1-1.. 1. . "1:11 ":1. $1.21” {3% 1 1 j" ‘13}; 11183:; ““1 . 11:11.31L Ml“ max; {11151193131 . $131913. 1‘ . €3.31 ‘3'“. v, _. I‘1‘1’1ju” ‘ h #115" 3'“ . 1 . 31312117 “5&1 $331333?ng fitvgfi-‘H 1" :n. 1 13m?" 1:121“ I f 11111:... AIL :13" 1:13“? 1;: 1;. my"! 1.1.1.1"qu ‘2' . L'11.1‘L. ‘ «1’31“ 1 ‘11 1.521!" 533314 1.1““ ‘31 ~ 1&1“ M1; “211 131.151 ' " 11:31:15, 311113111 $311. 1' T?” 1:13} . 1.111; Whig 11111; 1411131? ' Rmx~u311dxfl3k R36 111‘ ‘13.,31‘1‘14 1- . >557 $.13; ' 1.1%“ ”a“: Flt: “I ‘R‘E‘q‘hq‘ . 3 N 11 7"‘u1n " " E'k 1.11h s. . ’31:“: ' Mtge}: .. 131"”: 1..“ «"3111: “£11 3135 ":1. 111.1,}: .1 11111.1.)‘11L9LL 3““ .1 3333871 1121111 ‘13 [‘1\?3$ : 1 -.‘1. IV V3J£ 1‘1 ”1324-2: '1.“ ”2‘1 . 11:11! ‘ £31.; 3 LT: 1'2"":1.‘ . 5n ‘ '1: 311 11:11 m:‘|~\.‘ 111-141?» 31-359- ..:'m11"41:1§‘ 11?, 313311-11 Ira“? mm: iI‘-."1."_I" Cr. 5:331 . 1 1‘1‘1‘1} 11311. «1 . J" ‘1 3.41215“? 21 W153 . , 11$." 1 . 1. 113;. ' ' 1 311.1111 ~- .1- g . £131: .33» ~..;‘ 1111 111‘"... 131.. $.11. ~ #1331331 ~11 393‘33‘11. Q‘1,“ u‘hg‘vwfib 1 11 1 11 31. 1:331. , > . in. .3 11m “:2. QE.3,$§1S‘K_ 31"" ”13111513333131.14': 111 - L.~i‘ " . "‘1'" ‘t‘é‘i; :- .3 7 re :2): ’31:” ‘ “.1, ,1: ‘3 W731i: 3. '. .L 3 112121—231?» 1;;1“: I'm .752" " . :1 . .414 “‘31," r ‘1’. '45 .I‘ ”$1. 3%. -. k) 43> 1. L33. 1‘. 3,? r '.. #31 .1 :1 I 1‘: ,. x ,. .1 1 ‘, Zr 4 . ‘21 1? M; .1133. 4- u 2’ § 1 . f .1 7x - 3-51 "f~ 3‘ r151 . *1 MW‘M fight... I 1‘5; :‘g‘: JV 1 ( H~ ‘1 w. :1 ml . , :‘E‘Efi T3... "”115": 11 .. , 1' raw 7" .3111. ‘11,; 1‘ ~11. “ "anthems.“ :xmféié La 13'“ 1 ,- LVL‘Ir‘E‘I‘“ Mn 3“)“ 1““3. 1:21311‘». - V3: fl-‘u‘ -\§va \1 L1$~w:,‘$'~1:‘:¢ ‘3‘“ 2%“? 1131 U."Ju‘ 3-1-11“. 3911.. V». ‘11“ 1‘11: 4123' .J‘ {JP-1.1.4231: U...“ 11‘} "11, 31.1 .. 1 13‘“ 1 x’l‘fl' M3111 $€Lifiquggx .__.1' 3“31; 1.91:1; 139:5. 1.111 .111- ’ 351.112.1311 . 11-\ 111‘" ‘ “15.11% - '1 ~ . R. 32%? ‘1‘“ 113.1% rm» ”£11- 1-." V 3&3“ ” 5331 1.3L. W 2" | 41391;: This is to certify that the thesis entitled A Comparison of the Effects of Repeated Listening and Performance Experience on High School Band Students' Music Preference presented by Paul Gottlieb Dombroske has been accepted towards fulfillment of the requirements for Master ' 5 degree in Music Major professor [hue February 20, 1986 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES m RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. A COMPARISON OF THE EFFECTS OF REPEATED LISTENING AND PERFORMANCE EXPERIENCE ON HIGH SCHOOL BAND STUDENTS' MUSIC PREFERENCE By Paul Gottlieb Dombroske A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF MUSIC School of MUsic 1986 ABSTRACT A COMPARISON OF THE EFFECTS OF REPEATED LISTENING AND PERFORMANCE EXPERIENCE ON HIGH SCHOOL BAND STUDENTS' MUSIC PREFERENCE By Paul Gottlieb Dombroske The purpose of this inquiry was to measure high school band students' preferences for selected musical compositions, and to attempt to increase student liking for two compositions initially identified as less preferred. Repeated listening and performance experience were employed in attempts to increase music preference, and the relative efficacy of the methods was measured. In a pretest procedure, three groups of high school band students expressed preferences for ten recorded excerpts of band compositions. Two of the less-preferred compositions were performed repeatedly by one group, another group listened to them repeatedly, and a third group had no experiences with the compositions. Following the treatments, the students again expressed their preferences for the ten excerpts, and the effects of the differing treatments were compared. Students demonstrated robustly stable music preferences over time for compositions receiving no treatment. Performance experience markedly increased music preference; gains attributable to listening were much weaker. ACKNOWLEDGEMENTS The author wishes to thank the members of the guidance committee: Dr. Albert LeBlanc, Dr. Edgar Kirk, and Dr. Rosalie Schellhous for their assistance in the completion of the thesis. Special thanks are due to thesis advisers Dr. Melanie Stuart and Dr. Albert LeBlanc for their guidance and encouragement. Gratitude is extended to Our Savior Lutheran School; its Board of Education, Dr. Gary Knippenberg, Chairperson; and Mr. Ross Stueber, Principal. The author gratefully acknowledges the cooperation of the students, teachers, and administrators of the participating schools: Lansing Christian School, Lansing, Michigan; Mr. Glen Akers, Band Director; Dr. Robert Hill, Superintendent. Valley Lutheran High School, Saginaw, Michigan; Mr. David Britton, Band Director; Mr. David Krause, Principal. Lutheran High School Association of Greater Detroit, Dr. Herbert C. Moldenhauer, Superintendent. Lutheran High School East, Harper WOods, Michigan; Mr. Arthur Henne, Band Director; Mr. David Eifert, Principal. ii Lutheran High School North, Mt. Clemens, Michigan; Mr. Dean Kratz, Band Director; Mr. Dale Cooper, Principal. Lutheran High School Northwest, Farmington Hills, Michigan; Mr. Kurt von Kampen, Band Director; Mr. Clayton Hufnagel, Principal. iii TABLE OF CONTENTS Page LIST OF TABLES ..................................... vi LIST OF FIGURES ..................................... vii I. INTRODUCTION ................................. 