The development of a resistance to school music scale

Abstract

This study aimed to develop a valid and reliable measurement tool measuring secondary school students’ resistance to school music. Construct and content validity studies based on expert reviews, principal component analysis, confirmatory factor analysis, convergent validity, divergent validity, and Horn’s parallel analysis were conducted to provide additional evidence within the scope of the study. For reliability studies, Cronbach’s alpha and composite reliability analysis techniques were employed. A 15-item construct consisting of two factors, and explaining 63% of the total variance with their factor loadings ranging between 0.588 and 0.859 was obtained by evaluating the analyses conducted concerning the factor structure and Horn’s parallel analysis. When the sub-factors beneath the factor structure were examined, there were 11 items in the first sub-factor and four items in the second. In the model tested with the confirmatory factor analysis, the results showed that the convergent and divergent validity criteria were satisfied. Finally, it is thought that this research will contribute to the field of music education as the resistance of secondary school students to school music can be determined with the developed scale.

Keywords

resistance school music secondary school students scale development music teacher

From a holistic perspective, education is vital in supporting the cognitive, affective, and physical development of individuals. In addition, it is important to recognize that the affective characteristics students gain in the process bear a strong meaning. The affective domain is known to be important in education (Pierre & Oughton, 2007). For individuals to be capable of performing effective learning and sense-making in the cognitive domain, complementary skills (communication, value, organization, etc.) in the affective domain need to be taught and developed (Griffith & Nguyen, 2006). They hold an important place in students’ access to knowledge, achievement of course objectives, and their level of achievement at school (Kesici & Aşılıoğlu, 2017). In addition to contributing to academic achievement, affective elements are believed to shape individual behaviors regarding the teaching–learning process and are crucial elements to develop. These are the features that include individuals’ emotions and emotional dispositions such as interest, self-confidence, love, hate, and so forth (Biber, 2012; Gömleksiz & Kan, 2012).

Students develop many behaviors toward the elements of their school life in line with their affective characteristics. These behaviors can be positive as well as negative. When students exhibit behaviors that cause maladaptation in their school life, they are called resistance behaviors (Canoğulları & Sarı, 2015). In this context, it is thought that one of the most common negative affective behaviors encountered in teaching-learning processes is resistance. Resistance is mostly defined as student resistance in education. Student resistance is the behavior of challenging new social structures at school that do not fit with students’ thoughts, lifestyles, and expectations (Hendrickson, 2012; Yüksel, 2003). On the contrary, resistance represents the behavior of students in the face of a negative situation they experience in school, where learning opportunities are denied (Seidel & Tanner, 2013; Tolman & Kremling, 2017). In addition, it is a theoretical and ideological construct that provides a significant focus for analyzing the relationship between school and society (Giroux, 2001). When examining definitions of student resistance, researchers define it by dividing it into types. Considering both the negative and positive connotations of resistance, Burroughs et al. (1989) and Seidel and Tanner (2013) explained student resistance in terms of two categories of constructive or destructive behaviors. They argue that resistance activities performed to disrupt task behaviors in the classroom and limit students’ own and others’ learning are destructive resistance, whereas resistance activities performed to improve task behaviors and send constructive feedback for the development of the instructor are constructive resistance.

When the definitions are examined, it is thought that student resistance is an important affective element and it also enables instructors to have observations and information about the learning–teaching process. Student resistance provides a better understanding of the learning process by instructors in line with its characteristics (Mcfarland, 2001). In this context, detecting student resistance behaviors is important and should be examined in order for the learning–teaching process to be successfully carried forward by the instructors.

Music is a field of academic study where teachers support students’ cognitive, affective, and psychomotor learning (Lehmann & Davidson, 2002). Music can help students express their emotions, attitudes, and feelings by making their lives meaningful (Smith, 1994). As music provides rich experiences and contains emotions, it can contribute to students’ quality of life by affecting their thoughts and behaviors (Wolff, 1978). In line with the explanations made, it is thought that the affective domain is an important element in school music; the affective behaviors that students form toward school music are thought to have a decisive role. In this context, it is important to determine the resistance behaviors that students form toward school music.

