The interplay of possible language teacher selves in professional development choices

1 Possible selves

Among the first to contribute a cogent ‘self’ construct were Markus and Nurius (1986) and Higgins (1987); they termed these ‘possible selves’. Broadly defined, the possible selves of Markus and Nurius (1986) represent ‘individuals’ ideas of what they might become, what they would like to become, and what they are afraid of becoming’ (p. 954). Their ‘ideal’, ‘feared’, and ‘expected’ selves (Carver, Reynolds, & Scheier, 1994) build a bridge between cognition and motivation. Previous findings have established that, while holding a great deal of relevance for the present, the orientation of possible selves as mediators of action is anticipatory towards the future, influencing as-yet unrealized experiences and behaviors (Oyserman & Markus, 1990; Oyserman, Terry, & Bybee, 2002; Penland, Masten, Zelhart, Fournet, & Callahan, 2000).

Higgins, Roney, Crowe, and Hymes (1994) see three possible selves as particularly significant in motivating and directing behavior: ‘the ideal self’, ‘the ought self’, and ‘the feared self’. The ideal self is the positive future reference for the actual self. It is not only a positive vision of self drawing on hopes and dreams, but is also an affective experience of the individual in that desired future state. This ideal self contains the properties that an individual would ideally like to possess and aspires to attain. Unlike the ideal self, the ought self refers to the mental and experiential representation of the traits one believes one should, or ought to, possess (Higgins, 1987), based on someone else’s view of their duties, obligations, and responsibilities. The value of ought selves is unique in representing either positive or negative reference points in contrast to the actual self (Higgins, 1996), yet the primary distinction between the ideal and ought selves is that the latter is based on extrinsic expectations that are other-derived (Pizzolato, 2006) and may not be particularly appealing. The converse of the ideal self is the feared self (Ogilvie, 1987), which serves as a negative future reference for the actual self, and represents the undesired facets of the self in the future. Images that comprise the feared self are those that an individual wants to avoid at all costs yet fears becoming (Carver et al., 1994).

These three future self guides derive their motivational function from their ability to guide a person’s behavior towards an ideal, away from a fear, and either towards or away from an ought. Possible selves’ self-regulatory power – that of their promotion/approach or prevention/avoid functions – is posited to come from the proportionate desire to reduce the distance or mismatch between one’s actual self and ideal self, and to increase the distance or mismatch between one’s actual self and feared self (Hendrix & Hirt, 2009; Higgins, 1996); an individual may, therefore, possess a combination of these three for any given future domain.

Much more than mere abstractions of self in future states, these three possible selves contain distinct affective and experiential elements. To the extent that cognitive events are rarely absent of emotions (Boldero & Francis, 1999; Karademas, 2006; Kobal & Musek, 2001; Taylor, Pham, Rivkin, & Armor, 1998), and because affective states are related to conflicts or discrepancies within the self-concept (Higgins, 1987), negative or positive emotion about oneself is an unavoidable component of possible selves (Markus & Nurius, 1986). Rather than being restricted to images, possible selves are rooted in a vital aspect of agency and include experiences of one’s agency in a future state of self (Erikson, 2007; Markus & Kitayama, 1991). Partly because they are derived from and integrate schematic knowledge of past and current experienced states (Oyserman, Ager, & Gant, 1995), and partly because emotion requires an experienced state to exist (Boyatzis & Akrivou, 2006; Cantor, 1990), possible selves include an experience of what it would be like to be in that future state ‘from the inside’ (Erikson, 2007, p. 349).

While the model of possible selves grew out of work in mainstream psychology on the self and motivation theory, they have now become prominent in the literature on L2 motivation (e.g. Dörnyei, 2005, 2009, 2010; Dörnyei & Ushioda, 2011) and have even begun to advance into the field of language teacher motivation as well (e.g. Kubanyiova, 2009).

