AN EXPERIMENTAL STUDY OF PROFICIENCY IN SIGHT SINGING A SERIES OF ATONAL INTERVALS WITH ACCOMPANYING SONORITIES Thesis for the Degree of Phe D. MICHIGAN STATE UNIVERSITY GEORGE ADDISON WEST 1971 L 18 R A R Y Michigan State University This is to certify that the thesis entitled AN EXPERIMENTAL STUDY OF PROFICIENCY IN SIGHT SINGING A SERIES OF ATONAL INTERVALS WITH ACCOMPANYING SONORITIES presented by George Addison West has been accepted towards fulfillment of the requirements for Ph. D. Jemein Music Major professor Date September 24, 1971 1 0-7639 3" smiths av 5" HUAG 5 SONY 800K BINDERY INC. LIBRARY BINDERS SPRIIGPOIT. ”CIIGII MR 06 20029- ABSTRACT AN EXPERIMENTAL STUDY OF PROFICIENCY IN SIGHT SINGING A SERIES OF ATONAL INTERVALS WITH ACCOMPANYING SONORITIES BY George Addison West The investigation of music students' proficiency in sight singing series of atonally arranged intervals (atonal Interval Series) with accompanying atonal sonoroities con- stituted the main purpose of this study. Increased ability to sight sing unaccompanied tonally arranged intervals through atonal sight singing drill was also investigated as part of this experiment. Sixty first-year music theory students at the Uni- versity of Calgary participated in this experiment during the winter term of 1971. The students were enrolled in two class sections which were randomly designated experimental and control groups. Twenty-four original tape recorded Atonal Sight Singing Drill units were administered to the groups in a language laboratory. The experimental group performed drills comprised of Interval Series with accompanying atonal sonorities sustained by an electric organ. The control George Addison west group performed the same Interval Series drills unaccompa- nied. At the completion of the 24 drills, the remaining 22 experimental and 23 control group subjects experiencing all 24 drills were given Atonal Sight Singing Tests A and B and the Tonal Sight Singing Test (which had also been given as a pre-test). Atonal Sight Singing Test A corresponded in nature of content and presentation to the experimental group's drills and displayed a reliability of .83. Test B corre- sponded to the control group's drills showing a reliability of .86. The Tonal Sight Singing Test displayed a reliabil- ity of .80 and was similar in content to Test B but with tonal Interval Series. All three tests were originals with internal validity claimed due to the nature of content. This study was designed to test two primary and two secondary hypotheses related to atonal sight singing perfor- mance factors. To succinctly state the primary hypothesis, the experimental group's training experience (accompanied drill) is referred to as Method I; and the control group's training experience (unaccompanied drill) is Method II. All of the hypotheses, tested using analyses of variance and E tests, are stated with their results in the following null hypothesis form: Hypothesis 1 There is no difference in the ability to sight sing atonal Interval Series between students trained by Method I and Method II. Accepted. George Addison West Hypothesis 2 Progress in ability to sight sing atonal Interval Series based on sight singing drill of atonal Interval Series will not significantly affect progress in ability to sight sing tonal Interval Series. Rejected. Hypothesis 3 There is no significant difference between males and females in the ability to sight sing atonal Interval Series. Accepted. Hypothesis 4 There is no significant difference between persons with keyboard experience and persons without key- board experience in the ability to sight sing atonal Interval Series. Accepted. Although resulting data of this experiment indicated acceptance of Hypothesis 1, the F value yielded (3.02 with df 1/43) by the analysis of variance of Atonal Sight Singing Test B implies that accompanied atonal sight singing drill, though not more effective, is as effective a method of drill as is unaccompanied drill. The rejection of Hypothesis 2 by both the experimental and control groups at the .01 level of confidence decidedly indicates that drill in atonal sight singing will increase the subject's tonal sight singing ability. The equal ability of males and females to sight sing atonal Interval Series was not surprising, but interaction between subjects with and without keyboard experience within and between groups prevented a more accurate reading of that factor in Hypothesis 4. AN EXPERIMENTAL STUDY OF PROFICIENCY IN SIGHT SINGING A SERIES OF ATONAL INTERVALS WITH ACCOMPANYING SONORITIES BY George Addison West A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Music 1971 ACKN OWLE DGMEN TS In grateful acknowledgment of their contributions toward the realization of this dissertation, recognition is extended to: Professors Sherburn and Sidnell whose guidance was of invaluable aid; and Professors Harder, Campbell, Jones, Nathan and Reed whose approbation and encouragement validated this study. Susan, David and Robin who served as my motivation and experienced the utmost sacrifice. ***** ii TABLE OF Chapter I 0 THE P ROBLEM O O O O 0 Need for Study . . Purpose . . . . . . Hypotheses . . . . Definition of Terms Limitations . . . . Overview . . . . . II. REVIEW OF THE LITERATURE . . . . . . . . . . . CONTENTS Page 0 O O O O O O O O O O C O O CDQONU'IWI‘ \O The Development of Sight Singing Tests . . . 10 The Study of Factors Correlated with Sight Singing Ability . . . . . . . . . . . 13 Effectiveness of Various Types of Drill and Drill Procedures as Teaching Devices for Improved Sight Singing Ability . . . . 29 Atonal Music Investigations . . . . . . . . . 46 Pilot Study . . . . Summary . . . . . . III. DESIGN OF THE STUDY . Sample . . . . . . Drill Instrumentation . . . . . . . Interval Series Atonal Sight Singing Drills . Tonal Sight Singing Drill Sheets . . ......-'....... 53 ............. 58 . . . . . . . . . . . . . 61 . . . . . . . . . . . . . 61 . . . . . 62 . . . . . . . . . . . 63 . . . . . 63 . . 66 Tape Recording of Drill Sessions . . . . 66 Measures . . . . . Atonal Sight Singing Tests . . . . . Tonal Sight Singing Test . . . . . . Design . . . . . . Testable Hypotheses Analysis . . . . . . . . . . . . . . . . . . 68 . . 68 . . 70 . . . . . . . . . . . . . 71 . . . . . . . . . . . . . 78 . . . . . . . . . . . . . 79 IV. PRESENTATION OF THE DATA . . . . . . . . . . . 80 Review of Procedure Hypotheses . . . . . . . . . . . . . . . . . 81 . . . . . . . . . . . . . 81 iii Chapter Page Hypothesis 1 . . . . . . . . . . . . . . 81 Hypothesis 2 . . . . . . . . . . . . . . 82 Hypothesis 3 . . . . . . . . . . . . . . 83 Hypothesis 4 . . . . . . . . . . . . . . 84 Related Data 0 I O O O O O O O O O O O O O O 87 V. SUMMARY AND CONCLUSIONS . . . . . . . . . . . . 90 Summary . . . . . . . . . . . . . . . . . . . 90 Conclusions . . . . . . . . . . . . . . . . . 91 Discussion . . . . . . . . . . . . . . . . . 93 Suggestions for Future Research . . . . . . . 97 BIBLIOGMPHY O O O O O C O O O O O O O O O O O O O U C 9 8 APPENDIX A. TEST AND DRILL MATERIALS AND RAW DATA . . . . . 102 B. ATONAL SIGHT SINGING DRILLS FOR EXPERIMENTAL GROUP SWJECTS I O O I O O I O O O I O O O O 139 C. ATONAL SIGHT SINGING DRILLS FOR CONTROL GROUP SUBJECTS C O O O I O C C O O O O O O O 211 iv Table 1. LIST OF TABLES Analysis of Variance Summary Between the Experimental Group and Control Group on Atonal Sight Singing Test A . . . . . . . . Analysis of Variance Summary Between the Experimental Group and Control Group on Atonal Sight Singing Test B . . . . . . . . Analysis of Variance Summary Between the Tonal Sight Singing Tests (Pre-Test and Post-Test) of the Experimental Group . . . . Analysis of Variance Summary Between the Atonal Sight Singing Tests (Pre-Test and Post-Test) of the Control Group . . . . . . Summary of‘t Test Values Between Performance Scores of Males and Females in Both Study Groups on Atonal Sight Singing Tests A and B . . . . . . . . . . . . . . . . . . . Summary of Analysis of Variance of Fixed Factors Between Resulting Scores of Subjects With and Without Keyboard Experience on Atonal Sight Singing Test B . . . . . . . . Summary 0f.E Test Data Between Resulting Scores of Subjects With Keyboard Experience on Atonal Sight Singing Tests A and B . . . Summary of t Test Data Between Resulting Scores of Control Group Subjects With and Without Keyboard Experience on Atonal Sight Singing Test A . . . . . . . . . . . . . . . Summary of Correctly Sung Interval Pitch Averages of Male and Female Subjects of Both Groups on Atonal Sight Singing Test A . . . Page 81 82 82 83 84 85 86 87 89 Table 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. Reliability of Tonal Sight Singing Test as Determined by the Test-Retest Method . Raw Scores: Tonal Sight Singing Test . . Atonal Sight Singing Test A Raw Scores . . Raw Scores: Atonal Sight Singing Test B . Item Analysis, Atonal Sight Singing Test A Experinlental Group I O O O I O I O O O O 0 Item Analysis, Atonal Sight Singing Test A contrOl Group 0 O O O O O O O O O O O O 0 Item Analysis, Atonal Sight Singing Test B Experimental Group . . . . . . . . . . . . Item Analysis, Atonal Sight Singing Test B Control Group . . . . . . . . . . . . . . Difference Between Mean Scores on Male and Female Subjects on Atonal Sight Singing Tests A and B O O O O O I O O O I O O O 0 Difference Between Mean Scores of Subjects With and Without Keyboard Experience on Atonal Sight Singing Tests A and B . . . Hierarchy of Intervals as Derived from the Atonal Sight Singing Tests . . . . . . Octave Displacement of Interval Series Tones on Atonal Sight Singing Test A . . . vi Page 120 121 122 123 124 126 128 130 132 134 136 137 LIST OF FIGURES Page Treatment design for experimental group . . . 74 Treatment design for control group . . . . . . 75 Hierarchy of intervals resulting from both groups' performance of Atonal Sight Singing Tests A and B . . . . . . . . . . . . 88 vii CHAPTER I THE PROBLEM Need for Study Music composed during the final quarter of the nineteenth century displayed a gradual trend away from the established principles of tonality--initially through the increased compositional use of chromaticism and later through incorporations of modal, pentatonic and whole-tone scales. Twentieth century music carried this anti-tonal movement even further through utilization of such structural devices as dual modality, polytonality, pandiatonicism and quartal harmony. Leading into the second quarter of this century the music world witnessed the destruction of tonal- ity with the advent of the tone-row or twelve-tone technique of composition attributed to the genius of Arnold Schoenberg. Musical composition since then has proceeded in divergent paths, many incorporating in various ways some of the aforementioned techniques. Additional devices (some old--the majority new), considered by some to be bold, bizarre, and revolutionary, have opened new areas of musical exploration. Now in the eighth decade of this century, at a time when such devices as electronic synthesizers, computers, and multitrack tape recorders are being incorporated into the art of musical composition, the pedagogical aspects of music theory lag woefully behind. Many of the compositional tech- niques referred to above are already considered by contem- porary composers to be trite and passé. Yet, the basic concern of experimental research in theory pedagogy has been improvement of efficiency in learning the skills of tonal harmonic and melodic techniques. One of the very few studies concerning atonal (or non-tonal) materials was recently finalized by Sherman and Hill.1 A succinct statement based on the results of their study advises: Theory instructors should come to realize that the musical development of a university student need no longer be retarded because of a rather false and slavish reverence for the past or fear of the present. . . . Music is an ever changing, and today, a rapidly changing art that would most effectively be accompanied by correspondingly changing course of study.2 With increased emphasis on the use of tape record- ings, teaching machines, and programmed texts during the last two decades, much experimentation has been concerned with the aim of utilizing such teaching aids in certain 1Robert W. Sherman and Robert E. Hill, Jr., Aural and Visual Pergeption of Melodyin Tonal and Atonal Mus1ca1 EnvifOnments, Fifial Report of Project No. 2413, Office off Education, U.S. Department of Health, Education, and Welfare, Contract No. OE-4-lO-l77 (Muncie, Indiana: Ball State University, February, 1967). 2Ibid., p. 34. facets of music theory. Several programmed texts have been published in the area of music fundamentals as well as a text by Harder covering the part-writing principles of tonal harmony up through chromaticism.l Other programmed texts with accompanying tape recordings of ear-training exercises have also recently been marketed, including a set by Horacek and Lefkoff which also includes some recorded sight singing exercises.2 All of these aids to effective theory instruc- tion and increased student achievement have evolved through the efforts of experimenters dissatisfied with prevailing practices and limitations. However, the major emphasis of existing programmed materials remains on tonal music skills. Purpose Sight.singing has only recently been the subject of refined experimental study. The availability and versatil- ity of tape recording techniques has made this possible. There have been less than a score of documented experiments done in the area of sight singing and only one that deals directly with sight singing in an atonally structured music texture. That one study is the investigation of Sherman and Hill.3 1Paul O. Harder, Harmonic Materials in Tonal Music (Boston: Allyn and Bacon, Inc., 1968), two volumes. 2Leo Horacek and Gerald Lefkoff, Programmed Ear Training (New York: Harcourt, Brace & World, 1970), four volumes. 3Sherman and Hill, op. cit. The lack of research in the area of atonal sight singing, coupled with the interest among musicians related to atonal melodic perception, constitutes the basis for the present study. Of particular interest is whether or not an interaction exists in the performance of an accompanied atonal melody between the performer and the accompanying sonorities1 as exists in accompanied tonal performance. Edlund seems to indicate that one does not perceive such a relationship when posing the question: Does the music so far composed in the 20th century contain any such logical structured principles which could serve as a basis for a method of training the ear? Hitherto the methods have been based on the major/minor tonality with the triad as tonal basis. Is there any other equally clear tonal principle in 20th-century music, and any just as clearly discernible basic tonal design? The answer to this, of course, is "no."2 In his reference to "logical structural principles" though, Edlund does allude to other possibilities when describing the presentation order of material in his text. The study material presented . . . has been built up on a number of tonal and melodic figures which in the author's opinion have played some part in avoiding the major/minor- tonal limitations in 20th-century music.3 1See "sonority" in "Definition of Terms," Chapter I. 2Lars Edlund, Modus Novus (Stockholm: Nordiska Musikfdrlaget, 1963), p. 13. 3Ibid. In opposition to the stand of Edlund on the pre- viously stated subject is that of Thomson1 who, in the introduction to his recent sight singing text, states, One widely maintained (yet surprisingly undeveloped) position maintains that only a reliance on the discrete pitch interval can yield a reading technique that is useful for complex music. The validity of this argument dissolves as soon as its ramifications are made clear. Its most significant implication is that intervals can be apprehended and conceived as absolutes, exclusive of any background con- text. . . . But attempting to imagine the two pitches of an interval exclusive of any context beyond themselves is like trying to imagine two points in vision without a spatial reference. Rigorous attempts at either seem equally fruit- less; both must occupy positions in some form of auditory or visual background of space-time.2 This study does not attempt to investigate the possible predictive nature of accompanying atonal sonorities such that exists in tonal harmony. Instead, this study is designed to examine by comparison the affective nature of atonal accompaniment to performed atonal music. Hypotheses The hypotheses formulated for this study relate to the main tOpic of interest previously presented and areas of related secondary interest. These subjects of investigation stated as null hypotheses in general terms are: 1. No significant improvement in atonal sight singing ability will result from atonally accompanied drill 1William Thomson, Advanced Music Reading (Belmont, California: Wadsworth Publishing Co., 1969). 2Ibid., pp. ix—x. in atonal sight singing compared with unaccompanied atonal sight singing drill. 2. Progress in atonal sight singing ability, based on atonal sight singing drill, will not significantly affect progress in tonal sight singing ability. 3. There is no difference between males and females in the ability to sight sing atonal music. 4. There is no difference between persons with keyboard experience and persons without keyboard experience in the ability to sight sing atonal music. Definition of Terms 1. Atonal--a1though this term is often ambiguously used, for purposes of this study, the term refers to combi- nations of musical pitches lacking a devinite relationship to a tonal center that are not triadic or tertian in nature. 2. Atonal Sight Singing Test--a test comprised of Interval Series of five tones each, arranged in such a fashion as to not imply a definite tonal center. 3. Atonal Sight Singing Drills--drills comprised of Interval Series of five tones each, arranged in such a fashion as to not imply a definite tonal center. 4. Hanson's "six basic tonal series"--a "shorthand" method for visual recognition of the intervallic content of any given vertical pitch structure.l 1Howard Hanson, Harmogic Materials of Modern Music (New York: Appleton-Century-Crofts, Inc., 1960), pp. 7-16. 5. Interval Series--a series of intervals with no definite rhythmic notation--in the case of this study a series of five whole-notes. 6. Sonority--a term substituted for "chord," since the latter term implies a vertical structure of tonal harmonic usage. 7. Tonal Sight Singing Drills--drills comprised of Interval Series of five tones each, arranged in such a fashion as to imply a definite tonal center. 