Читать книгу Exploring Language Teacher Efficacy in Japan - Gene Thompson - Страница 13
Оглавление2 Teacher Efficacy as a Form of Self-Efficacy
This chapter helps the reader to locate teacher efficacy within its theoretical framework. As Borg (2003) has noted, definitional and conceptual clarity is an essential aspect of teacher belief research, and the lack of such clarity has been a recurring problem in teacher efficacy research (Klassen et al., 2011). The first section of this chapter introduces teacher efficacy with respect to its historical research context, shows how it is now viewed as a type of self-efficacy and discusses the theoretical underpinnings of efficacy beliefs by showing how teacher efficacy relates to social cognitive theory (SCT). The chapter then clarifies how efficacy beliefs differ from other self-beliefs, before introducing the reader to the process of efficacy belief development. The chapter concludes by providing a conceptual framework by which efficacy beliefs are assessed.
2.1 A Brief History: Two Views of Teacher Efficacy
Although teacher efficacy research now clearly reflects a self-efficacy orientation, it was born from two psychology theories related to control: Rotter’s (1966) locus of control and Albert Bandura’s (1977) self-efficacy theory. There was an early divergence in views about how the construct of teacher efficacy was viewed; one strand of research followed the locus of control distinction (i.e. the extent to which teachers perceived that they could influence student learning and achievement) while another strand of teacher efficacy research integrated Bandura’s theory of self-efficacy (i.e. teacher’s perceptions of capability to bring about the actions required for student learning and achievement to occur).
The first study to investigate ‘teaching efficacy’ was the 1976 Rand Corporation study (Armor et al., 1976) of reading programmes in Los Angeles. Two items were used to investigate differences between teachers regarding their outcome expectations:
(1) When it comes right down to it, a teacher really can’t do much – most of a student’s motivation and performance depends on his or her home environment
(2) If I try really hard, I can get through to even the most difficult or un-motivated students
(Armor et al., 1976: 73)
The researchers identified that a key teacher factor was their belief that ‘they could “get through” even to children with shaky motivation or home background’ (Armor et al., 1976: 38). They labelled this ‘personal efficacy’, and in a later Rand report, it was defined by Berman et al. (1977: 137) as ‘the extent to which the teacher believes he or she has the capacity to affect student performance’. The Rand studies demonstrated the value of teacher control beliefs, spurred action into teacher efficacy research and later studies using the two Rand items showed significant relationships between stronger ‘personal efficacy’ and student achievement (e.g. Ashton & Webb, 1986).
At the same time as the Rand study was being carried out, Albert Bandura (1977) was developing his theory of self-efficacy to discuss the role of cognition as a psychological process in learning and regulating behaviour. For Bandura (1997: 3), the crucial aspect of human agency was the individual’s beliefs in their personal efficacy, which he later defined as ‘beliefs in one’s capabilities to organize and execute the courses of action required to produce given attainments’. He argued that such beliefs act as cognitive mediators of behaviour, influencing action, effort, perseverance, resilience and stress coping strategies.
Bandura’s self-efficacy theory led to a reevaluation of the Rand items and a reinterpretation of the construct of teacher efficacy. The Rand items were designed to reflect a locus of control orientation, by examining the extent to which respondents perceived internal versus external control of what could be achieved by teachers (see Berman et al., 1977). However, the Rand study authors discussed ‘personal efficacy’ as a type of ‘confidence’, and noted that such teachers put forward extra (often special) efforts to help students because they were ‘confident that their teaching would yield positive results’ (Armor et al., 1976: 38).
A number of researchers (Ashton et al., 1983; Dembo & Gibson, 1985; Gibson & Dembo, 1984) used Bandura’s (1977) self-efficacy theory to reinterpret the two Rand study items, arguing that self-efficacy theory provided a clearer conceptual framework. To these researchers, Item 1 ‘When it comes right down to it, a teacher really can’t do much – most of a student’s motivation and performance depends on his or her home environment’ reflected outcome expectancy, while Item 2 ‘If I try really hard, I can get through to even the most difficult or un-motivated students’ oriented towards self-efficacy, where such ‘beliefs would be teachers’ evaluation of their abilities to bring about positive student change’ (Gibson & Dembo, 1984: 570).
Bandura (1977) had differentiated between ‘outcome expectations’ and ‘efficacy expectations’ by highlighting the difference between the effect on individual behaviour, where
outcome expectancy is defined as a person’s estimate that a given behaviour will lead to certain outcomes. An efficacy expectation is the conviction that one can successfully execute the behaviour required to produce the outcomes. Outcome and efficacy expectations are differentiated, because individuals can believe that a particular course of action will produce certain outcomes, but if they entertain serious doubts about whether they can perform the necessary activities such information does not influence their behaviour. (Bandura, 1977: 193)
In other words, the crucial aspect to Bandura was the extent to which individuals perceived that they have the necessary skills and competence to carry out actions in order to produce outcomes, rather than simply whether people believed that their actions produced outcomes. With respect to the two Rand items, teachers may reject Item 1, as they may believe that teachers can influence student achievement (i.e. strong outcome expectancy), but also disagree with Rand Item 2, as they may not have strong self-beliefs in their capability to effectively help lesser-motivated students (i.e. low personal efficacy). Accordingly, they may not put forth the effort to influence student learning.