1 Purpose ...................................... 2 Problem ...................................... 2 Definitions .................................. 3 Assumptions .................................. 4 Limitations .................................. 4 Need for the Study ........................... 5 II. RELATED LITERATURE ........................... 7 Overview .................................... 7 Familiarity and Repeated Listening ........... 7 Adult and Peer Influences .................... 10 Experimental Design .......................... 11 III. PROCEDURE .................................... 15 The Pilot Test ............................... 18 .Procedures for the Main Study; .............. . 21 IV. ANALYSIS or DATA ............................. 24 Group Test Results ........................... 24 Reliability Analysis ......................... 27 Effects of the Experimental Treatments ....... 29 Summary ...................................... 38 V. CONCLUSIONS AND RECOMMENDATIONS .............. 40 Summary ...................................... 40 Discussion ................................... 41 Conclusions .................................. 43 Recommendations .............................. 47 APPENDICES ................................... 49 Appendix A ............ - ....................... 49 Appendix B ................................... 50 Appendix C ................................... 51 iv Page LIST OF REFERENCES ................................. 52 Table Table Table Table Table Table 4-1. 4-2. 4~3. 4-4. 4-5. 4-6. LIST OF TABLES MPI Metronome Markings and Pretest Preference Results for All Groups ............................. Pretest Preference Results for Males and Females ...................... Pretest-Posttest Reliability by Group ............................... Wilcoxon Findings for Group A (Performance) .......................... Wilcoxon Findings for Group B (Listening) ............................ Wilcoxon Findings for Group C (Control) .............................. vi Page 25 28 3O 33 35 37 Figure 3-1. Figure 3-2. LIST OF FIGURES Compositions Used on the MPI ........... MPI Pilot Test: Preference Ratings and Rankings ........................... vii Page 17 20 CHAPTER I INTRODUCTION The teaching methods and objectives of every school music instructor are, in part, determined by the music preferences of his students. The responsible teacher encourages students to exhibit a tolerance to many kinds of music, and to seek active or vicarious music involvement for himself (Sidnell, 1973, p. 47; Leonard & House, 1972, p. 257). It is widely assumed that these broad goals of music education can be well served by school performance organizations such as orchestras, bands, and choirs. Sidnell (1973) stated, The purpose of the [performance] class is to contribute a vocabulary of musical skills, knowledge, perceptions, and attitudes to learners.. .Conductor/ teachers can provide for optimum musical development during the performance experience if they take a little care in planning total and varied musical experiences for students. (p. 14) 2 Having accepted (a) the goal of increasing student tolerance for many kinds of music, and (b) that this goal can be achieved in the performance class, the music educator must now choose effective techniques to accomplish it. In performance classes, time is typically spent rehearsing and performing music. Another method available to the performance class is listening to music, perhaps repeatedly and with instruction, reproduced with audio equipment. The listening technique has been thoroughly researched; live performance has not. The objective of the present inquiry was to compare the efficacy of repeated listening and performance experience as modifiers of students' music preferences. Purpose The purpose of this study was to measure high school band students' preferences for selected musical compositions, and attempt to increase student liking for two compositions initially identified as less preferred. Gains in preference were measured in three separate groups of subjects exposed to repeated listenings, repeated performance experiences, or no treatment. Problem It was hypothesized that repeated listening would increase students' preferences for a particular composition; that performance experience, as a form of 3 repeated listening, would increase students' preferences for a particular composition; and that performance experience would increase students' preferences for a particular composition more than would repeated listening. The study addressed itself to the following questions: 1. Is performance experience an effective method of inducing familiarity with the aim of increasing music preference? 