8. Tonal Sight Singing Test--a test comprised of Interval Series of five tones each, arranged in such a fashion as to imply a tonal center. Limitations Sixty students, enrolled in two laboratory sections of first-year music theory at The University of Calgary, were the participants in this study. One section was randomly selected as the experimental group and the remain- ing section designated as the control group. The duration of the experimental period was approximately thirteen weeks, equivalent to the winter term of study. Atonal Sight Sing- ing Drills, given in a language laboratory, were conducted twice each week over a twelve-week period with drill ses- sions approximately 22 minutes in length. At the conclusion of the twelve-week drill period an original two-part Atonal Sight Singing Test was individ- ually administered to all subjects. As a secondary study, a Tonal Sight Singing Test, which had been given to both groups as a pre-test prior to the twelve-week drill period, was readministered as a post-test. Scores from these tests were compared to measure relative group abilities to sight sing atonally-arranged and tonally-arranged Interval Series. Overview The following chapters present the main body of this study and conclusions drawn from its statistical results. A review of related literature, particularly of studies done in the area of sight singing during the last two decades and pertinent to this investigation, is presented in Chapter II. The design of this study with formal statements of the hypotheses to be tested and their analyses appear in Chapter III with the complete analysis, statistical documentation and interpretation in Chapter IV. The summation with conclusions drawn from the statistics and their implications for use by other experimenters in future studies are stated in Chapter V with the appendices following. For better comprehension of the purpose of this study one must be familiar with earlier investigations in the area of sight singing. Such information will enable one to grasp the logical manner in which it leads to the main hypotheses stated at the conclusion of Chapter III. CHAPTER II REVIEW OF THE LITERATURE One would expect that a disciplined skill dating back at least as far as the known history of music would be the subject of much experimental study. Yet the skill of sight singing, acknowledged by most musicians as a funda- mental tool, is one of the least researched areas in music. Furthermore, the application of sight singing to atonal music practices is relatively unexplored. The previously cited study of Sherman and Hill1 constitutes the only docu- mented information about sight singing in an atonal context, it being just one of fourteen areas covered by their exper- imentation. It follows, then, that the review of available existing literature must relate to the general area of sight singing, plus studies of perceptual abilities relevant to the experimental design of this study. In considering the historical development aspect of sight singing, much time could be spent commencing with the contributions of Guido d'Arezzo and tracing a forward prog- ress to the present. However, this has already been done in 1Sherman and Hill, op. cit. 10 1 in a comparable an extremely thorough manner by Ottman section of his dissertation. Therefore, repetition of that same anthological account seems unnecessary, and that excellent source is recommended to the reader. Recent research in sight singing seems to fall into three general areas of emphasis: (l) the development of sight singing tests, (2) factors that correlate with sight singing ability, and (3) the effectiveness of various kinds of drill as teaching devices for improved ability to sight sing. The last area also includes studies on the use of programmed learning materials, teaching machines and tape recorded exercises. The Development of Sight Singing_Tests The eternal dilemma concerning effective evaluation of sight singing performance has been, and remains, of con- stant concern to theory instructors. Some of the earliest research aimed toward arriving at satisfactory methods of measurement was done by Hillbrandz and M’osher3 during the 1Robert w. Ottman, "A Statistical Investigation of the Influence of Selected Factors on the Skill of Sight- Singing" (unpublished Ph.D. dissertation, North Texas State College, 1956), pp. 3-37. 2Earl K. Hillbrand, Measurigg Ability in Sight Sing- ing (Ann Arbor, Michigan: Edwards Brothers Pub., 1924). 3Raymond M. Mosher, A Study of the Group Method of Measurement of Sight Sin in (New York: Teachers College, Columbia University, Contributions to Education, No. 194, 1925). 11 mid-twenties of this century. It was not until very recently that this problem was again taken up as a subject of documented research. There have been, of course, many concerned educators who have drawn certain conclusions about this and other areas of sight singing through observation of their students; but this type of conclusion is generally considered to be quite suspect statistically. Reference is made to some of these observations made by respected theo- rists later in this chapter. Nelson1 compared the utilization of a sight singing test incorporating objective short items with the tradi- tional long-item test. Both groups of subjects tested received the same training in sight singing. This study revealed that,for the sample populations tested, the short- item method produced a significantly better evaluation of sight singing achievement than did the traditional long-item approach. In addition to this, three advantages of the short-item test over the long-item test were observed: (1) it contained a wider variety of performance problems with special reference to rhythm, (2) it was found more valid relating to a higher correlation with the dictation test which was also administered to both subject groups, and (3) it displayed greater efficiency in that administration and evaluation of the short-item test took one-third as much 1John C. Nelson, "A Comparison of Two Methods of Measuring Adhievement in Sight Singing" (unpublished Ph.D. dissertation, The University of Iowa, 1970). 12 time as did the long-item approach. Relative to the second advantage, the reliabilities of the sight singing test and the dictation test, computed in four ways, were found to be quite comparable with an approximate average of .95 for each test. However, such comparisons of two different types of tests is questionable. A performance test to evaluate the effectiveness of Kanable's "A Program for Self-Instruction in Sight Singing" was devised by Smith1 employing four matched groups of first-year theory students. No significant difference was statistically evident between any of the four groups--only one of which had experienced Kanable's programmed approach-- including the group that had received no sight singing training at all. Closer inspection of the results reveals certain areas of possible contaminating factors. The author himself questions the method of evaluation used which in- cluded a panel of judges observing live performances. Even so, the results obtained do not discredit the concept of using Kanable's self-instruction sight singing method as an effective substitute for classroom training by an instructor. IJames C. Smith, Jr., "A Performance Test of Kanable's 'A Program for Self-Instruction in Sight Singing'" (unpublished Ph.D. dissertation, The Florida State Univer- sity, 1968). 13 The Study of Factors Correlated with Sight Singing Ability Sight singing requires a synthesis of various musical performance factors which, when equally present, produce accurate musical interpretations. Researchers have approached investigations of these factors in various ways. Some experimenters chose to devise a criterion test which would include all factors and investigate their influence through performance comparison between standard test scores and criterion scores. Other studies have been accomplished through isolation of one or two factors and comparing them with other factors used as experimental variables. The former approach quite often produces interaction between factors causing confusion in interpretation of results. The latter approach receives criticism for placing musical fac- tors in an unmusical form. Statements by eminent persons connected with the field of music theory have, from time to time, been pro- nounced in relation to factors influencing an individual's aural and visual musical perception. Often these statements are products of personal theory or founded upon observations of others such as students in classes dealing with ear training or sight singing. Although not based on statis- tical data, these statements are deemed valuable from the standpoint of reflecting personal experience. Wedge, for example, states, 14 In sight singing there is no need of thinking the interval name or the size of the skip as long as the key is known. It is only when this feeling of key has been broken down by musical skips or a modulation that a knowledge of how to sing absolute or unrelated intervals is needed, and then only until the tonality has been reestablished. This statement has direct connotations toward the problem of sight singing atonal music. It implies that interval accuracy is of greater importance in sight singing atonal music; but of how great an importance is left Open to conjecture or research. MCNaught, referring to the perception of the inter- val itself, has said, "unless he possesses unusual capacity, [the student] finds himself continually distracted by . . . changes effected on intervals by the tonal relations of the notes."2 Farnsworth underscores this by commenting that “the interval as an abstraction or in isolation has only slight musical significance. What is of real importance is the interval in some specific context."3 Still another reference to the relative importance of interval recognition is that of Thomson“ previously cited in the Opening chapter (page 5) of this dissertation. 1George A. Wedge, Advanced Ear Training and Sight Singing (New York: G. Schirmer, Inc., 1922), p. 9. 2W. G. McNaught, "The Psychology of Sightsinging," Proceedings of the Musical Association, XXVI (1900), 35-55. 3Paul R. Farnsworth, The Social Psycholggy of Music (Ames, Iowa: The Iowa State Un1versity Press, 1969), p. 45. I’Thomson, op. cit. 15 In the broader sense of music perception, Hindemith has written the following commentary: One group catches primarily the general impression of a musical structure, and through further analytical thinking--with the aid of repeated apperceptions of the same material or of conclusions based on memory and logic-- discovers and adds the constituent parts, thus literally filling out the pre-established musical form. Of no less importance is the other method, which adds up great numbers of single aural impressions, till in the listener's mind the complete form is synthetically reconstructed. A well experienced musician will never rely on either of these two methods alone but will always mix them, using at any given moment (consciously or intuitively) the one that gives him the greater push forward.1 Early studies by Salisbury and Smith2 and Dean3 tried to determine what factors were correlated with sight singing ability. Salisbury and Smith found the main eval- uating factors to be dictation, pitch and tonal memory; whereas Dean, through determining the value of using the Seashore Measures of Musical Talent in the prediction of success in sight singing, found the Seashore pitch test and Seashore tonal memory test the most valuable in such a prediction. 1Paul Hindemith, Elementary Training for Musicians (New York: Associated Music Publishers, 1949), p. 182. 2Frank S. Salisbury and Harold B. Smith, "Prognosis of Sight Singing Ability of Normal School Students," Journal of Applied Psychology, XIII (1929), 425-439. 3Charles D. Dean, "Predicting Sight Singing Ability in Teacher Education," Journal of Educational Psyghology, XXVIII (November, 1937), 601-608. 16 Ortmann investigated the effect of melodic memory as part of aural perception.1 The subjects were required to immediately notate short melodic phrases of five notes each as they were played on the piano. The results of this study enabled Ortmann to list, in order of difficulty, certain characteristics of melodic material. Conjunct and disjunct motion with wide interval leaps were found to be most difficult to perceive. This study exemplifies the isolated factor approach which is similar to the present study. In the experimental study by Ottman a number of factors were investigated utilizing several standard pub- lished tests as well as tests devised by Ottman himself.2 The specific standard tests were The Seashore Measures of Musical Talent, The American Council on Educational Psycho- logical Examination for College Freshmen, and The Nelson- Denny Reading Test (music reading). The factors and/or variables examined and compared in this study included tonic memory, melodic modulation, melodic dictation with and with- out rhythm, identification of isolated intervals, spelling of isolated intervals, singing of isolated intervals, music literacy and two questionnaires--one for faculty evaluation 1Otto Ortmann, "Some Tonal Determinants of Melodic Memory," Journal of Educational Psychology, XXIV (September, 1933), 454-456. 2Ottman, op. cit., p. 93. 17 of individual student musicianship and one for students to furnish general information. The subjects for Ottman's study were three second- year music theory classes at North Texas State College (University) containing a total of 52 students. This group of students were not considered as a sample of the pOpula- tion and therefore, the "study should be considered as a descriptive investigation of this particular situation wherein certain avenues for further research may be made apparent."1 The criterion for measurement in this study was an unidentified melody by Luigi Cherubini which, because of its diverse musical content, was considered to be a suitable melody for such a purpose.2 The criterion pre-test dis- played a Spearman-Brown reliability of r = .861 and the post-test a Spearman-Brown reliability of r = .904. Of particular interest to this author was the use of an exact duplicate of the criterion melody in terms of pitches and key but notated completely in half-note values to eliminate the rhythmic element. Each of the fifty-two subjects was asked to sing this melody individually and errors in pitch were circled on the examiner's copy in the same manner as was done on the original criterion melody. Although this test was given before the post-test of the criterion melody and in spite of the fact that this version of 1Ibid., p. 93. 2Ibid., p. 239. 18 the melody was in the same key as the original, therefore causing all intervallic relationships to appear staff-wise the same as before, still none of the subjects appeared to recognize this rhythmless melody as the one they had sung before. . . . Most of the subjects made fewer pitch errors when the element of rhythmic read- ing was eliminated. Improvement in reading pitch took place for thirty-nine subjects, ranging in item improvement from one less pitch error to twenty-four fewer pitch errors, or from 2.86 per cent improvement to 100 per cent improvement.’ The statistical results of Ottman's study yielded no significant relationship between the criterion and any musical or mental element measured by the employed standard- ized tests. The only exception that might be considered in the preceding statement is the Seashore pitch test; its correlation coefficient with the criterion was .300. It was the lowest statistically significant coefficient obtained and it was significant at the .05 level only. Other results of this study, measured by original tests of Ottman, indicated: (a) "music literacy" (the ability of a sight-singer to hear mentally what is seen in the printed score) displayed the largest degree of relation- ship to the sight singing criterion being expressed with the high correlation coefficient of .733; (b) a correlation of r = .678 of the six interval tests as a group with the criterion; (c) scores of the Melodic Dictation, Melodic Modulation, and Tonic Memory tests all had positive cor- relations with the criterion scores and were statistically 11bid., pp. 88-89. 19 significant at the .01 level; and (d) of the seven correla- tion coefficients produced by the two questionnaires, four were statistically significant at the .01 levels: "Subjects' Self-Evaluation of Sight-Singing Ability (r = .662); Faculty Evaluation of Subjects' Musicianship (r = .517); Subjects' Active Musical Participation (r = .499); . . . and Subjects' Self-Evaluation of Musicianship (r = .352)."1 A.more recent investigation by Read, also conducted at North Texas State University, dealt with selected vari- ables in relation to sight singing ability.2 The subjects for this study were 125 undergraduate students selected for membership in the A Cappella Choir and the Chapel Choir at North Texas State University. The purpose of the study was the statistical comparison of twelve selected musical vari- ables with seven extra-musical variables. The criterion variable for the comparisons was the test score obtained from an original sight singing test given to each subject. The subject's music theory grades in ear training and keyboard technique, his permanent rec- ords in the Registrar's office, the subject's responses to an original questionnaire designed to obtain pertinent information about his environmental background and musical experience, and also the subject's placement on an original 11bid., pp. 223-24. 2Read, 0p. cit. 20 rating scale of music interests supplied data for the vari- ables under investigation. Three standardized tests--The Seashore Measures of Musical Talent, The Gordon Index of Musical Insight, and The American College Testing Programr-were also used to supply additional data along with a "Pure Tone Threshold Test" administered by the North Texas State University Speech and Hearing Clinic. Conclusions drawn from the results of the study indicate that the student excelling in sight singing ability tends to display more natural musical ability, has had more formal classwork in music and appears to be more intelligent than the student registering a low level of sight singing ability. The similarities of the design, selection of vari- ables, and their application to Read's study and that of Ottman cannot be dismissed as simply coincidence. Read's study was more or less a reapplication of Ottman's study to a different experimental population producing similar conclusions from the results. Replication is, of course, needed and appropriate. The effects of scale, harmony, and tonality related to interval accuracy in melodic sight singing was the subject of a study by Marquis.1 That experimenter also 1James H. Marquis, "A Study of Inteval Problems in Sightsinging Performance with Consideration of the Effect of Context" (unpublished Ph.D. dissertation, State University of Iowa, 1963). 21 investigated the relationship of scale, harmony, and tonality to the basic difficulties of interval quality. The element of rhythm was present in this study but controlled by simple usages. To accomplish this, Marquis approached the problem three different ways as stated in the following hypotheses related to first-year college-level sight singing students: 1. The percentage of errors made in singing a music interval will differ, depending on the context in which the interval appears. 2. Those who make errors in singing a given interval presented in melodic context will tend, also, to make errors in singing the same interval presented in isolation. 3. The percentage of errors made in singing an isolated interval at sight differs from that made in singing the same interval under different conditions of context.1 Two criterion tests were devised and tested for reliability and validity--the Sightsinging Criterion and the Isolated Intervals Criterion tests. These tests were designed to produce statistics which would test the above stated hypotheses. The Sightsinging Criterion was found to have a reliability of .979 (Spearman-Brown) , significant at .01 level of confidence. Validities of .776 and .828 were drawn from correlation coefficients with grade points in the sight singing and ear training course and scores on a semes- ter Aural Skills Dictator Test which evaluated interval scale lIbid., p. 5. 22 and chord perception; both were significant at .01 level. The Isolated Intervals Criterion was found to have a reli- ability of .830 (Spearman-Brown), significant at .01 level of confidence, and internal validity was claimed. Like interval items in the Sightsinging Criterion were classified, according to their contextual settings, into three categories: Simple, Moderately Complex, and Complex. Item to item comparisons were made of multiple sets of interval items with each set consisting of a single item taken from each categorized contextual setting. The analysis included 45 complete single-interval-item sets from each of the three contextual categories and 14 partial single-interval-item sets relating to only two contextual categories. The descending tritone, ascending minor sixth, ascending and descending major sixth, and the descending major seventh intervals could not be formed into sets, since their items were all present under the same contextual con- ditions in the test melodies. Therefore, they were not compared for melodic setting error differences. However, these intervals were included in the comparison with isolated interval sight singing. The subjects for Marquis' study were 52 first-year music theory students at the University of Iowa. The tests were administered individually and privately with all per- formances recorded on magnetic tape. In the interest of 23 having a student feel as relaxed as possible, the admin— istrator gave the tests rather than employing a structured tape-recorded presentation. There are merits to this approach, but it also allows the entrance of contaminating factors into the test situation. Also, when the subject made a mistake, he was stOpped, corrected by the adminis- trator and then allowed to continue on commencing at a point prior to the place of error. This is a form of reinforce- ment, but also a form of contamination. Because of the depth and breadth of the statistical results of this study, an attempt won't be made to reproduce the numerous facts, figures, and charts which Marquis derived from the investigation of each interval. However, interested readers are referred to that section of Marquis' dissertation.1 The results indicated acceptance of the first hypothesis, rejection of the second and acceptance of the third. In the words of Marquis this implied that: l. First-year college level music students are significantly affected in Sightsinging by differences in contextual settings of the intervals. 