Research findings have generally supported this distinction. Gibson and Dembo (1984) developed a 30-item questionnaire and tested this difference, using factor analysis to show that each of the Rand items loaded on separate factors. More recent studies (e.g. Ho & Hau, 2004; Skaalvik & Skaalvik, 2007) have subsequently shown perceived external control and teacher efficacy to be different constructs.
Finally, in a seminal article that brought the two teacher efficacy strands together, Tschannen-Moran et al. (1998: 22) argued that teacher efficacy should be conceptualised as a form of self-efficacy, defining it as the ‘teacher’s belief in his or her capability to organize and execute courses of action required to successfully accomplish a specific teaching task in a particular context’. Although the authors questioned whether the interpretation of the Rand Item 1 as outcome expectancy was appropriate, it seems clear that the interpretation of the Rand items as two distinct constructs started the realignment of the field towards a view of teacher efficacy as a form of self-efficacy.
Later, Tschannen-Moran and Woolfolk Hoy (2001) introduced the Teacher Sense of Efficacy Scale (TSES), an efficacy scale with long (24 item) and short (12 item) versions, from which exploratory factor analysis had identified three dimensions of teacher efficacy (Student Engagement, Instructional Strategies, Classroom Management). Developed by a team of researchers and graduate students at The Ohio State University (and originally called the Ohio State Teacher Efficacy Scale [OSTES]), this new means of assessing teacher efficacy would have a major impact on the field. Their article may be the most widely cited in the field (see Kleinsasser, 2014), and has stimulated research into teacher efficacy internationally (e.g. Cheung, 2006, 2008; Tsui & Kennedy, 2009), including studies of language teacher efficacy (e.g. Chacon, 2005; Swanson, 2010a; Yilmaz, 2011).
As research developed, it began to be recognised that teachers do not work alone; they work as part of teams. Bandura (1997) has suggested that efficacy beliefs influence the choices of individuals and organisations; therefore, teacher efficacy may be collective. Given the social structure and context of schools as social enterprises with numerous actors working together to help students learn and achieve, Goddard et al. (2000: 482) explained that ‘collective efficacy is associated with the tasks, level of effort, persistence, shared thoughts, stress levels, and achievement of groups’. In a study of teachers carried out in Norway, Skaalvik and Skaalvik (2007) further explained the potential importance of perceived collective teacher efficacy, stating,
teachers do not always work alone. In most Norwegian schools, teachers now work in teams sharing responsibility for a larger group of students. The actual instruction is partly done by individual teachers in smaller groups and partly by pairs of teachers in a larger group. Much of the organizing and the planning are done in teacher teams. The individual teachers’ self-efficacy may therefore be dependent on the functioning of the team. (Skaalvik & Skaalvik, 2007: 613)
Thus, teacher capability has both individual and collective aspects, incorporating ‘an organizational dimension’ (Goddard et al., 2004a: 4). Professional practice, teacher influence on instruction, text selection and materials development are examples of activities in which beliefs would be formed about the capability of the teaching faculty. These perceptions may affect the efficacy assessments of teachers, thus individual teacher efficacy may be influenced by team dynamics and beliefs about faculty capability.
Studies have demonstrated positive relationships between individual and collective efficacy (Goddard & Goddard, 2001; Skaalvik & Skaalvik, 2007), and indicated that collective efficacy may be crucial for curriculum change (Goddard et al., 2004b). There is also evidence that stronger perceived collective efficacy is related to student achievement (Goddard, 2002; Goddard et al., 2015; Tschannen-Moran & Barr, 2004). A series of meta-analyses (Hattie, 2012, 2015) have examined the relationship between student achievement and educational activities, identifying collective teacher efficacy to exert one of the strongest positive influences. Accordingly, this type of teacher efficacy appears to be a valuable area for future research and has many avenues for further exploration. Most collective teacher efficacy research has been carried out in the United States or in countries with similar cultural backgrounds (Klassen et al., 2011), thus a greater understanding is needed about the relationship between individual and collective efficacy, and the areas where individual and collective teacher efficacy beliefs operate, in international (i.e. non-Western) contexts.
2.2 Teacher Efficacy within Social Cognitive Theory
The construct of self-efficacy is central to Bandura’s (1986) SCT, which views individuals as active agents whose interpretations of the results of their performances inform and alter ‘their environments, and self-beliefs, which in turn inform, and alter their subsequent performances’ (Pajares, 1996: 542). SCT is based on a reciprocal relationship between behaviour, environment and personal factors (e.g. cognitive, biological), with each dynamically and bi-directionally interacting and influencing the others – what is known as ‘triadic reciprocal causation’ (Bandura, 1986).
Humans are capable of self-referent thought, with the capacity to regulate their effort and reflect on their activities. Thus, SCT rejects a behaviouralist view of the world, in which individuals primarily respond to their environments; it recognises that individuals can plan and have the capacity for self-influence. As Bandura (1997: 3) has stated, individuals’ behaviour (e.g. teaching practice) is dynamically influenced by personal factors (e.g. available pedagogic skills) and the environment they are in (e.g. the school context), where ‘people are contributors to, rather than the sole determiners of, what happens to them’. Simply put, individuals can shape and change their environments by using the skills they perceive to be available to them and choosing different courses of action. Thus, teachers can influence, but not determine, the learning that occurs within, and beyond, their classrooms.