2 Teacher development research

Following Mann’s (2005) distinction stemming from the contextually divergent perspective of the term ‘development’ (p. 104), I have chosen the term ‘continuing teacher development’ (CTD) for this study; the reasons for this follow. Development should be seen as a bottom-up process that is initiated by the language teacher himself/herself (Mushayikwa & Lubben, 2009); it is a process that extends beyond the initial teacher education or pre-service preparation received and must be seen as development that continues indefinitely throughout the entirety of one’s teaching career (Van Eekelen, Vermunt, & Boshuizen, 2006); at times deeply emotional (Wheatley, 2002; Zembylas, 2004), it serves both an enhancement function in relation to development to the teacher’s professional self, as well as one of repairing perceived inadequacies of the self (Clarke & Hollingsworth, 2002; Rowe & Sikes, 1989); it may involve a number of processes including pedagogic development, language development (Maley 1986, cited in Kubanyiova, 2006), undertaking research, engaging in mentoring (Halai, 2006), and reflection and collaborative problem solving (Dunn & Shriner, 1999).

1 Research questions and aims

The aim of this research project was to explore the following questions:

Previous empirical work on both teacher motivation (Hoy, 2008) and the mediating and regulatory influence of future self guides on behavior (Hoyle & Sherill, 2006; Martin, 2004) has demonstrated that large-scale positivistic, quantitative approaches have offered only superficial insight into these phenomena, leading researchers instead to build towards conducting modest-scale, contextually sensitive case studies facilitating a more profound understanding and awareness (Seferoğlu, Korkmazgil, & Ölçü, 2009).

What must also be underscored is the resonance between the chosen methods and my agency as researcher. The semi-structured interviews used in my case study were purposely chosen to explore, without prejudice, the particular stories of these teachers and the multi-faceted aspects of their language teacher possible selves. Seidman (1991) advises that, in essence, the end purpose of in-depth interviewing is ‘to have the participant reconstruct his or her experience within the topic under study’ (p. 9). The approach adopted was seen as being the most beneficial for the experience of the participants in reconstructing their own sense of worth.

2 Participants

This study used a blend of homogenous and criterion sampling. Initial recruitment of participants for this study was from a pool of 94 public school English teachers in the greater Seoul area who had taken part in a government-sponsored six-month training course during the 2007, 2008, and 2009 school years and had subsequently chosen to enroll in further postgraduate study in the country. Individuals were informed that the researcher was interested in their professional development choices, and were invited to participate in the research project without further inducements. The number of female teachers in this pool overwhelmingly outnumbered male teachers, a curiosity witnessed in sampling from other recent studies involving teachers (Alexander, 2008). The final sample was composed of seven teachers: one male and six female between the ages of 29 and 45; there was no attrition among these seven individuals.

Brief summaries of the participating teachers’ demographic information are given in Appendix 1. All names used are the initials of participant-chosen pseudonyms known only to the researcher and the individual participant.

3 Data collection and analysis

The primary data source was in-depth interviews conducted in English with the seven participants; secondary sources of data included a research journal, analytical memos and extensive annotations to the primary data. Having first established specific lines of questioning, a pool of over 20 interview questions was compiled on three broad topic areas. Participants were asked to talk about their actual and possible selves, their professional development choices, and their language teaching and learning experiences. Piloting of the interview schedule took place with four non-participating individuals.

Data collection took place over a series of three weeks during the spring of 2010. In order to decrease the perceived power distance between the interviewer and the interviewees (Ten Dam & Blom, 2006), interviews were conducted in either the institution where participants were studying or the workplace of the participants. Each of the interviews varied in length, yet answers given in every interview revealed significant within-case coherence and consistency resulting in corroboratory insights (Hayes, 2005, p. 171). Each hour of recorded interview data produced roughly 10–12 single-spaced pages of transcribed data (for a summary of interview data, see Appendix 2).