2. Ability or lack of ability in the singing of an isolated interval does not directly affect the singing of that interval in melody. 3. In melodic Sightsinging, ability to per- ceive the basic quality of intervals is lIbid., pp. 114-160. 24 considerably less important than ability to perceive the scalar, harmonic, and tonal changes across or surrounding intervals.1 One of Marquis' suggestions for further research concurs with one amplified in the present investigation as well; specifically, a comparison of sight singing material which is key-oriented with one group referring to key signatures and the other group using accidentals before the notes only. The findings of Marquis have direct bearing on the present study as the contextual environment is the prime variable investigated by both. A recent study by Buttram also deals with the subject of context in which an interval is presented; and, although the study does not directly involve sight singing, its implications most certainly relate to it.2 The subjects for this investigation were 1,678 freshmen and 1,071 sophomore theory students from 46 colleges and universities in 30 states. A study of such breadth and depth can only yield meaningful results with a high probability of its being a representative sample. Buttram investigated the influence of four factors-- interval quale, pitch distance, tonal context, and relative distinctiveness--on the identification of musical intervals 1Ibid., p. 173. 2Joe B. Buttram, "The Influence of Selected Factors on Interval Identification," Journal of Research in Music Education, Fall, 1969, pp. 305-15. 25 presented in a harmonic context. The term "quale," used in psychology, refers to the combination of two different musical tones resulting in a unified, distinctive effect possibly serving as a basis for interval identification. The feeling of harmony or tonality was created by the initial sounding of I-VPI chord progression before each interval was presented for identification. Then the tones composing the interval presented were drawn from those contained in the I, IV, or V chords of the same key. Influence of context was accomplished through the difference in selection of the two interval pitches as well as by comparison to the control. Therefore, the first and second presentations utilized different tones for each interval allowing comparison of results in two situations of context. The questions posed for this study were stated in null hypothesis form: 1. There is no difference in the accuracy of identification of intervals between the control version and one designed to reveal the influence of interval quale. 2. There is no difference in the accuracy of identification of intervals between the control version and one designed to reveal the influence of pitch distance. 3. There is no difference in the accuracy of identification of intervals between the control version and one designed to reveal the influence of tonal context. 26 4. There is no difference between the relative distinctiveness of intervals as identified in a control version and in versions de- signed to reveal the influence of quale, pitch distance, tonal context and melody.1 The test administered to the 2,749 subjects consisted of five versions of the same material--two pre- sentations of each interval contained in the octave. The same intervals appeared in all five versions in a varied order of presentation. Development of "key feeling" was avoided through juxtaposition of the intervals and the pitch range was confined to an octave above and below middle "C." Reliability, established by the split-half technique, for two repetitions of each interval for each of the five versions of the test ranged from .89 to .92 with interval validity also claimed. Four of the five versions related to the hypotheses previously stated with the fifth used as the control. To allow for widespread simultaneous testing and control of variables, the test was recorded on tape as were the instructions and practice examples. The subjects identified the intervals through written response. Results of the testing indicated the rejection of all four null hypotheses significant at the .01 level of probability. This was the case for both the freshman and sophomore groups as well as the total sample population. 'Ibid., p. 312. 27 Of particular interest are the results of the test versions examining context and melody. Although the highest accuracy, with a mean of 14.3 for the total group, was in the melodic version, the next highest occurred in the con- text (within a harmonic framework) version registering a mean of 13.2 for the total group. The order of difficulty of intervals, which was established in the control version, varied somewhat in the four experimental versions. Only the octave, shown to be the most distinctive interval, maintained its position. The conclusions to be drawn from this situation is that ordering or classification of intervals considered typical of any context is questionable. Buttram sums up the resulting conclusions of his study stating, Interval identification, as it occurred in this study, might best be described as a series of judgments based on a variety of characteris- tics of the interval and on the experience of the subjects with these characteristics.1 The factor of an accompanying duplicate pitch influencing pitch accuracy of a performer as opposed to an unaccompanied performance was investigated by Vbrce testing the following five hypotheses:2 lIbid., p. 315. 2Frederick W. Vbrce, Jr., "The Effect of Simulta- neous Stimulus on Vocal Pitch Accuracy" (unpublished Ph.D. dissertation, The Florida State University, 1964). 28 Subjects perform with greater accuracy when there is a time-lapse between the stimulus and the response than when the stimulus is sounded simultaneously. Subjects perform with a tendency in the direction sharpness (higher cps) rather than flatness (lower cps) when compared to the stimulus. There is not a consistent pattern of dif- ferences between individual subjects or groups. verbal conditioning produces a measurable difference in pitch performance in the direction of the conditioning process. Subjects do not perform with the same accuracy in each trial.1 The subjects, 40 music students at Florida State University, were randomly divided into five groups of eight subjects each--one group for each hypothesis to be tested-- and were asked to vocally respond to mixed combinations of two musical pitches, accompanied and unaccompanied. The two stimulus pitches were A = 440 and Eb = 330 with each pitch sounded simultaneously with the subject's response for the accompanied treatment. The unaccompanied treatment con- sisted of the pitch being sounded, then ceasing and, after a time delay, the subject responding. Other verbal direc- tions were given to influence conditioning for the different hypotheses tested. All tests were recorded on magnetic tape and were later analyzed using a Conn Chromatic Stroboscope. 1Ibid., p. 24. 29 The main conclusion from Vorce's experiment that is pertinent here is the indication that accompanied per- formances were highly superior to unaccompanied performances with all trial results significant at the .01 level of con- fidence. The vocal transposition of an octave by certain (male) subjects seemed to make no difference in the results, according to Vorce's data analysis. Both of these facts are pertinent to the present study related to the experimental treatment used. Effectivgness of Various types of Drill and Drill Procedures as Teaching Devices for Improved Sight Singing Ability Maximum achievement in the acquisition of skills related to music theory is of constant concern to the theory pedagogist. Much has been accomplished through the utiliza- tion of electronic tape recording devices in the field of ear training and more recently in the field of sight singing as well. The marketable pricing of electronic pianos with SOphisticated class instruction systems have revolutionized and accelerated the keyboard aspect of theory as well as the concept of class piano. Teaching machines can be programmed for instruction of music fundamentals up through any level of information desired. The one underlying benefit for both the student and the instructor is the limitless Opportunity for drill, the necessary ingredient to achievement of the highest possible level of any skill ability. 30 Experimenters and teachers have constantly searched for ways to aid and accelerate the teaching of skills at least as far back as the Guidonian Hand. Experimental studies related to different kinds of drill in sight singing have only recently increased in number, due mainly to tech- nological advancement. Earlier studies by Bean,l Stokes,2 and Christ3 have pointed the way to further experimentation. Examples of this are the several experiments done with the tachistoscope, a device used to accelerate skills in word- reading, employed by Stokes and Christ in their studies. The study by Cookson, conducted at Northwestern University in the summer of 1949, was an initial application of tape recorders as a teaching device for music skills.“ A class of sixteen subjects received ear training drill over a six-week period involving occasional reinforcement tests and a final placement test. Scores from the final test compared with scores of regular music theory classes indicated that 1Kenneth L. Bean, "An Experimental Approach to the Reading of Music," Psychological Monographs, L (1938), 1-79. 2Charles F. Stokes, "An Experimental Study of Tachistosc0pic Training in Reading Music" (unpublished Ph.D. dissertation, Teachers College, University of Cincinnati, 1944). 3William E. Christ, "The Reading of Rhythm Notation Approached Experimentally According to Techniques and Prin- ciples of Word Reading" (unpublished Ph.D. dissertation, Indiana University Music Library, 1953). I'Frank B. Cookson, Recordings and Self Tutoring (Cleveland: The Brush Development Co., 1949), pp. 5-8. 31 75 percent of the experimental class accomplished twice the amount of ear training material that would normally be expected of such a class. Spohn further investigated the potential of tape- recorded music drill, realized in Cookson's study, in its application to the skill of "aural comprehension."1 This experiment employed the classic two-group (control and experimental) design with results emanating from the com- parison of pre-test scores with post-test scores. The subjects for the study were music students at Ohio State University, who, through test scores on the Ohio State Psychological examination, music entrance tests, applied instrumental area, and sex were divided into two matched groups. The control group developed aural comprehension through conventional classroom methods, whereas the experi- mental group, using the same material, developed their aural comprehension utilizing specially prepared tape recordings outside of class. The time allotted to aural comprehension development was equal for both groups. The null-hypothesis for the study stated: 1Charles L. Spohn, Jr., "An Exploration in the Use of Recorded Teaching Material to DevelOp Aural Comprehension in College Music Classes" (unpublished Ph.D. dissertation, Ohio State University, 1959). 32 As a result of specially prepared tape recordings, to be used outside the classroom, the students of the experimental class will not develop better aural comprehension than will the control class who develop aural comprehension through conventional- classroommethods.l This null-hypothesis was rejected at the .05 level based on a comparison of decrease in the number of errors made between the pre-test and post-test Of both groups. The control group registered an average percentage decrease Of 57.68 percent, while the experimental group displayed an average percentage decrease of 80.33 percent leaving a difference of 22.65 percent between the two groups in percentage decrease. The Obvious conclusion drawn from these results is that the use of tape-recorded drills and exercises is a valid method for development of aural comprehension. Recomr mendations made by Spohn at the conclusion of the study have been the basis for experiments conducted by researchers since then. One of the follow-up experiments done in relation to the above study was also done by Spohn himself, assisted by Poland, under a United States Government funded grant.2 1Ibid., p. 81. 2Charles L. Spohn and William Poland, An Evaluation of Two Methods Using Magnetic Tape Recordings for_Programmed Instructippin the Elemental Materials of Music, Final Report Of Title VII, Project No. 876, NationalfiDefense Education Act Of 1958, Grant No. 7-34-0430-172 (Columbus, Ohio: The Ohio State University, November, 1963). 33 This study employed a four-group design representing four different treatments of equivalent materials. Three basic elements Of intervals, rhythms, and tone groups were iso- lated into three separate ten-week studies. All subjects were given the same pre-tests and post-tests at the begin- ning and conclusion of each ten-week period of training corresponding to the Fall, Winter, and Spring quarter-terms at Ohio State University. The treatment groups for all three elements inves- tigated employed the following treatment combinations: 1. Aural Stimulus-Written Response (treatment no. 1) 2. Visual Stimulus-Voice Response (treatment no. 2) 3. Aural Stimulus-Voice Response (treatment no. 3) 4. Visual Stimulus-Written Response (treatment no. 4).1 The second and third treatments are particularly relevant to the present study. They produced the following statis- tical results: 1. Interval Study--yie1ded F-ratios of 4.61 (significant at .05) for the stimulus and 12.30 (significant at .01) for the response, indicating both the stimulus and the response significantly affect the experimental results Of the second treat- ment. The third treatment results were complicated by significant interaction with an F-ratio of 4.36 1Ibid., p. 39. 34 (significant at .05). However, the stimulus yielding F 12.38 (significant at .01) and the response F = 5.24 (significant at .05) underscored the fact that there is no better training than the task itself. Rhythm Study--no significant differences were Obtained from the second and third treatments of this element. Tone Group Study-~the second treatment yielded an F-ratio of 6.35 (significant at .05) for the voice response indicating that it is superior to the written response treatments. The third treatment yielded no significant difference. The conclusion one can draw from the results of the study by Spohn and Poland, related to voice response, is that the treatment using a visual stimulus and calling for a vocal response can be considered valid in a similar environment of tape-recorded programmed instruction involv- ing intervals and tone groups. The tachistoscope, the device used by Stokes1 and Christ2 in their investigations, was also employed by 1Stokes, op. cit. 2Christ, op. cit. 35 Barnes,1 Hammer,2 and Ray3 in their investigations of selected drills and their influence on certain factors affecting sight singing ability. Barnes' study dealt with two primary purposes: (1) investigation of the effect of group drill in sight singing selected intervals upon the subject's ability to sight sing those intervals and (2) determination of the correlation between that intervallic sight singing ability and the ability to sight sing melodies composed of those same intervals. Secondary investigations of the study determined an order of interval difficulty and compared sight singing modal or modulating melodies with sight singing major, minor, or non-modulating melodies in relation to the ability to sight sing intervals. An original interval sight singing test and an original melody sight singing test were developed by Barnes and thoroughly checked for reliability and validity. The subjects for the experiment were 60 members of first-year theory class at Indiana State Teachers College and were 1James W. Barnes, "An Experimental Study of Interval Drill As It Affects Sight Singing Skill" (unpublished Ph.D. dissertation, Indiana University, 1960). 2Harry Hammer, "An Experimental Study of the Use of the Tachistosc0pe in the Teaching of Melodic Sight Singing," Journal Of Research in Music Education, XI (1963), 44-45. 3Harry B. Ray, "An Experimental Approach to the Reading of Pitch Notation" (unpublished Ph.D. dissertation, Indiana university, 1964). 36 placed into matched experimental and control groups. The two groups received the same classroom instruction, the only difference being an additional ten hours and twenty-five minutes of tachistoscopic drill in the sight singing of intervals over a ten-week period. Intervals used in the drills as well as the tests were limited to a fifth--major and minor seconds and thirds, perfect fourths and fifths. However, because of its diffi- culty, the tritone was excluded. A practical singable range of Bb to d2 was also decided upon. Notation for the inter- val drills and test was in half-notes with necessary sharps or flats placed before the notes eliminating the use of key signatures. Experimental drill sessions were accomplished, as a group, in a darkened room using a microfilm projector with a tachistoscopic lens attachment. Intervals were flashed upon a screen with the time interval of exposure varied from unlimited progressing tOrl/100 of a second. The ini- tial pitch of the interval was sounded with the students responding with both pitches. Several results produced by this experiment are relevant to the present study. The performance of the experimental group was significantly better than the control group in both the sight singing of intervals (mean differ- ence of 8.90 at the .01 level Of confidence) and of melodies (mean difference of 7.45 at the .05 level of confidence) 37 with the melodic sight singing ability being slightly less significant statistically. A high correlation (.916) was found between inter- vallic and melodic sight singing abilities, this being derived from the correlation of Interval Test scores and Sight Singing Test scores and found to be at the .01 level of confidence. However, the correlation between the im- provement in intervallic sight singing ability and the improvement in melodic sight singing ability, although very significant (.525) and also at the .01 level of confidence, was not as high as the correlation between intervallic and melodic sight singing abilities. The order of interval difficulty resulting from Barnes' study is of particular interest. Since it was derived from the same limitation of intervals employed in the present study, Barnes' chart1 is reproduced here for later reference. This interval hierarchy was determined by the results of 208 tests, in which each interval was presented five times in each direction, totaling 2,080 occurrences and, therefore, can possibly be considered significant. 1Barnes, op. cit., p. 71. 