of sixty-four subjects and partial sample of forty-four subjects . . . . . . . . . . . . 63 16. Analysis of the reading comprehension differ- ence scores between large print without magnification and standard print without magnification (Means, standard deviations and Tukey comparisons). . . . . . . . 64 17. Analysis of the reading comprehension differ- ence scores between standard print without magnification and standard print with magnification (Means, standard deviations and Tukey comparisons). . . . . . . . 65 18. Analysis of the reading comprehension differ- ence scores between large print without magnification and standard print with magni- fication (Means, standard deviations and Tukey comparisons) . . . . . . . . . 66 19. Distribution of the stated preference for four conditions . . . . . . . . . . 67 20. Analysis of the difference scores between preferred and alternate test conditions (means, standard deviations and t values) . 69 vii Table Page 21. Analysis of variance of reading proficiency scores for groups of upper and lower visually impaired elementary age children . 71 22. Analysis of variance of reading proficiency scores with visually impaired elementary children belonging to upper and lower intelligence groups. . . . . . . . . 73 23. Analysis of variance of reading proficiency scores with two groups of visually impaired elementary children having Visual acuity greater and less than 20/200. . . . . . 74 24. Numbers of children whose reading speed was greatest on large or standard print read without magnification when distributed according to eye defect and acuity level. . 80 25. Number of visually impaired elementary age print readers who were unable to use the magnifier distributed according to eye defect and acuity level . . . . . . . 84 26. Visually impaired elementary age print readers, whose reading speed was highest when using the magnifier as distributed by major eye defect and acuity level . . . . 85 27. Summary of hypotheses tested . . . . . . 87 viii LIST OF FIGURES Figure Page 1. Distribution of intelligence scores . . . . 45 2. Reading proficiency mean scores for low and high acuity groups . . . . . . 79 ix CHAPTER I INTRODUCTION Problem There is a continuing concern among educators for the development of learning environments which are conducive to the optimal development and maximum use of the poten- tialities which an individual may possess. In recent years there has been increased emphasis on the possibility of enabling the Visually impaired child to make greater and more efficient use of his residual vision. The realization that the majority of visually impaired children, including many in the legally blind category, are able to read print, has stimulated a lively interest in (a) direct vision training, (b) the use of low vision aids, and (c) the use of large type print materials. One of the primary objectives for individuals with low vision is to maximize their ability to receive and perceive greater quantities of the environment through the visual sense in the most effective manner.1 Three important considerations in any attempt to facilitate this process are, (a) the individual person, (b) the 1 characteristics of the stimulus, and (c) the nature of the stimulus. Since reading provides one of the chief avenues to information within the public school setting, it follows that printed materials occupy a position of major importance in the educational process. The quality of these materials and/or the manner in which they are utilized, may there- fore contribute significantly to the relative effectiveness of their use by the reader. Examination of current practices in programs for the visually impaired suggest two basic approaches toward modification of the stimulus in the reading process. The objective in both, use of large print and use'of magnifi- cation, is to effect an increase in the size of the retinal image. The use of enlarged print has been inter- mittently promoted since 1880 when it was first introduced "for children with weak sight whose eyes must be spared."2 However, the use of optical magnifications is a relatively recent occurrence.3 Though objective evidence is still somewhat lack- ing regarding the relative merits of standard print, large print and optical aids in facilitating the reading skills of visually impaired children, a new interest in this area of education appears to be developing. The changing philosophy in special education, towards integration of the handicapped child, has resulted in increased numbers of visually impaired children receiving their education alongside their sighted peers 5 This trend has, in some in the regular public schools. instances greatly extended the demand for an increase in the quantity and variety of large print materials. There is no doubt that many of the current practices in public schools are based on sound reasoning and an accurate assessment of children's needs but there remains some doubt regarding the efficacy of certain practices. Fonda6 indicates that the widespread use of large print books may be one of these. He contends that, "The demand for books in large print is great because of custom and 7 In his opinion it is reasonable to expect tradition." that most readers of large print would do just as well with standard print. Ophthalmologists8 at a major children's eye service center have recently suggested that further research is needed at the elementary level to determine the size of print required for acceptable reading on the part of the visually impaired child. The use of optical devices as aids in the utiliza- tion of standard print has been suggested as a possible 9'10 Minnerll states alternative to the use of large print. that he is convinced that most of the school children and others now using large print books could be using ordinary print if they were given appropriate aids. To the extent that magnification and large print perform a similar function, one might expect that they would be equally effective in facilitating the reading skills of visually impaired pupils. Proponents of optical aids indicate that the need for large print will diminish if more widespread use is made of optical aids.12 They express concern, however, that continual expansion of the large print field may actually discourage the increased acceptance of aids. Rootl3 expresses concern about this problem and emphasizes that the use of Optical aids by visually impaired pupils needs a great deal of mutual exploration by educators and ophthalmologists. She emphasizes that in this relatively new area of specialization there is much to be gained by sharing information regarding the efficacy of these aids in the classroom setting. Present emphasis on "sight utilization" as opposed to "sight saving" has raised numerous questions regarding the extent to which the use of standard or large print materials should be promoted in the elementary school.14 The lack of objective evidence regarding these questions underscores the need for research in this area. Batemanls emphasizes that the most striking con- clusion to be reached from a survey of the major research literature on the partially seeing, is that such research is almost non-existent. At the same time it has been 16 pointed out by others that traditional practice must be carefully examined and new knowledge and procedures deveIOped if the partially seeing child is to reach his potential development. The Purpose of The Study The purpose of this study was to seek objective information regarding the relative effectiveness of standard and large print, used with and without magnification, in facilitating the reading skills of visually impaired print readers at the elementary level. Additional information was sought regarding the possible influence of preference for standard or large print used with and without magnification on the reading performance of visually impaired print readers at the elementary level. And finally, exploratory evidence was sought regarding the possible effects of age, intelligence and visual acuity on the reading performance of Visually impaired elementary age print readers using standard and large print with and without magnification. Major Questions Specifically, this study sought to answer the following questions: 1. Is there a difference between the reading Speed scores obtained by visually impaired elementary age print readers on parallel forms of the Gray Oral Ireading test presented in standard and large print, with and without magnification. 2. Is there a difference between reading accuracy scores obtained by visually impaired elementary age print readers on parallel forms of the Gray Oral reading test presented in standard and large print, with and without magnification. 3. Is there a difference between reading compre- hension scores obtained by visually impaired elementary age print readers on parallel forms of the Gray Oral reading test presented in standard and large print, with and without magnification. Exploratory Questions These questions sought to explore the possible influence of additional factors. 1. Is there a difference between the reading scores obtained by visually impaired, early and late elementary age print readers on parallel forms of the Gray Oral Reading Test presented in standard and large print, with and without magnification? 2. Is there a difference between the reading scores obtained by visually impaired elementary age print readers of upper and lower intelligence ranges, on parallel forms of the Gray Oral Reading Test presented in standard and large print, with and without magnification. 