Reinventing the Dimensions of Reading Attitude and Motivation among Santri

Received Feb 27, 2018 Revised Apr 21, 2018 Accepted Apr 26, 2018 A large and growing body of literature has investigated students’ reading attitude and motivation either in elementary or secondary levels. Unfortunately, there remains a paucity of similar study in the context of santri or students in Islamic boarding school. This study, therefore, examined the reading attitude and motivation dimensions among santris. Three hundred and nineteen (319) students at pesantren-based senior high schools in Yogyakarta Special Region province participated in this study by completing reading attitude survey adapted from McKenna et al. and motivation for reading questionnaire (MRQ) adapted from Wigfield and Guthrie. Exploratory and confirmatory factor analyses confirmed that reading attitude was represented by two dimensions as suggested by the finding in previous studies. On the other hand, instead of eleven, only seven reading motivation dimensions manifested in this present study. Reliability coefficients for all the subscales were acceptable. Finally, correlations analysis revealed that each dimension of both reading attitude and motivation were correlated positively.


INTRODUCTION
Reading English texts plays a critical role in students' learning development as it does not only broaden knowledge and provide opportunities to self-developments [1,2]; but also become the primary source of second language input for students' language learning [3,4]. Specifically, reading English texts is also crucial for secondary school students as it is tested in the National Examination with approximately 70% questions in it [5].
With respect to the importance of reading at secondary school, a considerable amount of studies has been published on middle and high school student's reading attitude [6,7]; and motivation [8,9]. However, to date, reading attitude and motivation among santri at pesantren-based senior high schools or private Madrasah Aliyah (MA) have not been extensively investigated. Considering the big number of santri studying in private MA in Yogyakarta province, i.e. 6.776 students compared with 53.581 students of general senior high school, or approximately one ninth of senior high school students in Yogyakarta study in private MA in 2016 [10], there is an urgent need to address santris' attitude and motivation in reading English texts.
This study, therefore, sets out to investigate the reading attitude and reading motivation dimensions among santri in Yogyakarta. Specifically, this research seeks to address the following questions: (1) What are the dimensions of santris' reading attitude and motivation dimensions? and (2) Are there any correlation among the dimensions?

Reading Motivation
An early definition of reading motivation is proposed by Gambrell et al. [14] who suggest that reading motivation is an individual's self-concept and value on reading. Similarly, reading motivation is also defined as the individual's goals, values, and beliefs with regard to the topics, processes, and outcomes of reading [15]. Furthermore, reading motivation is also defined as individuals' intrinsic and extrinsic reasons for reading activities and achievement [16]. In this paper, the term reading motivation is defined as students' intrinsic and extrinsic motives, energy and drive to reading activities and achievement at school.

Santri
Santri is a term used for students of pesantren or Islamic boarding school [19,20]. Pesantren according to Dhofier [21] is an educational institution of religious teaching consisting of five elements, namely Kiai/Ustadz, santri, mosque, kitab kuning classical Islamic texts written in Arabic, and pondok or dormitory where all santri/students must stay inside of pesantren to learn the Holy Scripture and religious sciences.
Muhaimin [22] claims that Ministry of Religious Affairs of the Republic of Indonesia has categorized pesantren into four types, A, B, C, and D. Type-A pesantren maintains the traditional characteristics which offers only classical Islamic texts lessons only without any room for teaching general subjects. On the other hand, type-B pesantren offers both traditional teaching of classical Islamic texts and modern teaching of religion and secular subjects in several modern religious schools or madrasah. The madrasah has its own curriculum and/or adapts the curriculum from Ministry of Religious Affairs. Type C has similar characteristics as type-B pesantren which consists of a type-B pesantren plus public school. The public school is under the supervision of Ministry of National Education. Last but not least, Type-D pesantren provides only dormitory for its santri in which Kyai functions as a counsellor. The santri in this type of pesantren go to madrasah or other schools outside it as it has no madrasah there.
Santri, in this research, therefore, is defined as students, specifically senior high students, who live in type-B pesantrens. They go to madrasah in the morning to learn both religion and secular subjects and study classical Islamic texts from their Kyai in the evening.