38 ORDER OF DIFFICULTY OF INTERVALS INVESTIGATED AS DETERMINED BY ANALYSIS OF INTERVAL TESTS Number Number Rank of Times of Times Order of Interval Direction Presented Missed Difficulty Major third Down 1040 432 1 Perfect fifth Up 1040 394 2 Perfect fifth Down 1040 382 3 Perfect fourth Down 1040 372 4 Perfect fourth Up 1040 345 5 Major second Down 1040 342 6 Minor third Down 1040 315 7 Major third Up 1040 287 8 Minor third Up 1040 284 9 Major second Up 1040 257 10 Minor second Up 1040 161 11 Minor second Down 1040 107 12 The interval drill was found to be an effective aid to both sight singing of major, minor, and non-modulating melodies as well as modal and modulating melodies. However, because comparisons were made by dividing each group into highest and lowest quartiles, the small number of subjects in each quartile lessened the degree of significance in relation to the other results of the study. The results of Barnes' study indicate that tachis— tosc0pic drill was an effective method of teaching the sight singing of intervals. The studies of Hammer1 and Ray,2 on the other hand, yielded results that were divided as to ' 1Hammer, Op. cit. 2Ray, Op. cit. 39 the effectiveness of tachistoscoPic drilling of subjects in selected musical skills. Hammer used a tachistoscopic attachment on the lens of an overhead projector during tonal pattern practice ses- sions with a class of fourth-grade students. The results of his study showed tonal pattern practice with the tachis- toscope significantly superior to conventional methods of tonal pattern practice. Ray experimented with the tachistoscoPe comparing the results of two types of drill with pitch patterns. The experimental group, receiving the tachistoscope drill treat- ment, used pitch patterns of varying length and complexity and not identified with a particular key or mode. The con- trol group, drilling from.mimeographed sheets, used pitch patterns related to specific keys and modes. The subjects were first-year music students at Western Michigan Univer- sity. The measurements of the achievement of the two groups were based on scores derived from an original sight singing test administered both as a pre-test and post-test. The experimental groups received four half-hour drill ses- sions each week over a ten-week period while the control group received their practice in the conventional manner. Results of the statistical comparisons showed no significant difference between the groups and thereby questioned the value of the tachistoscope drills in this 40 particular experimental situation. However, one must con- sider the possibility of contamination in this study due to the differing nature of the musical materials used by the two groups. Duplicate materials might possibly have altered the results of Ray's investigation. Although not directly connected with the skill Of sight singing, the study by Ihrke which investigated the SOphisticated fusing of tape recorders, computers, and teaching machines into unified training systems for music theory skill development should be cited.1 A progress re- port on an initial study in rhythmic training, implemented at the University of Connecticut, described the combined use Of an electric organ, stereo tape recorder and an electrical device called a "rhythm monitor" in teaching rhythmic dicta- tion.2 The "rhythm monitor," the unique experimental fea- ture, employs error lights indicating earliness or lateness of responses. TWenty-four elementary education students partici- pated in the study with seven in the experimental group and seventeen in the control group. The resulting data indicated that "automated rhythm training" is a valid and effective training method. The study also indicated that 1Walter R. Ihrke, "Automated Music Training," Journal of Research in Music Education, XI, No. 1 (Spring, $63) I 3.200 2Walter R. Ihrke, Automated Rhythm Training, Council for Research in MuSIc Education Bulletin No. 7, Spring, 1966, pp. 35-37. 41 the experimental method of rhythmic perception training is as effective as traditional methods of training rhythmic perception. The combination of a programmed self-instructional sight singing method and the medium of the tape recorder was compared to the conventional teacher-classroom method by Kanable at Northwestern University.1 Using as subjects high school students enrolled in a three-week summer music program on the Northwestern University campus, control and experimental groups were randomly formed and given a pre- test in sight singing which was recorded on magnetic tape for scoring purposes. Both groups covered the same basic material and participated in the same amount of instruction time--twelve fifty-minute periods--with the only difference being the tape-recorded programmed self-instructional sight singing method applied to the experimental group. At the conclusion of the training period, the same sight singing test used as a pre-test was again administered to both groups as a post- test. The questions investigated by Kanable are stated as the following null hypotheses: 1Betty Mae Kanable, "An Experimental Study Comparing Programmed Instruction with Classroom Teaching of Sight Singing" (unpublished Ph.D. dissertation, Northwestern University, 1964). 42 1. There is no significant difference between sight singing scores of subjects as a func- tion or method of training. 2. There is no significant correlation between sight singing scores of experimental sub- jects and ability in tonal memory. 3. There is no significant correlation between sight singing scores of control subjects and ability in tonal memory.1 Results Of the measurement of scores, using an analysis of variance of error scores, indicated acceptance of the first null hypothesis. In other words, the tape- recorded sight singing instruction did not prove to be significantly more effective than the conventional classroom instruction. However, using a "t" formula for determining the ratio of the difference between related scores, results indicated t = 5.05 for the experimental group (significant at .01 level of confidence) and t = 3.02 for the control group (significant .02 level of confidence). The second null hypothesis was rejected using the Spearman rank order correlation in which the experimental group showed correlation significant at the .05 level of confidence on both the pre-test and post-test between scores of the Test of Tonal Memory and Error Detection and scores Of the Sight Singing Test. The last-mentioned tests were given to all students before the selection of the control and experimental groups as a screening procedure. 'Ibid., p. 17. 43 Thus, there was significant correlation between sight sing- ing scores of experimental subjects and ability in tonal memory. The Spearman rank order correlation was also used to test the third null hypothesis with the results indicat- ing its acceptance. The control group did show significant correlation at the .05 level between ability in tonal memory and pre-test sight singing scores, but correlation between tonal memory ability and sight singing scores on the post- test was not significant. Although the results were not significant enough in Kanable's study to claim complete support, they do indicate that the use of such a tape-recorded programmed self- instructional method is just as effective, if not moreso, as classroom instruction in sight singing, a fact which is quite relevant to the present experimental design. Contributing to the limitations of the present study are the conclusions derived from Helbling's study comparing the effects Of using the "whole" method and the "part" method in teaching sight singing.1 The subjects participating in the experiment were 26 selected undergrad- uate elementary education majors at Asbury College who had little or no previous formal study in sight singing. lDeVon W. Helbling, "An Experimental Study of the Relative Effectiveness of 'Whole' and 'Part' Methods of Teaching Sight Singing" (unpublished Ph.D. dissertation, Indiana University, 1965). 44 Using the parallel group technique, 13 class members of the Winter quarter section of the course were equated with 13 class members of the Fall quarter section to act as the experimental subjects. The Fall quarter experimental subjects were taught sight singing using the "whole" concept, and the Winter quarter experimental subjects learned sight singing through use of the "part" concept of teaching. Twenty class sessions were devoted to teaching each concept. An original sight singing test, containing five melodies, was constructed by Helbling. The melodies corre— sponded in type and level to those intended as the teaching- learning goal of the program of study. The test claimed content validity and, utilizing the test-retest method, was found to have a reliability of .97. This test was used as the pre-test, midterm test, and post-test to measure the relative progress of the two experimental groups. All tests were recorded on magnetic tape for uniform scoring purposes. The sight singing pre-test, midterm and post-test scores were compared measuring the relative progress of the two experimental groups. Using difference scores, the 3 test was employed to discern whether a significant differ- ence existed between the two sets of "difference" scores. The comparison of pre-test to post-test scores produced a mean difference of 2.01 in favor of the Whole Method Group, but was not statistically significant. The pre-test comparison with the midterm test scores yielded a mean difference of 6.61 in favor of the Part Method Group, 45 found to be significant at .05 level of confidence. The comparison of midterm scores with post-test scores showed a mean difference of 8.62 favoring the Whole Method Group and significant at .01 level of confidence. A general conclusion that may possibly be drawn from these results is that a combined use of the "whole" and "part" concepts might be a more effective approach to the teaching of sight singing with the "part" method emphasized in the early stages and the "whole" method employed later. This combined concept is similar to a method of approach advanced by Kunkle.1 If the "part" method is more effective in the beginning stages of sight singing training, it rein- forces the approach that has been incorporated in the design of the present study. More recent experiments with variation of drill procedures to study their effects have been done by Wilcox,2 Tucker,3 and Justus;“ but all have been basically concerned 1Robert F. Kunkle, "A Direct Approach to Sight Singing," Music Educators Journal, June-July, 1950, p. 41. 2Eunice A. Wilcox, "The Effects on Sight Singing of Voice Class Instruction Utilizing Variants of Traditional Vocalises" (unpublished Ph.D. dissertation, Michigan State University, 1968). 3Gerald L. Tucker, "The Influences of Isolated Rhythmic Drill on Growth in Sight Singing" (unpublished Ph.D. dissertation, The University of Oklahoma, 1969). I'Lane D. Justus, "Evaluation of an Innovative Instructional Design for Sight Singing" (unpublished Ph.D. dissertation, University of Arizona, 1970). 46 with the improvement of ability to sight sing music of a tonal nature; although Wilcox had hoped that drills devised for that study would aid the sight singing of atonal music as well. Unfortunately, significant results were not obtained in that investigation. Atonal Music Investigations The only documented direct investigation into atonal sight singing implemented so far has been by Sherman and Hill as one part of a fourteen—part investigation of melodic perception in tonal and atonal environments.1 This elab- orate study, made possible through a grant from the Office of Education of the U.S. Department of Health, Education and Welfare, was implemented on the campus of Ball State Univer- sity in Muncie, Indiana. The purposes Of the study were stated in three questioning hypotheses. / 1. Will progress in the aural and visual per- ception of music through instruction based upon atonal organization transfer to prog- ress in the aural and visual perception of tonal music. 2. Will students with diffefent levels of ability show similar attainment in the aural and visual perception of music under conditions of a. selected responses, and b. constructed responses? 3. Will students with different levels of musical ability respond with similar 1Sherman and Hill, Op. cit. 47 attitudes toward tape-recorded self- instruction?1 The first and third hypotheses are directly relevant to the present investigation as is this statement made in relation to the above hypotheses. Abundant evidence exists which points out that formal instruction in music is still conducted today primarily in tonal idioms, although con- temporary music tends to be atonal in nature. The most common justification for this situa- tion frequently centers around two issues: first, the majority of "classic" or serious music is in tonal idioms, much of it centuries Old; second, if fundamental instruction were given in contemporary atonal idioms only, stu- dents would be handicapped with respect tO the vast,majority of Older serious music which is tonal in nature. This project was directed primarily toward the second of these common justifications; i.e., will atonal instruction transfer to tonal idioms?2 To administer such a multi-faceted study, an elab- orate laboratory was constructed consisting of multiple student stations equipped with stereophonic tape recorders and head sets with a third Of the stations being insulated booths in which sung (vocal) responses were performed. An extensive examination, used as both the pre-test and post-test, was devised during the pilot year. It con- sisted of fourteen separate tests for which a wide selection of specially composed materials were stereophonically tape-recorded by various ensemble combinations of faculty 1Ibid., p. 35. 2Ibid., p. 35. 48 instrumentalists. The tests measured the subject's recog- nition of musical intervals and melodies accompanied by one, two, and four other voices set in approximately equal num- bers of tonal and atonal idioms. There were four types of responses called for in the pre-test (post-test) examination: (1) selected response-aural, (2) selected response-visual, (3) con- structed response-written, and (4) constructed response- vocal. The items also reflected varying degrees of diffi- culty relative to the context and complexity of accompanying voices as well as the relative difficulty of melodies com- pared to intervals. Nine degrees of complexity can be discerned beginning with the interval with one, two, or four other voices; the 3-4 note melody with one, two, or four other voices; and the 5-7 note melody with one, two, or four other voices. An outline of the fourteen tests which comprised the pre-test (post-test) examination of Sherman and Hill's study follows: PART I--SELECTED RESPONSE TO AN AURAL STIMULUS Test 1. Intervals Accompanied by One Other Voice 72 Tonal-72 Atonal Test 2. Intervals Accompanied by Two Other voices 72 Tonal-72 Atonal Test 3. Intervals Accompanied by Four Other Voices 72 Tonal-72 Atonal 49 Test 4. Three to Four Note Melodies Accompanied by: One Other Voice 10 Tonal-10 Atonal Two Other voices 10 Tonal-10 Atonal Four Other Voices 10 Tonal-10 Atonal Test 5. Five to Seven Note Melodies Accompanied by: One Other Voice 10 Tonal-10 Atonal Two Other Voices 10 Tonal-10 Atonal Four Other Voices 10 Tonal-10 Atonal PART II--SELECTED RESPONSE TO A VISUAL STIMULUS Test 6. Intervals Accompanied by One Other Voice 24 Tonal—24 Atonal Test 7. Intervals Accompanied by Two Other Voices 24 Tonal-24 Atonal Test 8. Intervals Accompanied by Four Other Voices 24 Tonal-24 Atonal Test 9. Three to Four Note Melodies Accompanied by: One, Two, and Four Other Voices 10 Tonal-10 Atonal Test 10. Five to Seven Note Melodies Accompanied by: One, Two, and Four Other voices 10 Tonal-10 Atonal PART III--CONSTRUCTED-WRITTEN RESPONSE TO AN AURAL STIMULUS Test 11. Intervals Accompanied by One, Two, and Four Other Voices 24 Tonal-24 Atonal Test 12. Three to Four Note Melodies Accompanied by: One, Two, and Four Other Voices 10 Tonal-10 Atonal PART IV--SELECTED RESPONSE TO AN AURAL STIMULUS Test 13. Isolated Intervals 72 Tonal-72 Atonal 50 PART VF-CONSTRUCTED-VOCAL RESPONSE TO AN AURAL STIMULUS Test 14. Intervals Accompanied by One, Two, and Four Other Voices 24 Tonal-24 Atonal.1 One hundred music students at Ball State University were assigned to four treatment groups of 25 subjects each. Predicted grades, based on final course grades, were used in assigning the students to the four sections in such a manner as to avoid systematic bias. Sections one and two received entirely atonal class instruction, and sections three and four received entirely tonal class instruction. Laboratory instruction was entirely atonal in section one, mixed atonal and tonal in section two, entirely tonal in section three, and mixed tonal and atonal in section four. The complete examination was administered to all four groups prior to their instruction period and administered again at the conclusion of the experimental period. Approximately 70 of the original 100 subjects remained at the conclusion of the experimental year. Rather than statistically relate the results of all fourteen tests, just the general findings are stated here, with the exception of the final test which relates directly to the present study. All 42 analyses of covariance of the data, except one, indicated retention of the null hypothesis, set at the 1Ibid., pp. G-l through G-4. 51 .01 level displaying no significant difference between tonal and atonal instruction. These results, then, infer that instruction in atonal music will transfer to tonal music. There were also indicated tendencies and trends somewhat favoring the atonal idiom over the tonal idiom. In reference to the second hypothesis, selected responses were found to be nearly perfect measures of perception, whereas, constructed responses illustrated a requirement of considerably more learning. The attitudes toward tape-recorded self-instruction in a laboratory situation did not significantly differ between higher achieving and lower achieving students. This comparison of attitudes and opinions was measured using a specially prepared semantic differential type of instrument, the results of which produced data registering at the .01 level Of confidence. The fourteenth test, the final one of the exami- nation series, involved a constructed-vocal response to a visual stimulus. There were 48 tonal and atonal medium range intervals, selected from tests one, two, and three, randomly arranged and notated on staff paper. Both the ascending and descending forms of each of the twelve inter- vals were selected, one each from tonal and atonal contexts. The students were required to sing these intervals, after being given the initial pitch of the interval, against re- corded accompaniments of one, two, and four other voices. 