3. Is there a difference between the reading scores obtained by visually impaired elementary age print readers with greater and lessor degrees of vision on parallel forms of the Gray Oral Reading Test presented in standard and large print, with and without magnifi- cation. This chapter has presented an introduction to the problems and outlined the purpose of this study. Chapter II includes a brief review of the related literature and a summary of the conclusions which may be drawn from it. The research procedures are given in Chapter III. Chapter IV deals with the research design and analysis of the results. The findings of the study and conclusions based on these findings are discussed in Chapter V. CHAPTER II REVIEW OF RELATED LITERATURE This chapter, devoted to a review of some of the related literature, will deal briefly with the visually impaired reader, the characteristics of print, optical aids, illumination and the reading process. Consideration will also be given to experimental design, measures of legibility and to related studies performed previously. The Visually Impaired Reader To describe visually impaired children is a difficult task. Their similiarities are few; they are, according to Nolan "quite heterogenous in regard to Visual disability, visual acuity and other characteristics."17 There is a wide variance in the degree and kind of functional loss among visually impaired children resulting from different pathological conditions.18 Clinically, subnormal vision is defined by literal interpretation of the acuity chart. Visual acuity in these terms usually indicates the degree of central or direct vision performed with the fovea, but in the absence of this, it represents peripheral vision.19 Central vision refers to our awareness of the objects at which we are looking directly, that is, those along or near the visual axis.20 Peripheral vision involves the ability to sense the parts of the visual field which surround the central part.21 Measures of visual acuity provide some knowledge regarding the size of the retinal image that can be appreciated by the eye.22 They are generally expressed in the form of a fraction, but this does not therefore represent the true percentage of normal vision.23 Visual acuity is based on the interaction of a complex variety of factors which include the presence or absence of optical defects, the size of the pupil, the state of light-dark adaptation, the part of the retina stimulated and the luminance of different parts of the test objects and the brightness contrasts.24 Simple mathe- matical representations of this dynamic function are there- fore subject to severe limitations.25 The accommodative power of a visually impaired person's eye is a matter of particular interest to a consideration of appropriate print size. Accommodation is the process by which the refractive power of the lens in the eye is adjusted so that both, distant and near objects may be distinctly imaged upon the retina.26 The stimulus to accommodation is thought to be a blurred image upon the retina which causes active contraction or 10 relaxation of the ciliary muscles. Accommodation arises from a variation in the thickness of the curvature of the lens.27 It is generally accepted that children tend to have relatively high powers of accommodation but that with an increase in age the "near point" of vision recedes at a fairly constant rate.28 The available power of accommodation becomes significant when the reader holds materials closer in order to increase the retinal image.29 Fonda30 indicates that the eyes fatigue more easily when a person is expend- ing maximum accommodation and interpreting a blurred retinal image. The relative amount of accommodation present would seem to be an important factor in limiting the reading distance. Traditional classification systems generally refer to children with significantly less than normal vision as either legally blind or partially sighted. Legal blindness is defined as vision of 20/200 or less in the better eye with best correction, or a visual field limited to twenty degrees or less. The partially sighted have a 31 measured visual acuity of 20/70 to 20/200. Hathaway32 points out that such criteria have proved to be inadequate in describing children for educational purposes. Bateman33 confirms this view and suggests that more adequate criteria need to be established. Rosenbloom34, while not offering it as a final solution, does indicate that more considera— tion should be given to the primary mode of reading as a 11 descriptive factor. Aschcroft,35 however, reminds us that the problems inherent in the use of such criteria have been adequately demonstrated by recent studies which indicate that mode of reading may be more closely related to the program in which the pupil is enrolled than to his actual visual ability. In spite of these limitations a classifi- cation, which emphasizes actual performance rather than physical disability, does offer an educationally useful alternative.36 Characteristics of Print Interest in the physical characteristics of printed material derives mainly from a concern over making this medium of communication more efficient and effective.37 Studies which examined the effectiveness and print character- istics prior to 1940, used readability and legibility interchangeably to mean "ease and speed of reading at a natural reading distance."38 More recently, with the advent of "readability formulas" this term has taken on an entirely new meaning. Readability now refers to content difficulty rather than print characteristics.39 Legibility of print, as defined by Tinker,40 is concerned with the ease, accuracy and efficiency of per- ceiving printed symbols while reading with understanding. In comparing legibility of varying print materials one wants to know to what extent the typography fosters ease, accuracy and Speed of reading.“"42 12 Research in this area has been conducted primarily with normally sighted adults.43 Only a limited number of studies have concerned themselves with the problems of the partially sighted.44 Much of the research with either of these groups has been focused on type face, type size and leading, although width of lines, inter-column spacing, weight of paper and other factors were also considered.45'46 Type face refers to the symbol which appears on the printed page. All type is divided into families, each family being composed of several "type faces". Every "type face" is a particular design given to each of the characters of the alphabet.47 Some are very ornate and light, others quite plain and bold. A number of type faces are designed with "serifs" - fine cross strokes at the top and bottom of each letter. "Sans serif type" faces, often used in advertising, are designed without the fine cross strokes. Type size is measured in "points". A point has been defined as l/72 of an inch with twelve points equal to one pica or 1/6 of an inch.48 In typography, point size indicates the vertical dimension of the body type. The actual face of the type will be somewhat smaller because of the space occupied by the shoulders on each side of the face, which produces a blank space between 49 If the lines are to be 50 the letters and the lines. separated further this may be done by "leading". "Leads" 13 are blank pieces of type metal which may be two, four, eight or twelve points in thickness inserted between the lines of type. In the letterpress process, type body size was fixed through standardization but type face size could vary within the same body size.51 The introduction of offset lithography, has contributed to some confusion in the area of type measure- ment. By this method any book which is optically enlarged may be labelled as a large print edition. This form of labelling has brought about a situation where there are "large print" books in existence whose actual print is smaller than the print in the regular edition of other books.52 To avoid confusion inherent in the lithographic process, Nolan53 has recommended a scale for measuring the height of capital letters as they appear in print. By this scale seventy—two, forty-eight, twenty-four, eighteen and twelve point print have capital letters whose respective heights are .750, .500, .250, .188 and .125 inches. This scale, already adopted by a number of publishers, also * served as a guide to type size in this study. Magnification Optical devices currently available to persons with low vision, range from simple hand magnifying lenses to * See Appendix A. 14 sophisticated tailor-made optical systems prescribed by the ophthalmic services.54 In attempting to delineate 55 points out the function of these optical aids, Gnade that the optical aid begins where conventional glasses end. The aid therefore is a plus value. Scholz also points out that the sole purpose of the convex lens in an optical aid is to produce a larger image on the retina by magnification.56 Many of these lenses have been borrowed from industry while others have been specifically developed for this purpose.57 Low vision optical aids can be categorized into 58'59 Apart from this, distance and close work aids. optical aids are generally described in terms of their focal length measured in diopters corresponding to their power of magnification. A diopter is the reciprocal of the focal length of a lens measured in meters. A four diopter lens measured in meters. A four diopter lens would have a focal length of one fourth meter. In general, it may be assumed (in the higher ranges) that every four diopters are equivalent to one power of magnification.60 Distance aids include telescopic lens glasses, clip on loupes, monocular and binocular field glasses. Close work or reading aids include microscopic lens glasses, telescopic lens glasses with a reading cap, head- born loupe, hand held magnifiers, focusable stand Inagnifiers, paperweight (Visolett) and projection magnifiers.61 15 The size of the image produced on the retina through the use of a magnifying device depends in part on the power of the lens and the manner in which it is utilized.62 The power of stand magnifiers, for example, can be increased by moving the eye away from the lens, but this procedure will also increase peripheral distortion.63 The use of stand magnifiers, according to Rosenbloom,64 frequently permits improved illumination of the printed page while relieving the person from the necesity of precise focusing required by the use of many other aids. The possible working distance and the extent of the visual field are both dependent on the power and size of the lens being used. With an increase in the strength of the lens of a given size there is also a continual decrease in the depth and perimeter of the field.65 With severe depth restrictions the object to be viewed or the paper to be read must be held quite steady because any tremor will move the print out of focus and make it unreadable.66 Magnification of the retinal image may also be achieved without the use of lenses. Non-optical magnifi- cation of 2X can be provided by reducing the distance IDetween the observer and the object being viewed, by half. TPhis principle is applicable to both distant and near 'Vision. Television viewed at two feet produces magnifica- ‘tion of 5X relative to a ten foot viewing distance. 