2. What difference, if any, is there in effect between the methods of repeated listening and performance experience in increasing music preference? Definitions Throughout this paper, certain words and terms will be used which need definition or clarification. They are the following: Music preference is a student's stated liking for specific musical examples. Performance experience includes full-ensemble recreation of specific musical examples in rehearsal or concert settings. Instruction refers to the teaching of historical style, form, and harmony, with increased aural discrimination of these elements as its goal. Not included for the purposes of this study is teaching aimed at increasing technical facility on a musical instrument. Assumptions The present study proceeded on the basis of the following assumptions: 1. It was assumed that increased preference for a particular composition or style of music will expand a growing body of student preference. Acquired preference for newly-introduced music need not cause the rejection of existing preferences. 2. It was assumed that performance experience is a form of repeated listening, and that one of its effects is induced familiarity. 3. It was assumed that a student's stated opinion is a valid measure of his music preference. Limitations The principal objective of this study was limited to an attempt to increase student preference for two musical selections. Conclusions and generalizations will necessarily be of a preliminary nature until the project can be replicated on a larger scale. Because of the time required to complete the treatments, it was necessary to entrust the instructions and responsibility for treatment to the regular classroom teachers. Every effort was made to adequately train and supervise the teachers to restrain the effects of possible extraneous variables. 5 The success of a direct comparison between listening and performance might be limited by the instrumental abilities of the group that must perform. In this study, every effort was made to select music well within the abilities of the performing group. Need for the Study School band instructors have long assumed that the performance experiences in their classrooms have helped increase students' tolerance for many kinds of music. Since repeated listening has been shown to increase music preference, it is likely that the repeated rehearsal of music also increases music preference. This is only an assumption, however, and there is a need for instrumental school music to demonstrate the effectiveness of its methods in accomplishing its stated goals. The typical school band devotes a vast majority of its time to rehearsal and performance. Some of the music is performed in public concerts, but the responsible teacher presumably introduces a repertoire of works aimed at increasing student preferences for many types of music. Much time is required to rehearse this repertoire adequately; a reasonably accurate performance is necessary for true aesthetic experience. The number and quality of experiences with unfamiliar music are limited in most classrooms by thme and the capabilities of student performers. If performance experience is indeed a highly 6 effective method of increasing music preference, it should by all means be pursued with its present intensity. If, however, performance experience is the mere equivalent of repeated listening in increasing student preference, the structure of instrumental music classes should be altered to include more listening. Music educators must consider their products to be the young musicians emerging from their classrooms, not concerts or contests. The enlightened, student—oriented educator will use the most effective methods to produce the best possible products. Traditions and widely-held assumptions are precarious means of selecting effective methods. Evidence is needed to confirm the value of performance experience as a modifier of music preference. CHAPTER II RELATED LITERATURE Overview In the past several decades, researchers have shown great interest in music preference. At least 18 variables affecting music preference have been explored and findings have frequently varied widely. Wapnick (1976) and Kuhn (1980) have comprehensively reviewed music preference research. LeBlanc (1980), 1982) has advanced a theory of music preference, and has developed a model demonstrating the interaction of variables. This chapter concerns itself with (a) variables thought to alter music preference: 'familiarity and repeated listening, and adult and peer approval, and (b) studies that lend insight to the