In line with the regulations determined by the Ministry of National Education in Turkey (MoNE), eighth graders (aged 14–16 years) are placed in appropriate high schools based on their scores on the High School Entrance Exam (HSEE). Music is taught in secondary schools but not included in the HSEE. Therefore, this may affect the interest and significance shown by students, who tend to base their study patterns on the HSEE, causing them to show negative behaviors toward school music courses. At the same time, potential negative thoughts that families attribute to music and the music course (being seen as unimportant compared with other courses, an unneeded area in life, etc.) can affect the students to whom they are parents, and thus, students can develop negative thoughts toward the music course and the music teacher. This situation may cause students to consciously and deliberately display negative behaviors toward the music course and the music teacher and develop resistance behaviors to the music course. As a result, it is of great importance to determine students’ resistance behaviors toward the music course by discovering the problems experienced in this course. There is no measurement tool measuring secondary school students’ resistance behaviors to school music in the national and international literature. Thus, the current study is significant.

Method

This study aimed to develop a measurement tool that determines secondary school students’ resistance to school music. In this context, a Likert-scaling technique was used in this study. This technique is one of the itemized rating scale techniques under non-comparative scaling. Non-comparative scaling techniques are a set of scaling methods in which items are scaled independently of each other and are of various types such as graphic rating scales, Likert scales, semantic differential scales, and Stapel scales (Swaminathan, 2010). In the scale development process, a summative rating approach of scaling was used. Of the different scaling approaches, the summative rating approach focuses on placing individuals in different places on the scale according to respondents’ responses to the items (Anıl & İnal, 2018; Crocker & Algina, 1986).

Participants

Two different study groups were used in the study. The principal component analysis (PCA) was performed with the first study group, and the confirmatory factor analysis (CFA) was performed with the second study group. The study groups included in PCA and confirmatory factor analyses were independent of each other. The PCA study group consisted of 454 students. As a result of examining the assumptions of the analysis, the number of observations was reduced to 316, and the analysis was continued with 316 observations. Tabachnick and Fidell (2013) stated that the number of observations should be a minimum of 300 in exploring the constructs. Another analysis conducted for the construct validity of the scale development process was CFA. In this phase, the final form created through PCA was administered to an entirely different group than the PCA group. In this context, the CFA study group (the second study group) consisted of 385 students. As a result of examining the CFA assumptions, the number of observations was reduced to 385 and the analyses were conducted with 385 observations. The distribution of the study group included in the PCA-CFA is shown in Table 1 by gender- and grade-level variables.

Table 1.

Distribution of the PCA-CFA Study Group by Gender and Grade Level.

		PCA		CFA
		Frequency	Percentage	Frequency	Percentage
Gender	Male	113	35.6%	163	42.4%
Gender	Female	203	64.4%	222	57.6%
Grade	5	87	27.6%	93	24.1%
	6	71	22.5%	82	21.3%
	7	70	22%	96	24.9%
	8	88	27.9%	114	29.7%
Total		316		385

Procedure

To obtain information about the psychological element addressed within the scope of the study, the scale development process began with a literature review. After the literature review, 85 secondary school students were invited to express their views regarding the variables addressed within the scope of the study through open-ended questions. In the meantime, information was obtained about potential resistance behaviors experienced to school music through focus group interviews with six music teachers. After the investigation, the item writing stage was initiated. In the item writing stage, the types of resistance and resistance expressions given in the international literature were taken into consideration (Burroughs et al., 1989; Chan & Treacy, 1996; Erickson, 1984; Seidel & Tanner, 2013). In this context, statements representing constructive, destructive, active, passive, conforming, sensible, and credible behaviors were included, and an item pool of 95 items, thought to reveal the level of resistance to school music, was prepared for expert validation.

One of the important stages of the scale development process is the expert validation study. In this context, an expert review form including 95 items was prepared based on a four-point rating format, and the experts were requested to express their suggestions, if any. The review form was presented to eight experts (three music teachers, two measurement and evaluation lecturers, two curriculum and instruction lecturers, and one music education lecturer). The analysis stage was initiated after obtaining the experts’ feedback. In the analysis stage, Davis’s (1992) technique was used to compute the content validity indexes of the items. The minimum value of 0.78 (α = 0.05) for items suggested by Yurdugül (2005) was considered as a criterion. As a result of the analysis, 20 items not yielding the minimum value (SVR (Scope Validity Rate) > 0.78, α = 0.05) were excluded. The remaining items were examined following suggestions provided by the experts. The researchers decided that 21 items measuring the same objective, repeating each other, and remaining out of scope should be excluded. As a result of these studies, a pilot form consisting of 54 items was prepared. In this study, the data were collected on a voluntary basis from the participant groups. The PCA study group data were collected through digital environments, whereas the CFA study group data were collected one-to-one through forms by the researchers.