Working up from the data began with a parallel-iterative process of analysing primary data for themes while examining the background literature, generating secondary data in the form of analytical memos, annotations and entries in my research journal, and then triangulating these multiple sources of data. In the second stage, I employed a three-tier coding process (Richards, 2005, p. 87–88) using NVivo 7. I began primary descriptive coding by attributing descriptive and contextual information to each of the cases being studied; I then assigned secondary topic coding to portions of text while saving them as free nodes, essentially what Strauss and Corbin (1990) have termed ‘fractur[ing] the data and … identify[ing] categories’ (p. 97); finally, I examined the data in these 19 topical categories, analysing how they were related to other ideas from the data, in order to draw more finely-tuned or combined conceptual categories that would result in ‘putting the data back together in new ways’ (p. 27).

During this tertiary level of analytical coding, I used Yin’s (2003) cross-case synthesis as an analytic technique: I looked for individual instances that corresponded with some uniform pattern among the participants, allowing me to draw cross-case conclusions from the findings. As a measure of validity, participants were asked to cross check the open topic coding and analytical coding of their data for verification. Throughout this study, I have chosen to report the participants’ words verbatim.

1 Patterns of CTD

Based on systematic data analysis, three outstanding incentives for these seven teachers to engage in continuing professional development emerged from this study. These were:

These findings illustrate that each type of possible language teacher self plays an integral role in initiating and sustaining in-service Korean public school English teachers’ motivation to engage in continuing teacher development.

For individuals whose feared language teacher self was the most dominant in their working self-concept, the primary driving force for CTD was repairing inadequacies of the self. Individuals who exhibited this pattern generated decisions for their CTD that were spurred by avoid tendencies. Three participants, NG, JH and PJ, were in this category. Their focus was on compensatory CTD – breaking even, so to speak – to avoid perceived matches between their actual language teacher selves and their feared language teacher selves:

I’m really afraid of going to […] meetings and speaking in public because when I speak English in front of people I cannot show my abilities and I’m intimidated. That’s because I’m worried about how people might look at me and how people will judge me. (JH, 6 May 2010)

In line with the principle of agency outlined in my review of the literature, these participants’ feared language teacher selves were concrete and experienced emotionally in real-time by the interviewees as they were being described:

I’ve seen many English teachers who are old so their English is getting worse, so they lose their confidence in the practical side of things and try to make up their lack of confidence in English with something else […] That’s the image I want to avoid. It’s not just that I want to avoid it, but that’s the image I’m afraid to be. (LA, 10 May 2010)

Individuals who had an ideal language teacher self as the most prominent element in their working self-concept reported engaging in CTD primarily for enhancement of the self. Four of the participants, AI, CE, HS and LA, were in this category. Their choices for CTD were driven by approach tendencies, and they reported a focus on supplementary CTD to increase consonance between their actual language teacher selves and their ideal language teacher selves.

Every participant in my study had a clearly elaborated ideal language teacher self, composed primarily of the positive future self images they possessed. These included positive future self images in two categories: self as expert language user, and self as expert teacher. For the most part, division into these two categories was anything but straightforward, with many data excerpts straddling an area related to both. The overwhelming majority of these positive future self images, however, relate to the participants ideals as expert language users. Curiously, these teachers tended to respond to questions about ‘self as a teacher’ with answers restricted to language proficiency issues, rather than strictly about teacher identity:

I always want to be as good as a native speaking teacher in a native speaking country […] One day I’m going to study abroad and I’m going to prove that I’m as good as the rest of them. (HS, 24 May 2010)

I imagine myself with a greater linguistic knowledge. Yeah, that would be perfect. (AI, 1 May 2010)

My image is that I can use English like it’s my mother tongue. […] I can see that. I can actually vision that right now. (JH, 6 May 2010)

It became clear that for all seven participants establishing confidence in their abilities as effective educators, their teaching self-efficacy, primarily involved establishing confidence in their being effective English language users, their language self-efficacy. This suggests that for these participants, and perhaps for language teachers in general, the two are inseparable.

Overall, the participants did not seem to differentiate between improving instructional practice and self-improvement, although this has been highlighted as a key distinction in CTD by others (Dunn & Shriner, 1999). As in other studies (Kwakman, 2003), self-interested factors exerted a more profound effect on CTD choices than any other factors in this context. The self in the above categories seems to be a more private self, one primarily related to the participants as L2 language users. This finding directly contradicts those of others in the teacher development literature to the extent that patterns of concern in teacher development customarily move outward, from self to other (Conway & Clark, 2003).