52 The vocal response tests were given individually in the specially constructed vocal response booths and were recorded on magnetic tape for scoring purposes. One person corrected both the pre-test and post-test which yielded the following data: ANALYSIS OF COVARIANCE Test 14 Tonal Sources df ss s ' ms' —— x 55y S y df y .Among 3 40.82 114.57 50.40 3 16.80 Within 64 1564.95 2246.65 1103.51 63 17.52 Total 67 1605.77 2361.22 1153.91 66 F = .9592 df = 3/63 Probability = .4176 Atonal Sources .Among 3 2.51 174.72 152.21 3 50.74 Within 64 1018.73 1526.22 662.15 63 10.51 Total 67 1021.24 1700.94 814.36 6 F = 4.8272 df = 3/63 Probability = .0043 Total Sources Among 3 61.50 555.34 323.37 3 107.79 Within 64 4709.39 6554.47 2435.48 63 38.66 Tota1 67 4770.89 7109.81 2758.85 6 F = 2.7882 df = 3/63 Probability = .0477 In these analyses of covariance, the pre-test score served as the covariate, and the post-test score served as the criterion. In this test, the F ratio of the atonal test (F = 4.8272) was most significant at less than the .01 level 53 Of confidence with the tonal test registering an unsignifi- cant F = .9592 and the total F ratio 2.7882 at the .05 level of confidence. The analysis of the atonal vocal responses item was the only one of the 42 analyses of covariance pro- ducing rejection of the null hypothesis (at the .01 level) that there is no difference between tonal and atonal music instruction. The above statistical results indicate that further experimental research is needed in the area of atonal sight singing. The fact that this test was one of only two--the other being Test 13--that was restricted to intervals calls for the incorporation of melodic forms of atonal intervals studied in an atonal environment. This conclusion directly influenced the experimental design and selection of musical materials used in the present study. Pilot Study The present experimental study was administered, in a pilot situation, to students attending the four-week ses- sion of Youth Music at Michigan State University during the summer of 1970. Attending participants were high school music students who ranged from 15 to 18 years of age. Two levels of theory were taught during Youth Music: Theory I presented basic fundamentals of music structure and Theory II introduced fundamentals of four-part writing and a heavy emphasis on sight singing and ear training. The Theory II classes were comprised of students who had 54 previously had Theory I and those who, on an initial place- ment test, scored in the upper 20 percent of that test. Two sections of Theory II, which met for five 50-minute periods each week, were selected as experimental and control groups for the pilot study. Because of scheduling problems involving performance group rehearsals, true matching or random selection was not possible. Therefore, the experimental group, which met at eight in the morning, was comprised mainly of string, key- board, and voice students. The control group, which met at two in the afternoon, was comprised basically of wind and brass students. An intensified program of tonal sight singing was conducted during the first week of classes in which the main emphasis was placed on drill sheets of tonally arranged interval series. Numbers were employed as an aid to sight singing with increased interval accuracy. These sheets progressed from smaller intervals (seconds and thirds) to larger intervals (fourths and fifths) in much the same sequence as was administered to the first year college theory classes in the year of actual implementation. No key signatures were used; accidentals were written by each note requiring alteration and all were in the treble (G) clef. At the conclusion of the week of tonal sight sing- ing training, a tonal sight singing test was administered by individual appointment and recorded on tape. The test 55 progressed in difficulty from intervals of a second up to a fifth (excluding the tritone) as did the drill sheets. The results of the Tonal Sight Singing Tests yielded a mean of 40.20, standard deviation of 25.29 for experimental group. The tests administered to the control group yielded a mean of 39.78, standard deviation of 5.31, and a variance of 28.19. The drill sessions in atonal sight singing were con- ducted in a language laboratory at Michigan State University and consisted of nine sessions in much the same manner as they would be administered to the college theory classes in the coming test year. Each drill unit contained twenty-five atonally arranged Interval Series printed on drill sheets. The implementation was executed in the same manner as planned for the actual experimental period. After nine drill sessions were completed, two Atonal Sight Singing Tests were given in the laboratory in groups of five students each (widely spaced throughout the room) with their responses recorded on test tapes. The first half of the test was administered within the same environmental conditions as the drill sessions. There were twelve Inter- val Series which were the same for both the experimental and control groups with the organ accompaniment (for the experi- mental group) being the only difference. The second half of the test was administered with the contexts reversed; that is, the organ accompaniment was removed from the experimen- tal group and given to the control group. As in the previous 56 test, the Interval Series were identical for both groups. Results of the Atonal Sight Singing Test A yielded a mean of 36.9, standard deviation of 8.31 and a variance of 68.99; and Test B yielded a mean of 31.40, standard deviation of 8.22, and a variance of 67.60 for the experimental group. Results of the Atonal Sight Singing Test A yielded a mean of 32.33, standard deviation of 10.30, and a variance of 106.00; and Test B yielded a mean of 32.44, standard devia- tion of 11.62, and a variance of 135.00 for the control group. Since the situation was such that a matching of groups could not be accomplished, no statistically signif- icant differences between the experimental and control groups were realized. However, changes in results of the Atonal Sight Singing Test B when compared with Test A did indicate effects of the treatment. As the preceding data indicate, the experimental group's proficiency dropped in all but one case when the accompanying sonority was removed, which might have been due to the limited number (9) of atonal drills possible in the time allotted to the pilot study. The control group, on the other hand, showed some increase (particularly with higher scoring subjects) when the accompanying sonorities ‘were added. A certain amount of confusion was evidenced in the results of the first two or three Interval Series (in most cases) when the accompaniment was added to the control group. These results would tend to indicate that the 57 accompanying sonority does affect (to an uncertain degree) proficiency in singing atonally arranged intervals. Along with the contamination due to unmatched groups, the time limiting factor in Youth Music schedules demanded that the final tests be given in groups of five each in the language laboratory. The subjects were sufficiently spaced throughout the room to control auditory transfer between subjects as much as possible. Even so, particularly in the unaccompanied sections of the test, subjects could conceiv- ably "cue in" on another subject's response although this was not evidenced during the evaluation of performance on the test tapes. The final tests in the actual experiment were, of course, given individually in as sound-proofed a room as possible using headphones and professional quality tape recorder. The results of the pilot study with additional information through observation and a student Opinion questionnaire indicated the following changes: 1. The tempo of the response (which, in the experimen- tal group, is indicated by the tempo of the organ tones being sounded) was reduced to CD= 30 or two seconds per note; it was observed that.C)= 60 was slightly fast for accuracy Of response. 2. The control group in particular was urged to go through the complete Interval Series without repe- tition or "stepwise testing" of intervals once they commence; and correction was attempted after singing 58 the series or during the playback by the piano. This applied to both the drill sessions as well as the final tests. Both groups were urged to anticipate vocally the correct note when singing with the playback of the correct Interval Series. In this way they could more accurately check their mistakes. The length of the drill units were shortened from 25 to 20 items as fatigue and boredom were observed during the last few items on each drill session. The upper range of the Interval Series was limited to d2 (a ninth above middle C); E-flat was found by many to be a difficult note to produce. The lowest note remained Bb (below middle C). The final tests were identical with only the order of administration of the test being changed, depend- ing on which group the subject being tested belonged. Summagy The information and commentary of this chapter pre- sented the most recent studies directly concerned with the skill of sight singing as well as pertinent statements by reputable theorists and educators. Since the research of Ottman has produced a thorough anthological survey of the history Of sight singing,l repetition of that aspect was deemed unnecessary in this presentation. 1Ottman, op. cit. 59 The constant striving for accurate and efficient instruments for measurement of sight singing ability has been, and still remains, a prime concern of music theory instructors and researchers. The studies surveyed indicated a trend toward the use of several shorter items rather than one or two all-inclusive items in sight singing examination in order to allow for greater flexibility and separation of problematic factors. The tape recording of sight singing examination performances has also proven to be of aid in controlling consistency in grading and increasing the reliability of the examination. Early studies to determine related factors influ- encing sight singing ability were accomplished by use of tests which combined these factors such as tonal memory, melodic memory, melodic modulations, interval identifica- tion, interval spelling, and interval sight singing. Scores from these tests were compared with scores of other pre- viously administered tests both standardized and original. This approach does retain the "musicality" of such tests, but also allows confusion of results through interaction of the factors. Ottman's realization of this fact is shown in his instrumentation of testing the same population with the rhythmic elements extracted from the original criterion melody. More recent studies have had success by limiting investigations to just one or two elements that might in- fluence sight singing ability. 60 Technological advancements have produced devices which have been incorporated into various approaches to instruction in ear training and sight singing as well as other areas of music theory. Several researchers, spanning almost four decades, have investigated various possible uses of the tachistOSCOpe. The sophistication achieved recently in the medium of tape recording and the improved versatility of the recording machines had supplied still another vehicle for improved instruction of skills as well as aiding more accurate and widespread study. The most direct influence on the present investiga- tion were the results and conclusions drawn from a similar study, comparing melodic perception in tonal and atonal contexts, by Sherman and Hill.1 Since that study uniquely explores the skill of sight singing in an atonal environment in its final test of fourteen, much space has been devoted, in this chapter, to the presentation and discussion of that study.‘ Factors derived from certain conclusions of all of the reviewed documentations have, in varying degrees, con- tributed to the experimental design, limitations, and instruments of measurement of the present study. 1Sherman and Hill, op. cit. CHAPTER III DESIGN OF THE STUDY Sample Sixty first-year music theory students at The University of Calgary participated in this experimental study during the winter term of 1971. The majority of these subjects were enrolled as music majors. However, because of a common first-year curriculum, some students were enrolled as tentative music majors. Only two subjects were defi- nitely non-music majors--one in the experimental group and one in the control group. The subjects' ages originally ranged from 18 to 53 years; but attrition changed the upper age to 42 years with the majority ranging from 18 to 20 years of age. There were 25 male subjects and 35 female subjects at the beginning of the experimental period. An attrition of 25 percent occurred throughout the term and 17 male subjects and 28 female subjects remained at the conclusion of the 24 drill sessions. The initial sixty subjects were assigned, by come .puter scheduling, to three class sections of second term first-year music theory. Since the 9 A.M. section contained only ten students, these students voluntarily adjusted their 61 62 class schedules to attend either the 10 A.M. or 12 noon class sections. Through a toss of the coin (the majority of seven tosses) the 10 A.M. class section with an enroll- ment of 35 students was designated the experimental group, and the noon class section with an enrollment of 25 students became the control group. Permission was granted by the head of The University of Calgary music department to conduct the Atonal Sight Singing Drills as part of the regular second term laboratory instruction in music theory. The drill sessions were admin- istered in a standard language laboratory. Each drill session lasted approximately 22 minutes and was given at the beginning of two of the three weekly laboratory class periods. Winter weather conditions and illnesses were the main causes of student absenteeism. To insure a large number of subjects receiving all 24 drill sessions, make-up sessions were held in the evening once every two weeks for those subjects missing drill classes. Thirteen experimental group subjects and two control group subjects failed to complete all 24 drill units in spite of the evening make- up sessions. Drill Instrumentation The uniqueness of this experimental study demanded instruments of drill and measurement which were also some- what unique. The approach used in this study is similar to 63 the design used in Barnes' study.1 All of the musical materials for tests and drills were originals devised with the approbation and guidance of the advisors to this study. Interval Series Interval Series containing five pitches (four intervals) were notated in whole-notes in the G (treble) clef. No key signatures were used. Upward and downward inflection of natural pitches was indicated using acciden— tals before the appropriate notes. Whole-note values were used to indicate an indefinite duration. The Interval Se- ries were limited to a vocal range of Bb up to d2. Major and minor seconds and thirds and perfect fourths and fifths, both ascending and descending, were the only intervals employed in the Interval Series. Since the tritone is gen— erally considered to be one of the more difficult intervals to perceive, it was excluded from linear use in the Interval Series, but did appear in the accompanying sonorities. Atonal Sight Singing Drills Twenty-four Atonal Sight Singing Drill units, com- prised of 20 Interval Series per unit, were constructed ‘within procedural guidelines determined by the experimenter and advisors to the study. It was decided that the experi- lnental drill sessions would involve a slow vocal arpeggia- 'tion, in broken-chord fashion, of the pitches of a five-tone 1Barnes, Op. cit. 64 sonority of varying degrees of dissonance. These degrees of dissonance were determined using Hanson's "Six Basic Tonal Series." Each Unit of 20 Interval Series progressed from lesser to greater qualities of dissonance within each unit.1 The Interval Series for each unit were written first. Then the experimental variable of the accompanying sonority for each Interval Series was vertically constructed within the normal SATB range using the same Interval Series pitches. The order of difficulty of intervals resulting from Barnes' study was used in reverse order constructing the Interval Series. This was done in order to observe what effect, if any, increased drill emphasis on the more difficult inter- vals would be evidenced in the resulting hierarchy of inter— vals of this study. The Atonal Sight Singing Drill units were constructed in the following four additive levels of interval content: Unit Numbers Intervals Employed I-II ' Major 3rd Down Perfect 5th Up Perfect 5th Down Perfect 4th Down Perfect 4th Up Major 2nd Down III-VII All of the Above Plus Minor 3rd Down Major 3rd Up 1Appendix B. 65 Unit Numbers Intervals Employed VIII-XVI All of the Above Plans Minor 3rd Up Major 3rd Up XVII-XXIV All of the Above Plus Minor 2nd Up Minor 2nd Down The drill sheets for the experimental group presented the Interval Series in the top of three staves with the atonal sonority notated in treble and bass clefs in the two lower staves.l The control group drill sheets presented only the Interval Series on a single staff.2 The Interval Series for both groups were written twice to facilitate self-evaluation by circling incorrectly sung intervals. The vertical sonorities accompanying the Interval Series of the experimental group's Atonal Sight Singing Drills have been generally termed as having an atonal (or non-tonal) texture. Needless to say, there are grounds for questioning some of the sonorities as actually having some- what of a tonal center, particularly sonorities with the perfect fifth as the bottom interval. However, it was felt that the dissonant quality of the sonorities would compen- sate for that type of structuring (further elaboration on this point has been stated by Reti).3 1Appendix B. 2Appendix C. 3Rudolph Reti, Tonality Atonality Pantonality (London: Barrie and Rockliff, 1958), pp. 38-411 66 Tonal Sight Singing Drill Sheets To prepare the subjects for the pre-test of singing tonally arranged Interval Series, four Tonal Sight Singing Drill units were devised within the limitations set for the Tonal Sight Singing Test.1 Four sheets of 25 Interval Series each were constructed with graduated interval pre- sentations of seconds and thirds, then fourths and fifths. These drill units were used in regular sight singing labo- ratory classes and were not tape recorded. After the com- pletion of the Tonal Sight Singing Drills, all fall term first-year music theory students were administered the Tonal Sight Singing Test. Tape Recording of Drill Sessions As has been corroborated by several researchers cited in the preceding chapter, tape recorded programmed music instruction is not only a valid method of teaching certain music skills, but is also a means for controlling contaminating factors. Therefore, the drill units were recorded on magnetic tape and were administered to the sub- jects in a language laboratory. This environment afforded the maximum amount of uniformity in drill presentations to both the experimental and control groups within the limita- tions of time and availability of suitable facilities. 1Appendix A, pp. 103-106. 67 The Atonal Sight Singing Drill units were recorded in a two-step process using stereo recording equipment. The accompanying sonorities for the Interval Series of each unit were recorded first. A Baldwin Orgasonic electric organ was used to sustain the sonority for the performance dura- tion of the Interval Series (12 seconds per series). The clarinet stop on the "solo" rank, with no vibrato or echo, was used at "soft volume" stop setting. Pitches below the "solo" rank range were played on the pedals with a combi- nation of flute and horn stops to balance the bass to the upper tones on the "solo" rank. A metronome equipped with a flashing light was used to insure uniformity of tempo. The second and final step of the taping process was the recording of the Interval Series on the right stereo track while the pre-recorded accompanying sonorities were transferred from the original tape recorder to the left stereo track on the second tape recorder. The Interval Series were played on a Wurlitzer electronic piano which was directly connected to the second tape recorder. Two model 800 Ampex stereo tape recorders were used to record the Atonal Sight Singing Drill units. The microphone used for recording the organ was a type D-20 Akustische-U. Kinogeréte Ges. M.B.H. made in Vienna, Austria. The organ sonorities were recorded at 7% I.P.S. They were combined on the second recorder with the electric piano at 3% I.P.S. to be compatible with the language laboratory equipment (DuKane language laboratory system). 68 Measures Atonal Sight Sipging Tests The Atonal Sight Singing Tests were composed of the same materials as the Atonal Sight Singing Drill units. Atonal Sight Singing Test A was comprised of twelve Interval Series accompanied by atonal sonorities, matching the drill experienced by the experimental group. Atonal Sight Sing- ing Test B contained twelve unaccompanied Interval Series, matching the drill experienced by the control group. The Interval Series for the tests were constructed so that all intervals included in the experiment were pre- sented four times, both ascending and descending, in each test. Three basic shapes were used in the melodic con- struction of the Interval Series in both tests. 1. Ascending--Descending 2. Descending--Ascending 3. Multi-Directional. Each shape was used four times in each test to retain uni- formity. Two Interval Series were deliberately constructed 'using consecutive fourths and one series using major thirds arranged in such a fashion as to imply a whole-tone context. Irhese interval arrangements were conceived to observe their eaffects in the resulting orders of interval difficulty. Atonal Sight Singing Test A was constructed first. fmhe accompanying atonal sonorities were structured so that 25 of the tones were within less than an octave of the 69 written Interval Series, and 23 tones were an octave or more below the Interval Series allowing observation of any effects of octave displacement. As in the drill units, the accompanied Interval Series were presented in gradual order of minimum to maximum dissonance quality.1 Originally, Atonal Sight Singing Test B was con- structed in the same manner as Test A but with Interval Series unrelated to Test A. It also contained the same equal number (4) of all included intervals. Tests A and B were tested for reliability using the test-retest method with a sample of 21 subjects. Test A displayed good reliability, but Test B did not. Consequently, a revision of Test B was constructed by inverting all but two of the Interval Series of Test A, and beginning them on appropriate pitch levels to fit the prescribed range limitations. The fourth and sixth Interval Series containing the consecutive fourths were retrogrades of those in Test A. The presenta- tion order of the Interval Series in Test A was retained in Test B for uniformity.2 After revision, Atonal Sight Singing Test A dis- played a reliability of .83 and Test B a reliability of .86. Internal validity was claimed for both tests due to the nature of content. 