A Study on Developing Learning Strategies in Violin Education

This study was conducted for the purpose of developing a valid and reliable learning strategies scale for students receiving violin education in Departments of Music at Fine Arts High Schools. The scale was applied to 391 violin students receiving education in the 11th and 12th grades in Departments of Music at Fine Arts High Schools in the provinces of Ankara, Eskişehir, Kayseri, Konya, Kırıkkale, Sivas, Niğde, Adana, Mersin, Isparta, Hatay, Osmaniye, İzmir, Aydın, Denizli, Kütahya, Manisa, Muğla, Bursa, İstanbul, Balıkesir, Çanakkale, Edirne, Tekirdağ, Samsun, Trabzon, Ordu, Bolu, Tokat, Malatya, Erzurum and Van. The 5-point Likert scale consists of 67 items. The data obtained after applying the scale were transferred to the SPSS Package Software. Explanatory factor analysis was then carried out on the basis of the data. As a result of the explanatory factor analysis, it was determined that the scale had six factors. In addition, the Cronbach’s alpha internal consistency coefficient of the scale was found to be 0.966.


Introduction
Music education, which is one of the most important aspects of arts education, is fundamentally the process of altering thoughts and behavior with regard to music and music-making.Music education can be separated into three main types: general music education, amateur music education and professional music education (Uçan, 2005, pp. 24, 30).
Professional music education aims for students to gain technical information, skills and knowledge regarding their field of choice.One of the most important dimensions of professional music education is instrument education.Fine Arts High Schools have an important role in teaching the core knowledge, skills and technical aspects required, as well as other areas of music, and in aiding students' transition from secondary education to higher education.Considered from this point of view, studies evaluating this process are of particular importance.One of the most important areas of instrument education is violin education.
Violin education is the process of permanently alteringthe psychomotor, cognitive and affective behaviors of individuals and helping them gain new behaviors in their lives through teaching themthe violin (Günay & Uçan, 1980, p. 8).
Violin education is a long and complex process in which psychomotor skills are intensively practiced intensively along side cognitive and affective aspects, and it involves many technical and musical difficulties.A failure to overcome the technical difficulties encountered in violin education and consequently to give the performance intended may not only decrease the desire of the student to study, but also cause her/him to develop negative attitudes towards the violin.Thus, it is very important to use approaches that increase the students' awareness in terms of playing both études and learning other works so that she/he can practice better and gain correct study habits.In this process, learning strategies play an important role in allowing the student to consciously follow her/his own learning process and evaluate herself/himself while learning the violin (Afacan & Çilden, 2017).Acquiring the basic skills to play the violin depends largely on students' ability to understand their own styles of learning and to direct their own learning (MEB, 2016).the process of learning and which are expected to affect their process of encoding information (Weinstein & Mayer, 1986, p. 315).Tay (2004) defines learning strategies as the students' efforts to make sense out of and appropriate the information that is presented to them during the learning-teaching process and in the personal preparations they make in their own minds.
Today, it is necessary that any knowledge gained is extensive and correct and the most important key to success is being effective and competent in learning.Some of the main elements or processes in effective learning are the learning strategies that are involved in "learning to learn".Learning strategies may enable the student to plan and direct her/his own learning and lead to easier and more permanent learning (Özer, 2002, p. 29).
Learning strategies make it easier for students to obtain, easily recall and transfer knowledge.A student who can use learning strategies in instrument education is likely to know how to obtain whatever knowledge is needed, and how to recall and use that knowledge while practicing her/his instrument.The student may transfer the knowledge gained from previously playing an étude or other work to a newpiece and thus achieve faster, easier and more permanent learning (Kılınçer, 2013).Learning strategies enable learners to make plans and carry out assessments, control their own learning, make decisions about learning processes and determine points of difficulty.Learners learn how to motivate themselves, how to remember and how to think by using learning strategies (Özkal & Çetingöz, 2006).
Examining the literature, it is seen that the most commonly used categorization of learning strategies today is that of Weinstein and Mayer.This study is thus based on their categorization since it is the one generally used in studies on instrument education.To this end, the scale used here was prepared by using the "Rehearsal (Repetition)", "Elaboration", "Organization", "Metacognition" and "Affective Learning" strategies in Weinstein & Mayer's (1986) learning strategies and the "Attention" strategy in Gagne & Driscoll's (1988) categorization.These strategies may briefly be explained as follows.
Rehearsal (Repetition) Strategies: Rehearsal strategies enable the student to intentionally choose and gain knowledge and are based on mental rehearsal (Özer, 1998).
Elaboration Strategy: Elaboration enables the student to associate new knowledge with former knowledge in her/his memory and encode it in long-term memory (Erden & Akman, 1995).
Organization Strategy: Organization enables the student to restructure new knowledge by using her/his prior knowledgein such a way that it will be more meaningful (Demirel, 2012, p. 140).
Affective Strategies: Affective strategies help to remove motivational and emotional obstacles to learning (Senemoğlu, 2005, p. 574).
Attention Strategies: Attention is the most important process for transferring the knowledge obtained from the environment into the short-term memory (Subaşı, 2000, p. 394).
Studies that have been conducted in our country have focused on the effect of learning strategies on students' success in instrument education (Akın, 2007;Ertem, 2003;Kılınçer, 2013;Şimşek & Balaban, 2010;Yokuş, 2009); attitudes (Özer, 2010); use of strategies (Akın, 2007;Ertem, 2003), and also on developing learning strategies in music for primary school students (Kocabaş & Sever, 2011).Studies conducted abroad, on the other hand, have focused on the learning strategies of conservatory students (Virkkula & Nissilä, 2017) and the learning styles and learning strategies of students receiving instrument education (Green, 2010).
In the literature, there is a limited number of studies regarding the use of learning strategies in instrument education and there is no valid and reliable assessment tool in studies aimed at the use of learning strategies in the national literature, especially with regard to violin education.As a result of this lack, this study aims to develop a valid and reliable assessment tool for the purpose of determining the use of learning strategies by violin students studying in Departments of Music in Fine Arts High Schools.