An SCT perspective indicates that teacher beliefs (and other personal factors), behaviour and context reciprocally influence each other. In fact, such a view aligns with conceptual frameworks of language teacher beliefs, such as Borg’s original (2003) and revised (Borg, 2006) models, which focus on the dynamic relationship between beliefs and contextualised behaviour. Thus, for language teachers, self-beliefs and perceptions of individual ability (e.g. of second language [L2] ability) may influence efficacy beliefs towards (or against) teaching tasks that require the teacher to act as a language model. The teacher’s interpretation of the results of the behaviour (i.e. whether they are perceived to be successful or unsuccessful) may then influence beliefs and other individual factors (e.g. stronger perceived capability to achieve the task in the future). Furthermore, both may also be influenced by context (e.g. working at a school with motivated students versus with students who have no desire or need to speak English). Thus, in the assessment of their efficacy beliefs, individuals are likely to weigh up the difficulty of carrying out the task (e.g. is it possible or valued?) and the expected outcome (i.e. will it be perceived to influence student learning and development?), alongside the skills they perceive available to use in achieving the task (see Figure 2.1).
Figure 2.1 Conceptual framework of self-efficacy assessment
Accordingly, self-efficacy beliefs are argued to influence agency, motivation and self-regulation, and to mediate states of anxiety. Agency is concerned with individuals’ capacity to control and coordinate their actions, beliefs and emotions to reach goals. Therefore, agency is the driver of intentional acts (as opposed to outcomes, which may be unintended). Bandura (2001: 10) claimed that ‘efficacy beliefs are the foundation of human agency’ and summarised their importance, stating ‘whatever other factors may operate as guides and motivators, they are rooted in the core belief that one has the power to produce effects by one’s actions’. As self-efficacy beliefs are perceptions of capability, these beliefs influence choice and effort towards goals, that is, individuals ‘regulate their level and distribution of effort in accordance with the effects they expect their actions to have’ (Bandura, 1986: 129).
Efficacy beliefs have also been consistently shown to have a negative relationship with anxiety (Mills et al., 2006; Pajares & Kranzler, 1995; Swars et al., 2006). Such a relationship is not surprising, as people who perceive a negative emotional and threatening response towards a task (e.g. anxiety towards teaching via English) may also believe themselves less capable of bringing about actions to complete it. Thus, anxiety and other affective responses are likely to be influences on the development of efficacy beliefs. However, research has also indicated that efficacy beliefs mediate the impact of anxiety on performance (Pajares & Kranzler, 1995; Siegel et al., 1985), as individuals with stronger capability to cope with task demands are also more likely to manage the negative influence of anxiety. This may be due to individuals responding to their self-doubts by ‘acquiring knowledge and skills’, which provide them with stronger perceived capability to manage the threatening situation (Bandura, 1997: 76).
Accordingly, self-efficacy beliefs are hypothesised to be key to motivation (Zimmerman, 2000), and their causal influence has been demonstrated in the wider psychological research literature (Pintrich & Schunk, 1996; Schunk, 1995) for language learning (Kormos et al., 2011; Oxford & Shearin, 1994; Tremblay & Gardner, 1995) and for teaching (Skaalvik & Skaalvik, 2014). In summary, individuals with stronger perceived capability for specific tasks are more likely to choose and expend effort on such activities.
Research within the field of teacher efficacy has supported these ideas, as teachers with stronger efficacy beliefs towards a subject (or subject area) are more likely to be more persistent and spend greater time teaching it (Enochs et al., 1995; Mulholland & Wallace, 2001; Woolfolk Hoy & Davis, 2006), and are likely to have stronger commitment to teaching (Chan et al., 2008; Chesnut & Burley, 2015; Ware & Kitsantas, 2007). On the other hand, low teacher efficacy towards a subject is associated with spending less time teaching that subject area (Enochs & Riggs, 1990). Furthermore, findings have also supported a link between teacher and student efficacy and student achievement (Bolshakova et al., 2011; Chang, 2015), indicating that teacher efficacy beliefs may influence not only teacher motivation and effort, but also student motivation.
2.3 How Efficacy Beliefs Differ from Other Self-Constructs
Although they are intertwined, self-constructs are generally divided by those that are primarily cognitive (i.e. centred on the knowledge and beliefs that individuals have about themselves and their abilities) versus those that are affective (i.e. primarily emotional and based on the feelings that people have about themselves, see Williams et al., 2015). This section briefly discusses how self-efficacy can be differentiated from other self-constructs by orientation, domain and character of evaluation.
As a form of expectancy belief, efficacy expectations represent perceptions of capability to carry out actions in the pursuit of outcomes. Firstly, efficacy beliefs have a future orientation (Zimmerman, 2000); they involve individuals carrying out an evaluation of their personal skills towards (potential) future behaviours. Second, efficacy beliefs are context and task dependent; the strength of these beliefs varies by the teaching task and the situation within which it is carried out (see Wyatt, 2018a). Finally, as explained in Section 2.6 (also see Figure 2.1), they involve individuals carrying out internal cognitive appraisals of the task conditions and the skills that are required for successful completion.
This contrasts efficacy beliefs with other self-constructs, such as self-concept and self-esteem, which do not have a future orientation, but rather a descriptive function concerned with evaluations of one’s competence (self-concept) and emotional self-worth (self-esteem). For example, Mercer (2008) explained that self-esteem has the strongest evaluative component, and is primarily affective, as it brings together the positive and negative feelings that people hold of themselves. Although self-esteem is also influenced by self-beliefs, these are primarily attitudinal (Rosenberg, 1965), in that self-esteem integrates a person’s self-beliefs of their value or worth. For example, global self-esteem evaluations can be shown by statements such as ‘I am a good person’ (Leary & Downs, 1995: 124).