16 Standard print (12 point) produces the same size retinal image at four inches as does large type (eighteen point) 67 Fonda68 held at 6.8 inches. suggests that this form of magnification is not only more versatile and less expensive than the use of lenses but also quite practical for class- room use. Illumination Illumination here is used to refer to the amount of light, its diffusion, distribution and direction. Appropriate illumination is a prime requisite for effective 69 seeing. Research has indicated certain relationships between light intensity, visual acuity, size of object 7o,71,72,73 Bright_ discriminated and brightness contrast. ness contrast refers to the difference in brightness between an object and its background such as print on paper, or between two objects in the visual field.74 Seagers states that, "the greater the contrast the more 75 Others76 have easily the visual task is performed." pointed out that the level of contrast between ink and paper will influence the amount of illumination required for efficient reading. Light intensity is also of significance when optical aids are being used since light intensity may affect the power of the lens required. Scholz77 indicates that within certain limits an increase of intensity may 17 permit the use of a lower power lens. In other instances it may produce an effect which eliminates the need for magnification. Seagers suggests that, "since visual acuity increases markedly with an increase in illumination, additional light is sometimes said to act as a magnifier."78 Many visually impaired persons require a higher level of illumination for reading, than persons with normal sight. It is also known, however, that certain eye defects necessitate reduced levels of illumination for 79 Luckiesh80 cites data which maximum visual efficiency. show that persons with subnormal vision have marked increases in acuity when illumination is increased from one, to ten, to one hundred foot candles. This improvement was present whether or not they were wearing their corrective glasses. The findings of Kuntz and Sleight81 suggest that persons with subnormal vision gain significantly more in visual acuity terms with an increase in target brightness than do persons with normal sight. On the basis of these and other findings Tinker82 suggests that illumina- tion, for sustained readings by visually impaired persons, should have a minimum brightness of forty footcandles. The Reading Process Reading is a highly complicated process. The fact that authorities, in this field, have never agreed on one definition of reading, attests to its complexity.83 Cohen84 suggests that reading may be described as the processing of a visual symbol into an oral-aural symbol 18 that elicits intellectual and emotional response drawn from experience. Heilman85 indicates that reading always involves the simultaneous application of a great number of mechanical and comprehension skills, all of which are influenced by the reader's attitudes, knowledge, and past experience. Reading, according to Faye, . . . is a complex brain function which begins with a visual image and ends with some sort of action. It is a high level associative function which, if intact, works in spite of imperfect vision, and, if impaired is not able to profit from a perfect visual system.86 The skills or abilities stressed in reading instruc- tion are (l) comprehension, (2) vocabulary, (3) speed and (4) accuracy. Each of these factors is obviously related to the other three.87 The development of one involves others to a greater or lesser extent. Normal reading growth is developmental. With adequate preparation most children continue to progress in basic reading skills through the primary grades.88 For the average eight year old, reading is generally a pleasur- able experience. He can deal with new words by phonetics or by context and he maintains the meaning of what he reads orally.89 During the intermediate grades most children continue to perfect the basic reading abilities acquired in earlier grades. This is accompanied by a greater emphasis upon the growth of Specialized abilities and skills 19 90 It is not needed for reading content subject matter. generally known to what extent materials influence this development. Research evidence does indicate that many of the techniques and skills employed in proficient silent reading 91 It follows, therefore, are also used in oral reading. that children do not learn two entirely different sets of reading skills but that proficiency in oral and silent reading are related. A pupil may, therefore, demonstrate his acquired skills through either oral or silent reading. Experimental Design The problem of developing effective experimental controls according to Hendersongz, is often a complicating and/or restrictive factor in research related to human behavior and learning. To develop or even to recognize adequate experimental controls is a difficult requirement in most social situations. The environment in a psychological or educational context is always complex, and total control well nigh impossible.93 In her review of the research performed previous to 1969, related to print size and visually handicapped readers, Shaw concludes that, "the 94 lack of decisive results" may be attributed to this very problem. The problem, however, need not prevent the development of meaningful enquiry if the importance of a good design is recognized.95 20 Evaluation Criteria Experimental evaluations of the relative effective- ness of various type faces and print sizes in promoting ease of reading have utilized a number of different criteria. Tinker96 has carefully reviewed some of the more commonly used criteria such as speed of perception, blink rate, visibility, fatigue, readers' opinions and rate of work. On the basis of his own extensive research and a review of the work of others he concludes that the most acceptable criteria for comparing the legibility of varying print is a measure of work performance. Criteria of work performance such as reading speed, accuracy and comprehension are currently recognized as being the most valid vriteria available.97 Research on Print Size Initial research on print size began in Cleveland, Ohio in 1913,98 when Irwin investigated eighteen, twenty- four, thirty and thirty—six point sizes in seven styles of type. According to his findings twenty-four point Century Schoolbook and Caslon Bold types proved most popular with teachers and pupils. Not enough detail is given in published descriptions to allow replication. It is known, however, that pupils were asked to read material aloud at a distance most comfortable to them. 21 The twenty—four point print which became the standard for most special textbooks in sight-saving classes as a result of Irwin's work, remained unchallenged until Fortner99 compared eighteen and twenty-four point print for ease of reading, in 1943. Fortner's research utilized eye blink rate and reader's opinion as criteria for ease J of reading. The findings of this study were statisticlly [ insignificant. Lu 100 Eakin, Pratt and McFarland in 1952 made an attempt to compare the readability of twelve, eighteen and twenty- four point type. The subjects in this study were divided into groups, but the matching of the groups was not well established and all groups did not receive the same stimulus. Reading distance was held constant at fourteen inches however illumination was not standardized. The investi- gators were unable to come to statistically significant conclusions although they did suggest that the twenty-four point size was preferrable since more children were able to read it at fourteen inches. A further clarification of the issue was sougt by Nolan101 in 1959. With a well designed study he compared eighteen and twenty-four point print in different type styles. He demonstrated that visually impaired children found common serif type (Antique Oldstyle) significantly more legible than sans serif type (Metrolite Medium) however, he was unable to find statistically significant 22 differences in legibility between eighteen and twenty- four point print. In a differential diagnosis of sixteen partially sighted children, ranging in age from seven to fifteen, Karnes and Wollershein102 found that the children in their study performed significantly better on sixteen point than on ten point print and concluded that it would be advisable to provide these children with reading materials in large print, "if for no other reason than that they may read with more comfort and find reading less fatiguing."103 In a study involving a sample of fifth and sixth 104 and his associates graders from fifteen states, Birch attempted to relate an examination of use of twelve, fifteen, eighteen, twenty-one and twenty-four point print to an evaluation of the academic achievement of visually impaired pupils. Experimental controls appear to have been weakened due to the necessity of conducting the testing through the mails. Controls for (a) differences in administration, (b) practice in taking tests, (c) levels of motivation of students were dependent on a spread of these effects. The report makes no mention of any attempt to control lighting or the scheduling of comparative tests. It may be for these and other reasons that the findings on print size were inconclusive. 