Development and validation of an English language teacher learning scale

Abstract Teacher learning is of significant importance in mainstream education and a number of attempts have been made to measure the quality of teacher learning across different contexts. The available instruments, however, have been originally developed to assess teachers’ perceptions of a single contributive element of teacher learning; therefore, multiple interacting variables which influence the complex process of teacher professional development in English Language Teaching (ELT) contexts have received little attention. This study aimed at developing and validating an instrument whose items are particular to ELT teachers and contexts. To this end, based on a detailed study of the related literature, interviews with ELT practitioners and teacher education professionals, a preliminary theoretical model was proposed. The model was verified and developed into an English Language Teacher Learning Scale. After developing the scale and administering it to 184 ELT teachers, it was validated through both exploratory and confirmatory factor analyses which resulted in a few alterations. This study portrays a more inclusive picture of teacher learning in ELT contexts in terms of factors related to teacher cognition and belief, teacher emotions, teacher motivation, and contextual variables.


Introduction
In the everlasting worldwide call for educational reform and change, the significance of teacher learning in influencing the educational context is beyond doubt. Teacher learning is of international importance and teachers are the portals through which any change and reform is realized (Kooy &

PUBLIC INTEREST STATEMENT
This paper presents the development and validation of an English Language Teacher Learning Scale (ELTLS). The scale takes into account multiple contributive factors which influence the complex process of teacher professional development in English Language Teaching (ELT) contexts. ELTLS provides English language teachers and teacher educators with an effective tool for assessing and reflecting on teacher learning in terms of factors related to teacher cognition and belief, teacher emotions, teacher motivation, and contextual variables.
Veen, 2012). In education literature, teacher learning refers to "a process of increasing participation in the practice of teaching, and through this participation, a process of becoming knowledgeable in and about teaching" (Adler, 2000, p. 37). Teachers themselves are considered as learners, developing their teaching practice and modeling for pupils the process of continual learning (Hagger, Burn, Mutton, & Brindley, 2008). In fact, improving student achievement requires improvements in teacher instructional practice and the capacity of schools to promote teacher teaching (Murray, 2013).
English Language Teaching (ELT), as an autonomous profession, is mainly concerned with the study of language teaching and learning with the purpose of bringing about change and improvements in the quality of language teaching and learning (Howatt & Widdowson, 2004). It has been argued that the set of standards for enhancing ELT may vary across different contexts since it is influenced by a host of sociocultural, political, economic, and historical factors (Mahboob & Tilakaratna, 2012). Therefore, ELT practitioners should develop their own teaching practices and adopt their own language teaching policies so as to ensure that such practices reflect the local conditions of their country or context. However, they need to be aware of the potential challenges in the process of developing ELT initiatives and interventions as well as the processes through which ELT programs are designed and delivered.
Teacher learning has been the focus of much investigation in the past decades (Endedijk & Vermunt, 2013;Feiman-Nemser, 2012;Jones & Dexter, 2014;Opfer & Pedder, 2011;Slavit & Roth McDuffie, 2013;Vermunt & Endedijk, 2011). These studies have acknowledged the importance of teacher attitude, attention, and awareness in professional development and examined how teachers acquire, generate, and use knowledge in their career through district-initiated, teacher-initiated, or teachers' independent learning activities. It is also argued that teacher learning is mainly influenced by the interaction of three subsystems (the teacher, the school, and the learning activity). Therefore, methodological practices must be adapted to focus on explanatory causality and the reciprocal influences of all three subsystems (Opfer & Pedder, 2011). In fact, teachers differ with respect to the learning patterns they adopt which are, in turn, influenced by the quality of teacher learning and professional development. Teacher professional development is enhanced when the educational environment supports and encourages teacher learning, supplies learning materials, engages teachers in conscious learning practices, and provides teachers with opportunities to improve the mental models of their teaching and learning practice (Shaw, 2010).