Research Design
The study was conducted using the survey model.The survey model is a research approach that aims to describe a past or a present situation as it is.The event, individual or object discussed in the study is described as far as possible with regard to the conditions in which it is found (Karasar, 2005, p. 77).

Target Population and Sample of the Study
The target population of the study consists of 11th and 12th grade violin students studying in Departments of Music in Fine Arts High Schools in Turkey in the school year 2017-2018.The sample consists of 391 violin students receiving education in the 11th and 12th grades in Departments of Music in Fine Arts High Schools in the provinces of Ankara, Eskişehir, Kayseri, Konya, Kırıkkale, Sivas, Niğde, Adana, Mersin, Isparta, Hatay, Osmaniye, İzmir, Aydın, Denizli, Kütahya, Manisa, Muğla, Bursa, İstanbul, Balıkesir, Çanakkale, Edirne, Tekirdağ, Samsun, Trabzon, Ordu, Bolu, Tokat, Malatya, Erzurum and Van.The first step in the scale development process was to examine previous scales, especially those found in foreign literature, regarding learning strategies in violin education (Güven, 2008;Kılınçer, 2013;Pintrich, Smith, Garcia, & McKeachie, 1991;Tay, 2002;Yokuş, 2009).As a result of the literature review it was determined that studies regarding learning strategies generally used the categorization of learning strategies by Weinstein & Mayer (1986) and Gagne & Driscoll (1988).In this study, the item pool of the scale was prepared by using "Rehearsal (Repetition)", "Elaboration", "Organization", "Metacognition" and "Affective" learning strategies from Weinstein & Mayer's (1986) categorization and the "Attention" strategy from Gagne & Driscoll's (1988) categorization.Then an item pool including these learning strategies as they apply to violin education was formulated.This item pool consisted of 111 items.Five options were presented next tothese items for the purpose of determining the level of students' agreement with the statements.The options were graded as follows: "Always" (5), "Frequently" (4), "Sometimes" (3), "Seldom" (2), "Never" (1).

Receiving Expert Opinions
Content validity relates to whether or not questions (items) in the assessment scale are appropriate for the assessment scale and represent the area it intends to assess, and is determined according to expert opinion (Karasar, 2005, p. 151).For this purpose, opinions were received from four experts for the external validity of the first draft of the scale consisting of 111 items regarding assessment and evaluation in violin education.
Regarding the statements, the expressions used and their spelling and punctuation, an expert working in Turkish education was consulted.The draft scale, corrected according to expert opinions, was reduced to 100 items.

Preliminary Application
In order to make the draft scale understandable to violin students in Fine Arts High Schools, four violin students in the 11th and 12th grades in the Department of Music, Kırşehir Neşet Ertaş Fine Arts High School were asked to read the "Learning Strategies Scale in Violin Education" and their opinions about the items were received.