Clearly, decisions for self-improvement were not driven by an underlying goal to help students’ improvement. They were much more selfish than that in this context. Those whose feared language teacher selves were most dominant in their working self-concept were driven to participate in CTD principally by their desire to repair perceived inadequacies of the self by avoiding a feared image of their future self as being an incompetent English language user. This corresponds with previous findings of prevention-focused self-regulation being more likely in instances when risk of an off-track possible self is high (Oyserman et al., 2006, p. 201).

2 L2 insecurity: Distracting or contextually congruent factor?

Substantiating mainstream claims of teachers’ experiences as learners influencing their teaching beliefs and practices (Borg, 2003), the participants’ experiences as L2 learners appear to have determined their respective possible language teacher self-concepts. Language self-efficacy emerged as a principal element of the participants’ language teacher self-efficacy. This language insecurity surfaced in the data as early as the opening responses, and exists as a primary constituent of the ‘actual language teacher selves’ of every single one of the participants; this is well-illustrated by CE’s extended response:

Doing well in English is the single most important thing in my career, more important than my teaching skills or anything, since I want to be recognized as a good English user before being recognized as a good technician for teaching English. I personally think teaching well is more crucial for my students’ improvement but people recognize a teacher’s own proficiency before his or her teaching ability. I do not want to be misjudged or under-evaluated due to my English. So, English speaking proficiency comes first before anything. (CE, 14 May 2010)

Prioritizing recognition as an expert language user above recognition as a teacher is intriguing, particularly because none of the participants achieved a level of proficiency lower than B2 on the CEFR, with all having scored between bands 5.0 to 7.5 on the IELTS in the year prior to data collection. Regardless, all participants exhibited a range of negative value judgments of their actual language teacher selves and presented evidence for what one participant called a language teacher ‘complex’:

A lot of English teachers these days say ‘I can’t speak English very well.’ It’s like English teachers have a complex. We genuinely think we’re bad at English and that we need to study more. (HS, 24 May 2010)

Sometimes I have what I call a ‘bad English day’. I make stupid mistakes which the students recognize right away and mock. If the first class is like that then for the rest of the day I don’t use English. (LA, 10 May 2010)

The data indicated that participants’ sense of teaching self-efficacy was likely to bear the brunt of these negative language self-efficacy beliefs. Numerous examples in the data, in fact, reveal that the primary perception – of teachers themselves and of their students, colleagues and society at large – of an English teacher’s obligatory role is that of a native-like linguistic expert who, almost as an afterthought, is also an educator:

Some students consider me like God and I have to know everything about English. (LA, 10 May 2010)

I don’t mention [to students] that I’m a learner. I always pretend to be an expert. (HS, 24 May 2010)

As language teachers, these seven individuals’ priority in CTD is naturally language. Given the participants’ constructed possible language teacher selves, improving language proficiency was seen as the most straightforward way to continue to develop as a teacher, and other empirical findings do confirm that non-native speaker English teachers in this context often have developing language proficiency as their foremost goal in professional development (Jeon, 2008; Kim, 2001, 2002, 2003, 2008).

These individuals are clearly engaged in a constant struggle on many different levels in which they experience an ongoing and deep-seated angst about their own English language skills:

My English ego is like my Korean ego but it has holes in it. I want to fill the holes so I can improve more and grow myself up. But I need to fill the holes first. (AI, 1 May 2010)

The participant teachers all either explicitly stated or demonstrated that they subscribe to the belief that:

[T]heir worth as an English teacher is dependent on their ability to speak the language, so they are not valuable as English teachers. (AI, 1 May 2010)

While this data may suggest that for some teaching was a part of realizing an overall ideal L2 self rather than language proficiency only being a component of an ideal teacher self, confidence as a language user is arguably one of the more crucial aspects of being a language teacher (Reves & Medgyes, 1994). If a teacher’s self-efficacy is defined as ‘situation-specific confidence’ (Moe et al., 2010, p. 1145) in one’s ability to help students learn, the subcomponents of that confidence must also be assigned relevance. Awareness of the limitations of a language teacher who is an expert pedagogue in every way yet lacks reasonable ability in the language being taught results in extreme anxiety about teaching self-efficacy (Chacón, 2005; Kang, 2008; Lee, 2002; Liu, Ahn, Baek, & Han, 2004; Medgyes, 1999; Park, 1992).