RESEARCH METHOD 2.1. Participants
Participants were drawn from senior high school students in grade X and XI of private Madrasah Aliyah (MA) who live in a type-B pesantren (N=319) in Yogyakarta Special Region province. Santris from this type of pesantren were chosen as the participants because they learn English in formal schools (madrasah) and learn Islamic scripture at the same time while students from other type of pesantren either just learn classical Islamic scripture or just learn secular subject.

Measures
Data were collected using two types of questionnaire: reading attitude questionnaire and reading motivation questionnaire. The initial set of reading attitude questionnaire consisted of 15 items which were adapted from McKenna et al. [6]. The fifteen items were divided into two constructs, i.e. academic reading and recreational reading both in print setting. Example items were "How do you feel about taking an English reading test?"; and "How do you feel about reading an English textbook?". The items that were retained after factor analysis are presented in Table 1.
The initial set of reading motivation questionnaire consisted of 45 items which were adapted from Wigfield and Guthrie [18]. The items were placed into eleven constructs, i.e. (1) reading efficacy; (2) reading challenge; (3) work avoidance; (4) reading curiosity; (5) reading involvement; (6) importance of reading; (7) competition in reading; (8) reading recognition; (9) reading for grades; (10) social reasons for reading; and (11) reading compliances. Example items were "I am learning to read in English merely because I would like to get good grades"; and "Learning to reading in English is important because it will be useful for my future education". The items retained after factor analysis are presented in Table 3. All items were measured on a 4point Likert scale (Strongly Agree, Agree, Disagree, and Strongly Disagree).

Statistical Analysis
The main statistical analyses used in this study were exploratory factor analysis (EFA), confirmatory factor analysis (CFA), reliability analysis, and correlational analysis. EFA was conducted to identify the dimensions of reading attitude and see if the data in this study fits the model of attitude and motivation constructs. Furthermore, CFA was conducted to confirm the factor model based on the EFA result. The reliability analysis was conducted to see the internal consistency for each factor. Finally, correlational analysis was performed to see how the factors underlying the dimensions of reading attitude and motivation are related to each other.
In running the EFA, there were several considerations to eliminate items and factors. First, items were deleted when they did not load on any factor. Then, items were also deleted if they contained factor loading smaller than .30. Third, when a factor had less than three items, it was deleted. It was because the minimum items per factor usually requires three factors [23].
EFA was conducted using both SPSS and Mplus while CFA was conducted using Mplus. Four indices with their own criteria were set to indicate a good model fit, such as Comparative Fit Index (CFI > .90), Tucker-Lewis index (TLI > .90), Root Mean Square Error of Approximation (RMSEA < .05), Standardised Root Mean Square Residual (SRMR < .05) (criteria cut-off scores indicated in brackets; see also Hooper, Coughlan, & Mullen [24]). Furthermore, chi-square (x 2 ) with its degrees of freedom (df) ratio values smaller than 2.00 were used as an acceptable model fit criterion [25].