Data analysis

PCA and CFA were conducted to test the construct validity of the data collected. In the CFA, convergent validity studies were conducted to obtain information concerning the relationships between the items under the factors. In addition, divergent validity studies were performed to examine the maximum values of the relationships by creating combinations between the factors. Horn’s Parallel Analysis was also used to provide additional evidence on construct determination.

Exploratory factor analysis was conducted to reveal the latent structure of the scale and the relationship between the items. Primarily, necessary assumption tests (missing data, normality, univariate and multivariate extreme values, and multicollinearity) were conducted to perform the PCA. In determining the extreme values, the observations were examined and excluded from the data set. However, when the extreme values were due to multicollinearity issues, the item causing the problem was examined and excluded from the data set. As a result of these inspections, six observations were found to have missing data, and the data were normal. When the univariate and multivariate extreme values were examined, three observations not having Z-values between −4 and +4 were univariate outliers. Also, an examination of the Mahalanobis distances (χ²₅₄, 0.001 = 80.37377) indicated that 129 observations were multivariate outliers. In line with these results, the observations were excluded from the analysis and the study proceeded with 316 observations. To determine multicollinearity problems, collinearity statistics were examined. In this context, multicollinearity problems were examined by looking at the Tolerance and Variance Inflation Factor (VIF) values (Tolerance > .20; VIF < 5). As a result, 27 items with multicollinearity problems were determined and excluded. Therefore, the study proceeded with 316 observations and 27 items in the observation set. After the examination of assumptions, Kaiser–Meyer–Olkin (KMO) and Bartlett tests were conducted to examine whether the data were suitable for PCA. The value obtained from the KMO test should be at least .80 (Alpar, 2014). As such, the Bartlett test should be significant for factor analysis suitability (Tabachnick & Fidell, 2013). PCA, which aims to maximize the variance explained regarding the feature under measurement, was determined as the factor extraction technique in determining the clustering patterns of factors (Tabachnick & Fidell, 2013). After examining the suitability for factor analysis, it was examined whether there was a relationship between factors. According to the results of the covariance matrix, no relationship was observed between factors, and thereby the varimax orthogonal rotation technique was preferred, a technique that maximizes the variance. The relationship between factors can also be examined based on a covariance matrix (Çokluk et al., 2018). After these investigations, it was found that the common variance of the items was .50, the factor loadings were .45, and the difference between the factor loadings was at least .10 in the data analysis (Büyüköztürk, 2011; Tabachnick & Fidell, 2013).

When deciding on the number of factors, the scree plot, the number of factors with eigenvalues of greater than 1 (Kaiser’s method), and the total variance explained were taken into account. Meanwhile, given that the scree plot involves subjective decisions of the researchers and risks of subjective evaluation (DeVellis, 2017), Horn’s parallel analysis, which allows determining the number of factors by making objective decisions through additional analysis, was also used. Horn’s parallel analysis enables getting an objective conclusion by creating a simulative data set that equals the number of participants and variables in the real data set and by comparing the eigenvalues between the two data sets (DeVellis, 2017). As a result of the factor analysis, suitable factor structures were obtained. However, one item (Item 54) under the second factor, causing problems in terms of naming the factor, was excluded from the factor structure, and the PCA was repeated. As a result of the analysis, the final form with suitable factor structures was revealed.

In addition, CFA was performed by presenting new evidence to strengthen and support the construct validity of the scale obtained as a result of the PCA. In this context, necessary assumption tests performed in the PCA were also performed for CFA. As a result, there was no univariate outlier, as the Z-values ranged between −4 and +4. Furthermore, the data had a normal distribution, one observation included missing data, and 37 observations were multivariate outliers in terms of Mahalanobis distances (χ²₁₅, 001 > 37.697). Thus, these observations were excluded from the analysis, and the study was continued with 385 observations. Tolerance and VIF values were examined for multicollinearity problems between the items (Tolerance > .20; VIF < 5). There was no multicollinearity problem, as the Tolerance values ranged between 0.394 and 0.571 and the VIF values ranged between 2.536 and 1.750. In this context, the CFA proceeded with 385 observations.