The Study of Validity
The "Learning Strategies Scale in Violin Education", which was prepared as a 5-point Likert scale, was examined in terms of both content validity and construct validity.Content validity relates to the assessment tool as a whole and to what extent each item in it serves a purpose (Tekin, 1993, p. 45).The content validity of the "Learning Strategies Scale in Violin Education" was provided by examining the contents of the "Violin 11" and "Violin 12" courses in Fine Arts High School Departments of Music and formulating items regarding knowledge acquisitions.Construct validity, on the other hand, explains the results and to what the results are related.In other words, it shows how accurately the assessment tool measures an intangible phenomenon (Tavşancıl, 2006, p. 45).In psychological tests, construct validity is of primary importance (Tezbaşaran, 1996, p. 51).Construct validity was assessed by using factor analysis.

The Study of Reliability
The decision toexclude items from the scale were made by considering factor analysis and item-total correlation values.The cronbach's alpha reliability coefficient for the scale was calculated with the remaining items.

Data Analysis
In the study, the final form of the draft scale was applied to a total of 391 violin students receiving education in the 11th and 12th grades in Fine Arts High Schools and validity and reliability analyses were then carried out.

Ethical Aspects of the Study
In the process of developing the "Learning Strategies Scale in Violin Education", the necessary permissions were obtained from the General Directorate of Secondary Education, Ministry of National Education to conduct the study and these permission letters, along with the scale, were sent to the Fine Arts High Schools that participated in the study.

"Development of the "Learning Strategies Scale in Violin Education"
The decision whether each item can be included in the final scale is made by applying various item analyses to the data obtained from the draft scale (Tezbaşaran, 1996, p. 51).In order to assess the initial "Learning Strategies Scale in Violin Education" prepared with 100 items, item analyses based on the difference of sub-super group averages, and total item correlation analyses were carried out.

Item Analysis Based on the Difference of Sub-Supergroup Averages
In order to determine the distinguishing powers of scale items, the t-value of the difference between the score averages of super group and sub group music literacy was calculated for each item.The learning strategies' scores of the students were primarily arranged from large to small; each sub-supergroup comprised 106 students constituting 27% of the entire scale.In the analysis, as the t-test result for item averages was p<.05; it was concluded that scale items contributed to assessing the condition that was intended to be assessed.

Total Item Correlation
First, the total item correlation values of each item in the 100-item learning strategies scale were surveyed.
Examining Table 3, it is seen that total item correlation values vary between 0.261 and 0.678 for all items in the scale.Büyüköztürk (2003) stated that scale items whose total item correlation coefficients were 0.30 and larger distinguished individuals very well, items remaining between 0.20-0.30might have to be included in the test or be corrected when necessary and items smaller than 0.20 should not be included in the test.It is an undesirable condition when total item correlations are negative or smaller than 0.30, because it is accepted that items with these values do not distinguish those that submit and do not submit positive views to scale items as well.Total item correlations are expected to be positive and larger than 0.30.Thus, the scale item for "2" learning strategies, whose total item correlation value was smaller than 0.30, was excluded.99 items remained in the scale.

Factor Analysis
It is possible to define factor analysis as multivariate statistics aiming to explore a limited number of conceptually significant new variables (factors, dimensions) by gathering numerous interrelated variables (Büyüköztürk, 2002, p. 472).The convenience of the data for factor analysis can be examined viathe Kaiser-Mayer-Olkin (KMO) coefficient and the Bartlett's Test.If the KMO is larger than 0.60 and the Bartlett's Test significant, it signifies that the data are convenient for factor analysis (Büyüköztürk, 2003, p. 120).
Testing the competence of the "Learning Strategies Scale in Violin Education" for factor analysis, the Kaiser-Mayer-Olkin (KMO) coefficient was found to be 0.948.Kaiser's values larger than 0.90 are described as "excellent" (cited by Afacan & Aydoğdu, 2006, p. 193).Similarly, Sharma (1996, p. 116) evaluated situations where the KMO coefficient value was 0.90 as "excellent" (cited by Eroğlu, 2008, p. 322).Thus, the learning strategies scale being prepared may be described as excellent.In addition, the Bartlett's Test (χ 2 =26555, 059; df=4851; p<.05) was found to be significant (Table 4).After conducting the item analyses and investigating the convenience of the scale for factor analysis, explanatory factor analysis, which is one of the factor analysis methods, was applied to determine the factor structure of the scale.