RESULTS AND ANALYSIS 3.1. Exploratory and Confirmatory Factor Analysis
From the result of EFA, there were 8 items retained from the initial set of 15 items of reading attitude questionnaire. These 8 items clustered into two factors with 4 items for each factor. The 2-factor model showed a good fit with fit indices of x 2 = 16.62; df = 13, x 2 /df = 1.28; CFI = .99; TLI = .99; RMSEA = .03; and SRMR = .02. When the same items were tested by CFA, the 2-factor model also showed an excellent fit to the data, with fit indices of x 2 = 29.56, df = 19, x 2 /df = 1.55, CFI = .99, TLI = .98, RMSEA = .04, SSMR = .03. Furthermore, factor interpretation based on the results of EFA and CFA is presented in Table 1. As shown in Table 1, there are two factors or dimensions of reading attitude namely reading attitude in academic print and recreational print.
This finding confirms the factors proposed by McKenna et al. [6,13] which covers academic and recreational reading. The researcher used two factors (i.e. academic reading and recreational reading in print setting) and excluded two other factors related to online setting as the respondents in the present study, based on the observation, were not allowed to bring any gadget and thus had limited access to the internet. The rule prohibiting them to bring any gadget is in line with the article written by Mahasin [26].
Factor one of the reading attitude is labelled as academic reading while factor two is labelled as recreational reading. The reliability coefficients ranged from 0.77 to 0.79 which show a good internal consistency [27]. This finding reflects the tendency that santris from type B pesantren did read for both their academic and recreational purposes. In accordance with the present results, previous studies have demonstrated that reading attitude dimensions comprise academic reading and recreational reading [6,7,13].   Table 2.
As shown in Table 2, there are seven factors or dimensions of reading motivation namely (1) Reading challenge; (2) Reading involvement; (3) Importance of reading; (4) Reading efficacy; (5) Competition in reading; (6) Compliances; and (7) Social reasons for reading. The loading factors for the three items range from .40 to .77. The reliability coefficient shows a value of .64 which can be considered as reliable. The loading factors for each factor are also considerably high, that is ranging from .62 to .75. The Cronbach's Alpha is also reasonably high with .76.
There are several reasons why the 11 dimensions as suggested by Wigfield and Guthrie [2 18] did not emerge in this study. A possible explanation for this might be that santris in this research did not have any reasons for committing reading avoidance in compensations during unsuccessful reading performance. They have been trained to obey whatever they were trained to do. It is a must for santri to show obedience to their teachers [20,28]. The discrepancy could also be attributed to the facts that since santris mostly read Islamic books during their stay in pesantrens. They might not develop the sense of curiosity as they were well-trained to memorize the verses and hadith during their study, so they did not have much time to read whatever they want to read. The case might be different from those staying in Type-D pesantren because santris in this pesantren could possibly attend the formal school under the supervision of Ministry of Education and Culture (MOEC) [22] in which the components of secular subjects weigh greater than religious subjects.  Table 3 shows correlations not only among reading motivation dimensions but also among reading motivation and reading attitude dimensions. This finding is consistent with the previous studies [18], [16], [29], [30], [31] claiming that all the reading motivation dimensions or factors are correlated positively and in the moderate range. Likewise, the finding from reading attitude is also in line with that of McKenna et al. [6] that claim the relationship between academic and recreational reading in print setting is comparably high. Furthermore, as can be seen from Table 3, both reading motivation factors and reading attitude factors are correlated positively. This result, therefore, supports the previous study from Baker and Wigfield [29] stating that motivated readers tend to have positive attitudes toward reading.

Correlations among the Dimensions
As shown in Table 3, all the correlations were positive, and ranged from low to moderately high. The strongest relations included those between Reading involvement and Recreational reading (r = .79), Reading involvement and Academic reading (r = .78), Academic reading and Recreational reading (r = .70), Reading challenge and Reading involvement (r = .61), Reading involvement and Competition in reading (r = .59), Reading involvement and Importance of reading (r = .57), Reading challenge and Recreational reading (r = .55), Reading challenge and Academic reading (r = .54), Importance of reading and Recreational reading (r = .54), Competition in reading and Recreational reading (r = .54), Competition in reading and Academic reading (r = .53), Importance of reading and Academic reading (r = .53), and Reading involvement and Reading efficacy (r = .51).
As these factors were correlated each other, there might exist a more general construct that underlies these factors. Therefore, a higher order confirmatory factor analysis was conducted to prove this [23].

Higher-order Confirmatory Factor Analysis
A higher-order CFA model was constructed with the 9 latent variables at the first-order level and 2 factors at the second-order level. The 2 second-order factors represent (a) reading motivation, and (b) reading attitude. The data fit of the second-order model was also good, with fit indices x 2 = 505.50, df = 367, x 2 /df = 1.38, CFI = .94, TLI = .93, RMSEA = .03, SSMR = .06. The second-order level factors include reading motivation which underlies 7 factors (i.e. reading challenge, reading involvement, importance of reading, reading efficacy, compliances, competition in reading, and social reasons for reading), and reading attitude which underlies 2 factors (academic and recreational reading). The second-order factorial structure is displayed in Figure 1.