23 The Library Association of London sponsored a research project in 1969 to determine appropriate print size and style for visually impaired readers which, although it focused mainly on adult readers, does have considerable relevance here. The study conducted by Shaw105 recognized the importance of experimental controls. In a testing format patterned closely after Tinker's extensive legibility research studies, an effort was made to control lighting, testing environment and test administration. Like Tinker, she used short paragraphs printed in varying type styles and print sizes. On the basis of her investigation, Shaw concluded that type size was one of the most important factors in print legibility for partially sighted adult readers. Her findings also indicated that print size was a significant legibility factor with children. The similarity in her findings with these two groups may, however, have been affected by the fact that, the majority (thirty-four out of forty-eight) of children in her sample were fourteen years of age or older. The most recent study, which addresses itself to the issue of print size, was performed by Sykes106 with a group of forty-one visually impaired students at the Michigan School for the Blind. The stated purpose of this investigation was to compare the effectiveness of standard and large print in enhancing the reading skills of visually impaired high school students. The mean age of the sample was seventeen and their mean I.Q. just over one hundred. 24 Two equivalent forms of the Davis Reading Test were administered in twelve and eighteen point print during separate sessions. One of the difficulties encount- ered by the investigator was a scheduling problem. This resulted in an average delay of eighteen days between tests. Illumination and reading distance were subject to individual preference. The format of the testing sessions allowed students to read the materials silently and then to respond to comprehension questions orally. The blink rate was used I as a basis for a visual fatigue score. Sykes determined, on the basis of his analysis, that the only advantage offered by large print was a reduction in the visual fatigue of partially sighted students. He concluded that standard print was as effective as large print in facilitating the reading speed and read- ing comprehension skills of both partially sighted and legally blind students. Research on Magpification The value of optical aids has been demonstrated in a survey conducted by Rusalem107 at the Industrial Home for the Blind, Brooklyn. The findings of the survey indicated that 340 persons in a group of 500 benefited from optical aids service. Of these, ". . . 238 had . . . 108 improved near v151on." 25 This review of the literature produced no evidence of any objective evaluation of various hand and stand magnifiers frequently found in schools. Indications are that such devices have been and still are being placed there largely on the basis of subjective judgements of sighted persons.109 Summary Statement The research on print size is limited. Few studies address themselves to the problems of the elementary school child. While objective evidence is limited, there are some indications that large print may be preferable at the elementary level, but that this is not necessarily so at the secondary level. Still lacking is objective evidence regarding the relative value of magnifiers which can be used to enlarge standard print through optical magnification to a size comparable to large print. Objective evidence is also lacking regarding the extent to which personal preference may be a useful guide in selecting appropriate print size for visually impaired elementary age print readers. There is also insufficient information available with respect to the relative importance of large print in facilitating the reading skills of visually impaired print readers at the elementary level. 26 Hypotheses Twelve hypotheses were developed as follows: Hl There is no difference between the reading speed scores attained by visually impaired elementary age print readers when reading large print without magnifi- cation and when reading standard print without magnification. H There is no difference between the reading 2 speed scores attained by visually impaired elementary age print readers when reading standard print without magnification and when reading standard print with magnifi- cation. H There is no difference between the reading 3 speed scores attained by visually impaired elementary age print readers when reading large print without magnification and when reading standard print with magnification. H4 There is no difference between the reading accuracy scores attained by visually impaired elementary age print readers when reading large print without magnifica- tion and when reading standard print with magnification. H There is no difference between the reading 5 accuracy scores attained by visually impaired elementary age print readers when reading standard print without magnification and when reading standard print with magnification. 27 H6 There is no difference between the reading accuracy scores attained by visually impaired elementary age print readers when reading large print without magnifica- tion and when reading standard print with magnification. H7 There is no difference between the reading comprehension scores attained by visually impaired elementary age print readers when reading large print without magnification and when reading standard print without magnification. H8 There is no difference between the reading comprehension scores attained by visually impaired elementary age print readers when reading standard print without magnification and when reading standard print with magnifi- cation. H There is no difference between the reading 9 comprehension scores attained by visually impaired elementary age print readers when reading large print without magnification and when reading standard print with magnification. H10 There is no difference between the reading speed scores attained by visually impaired elementary age print readers, reading under preferred test conditions and when reading under alternative test conditions. Hll There is no difference between the reading speed scores attained by visually impaired elementary age print readers, reading under preferred test conditions and when reading under alternative test conditions. 28 H12 There is no difference between the reading speed scores attained by visually impaired elementary age print readers reading under preferred test conditions and when reading under alternative test conditions. CHAPTER III METHODOLOGY The relative effect of large and standard print :3 and the use of magnification on the reading skills of visually impaired elementary age print readers was examined through the use of objective data based on individual reading performance under varying test conditions. The selection of factors to be tested and the choice of test procedures and instruments in this study required a compromise between (a) keeping the test within manageable time limits, (b) making the tests valid and (c) relating them to normal reading situations. Sixty-four visually impaired elementary age print readers were given four parallel forms of the Gray Oral Reading Test under four treatment conditions. The treat- ments consisted of reading standard and large print, with and without magnification. The reading speed, reading accuracy and reading comprehension was recorded for each subject under all four conditions. The discussion which follows will consider in greater detail, the selection of the sample, the test variables, the testing procedure and the materials and equipment used. 29 30 Sample The sixty-four children included in this study, were selected from the school districts of North York (Toronto) and London in the province of Ontario, the school districts of Warren, Livonia, Lakeshore, Flint and Grand Rapids in the State of Michigan and from classes for the partially sighted at the Michigan School for the Blind. To be included the children needed to be (a) visually impaired (20/70 or less with normal correction), (b) between eight and twelve years of age, (c) able to read at the grade one level and (d) able to read twelve point print without aids other than the use of ordinary glasses or holding the paper nearer to the eye. Teachers were asked to refer to the examiner all print readers who met conditions (a) and (b). The examiner made the final selection during the initial part of the test interview based on conditions (c) and (d). Independent Variables Print Size Print size was selected as one independent variable. Given the limited amount of objective evidence available regarding the possible effect of this variable on reading skills of visually impaired children at the elementary level, and the fact that most teachers of these children 31 need to make decisions with respect to appropriate print size from time to time, this seemed an important choice. An examination of a number of elementary texts indicated twelve point print as the size most commonly 110 used. Tillet, in reviewing current practice, confirms these findings. 