Statement of the problem and purpose of the study
The development of measurement instruments to assess teachers' perceptions of different aspects of teacher professional development has been the topic of a number of studies such as the ones by Ingersoll and Kinman (2002), (Teacher Candidate Performance Self-Assessment Instrument); Scott and Dinham (2003), (Teacher and School Executive Occupational Satisfaction Scale); and Uzunboylu and Ozdamli (2011) (Teacher Perception of Mobile-Learning Technologies). There are also a few studies which have evaluated teacher learning and education in various settings (Borko, 2004;Guskey, 2002). Teacher Candidate Performance Self-assessment Instrument is a screening instrument to assess pre-service teacher education candidates' self-perceived knowledge and ability. This instrument assesses candidates' self-perception of their abilities among four competencies labeled as teaching skills, classroom management skills, knowledge of children, and technology skills. Teacher and School Executive Occupational Satisfaction Scale investigates teacher and school executive career satisfaction, motivation, and mental health. However, these instruments have been originally developed to assess teachers' perceptions of a single contributive element of teacher learning and therefore by no means encompass the potential multiple variables which must be taken into account while studying the complex process of teacher professional development. The present study is motivated by the need to understand the underling variables which guide and influence the path of teacher professional learning in ELT contexts. Therefore, in order to extend prior work on the evaluation of teacher learning and to compensate for the lack of an existing assessment tool to evaluate language teachers' professional learning, the present study addresses the key issue of developing and validating an English Language Teacher Learning Scale (ELTLS) which takes the various elements of language teacher professional learning into account.

Theoretical conceptualization of the study
In order to construct an instrument which represents the distinctive characteristics of teacher learning in ELT contexts, it was necessary to develop a theoretical framework. The review of literature indicated that the most congruous model with the purpose of this study was the model of Language Teacher Conceptual Change (LTCC) which was proposed by Kubanyiova (2012). However, this model which was based on Gregoire's (2003) Cognitive-Affective Model of Conceptual Change (CAMCC) was not specific to ELT contexts. CAMCC took the role of emotions, appraisals, motivation, cognition, and contextual factors in teacher belief change into account and explained why teachers might resist changing their instructional practices. However, the model was criticized for its linearity and lack of empirical validation. Kubanyiova's LTCC model (2012) presented a robust and comprehensive account of language teacher conceptual change and explained its cyclical and dynamic nature. The model specifically investigated the main features which influenced teachers' intentional conceptual change, deep-level cognitive engagement, possible selves, and affective and motivational factors. The LTCC model (Kubanyiova, 2015), however, was basically a model for explaining teacher conceptual change and thus provided no means for the assessment of teacher learning and professional development.
To compensate for the lack of an existing tool to measure teacher learning in ELT contexts, first, factors influencing teacher learning were identified by the researchers through interviewing 25 ELT teachers and 7 ELT teacher educators as well as a thorough review of the related literature. Then, the factors identified at this level were integrated with general teacher learning components in the model proposed by Kubanyiova (2012). Later, using the empirical data gained from the interviews, a thorough review of the literature, and the main domains presented in Kubanyiova's LTCC model (2012), a conceptual model for teacher learning in ELT contexts was proposed. We suggest that teacher learning in ELT contexts is a function of a number of variables including teacher cognition and beliefs, teacher emotions, teacher motivation, and contextual variables. Finally, our proposed model led to generation of a pool of potential items to be included in the instrument. Table 1 represents the preliminary theoretical conceptualization of teacher learning in ELT contexts proposed in the present study.