Self-concept refers to a ‘general self-descriptive construct’ that integrates both ‘self-knowledge and self-evaluations’ (Zimmerman, 2000: 84). Generally, self-concept is viewed as a hierarchy (Marsh & Shavelson, 1985) with ‘global’ self-concept at the top, which brings together the self-beliefs people have towards their competency in different domains. Self-concept centres around a person’s evaluation of their abilities (Williams et al., 2015) and involves individuals comparing themselves with others – and against themselves – with respect to standards of competency and worth (Shavelson et al., 1976). Thus, while both self-concept and self-efficacy include the same central element (perceived competence), Mercer (2008) has explained that self-concept beliefs operate at a higher level, incorporating emotional and cognitive perceptions of competency towards a domain of activity (e.g. towards teaching). The normative assessment of evaluative competency means that self-concept is often assessed with items using ‘I am good at …’ (Pajares, 1996), reflecting beliefs about ‘personal qualities’ and competencies (Zimmerman, 2000: 83). A key component of self-concept involves the integration of comparative information about oneself or one’s abilities within a domain, either versus others or personal benchmarks. On the other hand, self-efficacy refers to the self-beliefs of capability people have towards tasks, which come about from an assessment of their skills to successfully complete them. As a result, although social comparison plays a part in efficacy development (often vicariously, see Section 2.4), the assessment of self-belief is more focused on skills towards the task rather than beliefs about oneself.
Generally, efficacy beliefs have been shown to be stronger predictors of behaviour in comparison to other self-concept measures (Bandura, 1997; Bong & Skaalvik, 2003; Pajares & Miller, 1994) due to their strong task-focused nature. Bong and Skaalvik (2003) have suggested that self-efficacy is a better predictor of performance due to its future orientation and stronger task-focus specificity. As Zimmerman (2000: 84) has explained, an efficacy item may ask students to ‘rate their certainty about solving a crossword puzzle of a particular difficulty level’ while a self-concept item may ask them ‘how well they expect to do on the puzzle in comparison to other students’. The first provides a measure of internal capability to complete a future task, while the second provides a perception of normative ranking. Accordingly, the relationship between the self-concept belief and actual behaviour may become less pronounced.
Efficacy beliefs can also be differentiated from self-confidence, as they reflect not only the strength of a belief, or of one’s general capability, but also perceptions of capability towards a certain task. Confidence primarily ‘reflects a degree of certainty about a perception, event, or outcome’ (Cramer et al., 2009: 322), thus confidence itself is primarily concerned with belief strength (Merkle & Van Zandt, 2006); it does not necessarily include what the belief is about (Bandura, 1997).
Individual ‘self-confidence’ represents the strength of an individuals’ belief in their personal capability or competence, and given the hierarchical nature of self-concept, self-confidence appears to be contributing towards general self-concept (Schunk, 1991). Thus, teachers with strong ‘teacher self-confidence’ may have a general belief that they have high competency as teachers (i.e. stronger teacher self-concept). However, such beliefs about self-confidence are not necessarily predictive of behaviour, as these do not consider the teaching task or context. In other words, such beliefs remain ‘nondescript’ (Bandura, 1997: 382) as they are not focused on capability towards a certain activity, nor are they embedded within a theoretical system that describes how such beliefs may influence behaviour. As a result, researchers (Bandura, 1997; Bong & Skaalvik, 2003; Cramer et al., 2009) have suggested that self-efficacy beliefs are a more appropriate means for investigating task-focused perceptions of perceived capability (e.g. towards L2 teaching activities) in comparison to self-confidence and other self-concept belief constructs.
Furthermore, when some researchers (e.g. Shrauger & Schohn, 1995: 257) have attempted to show how self-confidence at the domain (i.e. a higher-order self-concept belief) level is related to behaviour, they have created new divisions of hierarchy under self-confidence, with ‘situation-specific’ confidence used to describe confidence at the task level. Pajares (1992) warned that teacher cognition researchers should avoid using different names for existing constructs, and the task-focused situation-specific confidence that Shrauger and Schohn (1995: 257–258) refer to is difficult to separate from Bandura’s construct of self-efficacy. Others (e.g. Stajkovic, 2006) accept that self-confidence, as a feature of one’s self-concept, is reflective of various underlying beliefs, including one’s perceptions of capability to carry out tasks within specific contexts (i.e. self-efficacy). Indeed, it is now generally accepted that efficacy beliefs do contribute towards self-concept (Mercer, 2008; Williams et al., 2015) as individuals with greater perceived capability towards tasks within a domain may be more likely to have stronger perceptions of ability about their competence towards them. Also, as efficacy beliefs are theorised to contribute towards teacher agency, accordingly they may also influence wider, dynamic constructs such as teacher identity (Beauchamp & Thomas, 2009).
Given its original in psychology, it is important to note that teacher efficacy is a tightly controlled construct; it may not necessarily provide insights about how teachers view themselves or their general competence. The efficacy beliefs that individuals’ hold towards different tasks can be grouped together towards certain dimensions of activity (usually via the use of factor analysis on questionnaire data), providing researchers with information about individuals’ perceptions of their skills for achieving outcomes within different areas of teaching (e.g. towards instruction, towards discipline). However, the insights gained from such dimensions are tightly limited to the self-beliefs of capability towards carry out actions; they do not necessarily provide information about the value that teachers place on them, nor the extent to which teachers appraise them in their self-evaluations of esteem. Thus, a teacher could have strong efficacy for some teaching tasks but also consider themselves to be a poor teacher, due to the different ‘self-esteem reactions’ (Zimmerman, 2000: 84) that they attend to (e.g. the perceived value of the task) when constructing their self-concept beliefs about their ability as a teacher. As a result, while efficacy may contribute towards other self-beliefs, equally, substantial gaps may be present between efficacy expectations, self-concept beliefs, self-esteem evaluations and broader constructs such as teacher identity.