Analysis
The present study investigated the reading attitude and motivation dimensions among santri. In this investigation, reading attitude survey adapted from the model proposed by McKenna et al. [6], and reading motivation questionnaire adapted from Wigfield and Guthrie [18] were employed. Factor analyses in the forms of EFA and CFA were conducted to delineate the number of factors for each model. Factor analysis and computing reliability coefficients showed evidence that both reading attitude and reading motivation questionnaires were reliable and valid. From the results of EFA and CFA, it was found that there were 2 dimensions of reading attitude, namely, academic reading and recreational reading. This finding is consistent with the previous studies [6], [13]. The initial set of reading attitude questionnaire only covered these two dimensions in print setting and did not include those in online setting due to the reason that santri in this study were prohibited to bring any gadget both in pesantren (dormitory) and madrasah (school) thus, their access to be online during schooldays was limited [26]. This finding might be interpreted that reading attitude could be measured by the two constructs, namely, attitude towards academic reading (e.g. reading English textbook, reading assignment), and attitude towards recreational reading (e.g. reading for pleasure such as reading English novel). The reading motivation scale, based on the EFA and CFA, comprised 7 dimensions, i.e. reading challenge, reading involvement, importance of reading, reading efficacy, compliances, competition in reading, and social reasons for reading. Interestingly, four of the 11 dimensions did not emerge in this study. The reason might be that santri were trained to obey and respects their teachers [20], [28], [32], so that they had no reasons to avoid any reading tasks given by their teachers. They mostly read kitab kuning or classical Islamic texts either in pesantren or at madrasah. This fact was found by the researchers in all four different schools during the data collection process. In the classrooms, even, there must be either kitab kuning or Holy Qur'an on their school desks. Then, some of them, when they had finished completing the questionnaires, they started reading their kitab and writing some verses in Arabic on a piece of paper. As they had a tight schedule in the dormitory, they did not have much time to read things other what have been assigned to them. This statement was given by the santri themselves to the researcher. This reason might lead them not to have much curiosity to read. In the same vein, some santris admitted that English was not their top priority since their main purpose living and studying in pesantren was focusing on the study of religious texts and for some, memorizing verses from Holy Qur'an. This is in line with the statement from Dhofier [21] stating that the main reason for santri choosing pesantren life was to study Holy Qur'an, Hadith, and other Islamic scriptures. Therefore, they presumably neither care for their recognition nor their grades in English reading.
This finding might be different from those staying in Type-D pesantren or modern pesantren as these types of pesantren did not include the teaching of Islamic classical text to their santri. Furthermore, those staying in Type-D pesantren could attend any formal school outside the pesantren [22]. They could possibly choose a formal school under the supervision of Ministry of Education and Culture (MOEC) such as SMA (Sekolah Menengah Atas) which secular subjects weigh greater than religious subjects. Santris who live in modern pesantren with modern system of pesantren might also have different results as both Arabic and English were taught there. In contrast to the Type-B and other traditional pesantren, in some modern pesantren, even, santri were obliged to communicate in both Arabic and English in their dormitory [33].
Correlations analyses revealed that the seven factors correlated positively with each other. The results also showed that even there were positive correlations between reading attitude factors and reading motivation factors. This result reflects that of Baker and Wigfield [30] which found that motivated readers tend to have positive reading attitudes. Furthermore, a higher order confirmatory factor analysis was conducted to prove that there might exist a more general construct that underlies these factors since they correlated positively. The higher order CFA model revealed that the second-order level, consisting of two factors, was in a good fit model.

CONCLUSION
The present study has demonstrated, for the first time, which the two dimensions of reading attitude have emerged among santris' reading attitude. However, the eleven dimensions of reading motivation did not completely emerge among santris. These results imply that santris might have different reading motivation compared to those studying in public high schools (SMAs) and or vocational high schools (SMKs). This may suggest that teachers at pesantrens might need to adjust pesantren learning time to promote their learners reading attitude and motivation. The results of this research also suggest that the school authorities as well as the English teachers encourage santris to read more so that they would gain their reading curiosity, recognition, and grades. For example, students or santris are given a deeper understanding on the importance of reading English for final exam, so that they would consider their recognition and grades in reading, which is, could presumably improve their reading motivation.
This current research was limited in several ways. First, it has only examined the reading attitude and motivation among santri in Yogyakarta Special Region only. Second, the participants were only from pesantren-based senior high school. Future studies can also conduct the similar research in pesantren-based junior high schools or even in some pesantren-based elementary schools with children as the participants. Third, this present study was limited in Type-B pesantren only. While students living in other types of pesantren are defined as santri, more research using similar questionnaires can also be conducted in other types of pesantren.