Using journals to investigate the learner’s emotional experience of the language classroom

ELIA 6, 2005-6, pp. 63-91 USING JOURNALS TO INVESTIGATE THE LEARNER’S EMOTIONAL EXPERIENCE OF THE LANGUAGE CLASSROOM Sarah Mercer University of Graz, Austria This paper will report on a study conducted as part of a PhD project investigating affective factors in language learning. A study involving 73 advanced tertiary level learners was carried out over the course of one semester (approximately 4 months) in 3 parallel classes. The learners were asked to keep a journal following guidelines on content focusing on their emotional experience of the language classroom. On the basis of an initial analysis of the journals, a questionnaire was constructed to validate the data from the journals. In addition, the questionnaire provided meta-feedback on the use of journals from the learner perspective. This paper will discuss the use of journals as a tool for investigating learner beliefs and emotions and will present the preliminary results of the study. Key words: journals, learner beliefs, affective factors. 1. The Study of Affective Factors The humanist movement heralded a move away from a focus solely on cognitive factors to increasing attention to the rather neglected affective aspects of learning. However, critics of an overly-humanistic approach caution against championing affect in isolation. Nevertheless, as Kerr (2000) warns, humanism does not necessarily equate with a consideration of affect, although the two areas are clearly related. Stevick (1999, pp. 43-44) advises against viewing affect as the next teaching “philosopher’s stone” which will Estudios de lingüística inglesa aplicada 64 S. Mercer ELIA 6, 2005-6, pp. 63-91 serve as the magic answer to all possible teaching and learning problems, although he acknowledges its central role in learning. Many (cf. Arnold, 1999) have also cautioned against viewing cognition and affect as opposites and instead call for an integrated approach which would incorporate both cognition and affect in a balanced relationship without undue emphasis on one or the other. Common sense has told generations of teachers how important feelings and emotions are in learning but it has only relatively recently become a key issue for researchers, possibly due to the problems inherent in such research. Nowadays, it is widely accepted that feelings and emotions are a crucial source of information and act as a vital link between cognition, understanding, memory, motivation and learning (Damasio, 1994; LeDoux, 1998; Reeve, 2005). As LeDoux (1998, p. 39) states “minds have thoughts as well as emotions and the study of either without the other will never be fully satisfying”. It is perhaps important to clarify some definitions before continuing. Many researchers use the terms “emotion” and “affect” interchangeably whereas other disciplines view them as differing concepts (Hart, 1989). In this paper I use “affect” to refer to general evaluative feelings and emotions that refer to specific reactions to specific events. (For a more detailed discussion of the use of the terms, cf. Hart, 1989). This study was originally planned to explore learners’ reactions to specific in-class events, hence the use of the term ‘emotion’ was chosen for its familiarity and comprehensibility to the learners involved in the study. However, as will be seen in the results section, the learner data in fact also covers more general moods, evaluative reactions about themselves and their learning and a range of self-beliefs and hence, would be better classified as affective factors. 2. The Use of Journals to Investigate Learner Internal Factors Bailey (1990, p. 215) defines diary studies as being “a first-person account of language learning or teaching experience, documented through regular, candid entries in a personal journal”. Some researchers further distinguish between diaries in which the content is focussed more on feelings and Using journals to investigate the … 65 ELIA 6, 2005-6, pp. 63-91 personal entries, in contrast to journals which are seen as being more neutral and factual. In this paper I will use diaries to mean accounts which can contain any kind of information, objective or subjective, and as such I will use the terms ‘journal’ and ‘diary’ interchangeably (McDonough & McDonough, 1997). Diaries are, by nature, introspective, yet as MurphyO’Dwyer points out (1985, p. 100) they are in fact also retrospective accounts of events and when discussing them as a research tool it is important to bear this in mind. There have been a considerable number of studies in language learning which use diaries, particularly for investigating personal beliefs and experiences (Bailey, 1983, 1990; Murphy-O’Dwyer, 1985). Their use in applied linguistics began to gain popularity with Schumann and Schumann (1977) and researchers began to explore their potential for offering an insight into learner internal factors, such as thoughts, beliefs, feelings and emotions. Despite their increasing popularity, particularly in many areas of the social sciences and education research, journals have not been regarded favourably by all researchers. This is because of their supposed ‘soft’ data approach and the seeming preference within applied linguistics by some for a more quantitative or rationalistic approach. It is clear that diary data is unsuitable for generalisations given its unique self-reported nature. However, this form of research is often largely uninterested in generalisable truths but is more concerned with the uniqueness of individual learning experiences while