2.4 The Development of Efficacy Beliefs
Bandura (1977, 1986, 1997) posited that the development of self-efficacy beliefs is based on individuals’ interpretations of their experiences, others’ reactions and their emotions. He ascribed the development of self-efficacy beliefs to four factors: mastery experiences, vicarious experiences, social persuasion and physiological states.
Mastery experiences are based on perceptions of direct performances, where (perceived) successful performance of a task or activity leads to greater confidence in being able to complete the task proficiently in the future. It is argued that these are the strongest drivers of self-efficacy beliefs. The second factor is vicarious experiences; watching others or self-modelling one’s own performance may lead to increased efficacy. Social persuasion constitutes the third factor, as positive feedback or encouragement from others can be useful for overcoming self-doubt. Finally, an individual’s emotional and physical reactions and responses can influence their self-efficacy beliefs. Equally, while these factors may contribute positively (i.e. lead to stronger efficacy), negative attributions (i.e. perceptions of failure) can lead to weaker efficacy beliefs, as the interpretation of the event is crucial. Thus, for some teachers, attributions from perceiving a higher heart rate (i.e. a physical response) prior to a class may be interpreted negatively as a sign of nervousness, while for others it may signal excitement and be interpreted positively. For example, positive physiological states have been shown to be related to stronger teacher efficacy beliefs, with teachers feeling ‘excitement while teaching literature’ (Mills, 2011: 71) or ‘feeling energized’ (Morris & Usher, 2011: 238) after class, leading to stronger efficacy beliefs.
Within a teaching context, the development of teacher efficacy beliefs may therefore be influenced by the interpretations of teachers about their experiences as a trainee or teacher; self-modelling and observation of others (e.g. participation in seminars); feedback and encouragement from other teachers, students and parents; as well as affective processes such as anxiety or perceived stress. Research has provided support for the four factors theorised to influence self-efficacy beliefs, and generally substantiated the claim that mastery experiences tend to be the strongest influence on teacher efficacy beliefs (Morris & Usher, 2011; Mulholland & Wallace, 2001; Tschannen-Moran & Woolfolk Hoy, 2007). For example, in a study of ‘novice’ and ‘career’ teachers, Tschannen-Moran and Woolfolk Hoy (2007) found mastery experiences made the strongest contribution to teachers’ self-efficacy for teaching. However, Usher and Pajares (2008) also warned that while mastery experiences are usually most influential, sources may become more or less important in different contexts (e.g. for various domains of activity) or for different individuals (e.g. by gender). This point has also been noted by Klassen et al. (2011) in their review of teacher efficacy research, where the authors question the extent to which the strength of these four factors is culturally specific – a key issue for consideration in studies of language teacher efficacy where respondents are often from different cultural backgrounds.
Furthermore, efficacy sources appear to act in concert with each other. Particularly, mastery experiences and social persuasion have often been found to act simultaneously as sources of efficacy beliefs (Morris & Usher, 2011; Tschannen-Moran & McMaster, 2009), perhaps because many teaching training experiences involve practice teaching (i.e. a potential mastery experience) and feedback from senior teachers, trainers or students about competency (i.e. involving social persuasion). An example comes from Morris and Usher (2011), who used semi-structured interviews to investigate the sources of efficacy for university professors who had been recognised for ‘excellence in teaching’. Although the study found that vicarious experiences and affective states were positive influences on teacher efficacy beliefs, mastery experiences and social persuasion were identified to be the most influential sources of self-efficacy. The authors suggested that mastery experiences and social persuasion were closely integrated, as ‘in the context of college teaching, appraisals of past performance are almost always informed by social persuasions’ (Morris & Usher, 2011: 243). Thus, separating the individual sources and indeed measuring them (see Morris et al., 2016) remain difficulties for teacher efficacy researchers. In fact, other researchers have included perceived mastery experiences with social persuasion as one integrated source of efficacy information. For example, in a study of 198 pre-service elementary teachers in Greece, Poulou (2007) combined mastery experiences and social persuasion based on findings of previous studies (Anderson & Betz, 2001; Lent et al., 1996; Matsui et al., 1990), which had identified personal mastery experiences and social persuasion and emotional arousal to cluster together as one factor related to direct personal experience.
Vicarious experiences have also been shown to positively influence teaching efficacy beliefs. There have been mixed results about the influence of mentors and models, with some studies showing a significant influence (Knoblauch & Woolfolk Hoy, 2008; Morris & Usher, 2011; Siwatu, 2011a), but others showing no influence (Rots et al., 2007). Simulated modelling (imagining oneself teaching) has also been identified as a significant source of information influencing positive efficacy development in pre-service science teachers (Palmer, 2006), while Bautista (2011) found that (video) observation could influence teacher efficacy beliefs by providing teachers with new ideas and skills. Furthermore, as the different sources appear to be integrated, it seems that vicarious sources are most effective when they are followed by personal experience, leading to perceptions of mastery. For example, Tschannen-Moran and McMaster (2009) found that professional development lectures could lead to a significant development in teacher efficacy if they were also followed by practice and individual feedback.