development of experimental design. Familiarity and Repeated Listening The music educator may be chagrined at the resistance mounted by students faced with unfamiliar musical genres. It is apparent that, in many cases, most-familiar music is also most-preferred music. In a 1977 study, Gibbons found that elderly people (age 65-95) prefer popular music of their young adult years 7 8 to later popular music. Chalmers (1978) held that familiarity did not significantly affect university students' music preferences, but two similar studies (Keston & Pinto, 1955; Darling, 1982) discovered that recognition of musical examples significantly affected subjects' attitudinal responses. The widespread preference for the familiar was demonstrated by Burmeister (1955), who found that 60% of adults sampled felt that more familiar music -- popular, folk, and hillbilly -- should be taught in Missouri public schools. In a study otherwise characterized by indifferent responses to various types of music, Shehan (1979) found middle school students particularly inhospitable toward ethnic, or ”world” music, presumably because it was very unfamiliar. Since familiarity can apparently increase music preference, researchers have experimented with inducing familiarity through repeated listening. Instruction is frequently associated with repeated listening, but there is disagreement about the efficacy of such instruction. Schuckert and McDonald (1968) tested the music preferences of preschool children, and played the less preferred music during the children's play periods. After four listenings, 50% of the children changed their preferences. Getz (1963, 1966) developed an extensive string orchestra listening program for 9 seventh grade students. Five of the 40 selections were repeated 11 times during the four—week treatment period. Student responses were increasingly positive at each of the first 6 to 8 hearings, but subsequent hearings drew negative responses due to apparent fatigue. Some studies of repeated listening with instruction (Archibeque, 1966; Hornyak, 1966) asserted that an understanding of musical components was necessary to positively influence music preference. Bartlett (1973) found that awareness of structural elements and repeated listening increased university students' preferences for classical music. Preferences for ”best-liked popular" selections declined during the course of nine listening sessions. A series of three studies involving seventh grade students (Bradley, 1970; 1971; 1972) suggested that repeated listening is perhaps the single most effective modifier of music preference. While repeated listening increased the students' liking of specific compositions, preference for similar compositions (transfer) was accomplished only with analytical listening (instruction). Evans (1966) investigated the effect of repeated listening combined with instruction in junior high school general music classes. One treatment group combined listening experiences with lessons in musical structure; a second group experienced listening without lO instruction. A control group pursued other music related activities that included neither listening experiences nor lessons in musical structure. Instruction was found to have little or nothing to do with students' affective responses, but repeated listening apparently increased students' liking for art music. It is evident that, to increase students' tolerance for unfamiliar music, teachers may well pursue a program of repeated listening. The research suggests that instruction aimed at increasing aural discrimination would be helpful in expanding music preferences. Adult and Peer Influences The music educator interested in expanding students' music preferences must be aware of the influences exerted by adults and peer groups. Particularly with younger children, adult approval can affect music preference. Three similar studies (Greer, Dorow, & Hanser, 1973; Greer, Dorow, Wachhaus, & White, 1973; Dorow, 1977) involving preschool and elementary school children showed that music taught in a high-approval setting was increasingly selected by the children in their private listening. For fifth grade students, Alpert (1982) found that music teachers' and disc jockeys' approvals increased classical music listening, while peers' approvals decreased classical 11 music listening. Hughes (1980) discovered that peer models and peer approval were more effective than adult modelling and approval in shaping the music selection behavior of grade 12 students. Of particular interest to the present study, performing music was found to be an important vehicle for expanding music preference. In attempts to expand students' music preferences, the secondary teacher should be aware that his encouragement and modelling are less significant influences than the attitudes and examples of the peer