Explanatory Factor Analysis
The varimax rotation technique was applied in the explanatory factor analysis.After applying the first rotation, the scale was found to have six factors.A total of 17 items (13, whose factor load values were smaller than 0.30; and 91, 63, 86, 35, 55, 68, 31, 32, 72, 10, 17, 3, 85, 1, 89 and 20, whose difference between two high factor load values was smaller than 0.10) were excluded from the scale.After excluding these items, the varimax rotation procedure was applied for the second time and the items 70, 29, 33, 18 that were available in both factors were excluded from the scale.In addition, as items 100, 99, 95, 96, 88, 87, 90, 42, 14, 62 and 30 were found under different factors while naming the factors, they were excluded from the scale.Then varimax rotation was applied for the third time on the basis of the 67-item scale.Table 5 shows the results of rotated principal components analysis.The 67 learning strategies items that were included in the analysis were collected under six factors whose eigenvalues were larger than 1 (21.144, 5.627, 3.847, 2.143, 1.964, 1.522).These factors are shown in Table 6.(Kline, 1994;Scherer at al., 1988).With this variance value above 40%, the scale may be accepted as a 6-factor scale (Table 6).Figure 1 shows eigenvalues for the 6 factors.

FACTOR I
Factor I constitutes 12.158% of the total variance.Table 7 shows the data of Factor I, which is comprised of 17 learning strategies items whose factor load values vary between 0.483 and 0.686.Examining the 17 learning strategies items in Factor I, it was seen that there were items assessing the dimensions of metacognition.Factor I was named "Metacognition Strategies".The cronbach's alpha internal consistency coefficient of Factor I was found to be 0.919.

FACTOR II
Factor II constitutes 11.836% of the total variance.Table 8 shows the data of Factor II, which is comprised of 13 learning strategies items whose factor load values vary between 0.471 and 0.841.Examining the 13 learning strategies items in Factor II, it was seen that thelearning strategies involved items regarding attention.Factor II was named "Attention Strategies".The cronbach's alpha internal consistency coefficient of Factor II was found to be 0.938.

FACTOR III
Factor III constitutes 9.764% of the total variance.Table 9 shows the data of Factor III, which is comprised of 13 learning strategies items whose factor load values vary between 0.449 and 0.724.Examining the 13 learning strategies items in Factor III, it was seen that they consisted of items involvingorganization strategies.Factor III was named "Organization Strategies".The cronbach's alpha internal consistency coefficient of Factor III was found to be 0.924.

FACTOR IV
Factor IV constitutes 9.248% of the total variance.Table 10 shows the data of Factor IV, which is comprised by 8 learning strategies items whose factor load values vary between 0.680 and 0.793..71826 I understand the difference between harmonic features (tone alteration, modulation) used in the étude and work and harmonic features used in previously learned études and works.
.697 21 I understand the similarity between the string technique used in the étude and work and the string technique used in previously learned études and works.
.690 25 I understand the similarity between harmonic features (tone alteration, modulation) used in the étude and work and harmonic features used in previously learned études and works. .680 Examining the 8 learning strategies items in Factor IV, it was seen that they consisted of items containing elaboration strategies.Factor IV was named "Elaboration Strategies".The Cronbach's alpha internal consistency coefficient of Factor IV was found to be 0.934.

FACTOR V
Factor V constitutes 6.639% of the total variance.Table 11 shows the data of Factor V, which is comprised of 11 learning strategies items whose factor load values vary between 0.340 and 0.721.Examining the 11 learning strategies items in Factor V, it was seen that they consisted of items containing rehearsal (repetition) strategies.Factor V was named "Rehearsal (Repetition) Strategies".The Cronbach's alpha internal consistency coefficient of Factor V was found to be 0.859.

FACTOR VI
Factor VI constitutes 4.454% of the total variance.Table 12 shows the data of Factor VI, which is comprised of 5 learning strategies items whose factor load values vary between 0.556 and 0.621.Examining the 5 learning strategies items in Factor VI, it was seen that they consisted of items containing affective strategies.Factor VI was named "Affective Strategies".The Cronbach's alpha internal consistency coefficient of Factor VI was found to be 0.788.