Components and sub-components of teacher learning in ELT contexts
• Teacher cognition and beliefs (i.e. pedagogical knowledge, subject matter knowledge, values, awareness and sensitivity, self-reflections, beliefs and knowledge about teaching, students, and content, awareness of classroom problem-solving strategies, schooling, prior language learning experience, classroom practice, prior teacher education programs • Teacher emotions (i.e. emotional suffering, emotional freedom, emotional knowledge, emotional exhaustion and teacher burnout, school climate, social working environment, workload, emotional intelligence, cognitive self-regulation, hardiness, teachers' goal orientation, job satisfaction) • Teacher motivation (i.e. success expectancy, social prestige, working condition, intellectual fulfillment, salary, job security, autonomous motivation, teacher self-efficacy, goal internalization, uncertainty tolerance • Contextual variables (i.e. value consonance, supervisory support, relations with colleagues and parents, time pressure, discipline problems, time and resources, program flexibility, teacher education programs, teacher evaluation, materials and resources, classrooms observations, teacher learning strategies, job satisfaction Research on teacher cognition, belief, and competencies is not in its early days anymore (Borg, 2003;Golombek & Doran, 2014;Hoekstra, Brekelmans, Beijaard, & Korthagen, 2009;Kang & Cheng, 2013). Teacher cognition, defined as "unobservable cognitive dimension of teaching -what teachers know, believe, and think" (Borg, 2003, p. 81) is shaped by teachers' engagement in learning activities in workplace , teachers' experience of classrooms, language learning histories, classroom practice, and teacher education programs (Borg, 2003). In fact, it is argued that teacher belief change arises from several causes such as teacher's dissatisfaction with the existing conceptions, teacher's motivation and ability as well as characteristics of the learners (Gregoire, 2003). Research on teacher cognition has also indicated that what teachers learn during teacher education programs or classroom experiences is filtered by the teachers' prior experience and the sociocultural context (Kubanyiova, 2012).
Several publications have also appeared in recent years demonstrating teacher emotions as a key component of teacher professional and personal identity (Cowie, 2011;Golombek & Doran, 2014;Reio, 2005), teachers' reconstruction of professional understanding (Darby, 2008), teachers' commitment (Jo, 2014), teachers' successes and failures at schools (Zembylas, 2005), teachers' sensemaking of educational reforms (Schmidt & Datnow, 2005), and teachers' personal professional growth (Day & Leitch, 2001). These studies have indicated that teacher emotions are at the heart of successful teaching practice and have close connection with teachers' professional decision-making. On the other hand, teacher emotional exhaustion, depersonalization, and reduced personal accomplishment are all the sub-components of teacher burnout which are related to teacher emotions and can impair the quality of teaching practice and discourage teachers from continuing their profession . Furthermore, research on teacher cognition and teacher emotions has indicated an intrinsic and dialectical relationship between teacher cognition and emotions (Bullough, 2009;Zembylas, 2003), which has significant roles in teacher learning and professional developments (Golombek & Doran, 2014), teacher activities (Golombek & Doran, 2014), and teacher conceptual change (Kubanyiova, 2012).
Teacher motivation is also an important factor which has a significant role on educational reforms and teacher satisfaction (Neves de Jesus & Lens, 2005), quality of teaching practices, and teachers' involvement in professional learning activities (Thoonen, Sleegers, Oort, Peetsma, & Geijsel, 2011). Motivational factors, generally, constitute three main factors: expectancy, value, and affective components (Hascher, van der Veen, & Roede, 2005). Teachers' sense of self-efficacy, internalization of school goals, and teachers' tolerance of uncertainty are also among other factors which influence teachers' engagement in professional learning activities (Thoonen et al., 2011). Although the bulk of research on teacher learning concentrates on teacher cognition, assuming that teaching behavioral changes feed from changes in teacher competencies and knowledge, the role of contextual variables in facilitating or hindering the process of teacher professional development should not be overlooked.