Mulholland and Wallace (2001) carried out a longitudinal case study of a teacher making the change from pre-service to in-service teaching, and showed the difficulty of understanding how the various forces influence teacher efficacy beliefs, particularly whether such sources are always positive or negative. Using narrative inquiry, the authors examined the development of a novice elementary school teacher’s self-efficacy for teaching beliefs towards science, identifying mastery experiences and social persuasion to be, overall, the strongest sources of efficacy information. However, they also noted that the various forms of information (e.g. vicarious experiences, social persuasion) acted in different ways on the teacher, sometimes as a positive influence and sometimes in a negative fashion. The study showed the difficulty of assigning one direction (i.e. positive influence) to different factors, because each factor and each source is interpreted and assigned meaning by the individuals themselves.
The separation of these sources may become more difficult as time passes and technology becomes more integrated with teacher training, leading to virtual training opportunities. Generally speaking, simulated modelling is considered a type of vicarious experience (Palmer, 2006). However, when trainees act in a virtual teaching environment with ‘avatars’ (i.e. computer-generated virtual students), such experiences appear to be much closer to ‘real’ mastery experiences. Bautista and Boone (2015) suggested that such experiences contributed towards ‘cognitive pedagogical mastery’ or pedagogic knowledge that can be used during teaching as the simulated experience helped participants to imagine themselves as successful teachers. However, it could be argued these experiences are direct and personal teaching experiences (e.g. when the individuals are confronted by avatars who ask ‘why are we doing this?’ and the teachers must respond), thus it seems clear that definitional problems may become more apparent as technology progresses and the real and virtual worlds become further interconnected.
Debate continues about whether the original definitions and conceptions of the four sources should be expanded as self-efficacy is considered in different areas, such as teaching development. In order to preserve conceptual clarity, some (e.g. Morris et al., 2016) have argued that perceptions of school support and mentoring should not be considered ‘social persuasion’ unless the experience specifically involved evaluative feedback. However, others (e.g. Wyatt, 2015: 16) have argued that the source ‘verbal persuasion’ should be expanded to include any input received from others that teachers use to ‘reflect, conceptualize or plan’, rather than only evaluative feedback. Indeed, as the evaluations of others may be inferred from the ‘messages’ that teachers receive (Chong & Kong, 2012), rather than the actual words expressed, it seems clear that the focus should be on the extent to which teachers perceive the advice, support or feedback from others as an influence on their efficacy beliefs.
Finally, another key area for attention by teacher efficacy researchers is the extent to which the development of efficacy beliefs is dependent on cultural context (Klassen et al., 2011). Efficacy sources do not automatically influence efficacy beliefs, but are weighed and assessed according to the information that individuals select and deem to be relevant (Bandura, 1997). Some research has indicated that specific sources (such as social persuasion) may be considered more strongly by teachers in different cultures (Phan & Locke, 2015), and is a factor explored in this book. Chapter 10 discusses the ways in which cultural elements of the Japanese teaching environment appear to influence the attendance of teachers towards social factors as sources of efficacy information, likely due to cultural values that emphasise a collective orientation.
2.5 Additional Factors that May Influence Efficacy Beliefs
While there is strong evidence for the four theorised sources of efficacy beliefs outlined in Section 2.4, a number of other factors have been suggested as additional sources of efficacy, or as factors that may influence efficacy. One key source is the role of content and pedagogical knowledge. As Morris et al. (2016: 22) have noted, it appears that ‘teachers’ knowledge, and their beliefs about that knowledge, can play an important role in their development of self-efficacy’ as ‘knowing the material, and knowing how to teach it well, can improve teachers’ sense of efficacy’. The importance of (perceived) content or subject-matter knowledge has been clearly shown in the research literature, with studies of science teachers showing a positive relationship between the number of science classes they had taken and their teacher efficacy beliefs towards teaching that subject (Cantrell et al., 2003; Mulholland et al., 2004). With respect to language teaching (discussed further in Chapter 3), studies have also shown that foreign language proficiency was positively related to efficacy beliefs (Chacon, 2005; Yilmaz, 2011).
However, it is also worth remembering that ‘expectation alone will not produce desired performances if the requisite competencies are lacking’ (Bandura et al., 1977: 138) as efficacy beliefs are assessed against the skills that individuals perceive to have available to them (Bandura, 1977). In other words, a certain level of knowledge and competency is required for individuals to be able to bring about successful performances. Therefore, it seems natural that instruction about content or teaching strategies will influence efficacy beliefs, particularly for those beginning to teach, as these influence the skills that individuals can utilise and draw on. In fact, some researchers (e.g. Raudenbush et al., 1992) have interpreted self-efficacy as ‘a cognition that mediates between knowledge and action’, thus the key point is not whether instruction influences efficacy, but whether efforts to grow teacher knowledge are likely to have a stronger effect when such activities also attend to efficacy development (i.e. by providing opportunities for enactive mastery).
Furthermore, as Morris et al. (2016) have explained, knowledge itself is derived from experience. They argued that knowledge should not be considered a primary source of efficacy beliefs, but rather that it is developed via experience and is therefore a mediating factor appraised when individuals assess their efficacy beliefs towards a given task. Some examples, which align with the theorised sources of efficacy, are from strategies (i.e. knowledge) developed via personal teaching (i.e. an enactive mastery experience) or from learning teaching skills (i.e. strategies) by observing others or self-modelling (i.e. via vicarious experiences).