group. Experimental Design The extreme diversity of existing music preference research, while sometimes leading to difficulties in interpretation, has provided valuable insight in the area of experimental design. Some researchers have distrusted the validity of self-reported (verbal) measures of music preference (Cotter & Toombs, 1966; Greer, Dorow, & Hanser, 1973; Huges, 1980). These have used time spent listening as the indicator of preference, as measured by the Operant Music Listening Recorder (OMLR) or the music selection recorder (MSR). While the MSR does provide a very accurate measure of preference, Flowers (1980) and Alpert (1982) have found that verbal preference ratings compare favorably with data obtained with the MSR. The 12 effective usefulness of the MSR is apparently limited to studies with small populations, and the ranking of more than two selections cannot be accurately measured (Kuhn, 1980; Kuhn, Sims, & Shehan, 1981). It appears that a researcher may confidently use verbal measures of music preference. Testing the effect of gender on music preference may seem to be an anachronistic pursuit, but many studies support the notion that significant differences in music preference are associated with gender (Baumann, 1960; Broquist, 1961; MacGregor, 1968; Nolin, 1973; Rogers 1956; Schuckert & McDonald, 1968). There is a notable scarcity of research that tends to refute the gender factor (Noble, 1976; 1977). A manifestation of the gender effect was described by LeBlanc (1980): ”The respondent's sex can influence decisions of taste, as in the case of a listener especially sensitive to the sexual charisma of attractive performers of the opposite sex" (p. 32). Current research must continue to consider the gender of the respondent. The research stream provides two additional areas of information pertinent to the present study: (a) tempo and music preference, and (b) preferred music activity. In 1980, Wapnick examined the tempo preferences of undergraduate music majors. Using a Lexicon Corporation Varispeech II Speed/Time Compressor/Expander, subjects 13 were to adjust the speed of recorded music until the most appropriate tempo was reached. Subjects were found to consistently favor tempi faster than those of the original recorded performance. Getz (1966), LeBlanc (1981), and LeBlanc and Cote (1983) discovered that middle school students respond positively to music with fast tempi and strong beats. LeBlanc and Cote, however, found that tempo interacted with generic style and performing medium, and could report no clear finding for the effect of tempo alone. In a follow-up study, LeBlanc and McCrary (1983) restrained the effects of style and medium by using a test consisting only of instrumental traditional jazz music of various tempi. Four levels of tempo were identified, and it was discovered that each increase in tempo resulted in a corresponding increase in preference. Gibbons (1977) found, somewhat unexpectedly, that many elderly people prefer "stimulative” music to ”sedative” music. Blyler (1960), LeBlanc (1981) and Darling (1982) suggested that faster music was preferred to slower music. Music teachers can expect to find that music preference is influenced by tempo; faster music is more likely to be readily favored. Finally, two earlier studies (Broquist, 1961; MacGregor, 1968) found that elementary school children preferred playing instruments to other music class activities (listening, singing, dancing). These 14 findings may suggest that the primary involvement offered by instrumental music—making is itself a modifier of attitudes toward music. Clearly, more research is needed to explore the potential influence of performance experience in the development of music preference. CHAPTER III PROCEDURE A music preference inventory (MPI) was prepared using ten recorded examples of band music. The examples were chosen from professionally—recorded albums by the Eastman Wind Ensemble and the Michigan State University Symphonic Band. Two of the compositions were to be used in the treatment process; it was necessary that these be available in published editions. The remaining eight examples served both as distractor variables and as internal checks on the pretest-posttest reliability of the MPI. There was no need to obtain music scores or parts for these compositions. VIt was anticipated that students' initial preferences on the MPI would be formed largely on the basis of tempo and the presence of a strong beat (Blyler, 1960; Darling, 1982; Getz, 1966; LeBlanc,1981; LeBlanc & Cote, 1983; LeBlanc & McCrary, 1983), so a wide range of tempi and rhythmic characteristics was represented. Each excerpt featured a ”full band" timbre, to reduce the possibility that student preference for certain instruments would intervene as 15 16 an extraneous variable. The researcher attempted to make selections