The participants
In order to get a more comprehensive understanding of ELT teachers' views about factors influencing teacher learning in an ELT context and as a conceptual means for validating the scale, 25 ELT practitioners were purposefully selected to be interviewed. These teachers were heterogeneous in terms of age (between 25 and 47), gender (11 female, 4 male), degree (10 BA, 3 MA, and 2 PhD holders), teaching experience (3-11 years of experience), teaching context (3 teaching at schools and 12 at language institutions), and the proficiency level taught (8 elementary, 4 intermediate, and 3 advanced). Moreover, to gain more insight and know about experts' knowledge on the underlying elements of teacher learning in ELT contexts, seven teacher education experts (4 female, 3 male) with a PhD degree in TEFL were interviewed to learn about their expert opinions on the factors which influence teacher learning and the important issues related to the construct. These experts who were educational program observers, evaluators, and educators were selected based on their academic interests in teacher education.
After developing the new ELTLS (more details below), 45 English teachers at various proficiency levels (15 elementary, 15 intermediate, and 15 advanced) were purposefully sampled to participate in piloting the scale. Moreover, four researchers/teachers (2 male, 2 female) who were experienced in teacher education (5-7 year experience) revised the scale after the pilot testing phase. For final validation phase of the study, 184 English language teachers, dissimilar regarding their age (between 23 and 45), gender (98 female, 86 male), degree (73 BA, 67 MA, and 44 PhD holders), teaching experience (4-12 years of experience), teaching context (65 teaching at schools and 119 at English language institutions), and the proficiency level taught (93 elementary, 52 intermediate, and 39 advanced), constituted our sample. These participants were selected through availability sampling. Finally, it must be noted that all the participants were provided with full disclosure about the nature of the study and they all had a choice whether to participate in the study or not.

Data collection instruments
Twenty-five ELT practitioners were interviewed for their ideas about the potential factors influencing their learning as a teacher. Moreover, seven teacher education experts were interviewed to learn about their opinions on teacher learning in general and in ELT contexts in particular. This study also used the newly developed ELTLS, the development and validation of which is explained in the procedure section below.

Procedure
Based on the review of the literature, the theoretical conceptualization of the study (Table 1), and interviews with ELT teachers and teacher education experts, a pool of potential items to be included in the scale were generated. Initially, 25 ELT practitioners were interviewed to understand the potential experiences and factors which influenced teacher learning in ELT contexts. In order to make sure of the stability of the interviews, a checklist which included different factors which influenced teacher learning and was previously extracted from the literature was used (Appendix A). These interviews were audio-recorded and content analyzed to cross-check the information against the theoretical framework of the study.
Moreover, seven teacher education experts, who were all university professors, were interviewed to discover their opinions on teacher professional learning in ELT contexts in general, and the underlying factors which influence teacher learning in ELT contexts, in particular. There were seven openended questions (Appendix B) in the interviews based on the components of teacher learning which were outlined in Table 1. Each interview session lasted between 20 and 30 min and was audio-recorded to be later transcribed and content analyzed.
All the recorded and transcribed interviews with ELT practitioners and experts were meticulously read and analyzed in order to find their underlying themes. As an example, while asked about the factors which influenced teacher professional development in ELT contexts, a teacher responded: Well, I believe that there are many factors such as the learning resources which are provided for us, free access to the web-based learning materials, workshops and in-service teacher education programs. Moreover, teacher learning should be valued and encouraged by some strategies such as job promotions or rise in salaries.
Analyzing this response in terms of its underlying ideas, the researchers extracted three main themes: learning martial availability, teacher education programs, and motivation. All the interviews with ELT practitioners and teacher education experts were thematically analyzed the same way. The extracted themes along with the ideas gained through the review of the literature and the theoretical model formed the basis for the generation of a pool of 58 items as the candidates for eventual inclusion in the scale. These questions were on a Likert scale of one to five and required the English language teachers to indicate the extent to which each item applied to them.

Teacher cognition and belief 10
Teacher emotions 10 Teacher motivation 11

Contextual variables 19
Total 50 Next, the newly developed scale was reviewed and piloted before its administration to the intended participants. Four researchers/teachers who were experienced in teacher professional development were asked to review the scale and comment on its format, content, as well as wordings. This review process resulted in the revision of five items and deletion of eight items as to their repetitiveness, ambiguity, and irrelevance. Then, the scale, which now had 50 items, was piloted with 45 ELT teachers. Ultimately, to establish the construct validity of ELTLS, it was administered to 184 ELT teachers. The questionnaire was administered by the researcher in person. The data gained at this level were analyzed through both exploratory and confirmatory factor analyses. Factor analysis is a commonly used statistical procedure which specifically addresses the construct validation of tests and scales (Brown, 2015;Comrey & Lee, 2013;Westen & Rosenthal, 2003).