In a key study illuminating this discussion, Palmer (2006) investigated the teacher efficacy beliefs of 190 pre-service elementary school teachers at a university in Australia. The author brought together results from three surveys to identify two additional sources of teacher efficacy differentiated from enactive mastery and modelling (i.e. vicarious experiences). Palmer (2006: 349) argued that ‘cognitive content mastery’ (an understanding of science knowledge that developed from a learning experience) and ‘cognitive pedagogical mastery’ (a learning experience that influenced understanding of science teaching) were two ‘distinctive mastery experiences in their own right’ and requirements of enactive mastery. Students in the study directly referred to ‘the procedures’ or ‘great strategies’ (Palmer, 2006: 346) that they had learned from the lessons or activities in the course, leading the author to suggest that these should be considered separate, distinct sources of efficacy. However, this returns us to the discussion from the previous paragraph, where such experiences could also be interpreted as vicarious experiences that provided models later integrated into teaching practice. Findings presented in this book (see Chapters 7 and 8) indicate that past learning experiences may act as sources of efficacy-forming information; however, these appear to contribute towards knowledge that can be used in a task (i.e. language skills) or may indirectly influence efficacy via attitudes (i.e. affective states) towards different instructional approaches.
Finally, as Morris et al. (2016) also noted, not all sources of knowledge conveniently fall into the theorised sources of efficacy beliefs. They provided the example of independent reading as a source of pedagogic knowledge (identified by Buehl & Fives, 2009), which is not, in itself, clearly a modelling or (cognitive) mastery experience. Thus, it seems that there may be additional sources of information that can influence efficacy development, perhaps indirectly, by adding to the skills that individuals perceive to have available to them. In summary, it seems that a key area for teacher efficacy researchers is identifying other sources of teacher efficacy, and specifically gaining better understanding of the experiences that lead to the formation of content and teacher knowledge.
2.6 Assessment of Efficacy Beliefs
Efficacy beliefs can be considered with respect to level, generality and strength (Bandura, 1977, 1997, 2006) based on perceptions of personal skill towards specific teaching activities within different contexts. For example, teacher efficacy expectations may vary by level (e.g. carrying out teaching to help students achieve different degrees of examination success), generality (e.g. towards teaching a specific subject versus teaching overall) and strength (e.g. efficacy towards one task or dimension versus another). Accordingly, it seems apparent that different sources of information will be drawn on in the assessment of efficacy beliefs. This section brings the different factors together within a conceptual framework of teacher efficacy assessment based on SCT (Bandura, 1997), and by drawing together frameworks of other researchers (Gist & Mitchell, 1992; Morris et al., 2016; Tschannen-Moran et al., 1998; Wyatt, 2015).
Efficacy beliefs are considered against the level of task demand required, and the contextual situation (Bandura, 1997). Gist and Mitchell (1992) suggested that there are three processes that individuals draw on in the assessment of their efficacy beliefs. First, there is an analysis of the task, including consideration of what a successful performance entails and what is required for such a performance. Task familiarity is the second key attribute assessed, where individuals consider attributions of their experiences, drawing on the four sources of efficacy beliefs, before finally assessing personal (e.g. skill, knowledge, available effort) and situational factors (e.g. distractions) that may influence the skills they have available or the difficulty of the task. Studies have provided some support for the Gist and Mitchell (1992) model, showing the mediating influence of time pressure (Salanova et al., 2003) and task complexity (Bolt et al., 2001; Mangos & Steele-Johnson, 2001) alongside efficacy on performance. Tschannen-Moran and Woolfolk Hoy (2007) extended this model to the context of teaching, explaining that,
in assessing beliefs about their teaching capability in a particular context, teachers make two related judgments: the requirements of an anticipated teaching task and an assessment of their personal teaching competence in light of those requirements. (Tschannen-Moran & Woolfolk Hoy, 2007: 5)
In other words, teachers assess the personal skills and strategies they can utilise (including any perceived weaknesses, such as a lack of content knowledge) against any constraints within the teaching context (e.g. level of students, level of support) in assessing their efficacy expectations towards teaching tasks in the pursuit of student learning. These factors are presented in Figure 2.1.
The integration of these various sources is a personal process as it relies on each individual’s weighting of the different pieces of information (Bandura, 1997), where the ‘differential impact of each of these sources depends on cognitive processing what is attended to, what is remembered, and how the teacher thinks about each of the experiences’ (Tschannen-Moran et al., 1998: 18). Due to triadic reciprocal causation, the process is theorised to include attributions from behaviour (i.e. enactive mastery experience, vicarious experiences, attention to feedback from others and of one’s affective states); attention to contextual factors that influence perceptions of task difficulty and/or influence individuals’ knowledge of and capability to cope with the situation; plus perceptions of personal knowledge and skills that individuals can bring to the task.
Research has consistently shown that personal and contextual factors are related to efficacy beliefs. For example, significant relationships have been identified between efficacy beliefs and perceptions of support (Capa Aydin & Woolfolk Hoy, 2005), resources available (Tschannen-Moran & Woolfolk Hoy, 2007) and school setting (Chong et al., 2010; Knoblauch & Woolfolk Hoy, 2008; Siwatu, 2011b). Beyond content knowledge, other personal variables such as experience (Cheung, 2006; Fives & Buehl, 2009; Tschannen-Moran & Woolfolk Hoy, 2007) and gender (Cheung, 2006; Ross et al., 1996) have been found to be associated with efficacy beliefs.