Given its ubiquity in the data, the finding that language insecurity was a pre-eminent aspect of these Korean teachers’ underlying motivations for CTD was initially perplexing, yet attempts to dismiss this as a ‘distracting factor’ (Kubanyiova, 2006, p. 10) may be misguided. In this study language insecurity is the single most important factor guiding CTD, and thus warrants further investigation in other comparable contexts. Furthermore, the assumption that a teacher of any other subject would similarly attribute a great deal of importance to mastery of their given subject matter does not seem inappropriate. Past research findings are also implicit in the understanding that mastery of a given subject matter may be closely linked to stability of teacher expertise and, by extension, teacher efficacy (Rich, 1993).

It has been suggested that ‘teachers’ professional self perceptions constitute important criteria for evaluating the intended benefits of in-service programs’ (Rowe & Sikes, 1989, p. 139); an implication of this, therefore, is that CTD with a focus on improving both teachers’ language self-efficacy and teaching self-efficacy is the most advantageous in this and similar contexts (Lee, 2009, p. 1098). Rather than being seen as a distracting factor in CTD, perhaps we should embrace it as a factor congruent with the contexts and individuals in which it exists and occurs (Chacón, 2005; Guilloteaux, 2004; Horowitz, 1996; Horowitz, Horowitz, & Cope, 1986; Li, 1998; Medgyes, 1999).

3 Teachers’ beneficial lack of self-efficacy

Perceptions of oneself undoubtedly influence the way in which one acts (Bong & Skaalvik, 2003; Eren, 2009). Despite suggestions that lack of teaching efficacy may hinder the promotion of CTD (Guskey, 1988; Van Eekelen et al., 2006), my findings confirm that teachers’ self-efficacy doubts do indeed carry enormous potential for CTD. It was precisely these individuals’ lack of self-efficacy as language users, and therefore as language teachers, that incited them to undertake their respective paths of CTD:

There’s another aspect I’m afraid of, as a language teacher, my skills might become rusty and go down, and students don’t like old teachers especially when it comes to English because they think their English is really terrible. (CE, 14 May 2010)

If students or parents complain about my teaching then I think it’s going to really hurt myself […] I don’t want this kind of situation where people or students or parents feel my English teaching is not enough. I want to avoid those kinds of situations. I don’t want to be criticized in my job. (PJ, 28 April 2010)

Two paths, therefore, are likely possible: lack of teaching self-efficacy coupled with feelings of powerlessness (Schmidt & Datnow, 2005) and an absence of motivating future teacher selves – either negative or positive – ultimately leads to depression, decreased effort, and burnout; lack of teaching self-efficacy paired with feelings of potential for improvement due to motivating future teacher selves, however, impels teachers to undergo development as a route to achieving success as an educator (Wheatley, 2005), which may in turn lead to increased self-efficacy.

Given the overall lack of teaching self-efficacy in this context, the implications are that if the supportive capacity of power to improve is present, the end result will be teachers choosing to participate in CTD in order to attain greater levels of teaching self-efficacy (Freese, 2006). As long as it is accompanied by feelings of power to enact a change, low teaching self-efficacy may be beneficial.

4 CTD as an emotional endeavor

Positive self-enhancement is a fundamental human motive (Herbst, Gaertner, & Insko, 2003), and as emotions are present in all learning and developmental processes (Golombek & Johnson, 2004), in a profession such as teaching assumptions are that emotional vulnerability will be reduced through the potential for positive enhancement and fulfillment of the self (Kelchtermans, 2005).