Results and discussion
In the initial stage of this study, a theoretical framework of teacher learning, which captured the context-specific nature of language teaching in ELT contexts, was developed. Table 2 indicates the series of items within each component in the theoretical framework of the study. The reliability of the 50-item questionnaire, using Cronbach alpha, turned out to be .913, which was satisfactory. In order to examine the construct validity of the newly developed ELTLS in terms of its fundamental construction and the distribution of items under each component, both exploratory and confirmatory factor analyses were tested.
The outcome of exploratory factor analysis showed the existence of five initial factors, in contrast to the four factors we speculated in Table 1. As far as the factor loadings resulted from the exploratory factor analysis did not seem to be the most favorable, a number of confirmatory factor analysis were also examined on the data with different fixed factors ranging from two to eight. This was done using principal components analysis technique with varimax rotation. Based on confirmatory factor analysis, it was concluded that the best model of factor loadings was four factors which constituted 53.97% of the total variance (Table 3). Moreover, the KMO and Bartlett's Test of Sphericity for four factors was .76 (p < .05), which indicated that the original components were adequately correlated. It should be mentioned that the item loadings which resulted from the four factors were slightly different from what was assumed based on the tentative theoretical framework of this study, that is to say, a few items which were thematically related to different components in the theoretical framework loaded on the same factor. This led to small modifications in the original theoretical framework as indicated in Table 4.
On the basis of the results gained through factor analysis, five items were deleted from the 50item scale as far as they loaded on irrelevant factors. These items focused on "establishing a friendly relationship with students", "class size", "students' composition", "school networking", and "school reputation". After the omission of these items from the scale, the reliability was re-calculated through the Cronbach alpha and turned out to be .914 which was slightly higher than the reliability of the scale before the omission of the items. The finalized scale, shown in Appendix C, has now 45 items.

Finalized components of teacher learning scale
Teacher cognition and beliefs (factor 1) (items 1, 2, 3, 4, 5, 6, 7, 19, 37) Teacher emotions (factor 2) (items 8, 10, 11, 15, 20, 22, 27, 28, 31) Teacher motivations (factor 3) (items 9, 12, 16, 18, 21, 24, 25, 32, 39) Contextual variables (factor 4) (items 13, 14, 17, 23, 26, 29, 30, 33, 34, 35, 36, 38, 39, 40, 41, 42, 43, 44, 45)  It must be mentioned that that items 11, 13, 14, 27, 28, 36, and 39 were reverse scored. These items were about goal orientation, discipline, time pressure, emotional exhaustion, workload, teaching material, and success expectancy. Table 5 indicates the descriptive statistics of the finalized ELTLS. The mean and the standard deviation of each component and the total number of items were computed. The items were on a Likert scale of 1-5 and the value of each item was calculated. Moreover, the items which were reverse scored were recoded before computing the value for each item. ELT teachers who scored higher on any component or on the total are considered better teacher learners. As an example, an ELT teacher who scored 68 on the teacher cognition and belief component was considered a potentially better teacher learner in terms of teaching knowledge, expertise, experience, and background than an ELT teacher who scored 45. Also, an ELT teacher who scored 53 on the contextual factor was a better teacher learner than an ELT teacher who scored 42 on the same component since he/she worked in a more effective and appropriate educational system and context. In the same vein, a teacher who received a total score of 183 on the scale was generally considered a better teacher learner than a teacher who scored 134 since he/she was in a more advantageous position in terms of the host of factors which influenced the process of teacher learning as represented in theoretical framework of the study.
The result of this study indicated that the theoretical model of teacher learning in ELT contexts proposed for the purpose of this study was meaningful to the ELT teachers. Our framework suggests that teacher learning is a dynamic and cyclical process of structuring and re-structuring teacher cognition and teacher emotions which are, in turn, guided by a host of motivational and contextual factors. The qualitative data of the present study, which came from interviews with ELT practitioners and teacher educators, indicated that there was also a host of other factors which ELT practitioners and educators believed to have great impact on language teachers' learning. These factors, as mentioned in Appendix B, were mainly the challenges or threats that language teachers confronted within their actual teaching practice. For instance, one of the factors which stimulated less experienced language teachers to improve their professional learning was the threat imposed to the teacher by more linguistically advanced learners in language classrooms. These teachers reported that a number of classroom incidents such as being unable to answer the challenging questions asked by the learners and being corrected for the linguistic mistakes could threaten their professional identity as language teachers. Furthermore, language teachers reported that the challenge created by their more professional colleagues was another factor which encouraged them to be better teacher learners. They reported that they would try to be better teachers in order to be defined as popular teachers. They also wanted to learn about their profession so that they would be able to take part in informal discussions about teaching and learning issues with more professional collogues during the break times at work. In addition to learners and colleagues, instructional materials like textbooks could motivate language teachers to learn more and attend their classes fully prepared in order to avoid potential classroom teaching problems. In sum, there are a number of on-the-job challenges which have significant roles on language teacher learning and teaching preparation which, in turn, have to do with teacher professional identity at work. Language teachers try to learn more to be able to respond to learners' learning needs, take part in professional talks at work, and be defined as successful and popular teachers. Overall, a detailed review of the existing literature on teacher learning as well as the findings of the present study suggested that there were a number of interrelated factors which contributed to teacher learning in ELT contexts. We suggest that teacher cognition and teacher emotions are at the core of teacher professional learning which is, in turn, constantly being shaped by the contextual variables and motivational factors.