As a result of the factor analysis performed with 385 observations, the model-data fit was determined by obtaining the factor loadings, standardized values, t-values, and fit indices (Normed Fit Index [NFI], Non-Normed Fit Index [NNFI], Comparative Fit Index [CFI], Root Mean Square Error Of Approximation [RMSEA], and Standardized Root Mean Square Residual [SRMR]). In the meantime, Composite Reliability (CR), Average Variance Extracted (AVE), Maximum Shared-Squared Variance (MSV), and Average Shared-Squared Variance (ASV) were examined for convergent and divergent validity. In determining these values, the Maximum Likelihood (ML) method, which is based on covariance matrices when there are continuous variables in CFA, was used. Model data goodness-of-fit criteria were evaluated according to NFI > 0.95, CFI > 0.95 (Byrne, 2011), and 0.05 ⩽ SRMR ⩽ 0.10 (Schermelleh-Engel et al., 2003). As a result of the analysis, the model-data fit was achieved.

As CR is calculated from the explained variance and error variances based on standardized loadings obtained within the scope of the factor analytic model, it could be considered as an alternative reliability coefficient for congeneric measurements (Raykov, 1997). At the same time, convergent validity studies were conducted by examining the AVE, and divergent validity studies were conducted by examining the MSV and the ASV based on the CFA. The CR and AVE values were obtained based on the estimated standardized loadings, error variances, and the correlations between factors in multifactorial constructs of measurement models. In addition to these values, the MSV values were obtained for the proposed multifactorial model. In the conformity evaluation of the resulting evidence, the criterion of CR > 0.70, AVE < CR, and AVE of 0.5 or greater for convergent validity and MSV < AVE for divergent validity were used (Hair et al., 2014; Vinzi et al., 2010; Yaşlıoğlu, 2017).

Results

This section includes findings related to PCA, CFA, reliability, convergent, and divergent validity studies.

Results—PCA

After completing the necessary assumption tests (missing data, normality, univariate and multivariate outliers, and multicollinearity problem), the PCA was initiated with 316 observations and 27 items.

The KMO and Bartlett tests were conducted to check the suitability of the collected data for the factor analysis, and the explained common factor variances (communalities) were examined. When the KMO value (0.950 > 0.50) was examined, a perfect fit was achieved and the chi-square test was significant (χ² = 2991.096 p < .05) from Bartlett’s test. This result indicated that the data were suitable for the factor analysis. When the explained common factor variances were examined, the values ranged between 0.460 and 0.759. As a result of the analyses performed concerning the factor structure, a two-factor structure with eigenvalues greater than 1.00 was observed. The two factors accounted for 63% of the total variance, where the first and the second factors accounted for 56% and 7% of the total variance, respectively. The scree plot relating to the eigenvalues of the factors is illustrated in Figure 1.

Figure 1.

The Scree Plot of Factor Eigenvalues.

Figure 1 shows that the eigenvalues between the first and second factors plummet in the scree plot and then exhibit a horizontal move. In this context, a two-factor structure is observed. In addition to the scree plot, Horn’s parallel analysis was used to be able to make an objective decision concerning the number of factors through an extra analysis (Table 2).

Table 2.

Findings of Horn’s Parallel Analysis.

Factor	Real eigenvalue	Simulative eigenvalue
1	13.92	1.58
2	1.51	1.49
3	1.15	1.42
4	0.80	1.37

Table 2 shows the values obtained from the real and simulative eigenvalues. According to Table 2, the data produced from the simulative eigenvalues start getting larger than the real eigenvalues after the second factor. In Horn’s parallel analysis, the point where the eigenvalues obtained from the simulative data are greater than the eigenvalues obtained from the real values is the main criterion in determining the number of factors (Ladesma & Valero-Mora, 2007; O’Connor, 2000; Watkins, 2006). In this context, considering the scree plot of factor eigenvalues given in Figure 1 as well as Horn’s parallel analysis and the explained total variance (63%) given in Table 2, it was concluded that the resultant factor structure consisted of two factors.

Table 3 presents the factor loadings of items in the factor structure and the common variances explained.

Table 3.

Factor Analysis Results.

Factors	Items		Factor loadings	Common variance
Resistance to music course delivery	1	I skip class because I dislike the songs sung in music class.	.727	.582
	2	I do not come to class on singing days.	.693	.597
	3	I do not attend music class because I do not like to play an instrument.	.798	.681
	4	I change place when it is my turn to play an instrument.	.792	.663
	5	I make disruptive gestures when playing an instrument in class.	.732	.647
	6	I make excuses not to play an instrument.	.781	.707
	7	I do not bring my instrument to class on instrument-playing days.	.676	.650
	8	I make excuses not to participate in musical note reading/writing practices.	.692	.618
	9	I pretend to be reading in group musical note reading practices.	.667	.571
	10	I am not interested in music because the information I learn in music class will not be useful to me in the future.	.588	.460
	11	I pretend to be listening in music class because there are no questions from music class in the high school entrance exams.	.691	.637
Resistance to music teacher	12	I talk to my friend when the music teacher is teaching.	.682	.613
	13	I ask the music teacher irrelevant questions.	.648	.536
	14	I look outside when the music teacher is teaching.	.756	.721
	15	I scribble in my notebook when the music teacher is teaching.	.859	.759