that offered a diversity of historical styles while maintaining reasonably high musical standards. The MP1 included transcriptions of keyboard, orchestral, and orchestral/vocal works from the sixteenth and nineteenth centuries. Six selections were original twentieth century works for winds, ranging from circus marches and jazz to a neo-classical ”intermezzo” for band. Compositions chosen for the MPI are shown in Figure 3—1 with their composers and recording labels. Recorded examples were started at the beginning of each composition, and were one minute in duration with ten-second response intervals between excerpts. The one—minute duration was chosen to give students an accurate impression of each work and time to consider their responses without consuming too much class time or creating an atmosphere of boredom. The MP1 was recorded from vinyl discs to TDK Corporation Acoustic Dynamic (AD) cassette tape by the investigator, using a Sharp RT-150 tape deck and a Technics SL-B350 turntable, both operating through a Pioneer SX-303 stereo receiver. The same tape deck, receiver, and a pair of Sansui S-517 loudspeakers were transported to participating schools for each administration of the MPI. A student response form was devised consisting of 17 mmauoz xpmz pwsmssmfiaoz .m \Hepo> .o Aways money Emasvmm Homm> mam .> owomAHMm spouse: unoccmm .m .m wmocwm amuseme< uao co wuasm m symzme mmifiuz one: eoeo .4 \eumncwcco .s passages m.u0nmz snug was nonsuum>o moamzmemo ewomAHMm segues: aoumfim .2 came mmeauasse zoemHm mwommHMm museum: pmuhmm scum: Cummnaom mmm .HH wNomAHMm Apnoea: nouns .o \euxm .3 ouasm when emHHHaz N seem mmsam .H amaaoz sum: consumaeucmm .m muasm “New uaaonaesm oesaomHmozmm ofionmz m.t~0mxo mo oHHmm 659 .H mmomAHmm Apnoea: noose .o\ecsm .3 unanm eusm sesame: H omen Mam: mxmo .H ewOmAHmm Apnoea: nuances .m .m nausea amoauee< ado co muaam A samzzmm AwomAHmm museum: mafia .M mama msuuao ozHM HonE=z\HonmA tuouom Howcmuu<\pmwomaou mHuHH GOHuHmomaou Hmz ozu co tow: macauwmommmm .Him okmmmm 18 ten five-point response continua using a Likert-type scale (see Appendix A). Students were instructed to choose one of the following responses for each example: strongly dislike, dislike, undecided, like, or strongly like. Information concerning the respondent's age, grade, sex, major instrument, and years of experience was also collected on the response form. The author desired to compare individual students' pretest and posttest scores on the MPI; a means of identifying student response forms was required. To assure compliance with the Family Educational Rights and Privacy Act of 1974, numbered identification cards were used at the pretest, collected and stored by the classroom teachers until the posttest, after which they were returned to the classroom teachers for disposal. At no time did the researcher have access to students' names, and individually identifiable test responses were not made available to the classroom teacher. By using the identification numbers, the investigator was able to match individuals' pretest and posttest responses. The Pilot Test The MPI was pilot-tested in May and September 1984 with 64 high school band students at Lansing Christian High School, Lansing, Michigan, and Lutheran High School Northwest, Farmington Hills, Michigan. Preference ratings for this pilot-test group are shown 19 in Figure 3-2. BYRD 1 and BYRD 2, the two examples chosen for the treatment procedure, were ranked eighth and tenth out of ten, respectively. Because the objective of the treatments was to measurably increase preference for BYRD l and BYRD 2, initial low preference for these compositions was considered desirable. As expected, preference rankings of the pilot—test group were closely tied to the tempi of the MPI selections. With the exception of the best-liked composition, FARRAR, students' relative preferences decline in direct proportion to the tempo of each selection. Because FARRAR is in compound meter with many heavily accented eighth note passages, one might even argue that it was perceived as "fastest” by the students. If this is the case, student preference was tied to tempo with marked uniformity. The high and low mean scores of 4.00 and 2.03, as well as visual examination of the pilot test response forms, reveal a reluctance on the part of students to choose the extremes of the five—point response scale. LeBlanc (1979, 1981), LeBlanc and Cote (1983) and LeBlanc and McCrary (1983) reported good results with a seven-point Likert-type scale with only the extremes verbally anchored. The researcher elected to retain the five—point scale, verbally anchoring each point. In so doing, it was hoped that capricious or random responses would be discouraged. 