A profusion of negative and positive emotions encircles these teachers’ choices to engage in continuing teacher development, both prior to and parallel with their CTD paths. The overwhelming majority of teacher talk from the interviews was coded as instances of negative future self images, and this is reflected by one research journal entry written rather frustratedly:

One other phenomenon I’m observing in the data is the fact that there seems to be so much negativity and so many negative images and emotions surrounding these participants’ Possible Selves. Why is this? What does this say about them as individuals? (Research Journal, 17 June 2010)

Constructing an identity, whether on a personal or professional level, inevitably – and perhaps justifiably – involves emotions (Geijsel & Meijers, 2005). Particularly for the teachers in my study the initial awareness of dissonance between an actual and ideal self, and consonance between an actual and feared self, led them to establish a precarious and vulnerable position of self worth that was more affective than rational, and in some cases hardly representative of others’ perceptions of them:

It’s like going in circles […] I want to be a good teacher, and I want to be a good English speaker. I don’t want to be criticized by people who say ‘Oh, you are an English teacher but your English sucks.’ Oh, that’s terrible. (AI, 1 May 2010)

Throughout ongoing CTD, affect provided a propellant for the pattern of CTD motives that teachers aligned with. When a sense of an enhancing purpose of CTD was present, as was the case for teachers driven strongly by their ideal teacher selves, the affective tenants of their working self-concepts were primarily positive emotions. However, in instances when CTD purposes were reparatory in nature, as in the case of teachers driven by their feared teacher selves, it was frequently negative emotions that inhabited their working self-concepts. Yet, while research has previously highlighted the negative affect present in teachers’ learning endeavors (e.g. Moe et al., 2010), the reality is that positive emotions were also a vital part of these teachers’ CTD decisions and pathways. Not only are the participants’ CTD choices the main sources for their emotional vulnerability, they are also primary sources of positive self-esteem and emotional fulfillment:

Also as a teacher I will be able to say everything in English. Whenever I want to say something I can speak English. Even when I discipline the students I will be able to use English. (JH, 6 May 2010)

Perhaps fittingly, the implication of this undeniable affective nature of CTD in this context is that it results in a commitment to teacher-initiated, bottom-up development. The value of this rests in the assumption that from-the-ground-up CTD done by teachers is the most effective and sustainable form of development. Self-initiated growth, that is, may be the most potent from of growth there is.

5 Implications for stakeholders

The implications of this study affect the main communities invested in language teacher education: policy makers or teacher educators, and individual teachers. Lack of language self-efficacy emerged as the single most pronounced reason for initiating CTD in this study. Teacher educators must, therefore, accept a realignment of how this is seen: as indicative of authentic CTD needs and aims in this and similar contexts. In order to maximize their intended benefits, professional development initiatives should aim to develop expertise as well as content mastery; all indicators are that stakeholders in the Korean context have taken this finding to heart by advocating hybrid teacher development programs which target both.

The finding in this context that private, egocentric factors exerted more of an effect on teachers’ CTD choices than external factors and obligations suggests that policy makers would do well to encourage and promote self-initiated professional development. Conversely, top-down policy changes or teacher development initiatives are unlikely to be internalized and lead to real change if these are incompatible with individual teachers’ possible selves or the procedural paths they are undertaking on the way to their intended goals as professionals.

Emotional vulnerability outcomes are an inescapable part of teachers’ ongoing professional development, but this should not be viewed as detrimental by any means as it leads to a commitment to the most effective and sustainable form of CTD: the bottom-up, self-initiated kind. If individual teachers prime a possible self and translate their negative feelings of self-efficacy into greater participation in teacher development, this will undoubtedly lead to higher rates of success as educators and, perhaps in turn, lead to a higher measure of self-efficacy. Additionally, building a focus on the ‘software’ of teaching in to the teaching environment is likely to mitigate teachers’ feelings of deteriorating expertise and self-efficacy. This may be the most proactive way in which to reduce the need for large-scale teacher development interventions which are more reactionary in nature.