1Appendix A. pp. 116-117. 2Appendix A, pp. 118-119. 70 Atonal Sight Singing Test B was of particular importance since results from that test would indicate acceptance or rejection of Hypothesis 1. The true test of the experimental drill's effectiveness was the compared performance of the experimental group with that of the con- trol group on the test without the sonority accompaniment factor. Tonal Sight Singing‘Test The instrument of measurement for increased ability in sight singing tonally arranged intervals through practice in atonal sight singing was the Tonal Sight Singing Test.1 This test was used both as a pre-test and post-test. Since the subjects had received tonal sight singing training during music theory coursework in the fall term, this test was designed to measure subjects' ability to sing tonally arranged Interval Series notated without a key signature. The inclusion of all intervals to be investigated in equal numbers was unnecessary in the construction of the Tonal Sight Singing Test since the criteria for comparison ‘was Barnes' order of difficulty of intervals.2 However, the presentation of the Interval Series did progress from con- taining just seconds and thirds through gradual inclusion of fourths then fifths. The same vocal ranges, notation, clef, and number of Interval Series were used in the same manner 1Appendix A. p. 107. 2Barnes, op. cit. 71 as employed in the Atonal Sight Singing Tests. The test was recorded on magnetic tape using the same procedures followed in the recording of Atonal Sight Singing Test B. During the fall term, reliability of the Tonal Sight Singing Test was determined by administering it to 21 second-year music students at The University of Calgary. The test-retest method was used with the retest occurring after a time lapse of a week. The tests were given to groups of 5 or 6 students at a time widely dispersed in a large language laboratory. Comparison of the test scores with the retest scores using the Pearson Product-Moment Correlation indicated a high reliability of .8032.1 Internal validity was claimed for the test in view of the nature of its content. Design Testing of the two main hypotheses of this study required two different experimental designs. Both designs ‘were based on the use of an experimental group and a control group; but one hypothesis was related to an unfamiliar skill whereas the other hypothesis was related to increased performance ability of a familiar skill. Therefore, the investigation of the ability to sight sing atonally arranged series of intervals required the use of a "Posttest-Only 1Appendix A. P. 120. 72 Control Group Design."1 The investigation of increased ability to sight sing tonally arranged series of intervals through atonal drill called for the standard "Pretest- Posttest Control Group Design."2 The Atonal Sight Singing Drills were accomplished in a language laboratory. Although obviously not a com- pletely foolproof environment, the subjects' use Of head- phones in partitioned stalls did aid their concentration on the tape recorded drill units. It also helped control the factor of students "cueing in" on each other's vocal responses. A set of specific directions, occasionally repeated by the instructor, was given to the subjects which included the drill procedures to be followed by each group.3 The experimental group proceeded in the following manner for each cell pair of Interval Series: 1. (a) The accompanying vertical sonority is sounded on the organ with the lowest tone sustained initially and the remaining four tones added at the suggested performance tempo of O = 30 for each subsequent tone. (b) The initial tone of the Interval Series is sounded by the electric piano at the same performance tempo 1Donald T. Campbell and Julian C. Stanley, Egperimental and Quasi-Expgrimental Designs for Research (Chicago: Rand McNally & Co., 1969), pp. 25-31. 2Ibid., pp. 13-22. 3Appendix A, pp. 108-111. 4. 5. 6. 73 while the sonority remains sustained by the organ. The students vocally repeat the initial tone sounded by the electric piano and sing the remaining four pitches of the Interval Series at the suggested tempo implied by the organ ((3 = 30). The sonority is sustained for 12 seconds. After a four-second delay, the organ sounds the sonority (all five pitches simultaneously) and the Interval Series is performed in the established tempo by the electric piano. The students are urged to vocally anticipate the piano-performed tones to check their initial accuracy and circle incorrectly performed intervals as they proceed. Exact repetition of procedures in Step 1. Exact repetition of procedures in Step 2. Exact repetition of procedures in Step 3. The drill procedures for the control group were similar to those of the experimental group. The only difference between the two groups were the references to the organ accompaniments. Because the organ did not set the perfor- mance tempo for the control group, the laboratory instructor verbally directed them to perform the Interval Series at a very slow tempo (approximately 0 = 30). The Atonal Sight Singing Drills were conceived within the framework of the classic stimulus-response design illustrated by Figures 1 and 2. 74 .Qnonm Heunmaenemxo How nmflmmo unefiueoua .H munmwm .oeonnOm .ooonnom mownmm mmflHem He>nmunH He>HeunH mo onou mo mnou HeHanH HeHuAnH .omnaeumnm .oenfleumnm A mnemonom one aufluonom one n mnflmnemEOooe oonflepmnm mnflonmome mnwmnemeoooe omnflepmnm mnfionmome m nuw3 on 0» omonnOm nuflz on O» omonnOm D xoeo ommeam ma wonnflunoo anemonOm Moeo ooweam we monnflunoo wuflnonOm n meflumm He>neunH anemonom mo menoa mmflnom He>HounH wufluonom mo monoa hauomuuoonfl mauooHHoonfl omEHOmumm omEHomHOm A mae>uouna mHOHHO om nunv muwuonOm mHe>HmunH mHOHwO cm H nv muwnonOm n one xoenmeam ue mmfluem Henoue one xoenmeam we mmwumm Henoue D mnHuemHOHune He>nounH Hm>o mnfluemHOHune He>HeunH He>o m mmwnem He>nmunH mnflm mmwumm mmflumm He>uwunH onwm mmwuem H mnam muomflonm muoennnm He>umunH mnflm muomnnnm mpoomonm He>nmunH > nowuenae>m one emnommom .m HHnEHpm .v nowuenae>m one omnommmm .N Haneflum .H unmfimonomnamm .m unefieonomnflmm .m 75 .mnonm Aouunoo “Om nowmmo unmfiuemue .m OHDDHD oeonnOm ooonnom A meflumm mOAuom n Ae>nmunH Ae>umunH m xoen ommeam mA mo onou xoeo oeheam we mo mnou D meanem Ae>umunH AeAuAnH mefiumm Ae>neunH AefluAnH 4 hauomuuoonw hauomuuoonw omfinomnem mmeum omEHOmHmm wmeum A mAe>HOunA oaouwo om "AU MOAO mAe>HounA mAOHAO on u mean A one xoenmeam we mmfluem voeHD one xoeomeam De mefluem wagon» D mnAuemAOAune He>nounH mamnfim no mnAuemAOAune Ae>nounH OAmnAm no m meflumm Ae>HmunH mnwm mmwumm mmflumm Ae>HmunH mnfim mOAHmm H mnflm muoennnm mDOOnnnm Ae>nounH mnwm mpoenonm mavennnm Ae>nounH > noAuenAe>m one mmnommem .m AAnEADm .e nOAuenAe>m one mmnommem .m AADEAum .A #COETUHOM GHQ“ . 0 unmEmOHOMnAom .m 76 At the conclusion of the 24 Atonal Sight Singing Drill sessions, Atonal Sight Singing Tests A and B were administered to each subject individually. Only the in- structor and the subject were present in the room which was reasonably free from outside sound. The subject and instructor were situated so that there would be no visual contact during the period of testing. The tests, which were recorded on tape, were given using high quality Koss stereo headphones which also helped control any distracting outside sounds. The experimental group subjects were given Atonal Sight Singing Test A first, immediately followed by Test B. This was done in order to first test the subject in the environment familiar to him through the related drill expe- rience before switching to the environment on Test B. The control group subjects were given the tests in reverse order for the same reason, performing Test B first followed by Test A. Two tape recorded versions of the tests were made—- a version for each group--with the directions for taking each test.1 Both Atonal Sight Singing Test A and B were cali- brated in the same manner as the drill units using the stimulus-response order of presentation, but eliminating its repetition. Therefore, procedures of Test A related to blocks one through three of the experimental group drill 1Appendix A, pp. 112-115. 77 design diagram and Test B related to blocks one through three of the control group drill design (Figures 1 and 2).1 The Tonal Sight Singing Test, used as both pre-test and post-test, was administered to all sections of first- year music theory at the close of the fall term. The test was given individually as part of each student's final sight singing examination appointment. The same test was again individually administered as a post-test to the experimental and control group subjects at the conclusion of the 24 Atonal Sight Singing Drill sessions. The Tonal Sight Sing- ing Test was also recorded on tape following procedures similar to those of Atonal Sight Singing Test B. The tape recorded performances of the subjects accomplishing Atonal Sight Singing Tests A and B and the Tonal Sight Singing Tests were scored by the instructor. The resulting scores were checked for accuaracy through tape auditiOn by two other music theory faculty members at The University of Calgary. The performance scores were based on correct or incorrect singing of the individual intervals of the Interval Series rather than performance of the Inter- val Series as a whole. Allowance was made for a reasonable amount of intonation variance recognizing problems of voice control in students lacking vocal experience. 1Diagram, pp. 74-75. 78 Testable Hypotheses This study was designed to test two primary hypotheses and also two secondary hypotheses related to atonal sight singing performance factors. To succinctly state the primary hypothesis, the training experienced by the experimental group (accompanied atonal sight singing drill) is referred to as Method I; and the training expe- rienced by the control group (unaccompanied drill) is re- ferred to as Method II. All of the hypotheses are stated in the following null hypothesis form: Hypothesis 1: There is no difference in the ability to sight sing atonal Interval Series between students trained by Method I and Method II. Hypothesis 2: Progress in ability to sight sing atonal Interval Series based on sight singing drill of atonal Interval Series will not significantly affect progress in ability to sight sing tonal Interval Series. Of secondary interest, related to Hypothesis 1, are the following hypotheses: Hypothesis 3: There is no significant difference between males and females in the ability to sight sing atonal Interval Series. 79 Hypothesis 4: There is no significant difference between persons with keyboard experience and persons without key- board experience in the ability to sight sing atonal Interval Series. Analysis Scores from Atonal Sight Singing Tests A and B were transferred to mark sense scoring sheets and were processed by The University of Calgary Data Center. The mean, standard deviation, variance, mean item difficulty, and item choice distributions and percentages for both tests were furnished by this process. The resulting scores of the Atonal Sight Singing Tests were transferred to computer cards in preparation for statistical analysis by computer. A one-way design analysis of variance was used in comparing the results of the control and experimental groups on both Tests A and B. Further investigations into factors affecting atonal sight singing performance were accomplished through use of the p test and a two-way analysis of variance. The resulting scores of the Tonal Sight Singing Tests were also statistically analyzed by computer using the one-way design analysis of variance. CHAPTER IV PRESENTATION OF THE DATA Review of Procedure The investigation of proficiency in sight singing series of atonally arranged intervals with accompanying sonorities constituted the main purpose of this study. Related to this was the investigation of increased ability to sight sing tonally arranged series of intervals through drill in sight singing atonally arranged intervals. Sixty first-year music theory students participated in the experimental period which occurred during the winter term of 1971 at The University of Calgary. All sixty stu- dents had been given the Tonal Sight Singing Test as a pre- test at the close of the fall term. At the completion of the experimental drill period, the remaining 45 eligible subjects of the experimental and control groups were given Atonal Sight Singing Test A and B and the Tonal Sight Singing Test (post-test). The resulting test scores were recorded on computer cards and prepared for processing at the Data Center of The University of Calgary. A one-way analysis of variance was used to determine the acceptance or rejection of the two 80 81 primary hypotheses. Closer inspection of the factors of keyboard experience and sex was accomplished using a two- way analysis of variance of fixed factors and a small sample 3 test on resulting test data. Hypotheses Hypothesis 1 There is no difference in the ability to sight sing atonal Interval Series between students trained by Method I and Method II.1 The obtained F values for both Atonal Sight Singing Tests A and B indicated acceptance of the null hypothesis. The results are shown in Tables 1 and 2. TABLE 1 ANALYSIS OF VARIANCE SUMMARY BETWEEN THE EXPERIMENTAL GROUP AND CONTROL GROUP ON ATONAL SIGHT SINGING TEST A Source SS df MS F Between groups 22.33 1 22.33 0.51 Within groups 1899.32 43 44.17 Totals 1921.65 44 lSee "Testable Hypotheses," p. 78. 82 TABLE 2 ANALYSIS OF VARIANCE SUMMARY BETWEEN THE EXPERIMENTAL GROUP AND CONTROL GROUP ON ATONAL SIGHT SINGING TEST B Source SS df MS F Between groups 163.59 1 163.59 3.02 Within groups 2332.18 43 54.24 Totals 2495.77 44 Hypothesis 2 Progress in ability to sight sing atonal Interval Series based on sight singing drill of Atonal Interval Series will not significantly affect progress in ability to sight sing tonal Inter- val Series. The F values indicate rejection of the null hypoth- esis for both the experimental group and the control group. The results are shown in Tables 3 and 4. TABLE 3 ANALYSIS OF VARIANCE SUMMARY BETWEEN THE TONAL SIGHT SINGING TESTS (PRE-TEST AND POST-TEST) OF THE EXPERIMENTAL GROUP Source SS df MS F Between groups 491.11 1 491.11 9.93* Within groups 2076.77 42 49.45 Totals 2567.88 43 *Significant at the .01 level of confidence. 83 TABLE 4 ANALYSIS OF VARIANCE SUMMARY BETWEEN THE ATONAL SIGHT SINGING TESTS (PRE-TEST AND POST-TEST) OF THE CONTROL GROUP Source SS df MS F Between groups 515.56 1 515.56 13.49* Within groups 1681.39 44 38.21 Totals 2196.95 45 *Significant at the .01 level of confidence. A E_test was used to determine the possible exis- tence of any significant difference between experimental and control group progress in the ability to sight sing tonal Interval Series. The obtained E value of .289 indicated no such difference existed. Hypothesis 3 There is no significant difference between males and females in the ability to sight sing atonal Interval Series. A‘p test for small samples of subjects was used to determine whether or not there existed a significant differ- ence in performance scores on Atonal Sight Singing Tests A and B between males and females of both groups.1 Visual inspection of the data indicated that analysis of variance 1Appendix A, pp. 132-133. 84 would show little difference between males and females; therefore, this procedure was omitted. The null hypothesis was accepted as shown in Table 5. TABLE 5 SUMMARY OF E TEST VALUES BETWEEN PERFORMANCE SCORES OF MALES AND FEMALES IN BOTH STUDY GROUPS ON ATONAL SIGHT SINGING TESTS A AND B Group & Factor vs. Group & Factor Test A Test B Exp. Male Exp. Female = .421 t = .210 Exp. Male Con. Female = .278 t = .991 Con. Male Con. Female = .278 t = .281 Con. Male Exp. Female = .554 t = 1.220 Exp. Male Con. Male = .018 t = .743 Exp. Female Con. Female t = .866 t = 1.570 Hypothesis 4 There is no significant difference between persons with keyboard experience and persons without key- board experience in the ability to sight sing atonal Interval Series. A fixed factor analysis of variance was performed on the resulting scores of experimental and control group sub- jects with and without keyboard experience on Atonal Sight Singing Test B. The Atonal Sight Singing Tests were de- signed so that the comparative results of the experimental and control groups on Test B would indicate the effectiveness 85 of the experimental drill treatment. The F values obtained were not sufficiently high enough to be considered sig- nificant and therefore indicated acceptance of the null hypothesis. However, the F value indicating interaction reached near significance at .05 (p < .07) and called for further inspection of test results as shown in Table 6. TABLE 6 SUMMARY OF ANALYSIS OF VARIANCE OF FIXED FACTORS BETWEEN RESULTING SCORES OF SUBJECTS WITH AND WITHOUT KEYBOARD EXPERIENCE ON ATONAL SIGHT SINGING TEST B Source SS df MS F Keyboard factor 2.35 l 2.35 .04 Group factor 159.71 1 159.71 3.03 Interaction 4 213.48 1 213.48 4.05 Within 2160.77 41 52.70 A.p test for small samples of subjects was used to determine whether or not there existed a significant differ- ence in resulting scores of Atonal Sight Singing Tests A and B between experimental and control group subjects with and without keyboard experience.1 The obtained 3 values from the compared results of experimental and control group subjects with keyboard experience determined the sources 1Appendix A, pp. 134-135. 86 of interaction indicated by the analysis of variance. A summary of those particular E tests are shown in Table 7. TABLE 7 SUMMARY OF E_TEST DATA BETWEEN RESULTING SCORES OF SUBJECTS WITH KEYBOARD EXPERIENCE ON ATONAL SIGHT SINGING TESTS A AND B Group & Factor Mean S.D. df t Test Exp. Keybd. 39.08 6.60 Con. Keybd. 33.09 4.82 22 2.38* A Exp. Keybd. 38.15 6.45 Con. Keybd. . 30.27 6.83 22 2.80** B *Significant at .05 level of confidence. **Significant at .02 level of confidence. The obtained £_value from the compared scores of control group subjects with and without keyboard experience on Test A also proved to be significant at the .05 level of confidence as shown in Table 8. 87 TABLE 8 SUMMARY OF‘E TEST DATA BETWEEN RESULTING SCORES OF CONTROL GROUP SUBJECTS WITH AND WITHOUT KEYBOARD EXPERIENCE ON ATONAL SIGHT SINGING TEST A Group & Factor Mean S.D. df t Test Con. Keybd. 33.09 4.82 Con. No Keybd. 38.67 5.47 21 2.49* A *Significant at .05 level of confidence. Related Data The Atonal Sight Singing Tests were devised so that all intervals, both ascending and descending, included in the limitations of this study would be presented in equal numbers to render tables of interval hierarchy.1 Figure 3 contains the orders of interval difficulty for each test and the combined hierarchy of both tests. The interval position changes from one test to another are also diagrammed. Atonal Sight Singing Test A.was constructed so that accompanying sonority pitches would be approximately evenly divided between pitches within an octave (25) of the In- terval Series and pitches an octave or more below (23) the Interval Series. Table 8 shows the resulting averages of correctly sung interval pitches displaced within the sonority an octave or more below the Interval Series. 1Appendix A, p. 136. 88 Bppk_ Test A & B 1 Mi. 2 D 3 Ma. 2 U 4 Ma. 2 D 5 P. 4 U 6 Mi. 3 U 7 Ma. 3 U 8 P. 5 D 9 “Np. 4 D 10 P. 5 U 11 Mi. 3 D 12 Ma. 3 D ———— indicates general movement. ---- indicates movement possibly related to whole-tone arr. ---- indicates movement possibly related to consecutive fourth arr. D indicates downward direction of interval. U indicates upward direction of interval. Figure 3. Hierarchy of intervals resulting from both groups' performance of Atonal Sight Singing Tests A and B. (Interval rank order from least to most difficult.) 