111 of the National Accredita- Present recommendations tion Council for the Production of Large Print Books suggest eighteen point as the most appropriate size. A brief survey of large print books in use, showed that in practice eighteen point print was the most frequently occurring size. On the basis of these findings it was decided, albeit somewhat arbitrarily, that twelve point print and eighteen point print were to be designated as standard and large print respectively, for the purpose of this study. Magpification Magnification was chosen as a second factor for consideration in this study. It has been suggested that magnification may facilitate the reading process by allowing the child to use less of his accommodative power112 113 or compensate for the lack of it and thereby reduce fatigue. 32 The amount of magnification required to compensate for differences between standard and large print is relatively small. But it is not generally known to what extent the use of such magnification might provide a viable alternative to the use of large print or to what extent it might render standard print and large print more effective in facilitating an individual's reading skill. The ready availability of low power magnifiers and their currently increased introduction into classrooms114 influenced the choice of low magnification as the variable to be examined in this study. This, however, is not to say that high power magnification available in a variety of optical devices is not important and in as great a need of objective examination. Intelligence Score andithe Age Intelligence score and the age of the children were selected as additional factors of interest in order to carry out a further exploratory analysis of the experimental results. Stimson115 has suggested that intelligence plays an important part in seeing, because the mechanism used in seeing can be estimated to be half ocular and half cerebral. One might argue, therefore, that relative weakness in the cerebral function could possibly be offset by improvements in visual image, or ocular function. Conversely it might 33 be suggested that greater interpretive facility would compensate for more limited ocular functioning. It is not readily apparent whether differences in print size would produce differences in ocular stimulation great enough to cause differences in performance by brighter and less able children. The British Association for the Advancement of Science116 has made the recommendation that the print of books for younger children with normal sight should be appropriately enlarged. It would seem that a higher level of visibility and perceptibility brought about by increased letter size would be conducive to the development of reading skills. Barraga117 argues that reduction in the size of stimuli, once learning development has been established, is imperative if visual function is to reach its peak. For this reason one might expect that large print may be of greater importance at different age levels. Age, therefore, became a variable of interest about which additional information was sought. Materials The test materials utilized in the assessment of the relative effectiveness of varying print size and optical magnification have been identified as one of the most important variables. 34 In order to control for difficulty level of words and content, it was deemed important that reading materials be selected which had been standardized on these factors. The selection of forms A, B, C, and D of the Gray Oral Reading Test was based on the above considerations and the fact that the test provided for continuous text, closely related to normal classroom reading materials. These forms were reported to have coefficients of equiv- alence ranging .973 to .982119. Each of these forms contained a series of thirteen paragraphs of increasing difficulty. This permitted the subject to read paragraphs of the same difficulty level in each of the forms under different treatment conditions and at the same time gave opportunity for subjects of different abilities to read paragraphs appropriate to their ability level. The manual which accompanied this test provided for standardized scoring of speeding, accuracy and comprehension.120 The test, as published, was not available in the required print sizes. For purposes of this investigation the print size had to be modified. With written permission of the Test Division of Bobbs Merrill Company, the test ‘was reproduced by photo-offset lithography in twelve and _eighteen point print according to standards defined earlier in this study. According to Nolan,121 this procedure is employed in producing most of the large print texts currently available for classroom use. 35 Equipment A stand magnifier with a large lens* which would allow for binocular vision and would enlarge twelve point type by fifty percent in order to produce an optical image comparable in size to eighteen point print was con- sidered to be the most appropriate instrument for purposes of this study. Such an instrument allowed the subject complete freedom to concentrate on the reading material and required no physical effort on his part. The Ednalite Masterlens,122 selected for use in this study had a large viewing area and‘a high quality glass lens. This magnifier also had its own light source which could be focused directly on the reading material and thus permitted the standardized level of illumination to be maintained while materials were being viewed through the magnifying lens. The lens itself was mounted on a swivel which permitted the positioning of the lens and the reading materials at right angles to the reader's line of vision for greater viewing comfort. The positioning flexibility of the lens also permitted the materials to be placed at the appropriate distance from the lens to achieve fifty- percent magnification. * See Appendix B. 36 A perpendicular distance of seven and one half inches from the reading material produced this magnification and served as the standard in this study. The performance of each subject was recorded by means of a portable Model 7612 Realtone Cassette tape recorder. Separate cassettes were used to collect data for individual subjects. A standardized level of illumination (50 - 60 foot lamberts) was maintained by the use of portable light fix- tures with metal shades, equipped with two hundred watt incandescent work/study bulbs. The relative intensity of these lights was controlled by the use of a rheostat control. The level of illumination was measured with a General Electric type 213 light meter. Procedure The selection of the particular experimental controls used for the assignment of subjects in this study was based on the need to compare the influence of four different test— ing conditions on the reading process. The effective match- ing of sufficient numbers of visually impaired children with respect to the many variables affecting the reading process would be extremely difficult in View of the limited numbers of children available for study and the wide range of their individual differences. 37 To control for these individual differences the use of repeated measures seemed most appr0priate. Additional control was required in matching use of standard and large print, with and without magnification with each of the four forms of the Gray Oral Reading Test in order to achieve equal representation of the forms with each of the conditions. It was also considered important to control for possible I bias which might result from the order of presentation of the test forms and/or test conditions. Equal representation of test forms and test conditions as well as order of presentation were controlled by the use of a four by four graeco-latin square.123 Four rows of the letters A, B, C, D and four columns of numbers one through four were independently arranged at random. Letters were assigned to each of the four forms of the Gray Oral Reading Test and numbers to each of the four test conditions, standard and large print, with and without magnification. This procedure yielded the following four sequences of presentation. a) B3 C4 D2 A1 b) A4 D3 C1 B2 c) D1 A2 B4 C3 d) C2 B1 A3 D4 Within this set of sequences each of the four forms was paired with each of the four test conditions assuring equal representation of the forms and test conditions. 38 The sixty-four subjects were later assigned to the four sequences by means of a table of random numbers. These procedures gave some assurance that additional factors, not controlled or balanced, would not produce systematic bias.124 Control of environmental factors such as noise and interruption was achieved by conducting actual testing in a separate room with only the subject and the examiner present. Differential lighting, although a potentially important factor in the reading process, was not examined separately in this study. Care was taken to avoid glare and to maintain a standardized level of illumination. The level of illumination when reading with and without the magnifier was maintained at approximately 50 - 60 foot lamberts. This was slightly higher than the level suggested as adequate by Tinker.125 It did not, however, conflict with his further suggestion that higher intensities might be employed with safety, provided the diffusion of light was properly controlled.126 The entire work area was covered by desk blotter size sheets of light grey news print in order to avoid glare from polished desks and to standardize the levels of contrast in work area. The magnifier, tape recorder and all the materials to be used in an interview were arranged prior to the entry of each subject. The distance between the reading materials 39 and the magnifying lens was controlled by the use of a small reading stand placed at a distance of seven and one half inches from the lens. The subject's reading distance from the paper or the magnifying lens was not controlled. Individuals were free to choose their own reading distance. The Interview ‘ '.