When Table 3 is examined, factor loadings under the first factor with eleven items range between 0.588 and 0.798, and factor loadings under the second factor with four items range between 0.648 and 0.859. Considering the explained common variance, the values obtained vary between 0.460 and 0.759. According to all these findings, the results were determined to be highly valid. In this context, the construct measured by the items in the first factor was named “resistance to music course delivery” which accounted for 56% of the total variance. The construct measured by the items in the second factor was named “resistance to the music teacher,” which accounted for 7% of the total variance. The construct that the two factors measured together was named “resistance to school music” by the researchers, and the construct obtained accounted for 63% of the total variance.

Results—CFA

CFA was performed to provide additional evidence for the construct validity of the factor structure revealed by PCA. Standardized values and t-values yielded by CFA are shown in Figure 2.

Figure 2.

Path Diagrams With Standardized Factor Loadings and t-Values.

According to Figure 2, the factor loadings ranged between 0.60 and 0.78, and the t-values were significant (p < .001). In addition, there was a significant moderate relationship between the factors obtained (r = .68, p < .05). The results obtained also provide evidence for divergent validity. According to Kline (1998), the sub-dimensions of the scale are not required to measure the same feature but are expected to be consistent with each other.

As a result of CFA, when the fit indices of the related model were examined, it was determined that NFI = 0.95, NNFI = 0.95, CFI = 0.96, RMSEA = 0.10, and SRMR = 0.072. Given that the RMSEA value did not meet the acceptable fit criteria, it was evaluated along with the SRMR fit index. An overall evaluation of all findings yielded by CFA showed that the model-data fit was achieved, as the standardized factor loadings were high, t-values were significant, and model fit indices were within good model criteria.

Results—convergent and divergent validity

Convergent and divergent validity studies were conducted in line with the findings obtained from CFA within the scope of the study. For convergent validity, the CR and maximum AVE values for the relevant factors were examined. The CR value was 0.90 for the first factor and 0.90 for the second factor. Also, the AVE value was 0.50 for the first factor and 0.55 for the second factor. AVE values above 0.50 and lower than CR values (AVE > 0.50; AVE < CR) indicate convergent validity. This result reveals that the items under their respective factors are highly correlated with each other.

For divergent validity, the MSV and maximum ASV values were computed. The MSV and ASV values computed were equal to each other and 0.46 due to the two-factor structure. MSV and ASV values (MSV: 0.46; ASV: 0.46) smaller than AVE values (MSV < AVE; ASV < AVE) indicate divergent validity. This result shows that the factors are independent of each other and that single operations with these factors yield valid and reliable results.

Results—reliability

Cronbach’s alpha reliability coefficients and CR values were examined for the reliability study of the 15-item school music resistance scale obtained through the analyses. Cronbach’s alpha obtained for the whole factor structure through PCA procedures was 0.938. As per the results of investigations performed regarding the sub-factors, the Cronbach’s alpha values were 0.934 and 0.816 for the first and the second factors in the factor structure, respectively. In addition, the CR coefficient obtained through CFA procedures was 0.90 for the first factor and 0.83 for the second factor. All these results suggest that measurements made with the scale developed will be reliable.

Discussion

PCA

A measurement tool measuring secondary school students’ resistance to school music was developed within the scope of the study. In this context, factor analysis was conducted to determine the structure of the scale and the relationship between items. Factor analysis is grounded on bringing together variables related to each other within a general concept (Aksu et al., 2017). Considering the results of PCA, a scale consisting of 15 items with a two-factor structure was developed. The items are successfully clustered within the factors, the relationships between the factors and items are significant and strong, and the items successfully represent the psychological element desired to be measured. Considering the first factor, the items within the factor indicate statements concerning the music course teaching-learning process. In line with the statements indicated by the items, the first factor was named “resistance to music course delivery” by the researchers. The items under this factor represent behaviors that students have previously sketched out and mostly represent destructive, active, and passive resistance behaviors. In the meantime, 11 items of the 15-item scale are included in the first factor, which represents the general structure of the scale. In this context, students’ resistance behaviors to the music course seem to be mostly related to the teaching–learning process. Considering the second factor, the items under the factor consist of statements regarding the music teacher. Herewith, the second factor was named “resistance to the music teacher” by the researchers. There were four items under the “resistance to the music teacher” factor. It was observed that the items mostly include destructive, active, and passive resistance statements regarding the music teacher.