Discussion continues about how teacher self-efficacy beliefs grow and change. Wyatt (2014) argues that reflection is a crucial aspect neglected in models of efficacy change (such as the model of Tschannen-Moran et al., 1998). However, as Bandura (1997: 51) has noted, ‘efficacy beliefs are structured by experience and reflective thought’, where reflection is a key aspect related to the cognitive processing that individuals engage in when they consider their efficacy beliefs. Bandura (1997: 79) stated that information from efficacy sources ‘is not inherently enlightening. It becomes instructive only through cognitive processing of efficacy information and through reflective thought’. Much of this debate stems from another criticism of the Tschannen-Moran et al. (1998) model by Wheatley (2002), who has emphasised the role of ‘efficacy doubt’ as an influence on teacher beliefs and behaviour. Simply put, unsuccessful experiences can also inform future efficacy via reflection (Cheung, 2006; Wyatt, 2014, 2015). This is a valuable point for consideration, as the role of self-doubt appears to be valuable in driving reflective thought.
Particularly during the ‘skill development phase’ (e.g. for teachers developing new teaching skills, such as those discussed in Wyatt, 2010a, 2010b, 2013b), Bandura (1997: 76) has suggested that, ‘some self-doubt about one’s efficacy provides incentives to acquire the knowledge and skills needed to function successfully’. In other words, efficacy beliefs are not only driven by perceptions of success, but may also be mediated by perceptions of failure, anxiety and self-doubt. A number of researchers have highlighted the need for establishing a social environment (Cabaroglu, 2014; Karimi, 2011; Wyatt & Dikilitaş, 2015; Zonoubi et al., 2017) that provides teachers with opportunities for positive efficacy development (e.g. via action research and observation) when they experience such doubts.
Thus, reflection appears to be a key aspect of the cognitive processing that individuals engage in during the assessment of their efficacy beliefs, and is represented as central to the conceptual framework presented in Figure 2.1. Self-reflective thought involves the individual noticing and evaluating information from different sources. In other words, individuals carry out ‘strategic thinking about how to manage the environment’ and evaluate the ‘adequacy’ of their ‘knowledge, thinking skills, capabilities, and action strategies’ (Bandura, 1997: 5). Such considerations integrate the four sources of efficacy information, including the role of task knowledge and perceived skill (Gist & Mitchell, 1992; Morris et al., 2016; Tschannen-Moran et al., 1998), while also recognising that source information may be directly or indirectly related to experience with the task. Thus, an enactive mastery experience may directly influence efficacy beliefs via attributions of perceived success. It may also indirectly contribute to efficacy beliefs via information that is appraised by the individual, such as by providing knowledge of the task demands. The extent to which the source and knowledge are attended to (i.e. the cognitive process) is reliant on the individual, where ‘knowledge, competence, and various forms of self-knowledge and self-belief act in concert’ (Usher & Pajares, 2008: 790). As Bandura (1986, 1997) and others (Wyatt, 2015) have noted, while self-efficacy beliefs are key mediators of behaviour, they do not work alone and in isolation, thus other personal factors are likely to exert influence on efficacy beliefs (directly and indirectly). Finally, the conceptual framework develops on previous models by emphasising that beliefs and performances are located within and assessed against the contextual situation; significant changes in context are likely to lead to change in efficacy expectations (Wyatt, 2013b, 2014). This framework reflects findings presented later in this book, as efficacy sources, contextual factors and other personal factors (e.g. beliefs about language learning) appear to influence language teacher efficacy beliefs.
As Tschannen-Moran and Woolfolk Hoy (2001: 783) have explained, ‘teacher efficacy is a simple idea with significant implications’. Research has suggested that teachers with a lower sense of efficacy experience more stress, that teacher efficacy is negatively correlated with teacher burnout (Betoret, 2006; Brouwers & Tomic, 2000; Skaalvik & Skaalvik, 2007) and that teachers with a higher sense of efficacy are more likely to be committed to teaching (Chan, 2008; Chesnut & Burley, 2015; Ware & Kitsantas, 2007). Furthermore, teachers with stronger efficacy are likely to be more persistent (Enochs et al., 1995; Mulholland & Wallace, 2001; Woolfolk Hoy & Davis, 2006) and more open to innovation and change (Ghaith & Yaghi, 1997; Guskey, 1988; Skaalvik & Skaalvik, 2007). Accordingly, efficacy beliefs appear to act as predicted by SCT, by mediating the reciprocal relationship between self-beliefs, contextual factors and teaching behaviour.
This chapter has discussed the theoretical foundations of teacher efficacy as a form of self-efficacy, and has shown how teacher efficacy research has developed. It has introduced research findings that have provided support for the theorised ways in which teacher efficacy beliefs develop, and has provided a conceptual model for the assessment of teacher efficacy beliefs. The chapter has also highlighted a number of areas where further research is needed, such as developing a better understanding of the relationship between individual and collective efficacy, particularly in ‘international’ contexts such as many L2 teaching situations; understanding the development of teacher efficacy beliefs in different cultural contexts; and developing a better understanding of the types of information that may directly or indirectly influence efficacy beliefs. The research presented in this book contributes to information about these three areas, and is discussed in Chapters 6 through 10. Chapter 3 focuses specifically on the smaller field of language teacher efficacy. It reviews the knowledge that has been generated from previous language teacher efficacy studies, and suggests areas where future work is needed to develop greater understanding of the relationship between language teacher efficacy and other personal and contextual factors.