‘.-I 1 The interview was conducted in two parts. The first part of the interview served to familiarize the subject with his surroundings and the nature of the task to be performed. During this part of the interview an attempt was made to achieve a relaxed atmosphere. The subject was given colorful trinkets and printed material to examine with and without the magnifier. When the subject appeared to be ready the examiner began the second part of the interview and proceeded with the sequence below, which served as a guide for the remainder of the interview. (a) The subject was given a card containing a paragraph printed in standard print thought to be suitable to his reading level and asked to read the paragraph orally. He was then asked to respond to simple comprehension questions. Adjustments in level of difficulty were made as necessary. This step set the stage for procedures followed in the Gray Oral Reading Test and provided some assurance that the subject was able to read standard print before the actual testing began. 40 If the subject was unable to read, he was given alternate tasks to do in order that he would not feel rejected but was not included in the sample. Those children who were able to read then proceeded to the formal testing. (b) The subject was presented with the Gray Oral Reading Test forms in the sequence determined earlier by the research design. The subject was asked to read three consecutive paragraphs aloud. After reading each paragraph, the subject was asked to respond to standard comprehension questions supplied by the test manual for each paragraph. The subject was asked to read parallel paragraphs from each of four forms under each of the four test conditions. (c) The subject was asked to state which of the four test conditions he liked best. (d) The entire interview was recorded on tape in order to facilitate the timing and scoring of the test passages. The Scoring The scoring of the results was carried out according to the following criteria. (a) Reading speed was calculated in terms of the number of words read per minute. (b) Reading accuracy was based on the total number of errors. Subject's score was calculated by sub- tracting errors from 25. 41 (c) The comprehension score was determined by the number of correct responses. Correct answers received two points, partial answers one point and wrong answers no credit. Each treatment had a possible comprehension total of twenty-four points. Additional data such as age, intelligence and acuity level were collected by the examiner from existing records. CHAPTER IV FINDINGS OF THE STUDY Sample A total of eighty-two visually impaired children, eight to twelve years of age, with a visual acuity of 20/70 or less were referred to the examiner. Sixty-four of these children were able to meet the final criteria, ability to read twelve point print at the first grade level, during the pretest screening. The distribution of the children by age and sex may be noted in Table l. The reader will note that the distribution is somewhat skewed, both in terms of age and sex. Boys made up slightly more than sixty percent of the sample. Approximately nineteen percent of the children were under ten while forty-nine percent were more than ten years old. The average age for girls in the total group was 10.84 years while for the boys it was slightly less at 10.18 years. The average age for the total group was 10.45. Measures of intelligence based on an individual pychological examination were available for only forty— two of the children, included in the study. 42 43 m.mm s.oo mmmuamoumm ms.oa sw.oa ma.oa mommm>< we mm mm Hmpoe m.am ea m w NH N.mm ma m m Ha N.Hm om n ma OH m.oa n H m m m.n m H v m mmmucmoumm Hmuos mauflo mxom mom .xmm 6am mam an mamsmm may Lo coflusnfluumflouu.fi mamas 44 The available scores, as shown in Figure 1, ranged from a low of 63 to a high of 139. The mean intelligence quotient for the group of forty-two was 98.3. Since the criteria for inclusion in the study were the same for all sixty-four children the writer has no reason to believe that the twenty-two children for whom such scores were not available differed significantly in f intelligence from those for whom they were available. The a_ writer's own observations and other available information tended to support this contention. The available intelligence scores plotted in Figure 1 seem to indicate that the sample was fairly representative of a normal population as far as intelligence was concerned. The visual acuity scores presented in Table 2 indicate a spread in acuity from 20/70 to 20/400. The reader will note that half of the children have acuity levels which would allow them to be classified as legally blind. Another fact which is very apparent is that twenty of the remaining thirty-two children are at the lower end of the continuum for partially sighted. The reader will also want to be aware of the fact that forty of the children were wearing glasses at the time of testing. The eye defects of the children in the sample varied considerably. Findings in this regard are given in Table 3. A number of the children had several defects reported in their eye chart. Numerous combinations of the various 45 .mmuoom mocmmHHHmpCH mo cofluspfluumflonl.a musmflm omH ova oma omH OHH OHH om cm on ow om p p h r _ p L IN i4 4 4 4 4 4 4 4 4r. 1- . NH I VH magnum mo Hmnfisz 46 ‘III'IrII I .cmms soon on: poem now no m>flm mo mocwumflp mcflummu m muons pmunm>coo mumz mmuoom mpflsom one x. H.m m.s H.m m.mm s.am m.m m.m pgmonmm N m m mm om G o whomflnsm «mpflsom ooqxom oomxom osmxom oom\om ooa\om om\om os\om Hmsmfl> .muflsom anamfl>ll.m mamme 47 TABLE 3.-—Major eye defects of children in the sample. Number of Defect Students Lens Cataract 10 (2 with surgical aphakia) Iris Colomboma 1 HI Retina Macular degeneration Chorioretinitis Pigment degeneration Albinism I—II—awc. Optic Nerve Optic atrophy Ocular dystrophy AtrOpic fundus I—‘NU‘I col Eyeball in General Myopia Glaucoma NU! Neural and Muscular Nystagmus l Amblyopia Esotropia Exotropia thuam Other Undiagnosed Brain tumor F4» ml 48 defects might have been listed. The table shows only those which appeared to be major defects in each case. In some cases, however, these defects by themselves might not cause a disability severe enough to bring about significant visual impairment. It is often the combination of several factors which brings about visual impairment. The reader will also want to be aware of some of the characteristics of the eighteen children eliminated during the final sample selection. Table 4 provides some information with regard to age, sex and reason for exclu- sion. Thirteen of the children were lacking in sufficient reading skill. School records indicated that a number of these children were in the lower quartile of the intelli- gence range and this may have accounted for some of their difficulties. The sex ratio was quite similar to that of the group included in the sample. Two-thirds of this group were boys while just over sixty percent of the sample were also boys. The age distribution, however, of this group varies considerably from that of the sample. Eight of the eighteen children were eight years of age. The average age for this group was 8.8 years as compared to 10.4 years for the sample. It should also be noted that only three of the children were excluded because they were unable to see the materials and two because their inability to attend made it impossible for them to participate in the test. 49 whom H mHHHU H H whom H mHHHw v mmom H mHHHw H v whom N mHHHw HH OH ocmuua on anHHnmcH HmHuwumz HHme mchmmm mom on mHnmcD ucmHonwsmcH xmm :oHumcHEHHm MOM COmmmm .coHuomHmm mHmEmm Hmch mcHHsc pmumcHEHHm cmeHHQUII.v mqmde 50 Analysis of Data The analysis of the data was influenced consider- ably by one unexpected finding. It had not been antici- pated that twenty of the sixty—four children in the sample would not be able to use the stand magnifier at all. In View of this finding, the earlier consideration of an analysis by means of a graeco-latin square was discarded in favor of a two way analysis of variance with repeated measures since the latter is more versatile with respect to such problems. It was also decided that two separate analyses were to be made for the first nine hypotheses. The first analysis was based on the performance of the total group of sixty-four subjects with zeros entered for those who could not utilize the magnifier. The second analysis focused on the forty-four subjects who were able to read under all four conditions. A separate two-way analysis of variance with a subjects by conditions design, was carried out for each of the dependent variables in order to test the first nine hypotheses. Post hoc Tukey comparisons were performed when F values were significant. Dependent sample T-tests were used to test hypotheses ten to twelve. 51 TABLE 5.