Prior to analysis, the researchers included statements of resistance to school music, music course content, music teacher, and music course classmates, during the item pool preparation stage. However, when the items in the structure revealed by the PCA were examined, it was concluded that the main elements that formed the resistance behavior toward the course were the statements related to music course delivery and the music teacher. Thus, it may be thought that music course delivery and the music teacher are important factors in the students’ resistance behaviors toward school music. What teachers say and do during the class may influence student behaviors and thinking (Kearney et al., 1991). Moreover, teachers should pay attention to many variables that would affect classroom management and student behavior in order for students to actively participate in lessons (Gündüz & Bozkuş, 2016). At the same time, teachers’ classroom management has been proved to be effective in preventing undesired behaviors of students and encouraging their achievement (Burroughs, 2007). In this context, the communication that music teachers establish with students is important in forming an affective element of resistance to contents, methods, and techniques they choose during the course.

CFA

CFA was performed to provide additional evidence for the validity and reliability of the measurement model yielded by the PCA. CFA is a multivariate statistic that examines the conformity of the data obtained based on the structural equation modeling to the model determined by the researcher (DeVellis, 2017; Tabachnick & Fidell, 2013). CFA suggested that the construct obtained was a valid construct. At the same time, convergent and divergent validity studies performed with CFA show that the items within factors are correlated and diverge from each other. As per findings obtained by examining the CR, the model yielded a reliable and valid result and showed that the measurements made by the factors alone were valid and reliable. These results indicate that the researchers have successfully examined the construct and content of the psychological element addressed, the written items suit the psychological element addressed in the study, and the items validly cover the structure of the model. In the meantime, the results obtained through CFA show that the “Resistance to School Music Scale” developed may yield valid and reliable results in future studies.

Conclusion

As a result of the construct validity studies conducted within the scope of the study, a valid and reliable measurement tool measuring the secondary school students’ school music resistance behaviors was developed. This measurement tool consists of two factors and 15 items. The first factor of the scale includes 11 items, and the second factor includes four items. The items under each factor are highly correlated, and the factors in the scale have a structure that could be used in single operations. The researchers named the construct measured by the items in the first factor as “resistance to the music course delivery”, and the construct measured by the items in the second factor as “resistance to music teacher.” Furthermore, the scale developed through this study was named “Resistance to School Music Scale” (See in Appendices 1 and 2.).

Footnotes

Appendix 1

Appendix 2

School Music Resistance Scale.

(1) I skip class because I dislike the songs sung in the music class.

(2) I do not come to class on singing days.

(3) I do not attend the class because I do not like to play an instrument.

(4) I change my place when it is my turn to play an instrument.

(5) I make gestures that disrupt the integrity when playing an instrument as a class.

(6) I make excuses not to play an instrument.

(7) I do not bring my instrument to class on instrument playing days.

(8) I make excuses not to participate in musical note reading/writing practices.

(9) I pretend to read in group musical note reading practices.

(10) I am not interested in the course because the information I learn in music class will not be useful to me in the future.

(11) I pretend listening to the lesson because there are no questions from the music course in the high school entrance exams.

(12) I talk to my friend when the music teacher is teaching.

(13) I ask irrelevant questions to the music teacher.

(14) I watch outside when the music teacher is teaching.

(15) I scribble in the notebook when the music teacher is teaching.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

This study was approved by Mersin University Institute of Educational Sciences Research Ethics Committee.

ORCID iD

Cenk Akay

References

Aksu

Eser

M. T.

Güzeller

C. O.

(2017) Açımlayıcı ve doğrulayıcı faktör analizi ile yapısal eşitlik modeli uygulamaları [Structural equation modeling applications with exploratory and confirmatory factor analysis]. Detay Yayıncılık.

Alpar

(2014) Spor, sağlık ve eğitim bilimlerinden örneklerle uygulamalı istatistik ve geçerlik-güvenilirlik [Applied statistics and validity-reliability with examples from sports, health, and education sciences]. Detay Yayıncılık.

Anıl

İnal

(2018) Psikofizikte ölçekleme uygulamaları [Scaling applications in psychophysics]. Pegem Akademi.