20 SA SA SJN Sta. Sum 86 $6 wtm as“ oo.q ,mcwumm amoz mmcfixcmm tam mmnwumm monogamoum 21 Procedures for the Main Study Three Michigan Lutheran high school bands were selected to form the sample for the main study. The number of participants was 135; the number of valid cases was reduced to 127 by absence on either of the test days. The necessity for one treatment group to perform music made random assignment of subjects impractical; each school band comprised a treatment or control group. Decisions regarding the assignment to a specific treatment or control group were based on the pragmatic considerations of each group's available time and performance ability. Interviews with the band directors determined that none of the MPI compositions had been played by the participating bands in at least four years, and that few, if any, of the students would have heard any of the MPI excerpts. Two of the participating school bands received differing treatments (repeated listening or repeated performance); the third band served as the control group. All groups received pretest and posttest administrations of the MPI. The pretest and the posttest were identical, although the students were led to believe that they differed. LeBlanc (1979) had discovered that fifth grade students became impatient with a posttest procedure when they discovered it was identical to a previous test. To avoid such a reaction to the MPI posttest, students were told they were taking 22 a sflmilar test, and the appearance of the response form was altered by using paper of a different color, following LeBlanc's (1979) suggestion. During a treatment period of 15 school days, the group receiving treatment A (performance) rehearsed BYRD l and BYRD 2, two of the less-preferred MPI selections, exactly eight times. Additional repetitions would have improved the quality of performance, but Getz (1963, 1966) had suggested that additional repetitions might erode gains in preference due to fatigue. Treatment group B (listening) listened to vinyl disc recordings of BYRD l and BYRD 2 exactly eight times. To focus listener attention on the recordings, instrumental parts for the compositions were distributed to Group B. Group C (control) pursued its regular class activities, and was permitted to neither hear nor play the Byrd selections until posttest data had been collected. To disassociate the MPI and the researcher from the treatment procedures, pretesting took place 21 to 28 days before treatment began. The regular classroom teachers carried out the treatments in accordance with written instructions provided by the author (see Appendices B & C); students remained unaware of the relationship between the treatment and the MP1. The researcher returned to each school 21 days after treatment ceased to administer the posttest. 23 Pretest and posttest data were analyzed to (a) determine whether Groups A, B, and C initially shared similar preferences in music, (b) evaluate the stability of students' responses over time, and (c) measure any gains in preference and determine whether pretest and posttest scores differed significantly for any of the three groups. CHAPTER IV ANALYSIS OF DATA Test scores and descriptive statistics for this study were computed using a Statistical Package for the Social Sciences (SPSS) program. The presentation of data includes discussion of test scored for the treatment and control groups, reliability analysis of the pretest-posttest design, and evaluation of the experimental treatments. Group Test Results Pretest preference results for the total sample, treatment groups, and control group are shown in Table 4-1. Pretest preference rankings for all groups were combined for this exhibit, and are listed in rank-order of preference. Mean ratings are based on a scale of one to five, with five indicating greatest preference. Pretest mean ratings for all groups combined ranged from 2.43 to 4.54. Also shown in Table 4-1 are metronome markings (MM) for each MPI example. As expected, there appeared to be a strong relationship between tempo and music preference, with the faster compositions receiving higher preference ratings. Mean 24 25 OH N©.N OH NN.N OH mq.N oH mm.N Nq N Qmwm a wo.N m mm.N. m 0H.m m wN.N ooH ZOHmHm w ow.N w om.m m mm.N w mm.N NN H nmwm m NN.m m No.m m N©.N ,N oo.m NA mo H em.¢ m m®.m N wH.q H mN.q «SH m m.H mN.q H ON.q mwmoou mGOHuHmomEoo N .mcHumou coozuon mzmp Ho .oEHu Ho>o mmCNHNH some Ho zuHHHHmuw so momma mmB wcfiumMu muHHHanHomH Nma. ONm. MNm. .mmm m.amaumoam 0H mm.N oH N©.N I I I I I I I I N exam m ww.N m wo.N w Nm.N w Nm.N w oN.N N 0H.m ZOHmHm m qw.N w ow.N I I I I I I I I NH omwm N om.m N NN.m o mN.m N No.m N mN.N w No.N moot oumwcmum Ho muHca :H momcmzo mucouomoua asouw Ho moduHowma mnu HooHHmH mouoom NH MN. mH.H oH.q mm.q m o NN HQMm> Na. mo. om.m mm.m NH HH «H m Hemzzmm HN. Nm. MN.N No.N oH oH NH mo me. we. No.m mm.m HH NH NH m HHMZme mm. mm. mN.m No.m HH 0H mH mo ow. NH. Nm.m NN.N «H oH 0N m HHMZZMH mN. Hm. om.m NN.m N «H NN mo