89 The summary also shows the resulting percentages of male and female subjects of both groups. TABLE 9 SUMMARY OF CORRECTLY SUNG INTERVAL PITCH AVERAGES OF MALE AND FEMALE SUBJECTS OF BOTH GROUPS ON ATONAL SIGHT SINGING TEST A Number Correct Interval Group Sex of Subjects Averagea Experimental Male 8 17.37 Experimental Female 14 17.07 Experimental Both 22 17.18 Control Male 9 16.88 Control Female 14 16.57 Control Both 23 16.69 Both Both 45 16.93 aAverage of correctly sung interval pitches 8va or more below Interval Series pitch. The results indicated that octave displacement of tones within the sonority made no significant difference between males and females nor between groups. CHAPTER V SUMMARY AND CONCLUSIONS Summapy The primary purpose of this study was the investi- gation of proficiency in sight singing series of atonally arranged intervals accompanied by atonal sonorities. An additional related investigation of increased ability to sight sing tonally arranged intervals through atonal sight singing drill was also performed as part of the experiment. Sixty first-year music theory students at The Uni- versity of Calgary participated in this experiment during the winter term of 1971. The students had been given the Tonal Sight Singing Test, used as a pre-test and post-test, as part of the final sight singing examination of the fall term. Twenty-four Atonal Sight Singing Drill units were administered to two music theory laboratory class sections randomly designated as experimental and control groups. The drills were tape recorded and the drill sessions took place in a large language laboratory. At the completion of the 24 drills, the remaining 22 experimental subjects and 23 con— trol subjects who had experienced all 24 drills were given 90 91 Atonal Sight Singing Tests A and B and the Tonal Sight Singing Test (post-test). Atonal Sight Singing Test A corresponded in nature of content and presentation to the Atonal Sight Singing Drills given to the experimental group. It contained twelve atonal Interval Series presented over sustained sonorities comprised of vertical arrangements of the Interval Series pitches. Atonal Sight Singing Test B corresponded to the Atonal Sight Singing Drills given to the control group. It, like Test A, contained twelve atonal Interval Series but without sonority accompaniment. Each subject taking the Atonal Sight Singing Tests was first given the test that corresponded to his drill experience followed immediately by the remaining test. The tests, pre—recorded on tape, were given to each subject individually with the performances recorded on a second tape recorder for later evaluation. The Tonal Sight Singing Test, which had been given as a pre-test, was given individually to subjects of both groups as a post-test. It was administered in the same manner as the Atonal Sight Singing Tests. Its content was similar to Atonal Sight Singing Test B except that the Interval Series were tonal. Conclusions Two main hypotheses and two secondary hypotheses, stated in null form, were tested for statistical 92 significance producing the following research results and conclusions: Hypothesis 1 There is no difference in the ability to sight sing atonal Interval Series between students trained by Method I and Method 11.1 Accepted. Atonally accompanied sight singing drill does not significantly increase the subject's ability to sight sing atonal Interval Series over that of subjects drilled in unaccompanied atonal sight singing. Hypothesis 2 Progress in ability to sight sing atonal Interval Series based on sight singing drill of atonal Interval Series will not significantly affect progress in ability to sight sing tonal Interval Series. Rejected. Sight singing drill of atonal Interval Series will significantly increase the subject's ability to sight sing tonal Interval Series. Hypothesis 3 There is no significant difference between males and females in the ability to sight sing atonal Interval Series. Accepted. Sex makes no difference in a subject's ability to sight sing tonal Interval Series. Hypothesis 4 There is no significant difference between persons with keyboard experience and persons without key- board experience in the ability to sight sing atonal Interval Series. Accepted. 1See "Testable Hypotheses," p. 78. 93 Subjects with keyboard experience display no greater ability in sight singing atonal Interval Series than do subjects without keyboard experience. Discussion The main hypothesis and instrumentation of this experiment were based on the assumption that unaccompanied drill in atonal sight singing is superior to accompanied drill in teaching the skill of atonal sight singing. Although resulting data of this experiment indicated acceptance of that hypothesis, the F value yielded by the analysis of variance on Atonal Sight Singing Test B implies that accompanied atonal sight singing drill, although not more effective, is as effective a method of drill as is unaccompanied drill. The increase of scores on Atonal Sight Singing Test A over scores on Test B by the control group subjects was an additional implication of the effectiveness of accom- panied atonal sight singing drill. Eighteen of the twenty- three control group subjects scored higher on Test A than on Test B. Only five of the twenty—two experimental group subjects registered a drop in score from Test A to Test B. Although Tests A and B were environmentally different, the similarity and uniformity of interval presentation between the tests might justify such a comparison. Confusion, resulting from the change of context from that to which the subject was accustomed in the drills to 94 that of the opposite group, was evidenced in the subjects' performances on the Atonal Sight Singing Tests. This effect was particularly noted in performances by control group subjects in the initial Interval Series on Test A. The secondary investigations related to Hypotheses 3 and 4 compared scores of experimental and control group sub- jects categorized according to the investigated factors. The sex factor was investigated mainly from the standpoint of possible performance differences between males and females relative to octave displacement in the accompanying sonorities of Atonal Sight Singing Test A. Neither the sex factor nor octave displacement displayed any notable effect on the results. However, keyboard experience proved to be a significant factor in this group of subjects. The high F value indicating interaction in the analysis of variance of fixed factors performed on scores of subjects with and without keyboard experience called for further statistical investigation. A‘p test for small samples of subjects was used to compare mean scores within and between both study groups with and without keyboard experience. The p values shown in Table 7 attest to the significant performance of experimental group subjects over control group subjects, both with keyboard experience, on Atonal Sight Singing Test A and B. However, the p value shown in Table 8 indicated that the control group subjects without keyboard experience performed significantly better 95 than the control group subjects with keyboard experience thus exposing the source of the indicated interaction. The factor of singing experience was also consid- ered for statistical comparison; but it was found that virtually the same subjects with keyboard experience were also those with prior vocal experience. Consequently, further investigation of the vocal experience factor was unnecessary. The hierarchy of intervals resulting from the Atonal Sight Singing Tests displayed rankings which may have been influenced by the experimental treatment. The deliberate use of consecutive fourths in the same direction, in Inter— val Series four and six of both tests, seemed to influence their changes of position from Test A to Test B. The per- fect fourth upwards in Test A.was sung correctly 84 percent of the time; but with the sonority removed in Test B, it was sung correctly only 75 percent of the time moving it from third to fifth (less to more difficult) position. The per- fect fourth downward, shown to be more difficult, moved from tenth position in Test A to eighth position in Test B, but its percentage remained basically the same (approximately 66 percent correctly sung). The whole-tone arrangement of major thirds, which occurred only in the eleventh Interval Series on both tests, did not seem to display major position changes. However, the major third down, sung correctly 64.5 percent on Test A, 96 moved from eleventh position to the most difficult interval position (twelfth) on Test B with 52.25 percent. A general comparison of the hierarchy of intervals of this study with that obtained by Barnes shows that the interval rankings are quite similar with the exception Of the minor third downward. This interval consistently re- tained its position as being quite difficult in both tests, whereas Barnes' findings rank it in the lower half among the less difficult intervals. This might be indicative of its use in an atonal melodic arrangement as compared with its frequent triadic use in tonal melodic arrangements. Such questions emphasize the need for further research comparing influences of tonal and atonal contexts on melodic material. An observation that seemed to favor the experimental treatment was the greater attention span of the experimental group than that of the control group during the Atonal Sight Singing Drills. An obvious reason for this was the novelty of fitting the Interval Series pitches into the sustained sonority. The control group, lacking the sonority factor, was left to its own devices for singing the Interval Series other than suggestions included in the directions for drill. The Tonal Sight Singing Test as well as the Tonal Sight Singing Drills emphasized the key signature factor in sight singing performance. The traditional conditioning of musicians emphasizes reliance upon knowledge of the tonal center through use of the key signature. When this factor is removed from music and only accidentals indicating pitch 97 inflection are used before appropriate notes, determination of the tonal center becomes much more difficult for persons so conditioned. The rejection of Hypothesis 2 indicates that either method of atonal sight singing drill used in this study is a valid aid toward increased ability to sight sing tonal music without key signatures. Suggestions for Future Research 1. A replication of this study with modification of the Interval Series and inclusion of all intervals within an octave might produce additional useful information regarding Hypothesis 1. 2. A study similar to this design but using tonal sonorities and Interval Series would further investigate the factor of tonal center determination related to key signature in sight singing. 3. Experimental testing of large populations in sight singing atonal melodies with atonal accompaniments containing melodic line pitches compared with atonal ac- companiments not containing melodic line pitches would investigate the affective nature of such sonorities. The general lack of research in atonal musical practices leading to innovative pedagogical approaches sub- stantiates the need for such research. Such experimental research can lead to more effective teaching of the skills which aid in the performance, analysis, and understanding of twentieth century music. BIBLIOGRAPHY BIBLIOGRAPHY Books Apel, Willi. Harvard Dictionary of Music. Cambridge, Massachusetts: Harvard University Press, 1962. Campbell, Donald T., and Julian C. Stanley. Experimental and Quasi-Experimental Designs for Research. Chicago: Rand McNalIy and Co., 1969. Dallin, Leon. Techniques of Twentieth Century Composition. Dubuque, Iowa: Wm. C. Brown Co., 1964. Edlund, Lars. Modus Novus. Stockholm: Nordiska Musikarlaget, 1963. Farnsworth, Paul R. The Social Psychology of Music. Ames, Iowa: The Iowa State University Press, 1969. Hanson, Howard. Harmonic Materigls of Modern Music. New York: Appleton-Century-Crofts, Inc., 1960. Harder, Paul O. Harmonic Materials in Tonal Music. Boston: Allyn and Bacon, Inc., 1968. Hays, William L. Statistics. New York: Holt, Rinehart and Winston Co., 1963. Hillbrand, Earl K. Measuring Ability in Sight Singing. Ann Arbor, Michigan: Edwards Brothers Pub., 1924. Hindemith, Paul. Elementary Training for Musicians. New York: Associated Music Publishers, 1949. Horacek, Leo, and Gerald Lefkoff. Programmed Ear Training. New York: Harcourt, Brace & World, 1970. Reti, Rudolph. Tonality Atonality Pantonality. London: Barrie and Rockliff, 1958. Thomson, William. Advanced_Music Reading. Belmont, California: Wadsworth Publishing Co., 1969. 98 99 Wedge, George A. Advanced Ear Training and Sight Singing. New York: G. Schirmer, Inc., 1922. Periodicals Bean Kenneth L. "An Experimental Approach to the Reading of Music." Psyphological Monographs, L (1938), 1-79. Buttram, Joe B. "The Influence of Selected Factors on Interval Identification." Journal of Research in Music Education, Fall, 1969, pp. 305-15. Dean, Charles D. "Predicting Sight Singing Ability in Teacher Education." Journal of Educational Psychology, XXVIII (November, 1937), 601-608. Hammer, Harry. "An Experimental Study of the Use of the Tachistosc0pe in the Teaching of Melodic Sight Singing." Journal of Research in Music Education, XI (1963), 4 4" 430 Ihrke, Walter R. "Automated Music Training." Journal of‘ Research in Music Education, XI, No. 1 (Spring, 1963), 3-20. Kunkle, Robert F. "A Direct Approach to Sight Singing." Music Educators Journal, June-July, 1950, p. 41. McNaught, W. G. "The Psychology of Sightsinging." Proceedings of the Musical Association, XXVI (1900), 35-55. Ortmann, Otto. "Some Tonal Determinants of Melodic Memory." Journal of Educational Psychology, XXIV (September, 1933), 454-56. Salisbury, Frank S., and Harold B. Smith. "Prognosis of Sight Singing Ability of Normal School Students." Journal of Applied Psychology, XIII (1929), 425-39. Dissertations Barnes, James W. "An Experimental Study of Interval Drill as It Affects Sight Singing Skill." Unpublished Ph.D. dissertation, Indiana University, 1960. Christ, William E. "The Reading of Rhythm Notation Approached Experimentally According to Techniques and Principles of Word Reading." Unpublished Ph.D. dissertation, Indiana University Music Library, 1953. 100 Helbling, DeVon W. "An Experimental Study of the Relative Effectiveness of 'Whole' and 'Part' Methods of Teaching Sight Singing." Unpublished Ph.D. dissertation, Indiana University, 1965. Justus, Lane D. "Evaluation of an Innovative Instructional Design for Sight Singing." Unpublished Ph.D. disserta- tion, University of Arizona, 1970. Kanable, Betty M. "An Experimental Study Comparing Pro- grammed Instruction with Classroom Teaching of Sight Singing." Unpublished Ph.D. dissertation, Northwestern University, 1964. Marquis, James H. "A Study of Interval Problems in Sight- singing Performance with Consideration of the Effect of Context." Unpublished Ph.D. dissertation, State University of Iowa, 1963. Nelson, John C. "A Comparison of Two Methods of Measuring Achievement in Sight Singing." Unpublished Ph.D. dissertation, The University of Iowa. Ottman, Robert W. "A Statistical Investigation of the Influence of Selected Factors on the Skill of Sight- singing." Unpublished Ph.D. dissertation, North Texas State College, 1956. Ray, Harry B. "An Experimental Approach to the Reading of Pitch Notation." Unpublished Ph.D. dissertation, Indiana University, 1964. Smith, James C., Jr. "A Performance Test of Kanable's 'A Program for Self-Instruction in Sight Singing.'" Unpublished Ph.D. dissertation, The Florida State University, 1968. Spohn, Charles L., Jr. "An Exploration in the Use of Recorded Teaching Material to Develop Aural Compre- hension in College Music Classes." Unpublished Ph.D. dissertation, Ohio State University, 1959. Stokes, Charles F. "An Experimental Study of Tachistoscopic Training in Reading Music." Unpublished Ph.D. disserta- tion, Teachers College, University of Cincinnati, 1944. Tucker, Gerald L. "The Influence of Isolated Phythmic Drill on Growth in Sight Singing." Unpublished Ph.D. disser- tation, The University of Oklahoma, 1969. Vorce, Frederick W., Jr. "The Effect of Simultaneous Stimulus on Vocal Pitch Accuracy." Unpublished Ph.D. dissertation, The Florida State University, 1964. 101 Wilcox, Eunice A. "The Effects on Sight Singing of Voice Class Instruction Utilizing Variants of Traditional Vocalises." Unpublished Ph.D. dissertation, Michigan State University, 1968. Special Studies Ihrke, Walter R. Automated Rhythm Training. Council for Research in Music Education, Bulletin No. 7, Spring, 1966, pp. 35-37. Sherman, Robert W., and Robert E. Hill, Jr. Aural and Visual Percgptipn of Melody in Tonal and Atonal Musical Environments. Final Report of Project No. 2413, Office of EducatiOn, U.S. Department of Health, Education, and Welfare, Contract No. OE-4-10-l77, Muncie, Indiana: Ball State University, February, 1967. Spohn, Charles L., and William Poland. An Evaluation of Two Methods Using Magnetic Tape Recordings for Programmed Instructiop in the Elemental Materials of Music. Final Report of Title VII, Project No. 876, National Defense Education Act of 1958, Grant No. 7-34-0430-172, Columbus, Ohio: The Ohio State University, November, 1963. Miscellaneous Cookson, Frank B. Recordings and Self Tutoring. Cleveland: The Brush DevelOpment Co., 1949. / Mosher, Raymond M. A Study of the Group Method of Measure- ment of Sight Singipg. New York: Teachers College, Columbia University, Contributions to Education, No. 194, 1925. APPENDIX A TEST AND DRILL MATERIALS AND RAW DATA 102 ATONAL SIGHT SINGING EXPERIMENT SUBJECT QUESTIONNAIRE Name Sex (male) (female) University year level Class (9:00)(10:00)(12:00) Performance major Years of study ARCT or Western Regional grade level Performance minor Years of study ARCT or Western Regional grade level Keyboard experience (other than above); if so, years of study Formal ear training or sight singing experience before university level? If so, explain briefly the situation in which you received such experience. Subject's signature DRILL SESSION ATTENDANCE RECORD UNITS: I II III IV V VI VII VIII IX X XI XII XIII XIV XV XVI XVII XVIII XIX XX XXI XXII XXIII XXIV (circled units indicate absences) TEST RESULTS ATONAL SIGHT SINGING TEST A Score: ATONAL SIGHT SINGING TEST E Score: TONAL SIGHT SINGING TEST A Score: TONAL SIGHT SINGING TEST B Score: IO} TONAL SIGHT SINGING DRILLS SHEET I IIIII 104 row SIGHT SINGING DRILLS SHEET I I IIIII .IAIA AAAAA \\\\..‘ \\\.\\ \\\\\ \\\\\ \\\\\ 105 TONAL SIGHT SINGING DBILIS SHEET Irif IIIII 106 TONAL SIGHT SINGING DRILLS SHEET IV Name 107 TONAL SIGHT SINGING TEST 7o 1. ' J J j I A .n I “ - - .lo- .__. I "“u' - ' -' . .n'J-w—I_Ha -x--I \'_.m 19_'.L‘ - o—r—I . - ' ‘ q. ’00 "c ‘1, I I i A __A 1 I I__1 . a r 1s .71 D I v Aur -- I'm 1..