Biber

(2012) Duyuşsal özelliklerin probleme dayalı öğrenme sürecinde öğrencilerin matematiksel kazanımlarına etkisi [The effect of affective characteristics on students’ mathematical achievements in the problem-based learning process] [Doctoral dissertation, Dokuz Eylul University]. Dokuz Eylul University Institutional Academic Open Archive. https://acikerisim.deu.edu.tr/xmlui/handle/20.500.12397/6767

Burroughs

N. F.

(2007). A reinvestigation of the relationship of teacher nonverbal immediacy and student compliance-resistance with learning. Communication Education, 56(4), 453–475. https://doi.org/10.1080/03634520701530896

Burroughs

N. F.

Kearney

Plax

T. G.

(1989). Compliance-resistance in the college classroom. Communication Education, 38(3), 214–229. https://doi.org/10.1080/03634528909378758

Büyüköztürk

. (2011) Sosyal bilimler için veri analizi el kitabı- istatistik, araştırma deseni, spss uygulamaları ve yorum [A data analysis handbook for social sciences: Statistics, research design, SPSS applications, and interpretation (15th Ed.)]. Pegem Akademi.

Byrne

B. M.

(2011) Structural equation modeling with AMOS: Basic concepts, applications, and programming. Routledge.

Canoğulları

Sarı

(2015). Öğretmenlerin okullarındaki öğrenci direnç davranışlarına ilişkin algılarının incelenmesi [Investigations of teachers’ perceptions of student resistance behavior in their schools]. International Journal of Social Sciences and Education Research, 5(2), 219–232. https://doi.org/10.24289/ijsser.541120

10.

Chan

Treacy

M. J.

(1996). Resistance in multicultural courses. American Behavioral Scientist. 40, 212–221.

11.

Çokluk

Ö.

Şekercioğlu

Büyüköztürk

. (2018) Sosyal bilimler için çok değişkenli istatistik: SPSS ve LISREL uygulamaları [Multivariate statistics for social sciences: SPSS and LISREL applications]. Pegem Akademi.

12.

Crocker

Algina

(1986) Introduction to classical and modern test theory. Holt, Rinehart and Winston Inc.

13.

Davis

L. L.

(1992). Instrument review: Getting the most from a panel of experts. Applied Nursing Research, 5(4), 194–197. https://doi.org/10.1016/S0897-1897(05)80008-4

14.

DeVellis

R. F.

(2017). Scale development. Sage.

15.

Erickson

(1984). School literacy, reasoning, and civility: An anthropologist’s perspective. Review of Educational Research, 54(4), 525–546. https://doi.org/10.3102/00346543054004525

16.

Giroux

H. A.

(2001). Theory and resistance in education: Towards a pedagogy for the opposition revised and expanded edition. Greenwood Publishing Group.

17.

Gömleksiz

M. N.

Kan

A. Ü

. (2012). Eğitimde duyuşsal boyut ve duyuşsal öğrenme [Affective dimension in education and affective learning]. Electronic Turkish Studies, 7(1), 1159–1177. https://doi.org/10.7827/TurkishStudies.3127

18.

Griffith

K. G.

Nguyen

A. D.

(2006). Are educators prepared to affect the affective domain? National Forum of Teacher Education Journal, 16(3), 1–4. https://citeseerx.ist.psu.edu/document?repid=rep1&type=pdf&doi=5e18850606eab2c0d8586abcebcd1e0172488e92

19.

Gündüz

Bozkuş

(2016). Öğretmenlerin Sınıf Yönetimi İlkelerine Uyma Düzeyleri [Teachers’ adherence levels to the classroom management principles]. Milli Eğitim Dergisi, 45(209), 34–46. https://dergipark.org.tr/en/download/article-file/441315

20.

Hair

J. R.

Hult

G. T. M.

Ringle

Sarstedt

(2014). A primer on partial least squares structural equation modeling (PLS-SEM). Sage.

21.

Hendrickson

K. A.

(2012). Student resistance to schooling: Disconnections with education in rural Appalachia. The High School Journal, 95(4), 37–49. https://www-jstor-org-443.web.bisu.edu.cn/stable/23275416

22.

Kearney

Plax

T. G.

Hays

E. R.

Ivey

M. J.

(1991). College teacher misbehaviors: What students don’t like about what teachers say and do. Communication Quarterly, 39(4), 309–324. https://doi.org/10.1080/01463379109369808

23.

Kesici

Aşılıoğlu

(2017). The effect of secondary students’ affective features towards mathematics and the stress they experience before the TEOG exam (the exam for accessing to various types of high schools) on their mathematical success. Ahi Evran Üniversitesi Kırşehir Eğitim Fakültesi Dergisi, 18(3), 394–414. https://dergipark.org.tr/en/download/article-file/1487200

24.