--Overview of hypotheses to be tested. Independent Dependent Variables Variables Speed Accuracy Comprehension Lge print vs Std print H1 H4 H7 Std print vs Std print mag H2 H5 H8 Lge print vs Std print mag H3 H6 H9 Preferred vs Alternate H10 H11 H12 The main analysis relating to the twelve hypotheses was performed with subjects' raw scores* on reading speed, accuracy and comprehension. For the eXploratory analysis these scores were converted into standard scores and combined by equal weighting into a reading proficiency score. The raw score means for reading speed, reading accuracy and reading comprehension are shown in Table 6 for both the total sample of sixty-four and the partial sample of forty-four subjects. Two of the most readily apparent features of Table 6 are, (a) the tendency of performance on large print to exceed performance on standard print and, (b) the negative effect of magnification on performance scores. These trends, although not significant, appeared in both the total and partial sample means. The only exception that was noted * See Appendix C. 52 .mmmcommmu pomnuoo ucmmmummu csocm mmuoomo .mm Eouw pmuomanSm muouum wo Hogans on“ “communes czonm mmuoomn .muscHE Hod mcuozm .mocmEHOMHmd Hmpumn muMOHch mmHoom HmzmHm so.HN mH.H~ mm.om em.om we ucon Icmnmumfiou Gs.sH GN.HN mH.HH ms.om as mcHemmm mm.sH os.mH mm.sH Hs.sH as nmomusoos ov.HH om.mH GN.OH os.nH so maHemmm om.mm mm.sa mm.ms as.ma «a powwow sm.mm om.em AH.Nm so.om as manemmm coHumoHMHcmmz soHumoHMHcmmz COHumoHMHcmmz coHuMOHMHcmmz guHs usonuHs spas usoguHs muowhnsm HHme mcHemmm ucwum mmumq ucHHm mmumq ucHum pumpcmum ucwum cnmpcmum . . .muomeSm HSOMIMHHOM mo mHmEMm HMHuHmm can muomflQSm unculmume mo onEmm Hmuou now concmsdeEoo can homusoom .pmmdm mchan so mmpoow 3mm smmzll.m MHmdB 53 was a very slight positive gain on reading comprehension when standard print was magnified for those children who were able to use magnification. Twelve separate tests (nine analysis of variance and three T-tests) were conducted to examine the major research and the eXploratory questions. A .01 level of I significance was used with all tests. This gave the [ total experiment a .12 level of significance. (JP-2'. . In order to facilitate a presentation of the results and to allow the reader to make comparisons more readily, the hypotheses will be dealt with in groups of three as they relate to each of the dependent measures. Hypotheses l, 2 and 3 relate to reading speed. The overall analysis of variance for reading speed with sixty—four subjects as presented in Table 7 shows a significant F value of 44.18. The overall analysis of variance for forty—four subjects given in Table 7, also showed a significant, although somewhat lower F value of 17.13. Hypothesis I Hypothesis one postulated that there would be no difference between the reading speed scores of visually impaired elementary age print readers when reading large print without magnification and when reading standard print without magnification. Table 8 presents the mean difference 54 Ho. v a vs.an mmH AHV coHpomnmucH «MH.nH H\U mm.wmmm m AUV mQOHqucou muomhbsm Hm.ommm ms Amy muommnsm HsoL-sunom mm.mno mmH AHV coHuompmch «mH.qw H\O om.vmmmm m ADV mcoHpHpcoo . muomnbsm m~.mamm me Ame muomflnsm “sowusanm m mumsqm com: mo mocmHHm> mo moonsom .mpomflbsm HSOMISHHOM mo deEmm HmHuumm can muoanSm “50mlmume mo deEMm Hmuou new monoom Ummdm mchmmH mo mocmHHm> mo mHmmHmc¢I|.n mHm mo mmousom .mpommbsm HSOMISDHOM mo deEMm HMHunmm can muownbsm nsowuhume mo deEMm Hmu0p How mmpoom momusoom mchmmH mo mUQMHHm> mo mHmmHmc¢I|.HH mHmde 60 TABLE 12.—-Analysis of the reading accuracy difference scores between large print without magnification and standard print without magnification (Means, standard deviations and Tukey comparisons). Sixty-four subjects Forty-four subjects Lge print 18.56 Lge print 18.40 Std print 17.40 Std print 17.61 D = 1.16NS D = .79NS Sd = 5.23 Sd = 3.06 q = 2.82 q = 2-03 NS ' Not significant Hypothesis 5 It was stated in hypothesis 5 that there would be no difference between the reading accuracy scores attained by visually impaired elementary age print readers when reading standard print without magnification and when read- ing standard print with magnification. The difference scores presented in Table 13 show that in comparisons of the means for both the sixty—four and the forty-four subjects, a significant difference was found. The difference for the first group was 7.14 and the difference for the second group was 2.68. The hypothesis that there is no difference in the reading accuracy scores of visually impaired elementary age print readers when reading standard print with magnifi- cation and when reading standard print without magnification is therefore not accepted. 61 TABLE 13.--Analysis of the reading accuracy difference scores between standard print without magnification and standard ' print with magnification (Means, standard deviations and Tukey comparisons). Sixty-four subjects Forty-four subjects Std print no mag 17.40 Std print no mag 17.61 Std print mag 10.26 Std print mag 14.93 * * D = 7.14 D = 2.68 Sd = 5.23 Sd = 3.06 q = 2.82 q = 2.03 'k p <.Ol Hypothesis 6 It was postulated in hypothesis 6 that there is no difference in the reading accuracy scores of visually impaired elementary age print readers when reading large print without magnification and when reading standard print with magnification. The presentation of the difference scores in Table 13 from a comparison of the means on reading accuracy for sixty-four and forty-four subjects shows that both difference scores are significant. The difference for the first comparison is 8.30 and the difference for the second comparison is 3.47. The hypothesis that there is no difference between the reading accuracy scores of visually impaired elementary age print readers when reading large print without mangifica- tion and when reading standard print with magnification, is therefore not accepted. 62 TABLE l4.--Analysis of the reading accuracy difference scores between large print without magnification and standard print with magnification (Means, standard deviations and Tukey comparisons). Sixty—four subjects Forty-four subjects Lge print 18.56 Lge print 18.40 Std print mag 10.26 Std print mag 14.93 'I: * D = 8.30 D = 3.47 Sd = 5.23 Sd = 3.06 q = 2.82 q = 2.03 ‘k p <.01 Hypotheses 7, 8 and 9 are related to reading compre- hension. The results of the overall analysis of variance on reading comprehension for the group of sixty-four are given in Table 15. The reader will note that the F value of 23.43 is significant, indicating a difference between the test conditions. However the results of the overall analysis of reading comprehension for the group of forty-four subjects given in Table 15, shows an F value of .788 which is not significant. Hypothesis 7 It was postulated in hypothesis 7 that there would be no difference between the reading comprehension scores attained by visually impaired elementary age print readers when reading large print without magnification and when reading standard print without magnification. 63 ucmoHuHcmHm uoz mz Ho. v Q a» mm.s mNH AHV coHuomumpaH mmen. H\U mo.m m HOV mcoHqucoo mpomflosm mm.mm ms Amy mpomflnsm usomusunom mo.mm mmH AHV COHuomHmucH *mv.mm H\U oH.va m ADV mQOHpHpcou mpomhnsm mm.MHH mo Amy muommnsm Hsomnhume m mnmsqm com: mo mocmHnm> mo mmoHsom .muomflQSm HDOMI>HHOM mo deEmm HmHuumm can muomnbsm HSOMISHme mo mHmEmm Hmqu “Om mmuoom COHmcmandEOo mchmmH mo mocmHHm> mo mHmmHmcmll.mH mHmmB 64 The difference score of .844 on reading comprehen— sion shown in Table 16 was found to be insignificant. Therefore the hypothesis that there is no difference between the reading comprehension scores of visually impaired elementary age print readers when reading large print with— out magnification and when reading standard print without magnification was accepted. TABLE l6.—-Analysis of the reading comprehension difference scores between large print without magnification and stan- dard print without magnification (Means, standard deviations and Tukey comparisons). Sixty—four subjects Lge print 21.26 Std print 20.42 D = .84NS Sd = 6.24 q = 3.36 NSNot significant Hypothesis 8 It was further postulated that there would be no difference between the reading comprehension scores attained by visually impaired elementary age print readers when reading standard print without magnification and when reading standard print with magnification. 65 Table 17 shows a difference score of 6.24 based on the comparison of means from the performance of sixty— four children. This difference was significant. Therefore the hypothesis that there is no difference between the reading comprehension scores of visually impaired elementary age print readers when reading standard print without magnification and when reading standard print with magnifi- cation was not accepted. It should be remembered, however, that the overall analysis of variance for reading comprehension, based on the scores of forty—four subjects given in Table 15, showed an F value which was insignificant. TABLE l7.--Analysis of the reading comprehension difference scores between standard print without magnification and standard print with magnification (Means, standard deviations and Tukey comparisons). Sixty-four subjects Std print no mag 20.42 Std print mag 14.18 3': D = 6.24 Sd = 6.24 q = 3.36 * p <.01 66 Hypothesis 9 It was stated in hypothesis 9 that there would be no difference between the reading comprehension scores of visually impaired elementary age print readers when reading large print without magnification and when reading standard print with magnification. The difference score given in Table 18 was based on the means for sixty—four subjects. The difference of 7.08 was found to be significant. The hypothesis that there is no difference between the reading comprehension scores of visually impaired elementary age print readers when reading large print with- out magnification and when reading standard print with magnification therefore was not accepted. It should, however, not be overlooked that the overall analysis on reading comprehension of the scores for forty-four subjects shown in Table 15 produced an F value which was insignificant. TABLE 18.