Findings Regarding the Correlation between the "Learning Strategies Scale in Violin Education" and Its Sub-Factors
Table 13 shows the results regarding the correlation of sub-factors with each other and the entire scale determined after factor analysis.According to the correlation analysis that was conducted for determining the relationship between general averages of items in the entire scale and the factors, it is seen that correlation coefficients vary between 0.364 and 0.845.Büyüköztürk (2003) interpreted a value ofbetween 0.70-1.00 to have an absolute value as a "high level" relationship; on the other hand, a value between 0.70-0.30indicates a "moderate level" relationship.It is possible to state that there is a positively moderate and high relationship between the "Learning Strategies Scale in Violin Education" and its sub-factors.This result shows that each factor has a common purpose (p.32).

Findings Regarding the Reliability of the Internal Consistency Level of the "Learning Strategies Scale in Violin Education"
After the factor analysis, the reliability of the 67-item "Learning Strategies Scale in Violin Education" was calculated.The Cronbach's alpha internal consistency coefficient of the learning strategies scale was found to be 0.966.This value may indicate that the scale is "highly" reliable (cited by Tavşancıl, 2006).The reliability analysis of the "Learning Strategies Scale in Violin Education" according to factors was conducted using the Cronbach's alpha reliability coefficient.Table 14 shows reliability analysis values for each factor.

Results and Discussion
This study was conducted for the purpose of developing a valid and reliable learning strategies scale for students studying violin in Departments of Music in Fine Arts High Schools.391 violin students receiving education in the 11th and 12th grades in Fine Arts High School Departments of Music in the provinces of Ankara, Eskişehir, Kayseri, Konya, Kırıkkale, Sivas, Niğde, Adana, Mersin, Isparta, Hatay, Osmaniye, İzmir, Aydın, Denizli, Kütahya, Manisa, Muğla, Bursa, İstanbul, Balıkesir, Çanakkale, Edirne, Tekirdağ, Samsun, Trabzon, Ordu, Bolu, Tokat, Malatya, Erzurum and Van participated in the study.The 100-item "Learning Strategies Scale in Violin Education" that was prepared as a draft was reduced to 99 items with item analysis, which was among the data obtained as a result of pilot study and was based on the difference of sub-super group averages and total item correlation values, and then to 67 items with the varimax rotation technique that was applied three times as a result of factor analysis and exclusion of items in different factors.As a result of the explanatory factor analysis, it was determined that the "Learning Strategies Scale in Violin Education" consisted of 6 factors.These were "Metacognition Strategies", "Attention Strategies", "Organization Strategies", "Elaboration Strategies", "Rehearsal (Repetition) Strategies" and "Affective Strategies".The reliability coefficients of the aforementioned factors were 0.919, 0.938, 0.924, 0.934, 0.859, and 0.788 respectively.In general, the Cronbach's alpha reliability coefficient of the scale was determined to be 0.966, which indicates that the scale is "highly" reliable.
Learning strategies are defined not only as the specific ways students acquire the intended learning, but also the ways used by students to encode information in their short and long-term memories (Yüksel, 2011, p. 151).
It is important to enable individuals to develop their learning strategies and that they are able to practise these strategies in order to achieve the success they hope for.Determining which learning strategies are used by students and how often they use them, revealing whether their academic achievement varies according to the learning strategies used, and determining effective learning strategies, all constitute important tasks.These activities should be constantly updated and strengthened in the literature with regard to different courses and at different educational stages so that education goals can be achieved (Şahin & Uyar, 2013).It is recommended that research be conducted to determine to what extent the learning styles of Fine Arts High School students affect their learning strategies.

Table 2 .
T-test results for item averages of 27% sub and 27% supergroups of the scale

Table 3 .
Total item correlation values

Table 4 .
Result of the KMO and the Bartlett's Test

Table 5 .
Rotated principal components analysis

Table 6 .
Eigenvalues of the factors of the six-factor "Learning Strategies Scale in Violin Education", number of items in the factor, factor variances, variance values that increase as factors are added The variance explained by these six factors regarding the scale is 54.099%.At least 40% of the general variance of the scale is explained is considered competent in terms of behavioral sciences

Table 7 .
Learning strategies items and factor loads in Factor I

Table 8 .
Learning strategies items and factor loads in Factor II

Table 9 .
Learning strategies items and factor loads in Factor III

Table 10 .
Learning strategies items and factor loads in Factor IV

Table 11 .
Learning strategies items and factor loads in Factor V

Table 12 .
Learning strategies items and factor loads in Factor VI Table13.Correlation regarding the "Learning Strategies Scale in Violin Education" and its sub-factors

Table 14 .
Reliability analysis values for the "Learning Strategies Scale" in violin education and its factors