Kline

R. B.

(1998). Structural equation modeling. Guilford.

25.

Ladesma

R. D.

Valero-Mora

(2007). Determining the number of factors to retain in EFA: An easy-to-use computer program for carrying out parallel analysis. Practical Assessment, Research, and Evaluation, 12(1), 1–11. https://doi.org/10.7275/wjnc-nm63

26.

Lehmann

A. C.

Davidson

(2002). Taking an acquired skills perspective on music performance. In Colwell

Richardson

(Ed.), The new handbook of research on music teaching and learning (pp. 542–560). Oxford University Press.

27.

McFarland

D. A.

(2001). Student resistance: How the formal and informal organization of classrooms facilitate everyday forms of student defiance. American Journal of Sociology, 107(3), 612–678. https://doi.org/10.1086/338779

28.

O’Connor

B. P.

(2000). SPSS and SAS programs for determining the number of components using parallel analysis and velicer’s MAP test. Behavior Research Methods, Instruments, Computers, 32(3), 396–402. https://doi.org/10.3758/BF03200807

29.

Pierre

Oughton

(2007). The affective domain: Undiscovered country. College Quarterly, 10(4), 1–7.

30.

Raykov

(1997). Estimation of composite reliability for congeneric measures. Applied Psychological Measurement, 21(2), 173–184. https://doi.org/10.1177/01466216970212006

31.

Schermelleh-Engel

Moosbrugger

Müller

(2003). Evaluating the fit of structural equation models: Tests of significance and descriptive goodness-of-fit measures. Methods of Psychological Research Online, 8(2), 23–74. https://www.stats.ox.ac.uk/~snijders/mpr_Schermelleh.pdf

32.

Seidel

S. B.

Tanner

K. D.

(2013). “What if students revolt?”—Considering student resistance: Origins, options, and opportunities for investigation. CBE—Life Sciences Education, 12(4), 586–595. https://doi.org/10.1187/cbe-13-09-0190

33.

Smith

B. A.

(1994). Understanding more about the affective domain in music [Master’s Thesis, University of Alberta]. ERA: Education and Research Archive. https://era.library.ualberta.ca/items/8c4aac07-f332-41bf-962d-852dc97578f1/view/b1168323-727a-4be8-a4e2-0212766c2688/MM11374.pdf

34.

Swaminathan

(2010). Noncomparative scaling technique. In Sheth

Malhotra

(Eds.), Wiley international encyclopedia of marketing. John Wiley & Sons Ltd. https://doi.org/10.1002/9781444316568.wiem02048

35.

Tabachnick

B. G.

Fidell

L. S.

(2013). Using multivariate statistics. Pearson.

36.

Tolman

A. O.

Kremling

(2017). Why students resist learning: A model for understanding and helping students. Stylus Publishing.

37.

Vinzi

V. E.

Chin

W. W.

Henseler

Wang

(2010). Handbook of partial least squares. Springer.

38.

Watkins

M. W.

(2006). Determining parallel analysis criteria. Journal of Modern Applied Statistical Methods, 5(2), 344–346. https://doi.org/10.22237/jmasm/1162354020

39.

Wolff

K. I.

(1978). The nonmusical outcomes of music education: A review of the literature. Bulletin of the Council for Research in Music Education, 55, 1–27. http://www.jstor.org/stable/40311603

40.

Yaşlıoğlu

M. M.

(2017). Sosyal bilimlerde faktör analizi ve geçerlilik: keşfedici ve doğrulayıcı faktör analizlerinin kullanılması [Factor analysis and validity in social sciences: Application of exploratory and confirmatory factor analyses]. İstanbul Üniversitesi İşletme Fakültesi Dergisi, 46, 74–85. https://dergipark.org.tr/en/download/article-file/369427

41.

Yüksel

(2003). Öğrencilerin öğretme- öğrenme sürecine yönelik direnç davranışları [Resistance behavior to teaching and learning process of students in schools]. Uludağ Üniversitesi Eğitim Fakültesi, 17(1), 235–251. https://dergipark.org.tr/en/download/article-file/153227

42.

Yurdugül

(2005, September 28–30) Ölçek geliştirme çalışmalarında kapsam geçerliği için kapsam geçerlik indekslerinin kullanılması [Using content validity indexes for content validity in scale development studies] [Congress session]. XIV. National Educational Sciences Congress, Denizli, Turkey.