--Analysis of the reading comprehension difference scores between large print without magnification and standard print with magnification (Means, standard deviations and Tukey comparisons). Sixty-four subjects Lge print 21.26 Std print mag 14.18 * D = 7.08 Sd = 6.24 q = 3.36 * p <.Ol 67 Hypotheses 10, 11 and 12 are concerned with the reading performance of visually impaired print readers when reading under conditions for which they have stated a preference as opposed to reading under those conditions they had not selected. Before proceeding to the statistical findings it seemed appropriate to interject additional observations of the examiner here. The examiner noted that very few children contem- plated their choice for any length of time. They appeared to be certain of their choice and not swayed by momentary novelty of one or the other choice. The reader will note from Table 19 that the fifty children who preferred conditions without the magnifier were almost equally divided with twenty-seven preferring standard print and twenty-three large print. Those prefer- ring conditions involving the magnifier showed a similar split with eight choosing large print and six standard. TABLE l9.--Distribution of the stated preference for four conditions. Condition Number Standard print without magnification 27 Standard print with magnification Large print without magnification 23 Large print with magnification 68 Hypothesis lO It was postulated in hypothesis 10 that there would be no difference in the reading speed of visually impaired elementary age print readers reading under preferred test conditions and when reading under alternative test conditions. The analysis of the difference scores presented in Table 20 shows that the difference of 2.3 was not significant. Hypothesis 10 is therefore accepted. Hypothesis ll Hypothesis 11 stated that there would be no difference in the reading accuracy scores of visually impaired elementary age print readers under preferred test conditions and when reading under alternate test conditions. It was noted in Table 20 that the difference score of .44 was not significant. Hypothesis 11 is therefore accepted. Hypothesis 12 It was stated in hypothesis 12 that there would be no difference in the reading comprehension scores of visually impaired elementary age print readers reading under preferred test conditions and when reading under alternate test conditions. The findings presented in Table 20 indicate that the difference of .50 is not significant. Therefore hypothesis 12 is accepted. 69 pcmonHcmHm poz mz m2mv.H mm.m om. om.om oo.Hm concmnmumsoo mchnmm mzHa.H ma.m we. NH.EH sm.kH somusooa scaemmm mst.H mm.NH om.m om.sm om.mm emmmm mcHemmm wumcuwqu cmuummmum D em m mammz .AmmsHm> u can mcoHpmH>mp pumpcmum .msmmEV mcoHqucoo pump mumcumpHm can pmnumwmpd comzpmp mmuoom mocmHDMMHp may mo mHmmHDC£II.om MHmHB 7O Exploratory Analyses The exploratory analyses were undertaken in order to examine the possible influence of age, intelligence and acuity level on reading performance with large and standard print. The focus of these analyses was on the comparison of print size alone. Magnification was not considered. This made it possible to include all those subjects who were partially excluded earlier because they could not use the magnifier. The analyses here made use of a combined reading proficiency score based on the combination of reading speed, reading accuracy and reading comprehension. It was anticipated that there might be a difference in the reading proficiency scores of visually impaired upper and lower elementary age print readers. To test this notion the sample was divided into an upper and lower age group of thirty-one children each. The first group was comprised of eleven and twelve year olds, the second of eight, nine and ten year olds. Two older subjects were randomly discarded to balance the size of the groups. The results of the analysis of variance given in Table 21 show that the groups by conditions interaction was not statistically significant indicating that the groups performed in‘a similar manner. 71 cmoH Human o u .m. . u Zmz GAGH. mm. Amsmunmv mosouo cHrpHs muomflnsm x ucHum ommo. H A0 x mv mmsouw x ucHum enom. H Amy ucHum x mszo. HEW mow m m mmmH. Ho onvmv masons casuaz mDomHDsm cams. H ADV manage m mumsow saw: me DUGMHum> mo mmomsom .cmanHco mmm mnmucmEmHm pmHHMQEH NHHmsmH> HmBOH cam Homo: mo mmsoum MOM mmuoom wocmHOHwoum mGHpmmH mo mocmHum> mo mHmmHmc mo mmousom .mmsoum mocmeHHmucH HDBOH can Heads 0» mchconn cchHHno humucmEmHm pmHHmmEH wHHMSmH> SDHB mmuoom hocmHonon mchmmH mo mocmHHm> mo mHthmc<|I.mm MHmde 74 CDOH Hcmam o p .m. . u zmz meH. HG xxovm x no masono cHrDHz mnomflnsm x ucHDm ommo. H no x as masonu x DcHum NHGH. H Lav DcHum x - m2mmo. "Axowmw m m samH. mm HHovmv masons cHrDHz mDomHnsm Hooo. H Hoe masouo m mumsqm com: mo mosmHHm> mo mmousom .oom\om cmcu mme pcm Hmpmmpm mpflsom Hm5mH> mcH>m£ cmHUHHco mumucmEmHm pmHHmmEH mHHmDmH> wo mmsoum 03p suHB mmuoom xocmHonoHQ mchmmH mo mocmHHm> mo mmeHmc¢II.mm MHmHmz OHumO m AHV m Amy mcHumm OH m Amv m mcmH Hmnoe oomxom v HDHDDH oom\om A suHsua summon .Hm>mH muHDom can powwow mxm ou mchHooom couscHHpme cmc3 coHumoHMHcmmE usocpHB pomp ucHHm pumpcmpm Ho mmHmH co pneumonm mp3 commm mchmmH mmoca cchHHno mo mHmQEDZII.¢m MHmHB 81 The comparison also indicates that acuity level bears some relationship to performance on a particular print size. It is readily apparent that the majority of the children (26 out of 32) with acuities of 20/200 or less had their highest reading speed when using large print. It should be noted, however, that the same is not true for children with acuities greater than 20/200 where a majority (19 out of 32) had their highest reading speed score on standard print. Table 24 does not, however, appear to reveal any well defined patterns which might indicate that a particular eye defect would favor one print size over another. One possible exception may be that eight out of ten children with a lens defect favored large print. The supposition that acuity level does appear to be related to performance on different print sizes, may offer another and perhaps the most plausible explanation for the insignificant differences in the total group. If in fact the acuity level is important, then the performance of children of different acuity levels would tend to have a leveling effect on overall performance on large print. As noted earlier, if children with higher acuity levels are restricted in performance on large print while those with lower acuity perform better, the net effect will be a compromise somewhere near their combined per- formance on standard print. 82 From the available evidence it was concluded that large print and standard print were equally effective in facilitating the reading skills of visually impaired elementary age print readers when no further differentiation for type and severity of defect is made. This conclusion, however, must be interpreted with some caution because of the additional findings. Magnification The hypotheses 4, 5 and 6 all state that there will be no difference in reading performance of visually impaired elementary age print readers when reading standard print with and without the magnifier. These hypotheses were all rejected since significant differences were found to exist in reading speed, reading accuracy and reading comprehension between the two conditions. The large significant differences were brought about by the fact that twenty children were unable to use the magnifier and obtained zero scores. But even when allowance was made for this fact, it was found that mean scores for reading speed and reading accuracy were still significantly different under the two conditions. In order to explore possible relationships between eye defects and ability to use the magnifier a distribution of the twenty children who could not use the magnifier was made in relation to eye defect and acuity level. 83 The figures in Table 25 show no particular bias with respect to any one eye defect. It will be noted, however, that fourteen of the twenty children had acuities of 20/200 or less. Observations made by the examiner during the interview seemed to indicate that a number of these children might have been able to use greater magnification. In some further post-test discussion they indicated an ability to use higher power hand magnifiers. Figures in Table 25 suggest that severity of visual loss played an important role in deciding whether or not an individual could use the magnifier. It was interesting to note, however, that six children had their highest reading speed scores when using the magnifier. This number might have been greater if more practice time had been available to the children before the actual testing. It was also found that of the six children whose best reading performance was with the magnifier, five had acuity levels better than 20/200. No relationship between eye defect and use of the magnifier, as indicated in Table 26 however, was readily apparent. The evidence in this study clearly leads to the conclusion that the magnifier used with standard print is not as effective as is standard print alone, in facilitating the reading skills of visually impaired elementary children. 84 H H Hmnuo m H N Hmnmcmw CH HHwnmmm m m H m>nmz OHHQO m w H HmHsomDS can Honsmz v v msHumm H H mHHH m H m mcmH Hmuos oom\om v HuHsoa oom\om A HDHDDH uommmo .Hm>mH qusom cam powwow who Op mchHooom cmuanHpme HDHMHsmmE can own on mHnmss mums 0:3 mumpmmn ucHHm mmm humpcmEmHm pmHHdeH mHHmsmH> mo HDQESZII.mN mqmme 85 m H v Ml II. .II Ummocmchco H H coHumHmcmmmo HcHsooz H H . mHmomz H H DHQOHMODM H H mHuHcHumHoHnoco H H msEmmummz Hmuoe oom\om v HDHsoa oom\om A HDHsoa Dommma mam .Hm>mH huHsom 0cm uommmp mam Momma hp pmuanHume mm HwHMHcmmE mcp mchs cmnz ummann mp3 pmmmm mchmmn mmocs .mumcmmu ucHum mom mumucmEmHm cmHHdeH hHHmsmH>II.mN mHm