- [I . - m- _ U 108 ATONAL SIGHT SINGING DRILLS DIRECTIONS FOR EXPERIMENTAL SUBJECTS Drill sheets with 20 cells of atonal Interval Series will be distributed to you at the beginning of each labora- tory drill session. Sign your name and today's date at the tOp of the first page. Put on your headsets and adjust them so that they comfortably fit your ears. Adjust the knob labeled "program" to hear the recorded drills. No other controls require any adjustment. Each Interval Series to be sung is presented twice within each cell. The Interval Series are written in whole- note notation, without key signatures, and appear on the tOp staff of each cell. The bottom two staves contain the same pitches as are in the Interval Series but arranged in a vertical sonority. An electric organ will sound the individual pitches of the accompanying sonority, beginning with the lowest pitch, and will sustain them as they are sounded. Then an electric piano will sound the first pitch of the Interval Series to be sung. You.will sing that same pitch and then proceed through the remaining four pitches of the series. The approximate tempo at which you should sing is that at which the pitches were individually sounded (CD = 30). Once you have begun, go completely through the five pitches 109 without repeating notes. As you sing, try to blend each pitch into the sustained sonority. After a slight pause, the sonority will be sounded by the organ and the electric piano will play (correctly) the Interval Series back to you at the previous tempo. As the notes are played, sing with the piano, vocally antici- pating each note, and circle the intervals (pairs of pitches) on which you errored. The same drill sequence will again occur and you will vocally respond in the same manner. Try to keep in mind the intervals incorrectly sung on the first performance while accomplishing the second. Do not sing openly--hum! Headsets restrict your hearing openly through the ears. The tones you will hear are transmitted by the skull bones. Therefore, little is gained from singing loudly which will distract others. Remember to sing very slowly and try to blend your sung pitches into the sustained sonority. Leave all drill sheets in your booth at the conclusion of the drills and be certain you have signed your name and today's date. 110 ATONAL SIGHT SINGING DRILLS DIRECTIONS FOR CONTROL SUBJECTS A drill sheet with 20 cells of atonal Interval Series will be distributed to you at the beginning of each laboratory drill session. Sign your name and today's date at the top of the sheet. Put on your headsets and adjust them so that they comfortably fit your ears. Adjust the knob labeled "program" to hear the recorded drills. No other controls require any adjustment. Each Interval Series to be sung is presented twice within each cell. The Interval Series are written in whole- note notation and without key signatures. An electric piano will sound the first pitch of the Interval Series to be sung. You will sing that same pitdh and then proceed very slowly through the remaining four pitches of the series at approximatelyCD = 30. Once you have begun, go completely through the five pitches without repeating notes. To determine each interval, use previously instructed techniques such as numbers or solfege; refrain from singing wide intervals in steps and half-steps. After a slight pause, the electric piano will play (correctly) the Interval Series back to you at the approx- imate tempo at which you should sing ((D== 30). As the notes are played, sing with the piano, vocally anticipating 111 each note, and cirlce the intervals (pairs of pitches) on which you errored. The same drill sequence will again occur and you will vocally respond in the same manner. Try to keep in mind the intervals incorrectly sung on the first performance while accomplishing the second. Do not sing openly--hum! Headsets restrict your hearing openly through the ears. The tones you will hear are transmitted by the skull bones. Therefore, little is gained from singing loudly which will distract others. Remember to sing very slowly and accurately. Leave your drill sheet in your booth at the conclusion of the drills and be certain you have signed your name and today's date. 112 ATONAL SIGHT SINGING TESTS (TAPE RECORDED DIRECTIONS FOR EXPERIMENTAL GROUP SUBJECTS) This final test in atonal sight singing is in two parts labeled A and B. In Test A, which you will experience first, you are to proceed just as you have been doing in the drill units with the exception that you will only give one response of the five tones to the given pitch. The Interval Series will be played back to you in its correct form and you may vocally anticipate the notes as you did in the drills; but then you will proceed directly to the next row executing it in the same manner. Remember to sing very slowly taking time to accurately recognize each interval and blend it into the background sonority as well as you can. Do not back up, repeat, or vocally test between the notes of the intervals. Once you start each row go completely through to the last note, but remember to do so slowly and methodically. Directions for Test B will follow Test A. Please state your name. (TEST A) Atonal Sight Singing Test B will be accomplished in the same manner as Test A but with one difference: there will be no accompanying sonority on the organ. The begin- ning tone will be sounded by the electric piano which you 113 will repeat and then slowly proceed through the remaining tones. As before, go completely through the five tones without repetition or testing of intervals. (TEST B) 114 ATONAL SIGHT SINGING TESTS (TAPE RECORDED DIRECTIONS FOR.CONTROL GROUP SUBJECTS) This final test in atonal sight singing is in two parts labeled A and B. You will be administered Test B first in which you are to proceed just as you have been doing in the drill units, with the exception that you will give only one response of the five tones to the given pitch. The Interval Series will be played back to you in its cor- rect form and you may vocally anticipate the notes as you did in the drills; but then you will proceed directly to the next row executing it in the same manner. Remember to sing very slowly taking time to accurately recognize each interval. Do not back up, repeat, or vocally test between the notes of the intervals. Once you start each row, go completely through to the last note, but remember to do so slowly and methodically. Directions for Test A will follow Test B. Please state your name. (TEST B) Atonal Sight Singing Test A will be accomplished in the same manner as Test B with one difference: there will be an organ sustaining an accompanying chord which will be composed of the same five tones you will sing. As you sing the tones, try as best you can_to blend them within the 115 chord. The organ will arpeggiate the chord from the lowest note upwards in approximately the same tempo in which you should respond. Then the first tone of the Interval Series will be sounded by the electric piano and you will proceed as usual singing the five tones. The tones will be played back to you in their correct form with the organ accompa- niment. As before, go completely through the five tones without repetition or testing of intervals. (TEST A) 116 ATONAL SIGHT SINGING TEST A Name 117 ATONAL SIGHT SINGING TEST A Name OCTAVE DISPLACEMENT, INTERVAL DESIGNATION AND SONORITY ANALYSIS Arrows indicate octave displacement (23 pitches) in relation to male v01ce range. 1 1 3H J. :flififinlc 5 vv £> I0 u [2. [3 I ‘5 lb P2M3n2 Sd2 P2M2n33dt P3Mn232dt PzMznZSdzt ._J;l.__ 5. 'z 1. 29 30 3! 31 P2M3n2d2t P2M2n28d2t P3Mn82d2t PZMnZSZdzt 7. (b. (L 33 3! 35'36 4! N2 93 9‘! ”I5 .6 ‘07 '05 PMn3SZd2t P2M2n3d3t M2n283d2t PMZnZSZdtZ 118 ATONAL 510m SINGING TEST 119 ATONAL SIGHT SINGING TEST 3 Home INTERVAL DES IG NAT ION 0' 1: 910 u i1 :3 u) (516 “as... out-’n-cn ..----.o -.—.-— c.---- ....-o——-_ -. -§.. mo...— . O... “-*.——- ----~o c--*“o—- ' ’00 "I '10 90--....— -......- «nu—I {’9 F 33 an 35' 36 37 as 39 40 In «2 '03 «q as 45 on 1,9 120 TABLE 10 RELIABILITY OF TONAL SIGHT SINGING TEST AS DETERMINED BY THE TEST-RETEST METHODa Subject Test Retest L.B. 26 '28 D.C. 35 37 J.C. 41 42 A.D. 22 21 PWE. 22 ' 23 L.F. 44 45 S.F. 31 32 B.G. 38 40 C.H. 28 29 M.J. 32 33 B.K. 30 30 R.N. 24 26 3.1:. ‘ 33 35 O.R. 25 24 A.S. 42 41 G.S. 39 39 T.S. 25 27 L.T. 23 25 R.T. 36 38 B.W. 35 34 N.W. 32 34 aScore = number correctly sung of 48 intervals; Test-Retest reliability = .8032 (Pearson Product—Moment). 121 TABLE 11 RAW SCORES: TONAL SIGHT SINGING TEST Experimental Group Control Group Subjects Pretest Posttest Subjects Pretest Posttest B.A. 37 43 K.B. 35 45 A.B. 28 38 M.B. 23 36 D.B. 36 41 V.B. 24 28 J.B. 30 34 L.D. 34 39 I.C. 39 46 J.B. 27 36 J.C. 39 47 J.F. 32 41 O.C. 29 37 D.G. 30 40 D.B. 29 36 P.H. 39 41 E.F. 29 36 N.K. 38 44 P.F. 19 ' 26 T.L. 38 43 S.H. 27 37 B.M. 25 26 E.J. 41 45 B.M. 22 31 B.L. 40 45 S.M. 31 45 D.M. 19 28 L.P. 18 28 N.M. 38 42 D.R. 43 47 T.M. 30 35 P.R. 37 39 M.P. 36 42 R. Su. 42 46 B.R. 29 35 B.V. 26 35 B.R. 33 42 C.W. 38 44 G.S. 32 41 K.W. 40 47 V.S. 34 39 B.W. 44 47 R.Sc. 23 29 M.Y. ‘_36 ‘_41 R.M.T. _40 _41_ Number of Subjects 22 22 23 23 Mean 32.82 39.50 32.04 38.74 Standard Deviation 7.50 6.53 6.43 5.93 ATONAL SIGHT SINGING TEST A RAW SCORES 122 TABLE 12 Experimental Group Control Group Subjects Scores Subjects Scores B.A. 40 K.B. 33 A.B. 32 M.B. 30 D.B. 39 V.B. 29 J.B. 32 L.D. 45 I.C. 40 J.B. 29 J.C. 44 J.F. 34 O.C. 28 D.G. 27 D.E. 36 P.H. 43 E.F. 30 N.K. 43 P.F. 28 T.L. 43 S.H. 33 B.M. 31 E.J. 45 B.M. 30 B.L. 47 S.M. 40 D.M. 30 L.P. 28 N.M. 43 D.R. 47 T.M. 39 P.R. 43 M.P. 38 R.Su. 41 B.R. 28 B.V. 31 E.R. 37 C.W. 44 G.S. 3o K.W. 44 V.S. 33 S.W. 48 R.Sc. 30 M.Y. _4_o R.M.T. _43 Number of Subjects 22 23 Mean 37.41 36.00 Standard Deviation 7.11 6.17 123 TABLE 13 RAW SCORES: ATONAL SIGHT SINGING TEST B Experimental Group Control Group Subjects Scores Subjects Scores B.A. 40 K.B. 32 A.B. 30 M.B. 30 D.B. 40 V.B. 29 J.B. 31 L.D. 43 I.C. 40 J.E. 29 J.C. 41 J.F. 32 O.C. 22 D.G. 24 D.B. 33 P.H. 44 E.F. 32 N.K. 41 P.F. 24 T.L. 43 S.H. 24 B.M. 33 E.J. 43 E.M. 27 B.L. 45 S.M. 41 D.M. 25 L.P. 26 N.M. 38 D.R. 42 T.M. 33 P.R. 42 M.P. 33 R.Su. 46 B.R. 21 B.V. 31 E.R. 32 C.W. 45 G.S. 30 K.W. 43 V.S. 35 S.W. 47 R.Sc. 23 M.Y. _3_§ R.M.T. _4; Number of Subjects 22 23 Mean 36.73 32.91 Standard Deviation 7.41 7.32 124 TABLE 14 ITEM ANALYSIS, ATONAL SIGHT SINGING TEST A EXPERIMENTAL GROUP (N = 22) Item Index of Pt. Biserial Number Difficulty Correlation l 9 .04 2 50 .48 3 32 .59 4 55 .70 5 0 .00 6 27 .04 7 27 .33 8 41 .70 9 0 .00 10 14 .44 11 14 .16 12 18 .52 13 27 .40 14 41 .51 15 23 .31 16 23 .67 17 18 .55 18 0 .00 19 36 .56 20 14 .29 21 9 .36 22 36 .51 23 9 .09 24 23 .13 25 5 .11 26 .00 27 23 .09 18 .33 N 00 125 TABLE 14--Continued Item Item of Pt. Biserial Number Difficulty Correlation 29 9 .41 30 14 .37 31 23 .ll 32 14 .50 33 18 .30 34 18 .14 35 41 .51 36 32 .42 37 9 .29 38 18 .49 39 36 .48 40 41 .51 41 27 .54 42 14 .07 43 23 .56 44 36 .71 45 14 .46 46 41 .34 47 18 .38 48 23 .52 Reliability = .86. 126 TABLE 15 ITEM ANALYSIS, ATONAL SIGHT SINGING TEST A CONTROL GROUP (N = 23) Item Index of Pt. Biserial Number Difficulty Correlation 1 26 .16 2 65 .57 3 39 .62 4 35 .24 5 9 .36 6 43 .45 7 9 .36 8 43 .46 9 4 .28 10 4 .28 ll 13 .ll 12 30 .38 13 35 .23 14 43 .17 15 39 .44 16 35 .48 17 30 .41 18 9 .36 19 26 .43 20 22 .43 21 9 .36 22 30 .36 23 26 .46 24 57 .61 25 .ll 26 .36 27 17 .17 26 .31 N no 127 TABLE 15--Continued Item Index of Pt. Biserial Number Difficulty Correlation 29 4 .14 30 13 .04 31 13 .43 32 26 .43 33 26 .07 34 30 .38 35 30 .47 36 26 .15 37 0 .00 38 17 .53 39 22 .30 40 30 -.02 41 22 .35 42 17 .30 43 17 .19 44 43 .23 45 0 .00 46 39 .38 47 35 -.02 48 48 .36 Reliability = .7988. 128 TABLE 16 ITEM ANALYSIS, ATONAL SIGHT SINGING TEST B EXPERIMENTAL GROUP (N = 22) Item Index of Pt. Biserial Number Difficulty Correlation 1 32 .58 2 45 .74 3' 23 .61 4 45 .18 5 5 .20 6 32 .46 7 18 .24 8 32 .24 9 5 .38 10 23 .52 ll 9 .36 12 18 .49 13 14 .19 14 9 .45 15 32 .57 16 18 .24 17 18 .47 18 0 .00 19 55 .43 20 14 .28 21 .01 22 0 .00 23 14 .24 24 41 .63 25 27 .65 26 14 .35 27 18 .52 .40 N 00 \D 129 TABLE l6--Continued Item Index of Pt. Biserial Number Difficulty Correlation 29 36 .64 30 9 .01 31 23 .34 32 23 .41 33 55 .30 34 9 .01 35 18 .11 36 27 .53 37 0 .00 38 32 .58 39 32 .65 40 36 .27 41 45 .42 42 14 .44 43 32 .26 44 18 .21 45 9 .32 46 32 .41 47 S9 .48 48 41 .42 Reliability = .87. 130 TABLE 17 ITEM ANALYSIS, ATONAL SIGHT SINGING TEST B CONTROL GROUP (N = 23) Item Index of Pt. Biserial Number Difficulty Correlation 1 22 .35 2 43 .32 3 22 -.05 4 61 .02 5 13 .14 6 35 .60 7 22 .44 8 13 .41 9 9 .28 10 35 .44 ll 4 .30 12 22 .33 13 26 .35 14 9 .41 15 26 .32 16 26 .26 17 13 .23 18 0 .00 19 78 .32 20 52 .32 21 13 .18 22 22 .42 23 57 .72 24 30 .47 25 30 .32 26 30 .36 27 22 .38 39 .53 N 00 131 TEST l7--Continued Item Index of Pt. Biserial Number Difficulty Correlation 29 48 .56 30 30 -.01 31 52 .46 32 30 -.06 33 65 .48 34 26 .49 35 35 .32 36 26 .14 37 9 .38 38 35 .44 39 39 .29 40 52 .41 41 48 .33 42 17 -.01 43 48 .80 44 48 .34 45 9 .41 46 35 .67 47 48 .32 48 35 .41 Reliability = .84. 132 TABLE 18 DIFFERENCE BETWEEN MEAN SCORES OF MALE AND FEMALE SUBJECTS ON ATONAL SIGHT SINGING TESTS A AND B Group & Factor Mean S.D. df t Test Exp. Male 36.50 7.48 Exp. Female 37.92 6.61 20 .42 A Exp. Male 36.25 8.06 Exp. Female 37.00 6.71 20 .21 B Exp. Male 36.50 7.48 Con. Female 35.71 6.40 20 .23 A Exp. Male 36.25 8.06 Con. Female 32.57 7.62 20 .99 B Exp. Male 36.44 5.47 Con. Female 35.71 6.40 21 .28 A Con. Male 33.44 6.38 Con. Female 32.57 7.62 21 .28 B 133 TABLE 18--Continued Group Factor Mean S.D. df t Test Con. Male 36.44 5.47 Exp. Female 37.92 6.61 21 .55 A Con. Male 33.44 6.38 Exp. Female 37.00 6.71 21 1.22 B Exp. Male 36.50 7.48 Con. Male 36.44 5.47 15 .02 A Exp. Male 36.25 8.06 Con. Male 33.44 6.38 15 .74 B Exp. Female 37.92 6.61 Con. Female 35.71 6.40 26 .87 A Exp. Female 37.00 6.71 Con. Female 32.57 7.62 26 1.57 B 134 TABLE 19 DIFFERENCE BETWEEN MEAN SCORES OF SUBJECTS WITH AND WITHOUT KEYBOARD EXPERIENCE ON ATONAL SIGHT SINGING TESTS A AND B Group & Factor Mean S.D. df t Test Exp. Keybd. 39.08 6.60 Exp. N-Keybd. 35.00 6.70 20 1.34 A Exp. Keybd. 38.15 6.45 Exp. N-Keybd 34.55 8.12 20 1.05 B Exp. Keybd. 39.08 6.60 Con. N-Keybd. 38.67 5.47 23 .16 A Exp. Keybd. 38.15 6.45 Con. N-Keybd. 35.33 6.60 23 1.04 B Con. Keybd. 33.09 4.82 Exp. N-Keybd. 35.00 6.70 18 .68 A Con. Keybd. 30.27 6.83 Exp. N—Keybd. 34.55 8.12 18 1.19 B Con. Keybd. 33.09 4.82 Con. N-Keybd. 38.67 5.47 21 2.49* A *Significant at .05 level of confidence. 135 TABLE l9--Continued Group & Factor Mean S.D. df t Test Con. Keybd. 30.27 6.83 Con. N-Keybd. 35.33 6.60 21 1.74 B Exp. Keybd. 39.08 6.60 Con. Keybd. 33.09 4.82 22 2.38* A Exp. Keybd. 38.15 6.45 Con. Keybd. 30.27 6.83 22 2.80** B Exp. N-Keybd. 35.00 6.70 Con. N-Keybd. 38.67 5.47 19 1.27 A Exp. N-Keybd. 34.55 8.12 Con. N-Keybd. 35.33 6.60 19 .23 B *Significant at .05 level of confidence. **Significant at .02 level of confidence. 136 .mmaoum suon mo mam>umucfl mcdm manomuuoo mo mmmucwonmm mumowpcfl monamflmm mnm.mm D .82 m~.Nm D m .mz mw.mm D m .Hz NH + umoz mnm.mm D .Hz oo.mo D m .Hz om.vm D m .82 Ha oom.mm D .m mh.mm D m .m mh.mw D w .m ca 9 mNm.mm Q .m om.¢m D m .m mm.hm D m .m m A mnm.on D .m om.mm G v .m om.¢h D m .02 m D mmm.~> D .mz mh.o> D m .62 mm.hh D m .m n O oom.mh D .Hz mm.mn D m .HZ mh.mh D m .flz m H oom.mh D .m oo.mh D v .m mN.Hm D N .HS m m oom.am D .82 m~.mh D m .m: mh.mm D m .82 v m mnm.vm D .mz mm.mm D N .m: oo.¢m a a .m m H mma.mm D .flz mh.mm D m .flz om.¢m D m .m: m D oom.mm Q N .HS oo.mm D N .HS mN.om D N .wz H + ummmn mucmoumm m w m umma mucmoumm m umms mucmoumm d umma Hmpuo. mBmMB OZHOZHm EDUHm Adzoafl mma 20mm Qm>Hme md wfl¢>mmBZH m0 NDUMdmmHm om mqmda 137 TABLE 21 OCTAVE DISPLACEMENT OF INTERVAL SERIES TONES ON ATONAL SIGHT SINGING TEST Aa Experimental Group Control Group Female 8va Female 8va Subjects Displ. Total Subjects Displ. Total K.B. 14 33 A.B. 15 32 V.B. 13 29 J.B. 16 32 L.D. 20 45 J.C. 21 44 J.F. ll 34 E.F. 12 30 P.H. 18 43 P.F. 15 28 N.K. 21 43 S.H. 13 33 B.M. 14 31 E.J. 22 45 B.M. 13 30 B.L. 22 47 S.M. 18 40 M.P. 19 38 L.P. 13 28 B.R. 12 28 P.R. 22 43 E.R. 18 37 K.W. 19 44 V.S. 13 33 S.W. 23 48 R.Sc. 14 30 M.Y. _2_g fl R.M.T. _2_o_ _43 Totalsb 239 531 232 500 aCorrect interval averages for these scores are shown in Table 13. bThe above scores indicate correctly sung intervals. 138 TABLE 21-—Continued Experimental Group Control Group Male 8va Male 8va Subjects Displ. Total Subjects Displ. Total B.A. 20 40 M.B. 14 30 D.B. 17 39 J.E. 12 29 I.C. 19 43 D.G. 15 27 O.C. 15 28 T.L. 20 43 D.E. 16 36 R.Su. 18 41 N.M. 19 43 B.V. 17 31 T.M. 17 39 C.W. _3_l_ _5_4 G.S. __1_3_ 39 Totalsb 139 292 152 328 Female Totals 239 531 232 500 Male Totals _132_ _222. ‘_152 _328 Comb. Totals 378 823 384 828 aCorrect interval averages for these scores are shown in Table 13. bThe above scores indicate correctly sung intervals. APPENDIX B ATONAL SIGHT SINGING DRILLS FOR EXPERIMENTAL GROUP SUBJECTS 139 ATONAL SIGHT SINGING DRILLS UNIT I Ila-o Date 140 Na I3. 141 I l I 11 til-III! '2' III-All! Ii 142 ATONAL SIGHT SINGING DRILLS UNIT II In. Dat a 7.2.1.3,» «2.3.? II. _____ 1-11:2; 1:2-.. :1: \\\\\ ...,\\.\\\. .\\\\\. IN\.\\\ U" 2 143 L.D., nJ..+.~wu. (44”,..4. ~..-L_ ~~..L -- -~D\»«. Exxxxx. \\\..\ U " 3 144 145 ATONAL SIGHT SINGING DRILLS UNIT III I... Date 1:3- 1.33.. I:_ ___: 5112. 1.3: 4:: \EI tité (FCC. - x \ xxx“: 4411111] m - 2 146 IO. II. I 2. 13. m. ‘ us. :L: li4~§a \L+~I. 7:: :L: .i-»‘\:. .xxxxx- \\\\\ III - 3 147 148 ATONAL SIGHT SINGING DRILLS ON IT IV In” Date 112.. 212.2,. 2:. .2: 9-21:. 42:. 4:: \:: EFCC. \CQ. ...,.......\....._. IV - 2 149 ID. IL I 2. IV- 3 150 l6. (‘7, If. 151 ATONAL SIGHT SINGING DRILLS UNIT V Halo Date HH .l 1 ill 11‘“ III! III“. { ‘3. I": ’5, V-3 153 154 ATONAL SIGHT SINGING DRILLS UNIT VI Name Date VI— 2 155 /:/’ JVZDZ- .:.I. I... I... IL: ::;. .::\ VI - 3 156 157 ATONAL SIGHT SINGING DRILLS UNIT VII Nan. Date 53. (‘1. l5. VII - 3 159 -- --- v“; —. ..-- ‘ 7w .—-.* “h.“ v—v - -——— “ "“—-"‘ ———~ .———~ 160 ATONAL SIGHT SINGING DRILLS UNIT VIII Date Name VIII - 2 161 ‘HH VIII - 3 ‘ .J,. a... --<-- - o...- a...— ~--- - ~"‘.—_ _,—_ -..-—-—. -’_- ”4 Ohm -___.,.. -———-—.——-——. .4..- 163 ATONAL SIGHT SINGING DRILLS UNIT IX Name Date _.--_——-__-_- v— — —- --—~——-—-.—.. " .- —.—.-- _ _ __ -—--—-—— w... - .-——._‘_---—-—-—_. - _._.—. -. ’31 I”! [gt IX“ 3 165 (0- (7: ,6! VIII- 1:12. .337 \\\\\ - \\\.\K. \\\\\..... \..\....\...m\wa. .\\\\\ I VITO.“ 166 ATONAL SIGHT SINGING DRIILS UNIT I Ran. Data .w“-~-. ---~ - w“ CAL. -::L. :Z. I... 57....-.2- (222.! :2. :2. -ZIGI #2:; 3:: -I- I!’ X~2 167 HM X-B 168 (6. I7. IS, 169 ATONAL SIGHT SINGING DRILLS UNIT XI Date Name XI- 2 170 :0. u, _ I2, ‘3’ H. I5. XI - 3 171 172 ATONAL SIGHT SINGING DRILLS UNIT XII Nam XII - 2 173 I3. I‘I. IS. XII '- 3 174 ((9., (7a ’8! I ‘7. 20. 175 ATONAL SIGHT SINGING DRILLS UNIT XIII Nam Dat n 176 XIII - 2 l3. H. ’5' 177 1111 - 3 lb . I7. 16 , 178 ATONAL SIGHT SINGING DRILIS UNIT XIV Nana ' Date K—rr.‘ 179 XIV -- 2 (O. H . ll, 180 11v - 3 lb. (7. (8'. 181 ATONAL SIGHT SINGING DRILLS UNIT XV Name Date XII-2 182 ‘3' (all (5, XV- 3 183 16. - 17. I8. 184 ATONAL SIGHT SINGING DRILLS UNIT XVI Date Name XVI - 2 185 -- - Ir. —.— I’—' I; I_— - —-oo n~wd~ onc— bi 13, H. I5. XVI - 3 186 Iquu, z—auad s-- \ x ‘ \ . xxxxx \\\\4; \\\\\ \\\\\ 187 ATONAL SIGHT SINGING DRILLS UNIT XVII Name Date 2. 3, I I] J h I . n F u 1 A I WWW—W MUD l j) I D 1] D A I D A_. " ' D 9- 6 "' V XVII - 2 188 I9. ”I ' (1' ”a.“ *‘*-——..... - '3, M. I5, XVII ~ 3 189 (G. _ I7. 18'. o4---——. “no...“ ‘ ta“ -—_—.-..—-.g-..;...4 on. 1...— _-—~——.— —~.—.--_._ *um- 4L —- ‘“————~ . ._ L.;.....+. V-‘+*g.‘. ‘axga \\_\\\ e. \xx‘x. \\\\\. \\\\\ \.\\\\ 190 1705.11. SIG‘Z‘I‘ SINGING DRILLS UFI'I‘ XVIII 253.): Date ..... . . . 1" III! ‘I Ii . XVIII - 2 191 .— --~—-a.. -n—m" .— ~--.—.¢---——_— — I3. [1! 0 l5, ."wow -....a~- :2, 1:.Cl.. ......ll. ..... 2.: L.F..-.ggi ENC... :LK\ XVIII -«- 3 192 (G. ‘7 c l 8’. --—. ---.—--.-— -—-—-‘ --— -0- *_H ‘- -——.-. -- _.—-—_..—..“. —-.-a w--.“ v w ._ —-w—-_———-.- 193 ATOM SIGET SI‘EGING DRILIS UNIT 13 lien s Dnt o 111- 2 194 :3, . H. IE. XIX - 3 195 196 ATONAL SIGHT SINGING DRILLS UNI!.XX Nil. Data xx-2197 ’3 "In ‘5, 11'3 -...... ...s... ..... ..... . xx.\\‘ k\\\\\. \\\\\ \\\\\ 199 ATONAL SIGHT SINGING DRILLS UNIT XXI Name Dat e XXI~2 200 :3, H. :5, XXI - 3 201 202 ATONAL SIGHT SINGING DRILLS UNIT XXII Nana Date ' u " —,\v ..... IIIIIIIIIlIIl-l. I XXII - 3 204 ----——- 205 ATONAL SIGHT SINGING DRILLS UNIT XXIII Name ' Eat a XXIII - 2 206 '30 l”: ’5’ xxm - 3 207 , :.......M.IIIE.,»_. {xiii LEE! I. 1...th is; >22 208 ATONAL SIGHT SINGING DRILLS UNIT.XXIV Name Date $3.74 344~9J« 7.~._ ..._. L.D.Iméuv -.H-%~‘., 12...- I... .....::..7 it}? \\\\\ XXIV - 2 209 z a float; VIII" XXIV - 3 210 H‘H APPENDIX C ATONAL SIGHT SINGING DRILLS FOR CONTROL GROUP SUBJECTS APPENDIX C ATONAL SIGHT SINGING DRILLS UNIT I Nan. Date 211 212 ATONAL SIGHT SINGING DRILLS UNIT II I... but. 213 ATONAL SIGHT SINGING DRILLS UNIT III III. Dita /0. IL '2. M k k 214 ATORAL SIGHT SINGING DRILLS UNIT IV Inn. Date ___ _K ¥ h- III i III I|l I{.ll [III {Ill-I'll. __.... ..:..._. ......x ...\\\\x. ..\.\.\.\.\. 1 \\\\\. \,\\\\ I 215 ATONAL SIGHT SINGING DRILLS UNIT V Name DISC I3, "I: (5, lb. '70 I8! 216 ATONAL SIGHT SINGING DRILLS UNIT VI Nam Date ~..- .let \V\ 217 ATONAL SIGHT SINGING DRILLS UNIT VII Nana Dat. I3. [”0 F5, lb. I): It. 218 ATUhrL LIGNT SILGIHG LilLLi UNIT VIII Name Lrte ((9, I7. l8, m-. _—‘.—~.——--—-—.—-— IIIII 219 hTONnL SIGFT SINGING DRILLS UNIT IX Lame Date 1’3, H. IS [(9. I7. 1‘6, ——--_—.. .__.- .____ -- -————— -— —-—.<- .—---_ 220 ATONAL SIGHT SINGING EEILLJ UNIT X Date Name I 6. I7: (‘8', 221 ATONAL SIGHT SINGING DRILLS UNIT XI Nan. Dat e --. ~ ~— . --—. ———- ATCNAL SIGHT SIIIGZI‘NG DRILLS UNIT XII Name Date ’6' ‘7: lg! wot-v. q-o—o—r- 223 ATONAL SIGHT SINGING “PILLS UNIT XIII Date Name I) I?! I6, 224 ATONAL SIGHT SINGING DRILLS UNIT XIV Bane Data 8' H. 11;, (6‘ I7. )9, 225 ATONAL SIGHT SINGING DRILLS UNIT XV Nana Dat e (3. “In IS. 19. 10. 226 ATONAL SIGHT SINGING DRILLS UNIT XVI Name Date 6. H. "3. I6. I70 I“: 227 ATONAL SIGHT SINGING DRILLS UNIT XVII Name Dat o ’0' ”0 I2: I3. I‘Ia I5 ’Q- 17, n. . 228 ATONAL SIGHT SINGING DRILLS UNIT XVIII Na no. Dal: P ‘6- n. 18. 229 ATONAL SIGHT SINGING DRILLS UNIT XIX In. Dat o 230 ATONAL SIGHT SINGING DRILLS UNIT II an0 Date '6. I7. 1%, 231 ATONAL SIGHT SINGING DRILLS UNIT XXI Nana But a 232 ATONAL SIGHT SINGING DRILLS UNIT XXII Name Dat o '3’ I“, ';p 233 ATONAL SIGHT SINGING DRILLS UNIT XXIII Naue Date 234 ATONAL SIGHT SINGING DRILLS UNIT XXIV Nana Date .— —_ ¥ h IIIIIIIIIIIIIII?! '