Conclusion
English language teaching profession has become increasingly aware of the prominent role of teacher learning in determining the quality of language instruction. However, research on ELT teacher learning does not provide us with an assessment tool to evaluate the underlying factors which can either promote or obstruct teacher professional learning. These underlying elements make it necessary to come up with a different conceptualization of teacher learning which reflects the unique characteristics of teacher learning in ELT settings. It must be noted that the present study was conducted in an EFL setting where both teachers and learners had limited exposure to English outside language classrooms. Therefore, investigating the underlying factors which contribute to teacher learning in ESL contexts can greatly add to our knowledge about the impact of context on language teachers' conceptualization of their learning.
Through a detailed review of the literature and interviews with ELT practitioners and educators, the present study developed and validated an ELTLS. The scores received by the English language teachers in the validation stage of the study (Table 5) indicated that the constructed items and our conceptualization of teacher learning in ELT contexts were meaningful to ELT practitioners and indicative of the determining elements of their learning. Context-sensitive measurement scales, like the one developed for the purpose of this study, are even more efficacious, meaningful, and practical. ELTLS provides English language teachers, teacher students, and teacher educators/observers with a systematic, meaningful, and precise tool to assess the underlying factors which contribute to teacher professional learning. It portrays a more detailed picture of teacher learning and presents a more effective means of providing feedback on the quality of teachers' job. The scale can also be used for planning remedial training programs and making necessary modifications in educational policies which will, in turn, lead to better teacher education programs and more qualified teachers in general. Further research can undoubtedly lead to improvements in the present scale. The present study

Appendix B
The open-ended Interview Questions (1) What factors motivate or demotivate you to be a better language teacher?
(2) What are the characteristics of a good teacher learner?
(3) What contextual factors influence your desire to be a better teacher?
(4) How do your emotions influence your own learning as a teacher?
(5) What individual and unique features do you think you have which have made you a better teacher learner?

Appendix C ELT learning scale (ELTLS)
Dear teacher This scale is intended to measure teacher learning in ELT contexts. The information will be kept strictly confidential and used for research purposes. Your sincere participation will be appreciated. Name: ___________ Age: __________ Gender: male ⎕ female ⎕ Years of teaching experience: _____ Degree: __________ Major: __________