recognising the powerful influence of the complexity of the real-life context. Simons (1996, p. 231) when discussing the value of case study research suggests that in fact “by studying the uniqueness of the particular, we come to understand the universal”. These issues are aspects of a debate beyond the scope of this paper but they are considerations which the researcher must be aware of, if such a methodology is chosen. Nevertheless, journals offer an excellent method particularly for investigating learner’s internal beliefs and emotions and they have already been used in several studies in this area. Nunan (1992, p. 121) stresses that through diary studies “affective factors emerge as being particularly significant in language learning”. As a research tool for investigating such sensitive learner internal areas, such as emotions, feelings and beliefs, diary studies have the advantages of allowing for a wider sample, there is less 66 S. Mercer ELIA 6, 2005-6, pp. 63-91 chance of ‘reactivity’ and they can provide a more holistic picture of learning in context than other research tools. Diaries provide access to data that would not otherwise be accessible to the researcher and provide a valuable learner account of the language learning classroom. As Murphy-O’Dwyer (1985) points out it is the learnercentred nature of such a diary that is one of its greatest advantages, as it allows researchers to gain an insight into their view of the classroom and their own personal experience of it. Journals set the learner accounts within the learner’s own personal learning context and allow for a more situated perspective. A diary also has the advantage of revealing the development of the learners’ experience as it is recorded over a longer period of time as opposed to a ‘snapshot’ of the class at one single moment in time. In fact, the temporally organised longitudinal nature of a diary study is one of its greatest assets, enabling the researcher to observe the progression of states and emotions over time. As Bailey (1983, p. 98) succinctly puts it, “diary studies allow us to see the classroom experience as a dynamic and complex process through the eyes of the language learner.” However, there are some practical problems with diaries as a source of data. A key problem that concerned me was the considerable amount of commitment required of the participants to such a long-term project involving a lot of their time and effort over an extended period. As a researcher, much of my work has been influenced by principles underlying Exploratory Practice (Allwright, 2003) which suggest that learners should also benefit from the experience of taking part in a research study. A study should be mutually beneficial to both learner and researcher with as little intrusion as possible into learners’ lives and learning. It was hoped that despite the extra effort and time necessary for keeping the journals that learners would feel that they benefited; firstly in terms of writing regularly in English for practice and secondly in terms of increased self-awareness of themselves as language learners. A further problem associated with diaries is their intrusion into an individual’s private sphere; however, this is less of a problem when diaries Using journals to investigate the … 67 ELIA 6, 2005-6, pp. 63-91 are kept specifically for research purposes in which the purpose of the diary is made clear to all beforehand. Participants will tend to self-select what they wish to include and it must be remembered that “although the diary remains a personal account, the domain is quite tightly specified by the researcher” (McDonough & McDonough, 1997, p. 122) which consequently affects the content and focus of the entries. Nevertheless, ensuring confidentiality and anonymity can help reduce potential feelings of inhibition. With journal entries, as with other research data collection methods which are dependent on learner self-reporting, there is a danger that respondents will ‘show off’ to some extent in their journals and use ‘selfflattery’. As a researcher, one can explicitly request respondents to be as honest as possible and give direct assurances of confidentiality and anonymity to encourage participants to be as candid as possible, but the possibility of distortion must still be kept in mind during analysis. Similarly, the extent to which the journal entries reflect what actually happened or what learners consider to have happened is difficult to ascertain. Some researchers argue that such data are only a reflection of what the learner perceives as being important but in this study it is the salient learner perceptions that are of importance. Further, this point leads to an issue that is, as yet, unresolved in research on emotions, namely the discussion about certain emotions being unconscious (LeDoux, 1998) and consequently some entries will be incomplete as participants can only report what they can consciously access. Although a full discussion of this topic is beyond the scope of this paper, it should be noted that this study will intentionally focus on those emotions which the learner is aware of and can articulate, and by implication those which can be more easily influenced. As with any form of research, some of the respondents will be better than others at providing valid, detailed and complete self-report data. Some participants are better able to provide more honest and rich answers, and others can reflect more deeply. Important too to keep in mind that some are better suited to written expression whereas others excel at verbal communication. This variation in the quality, both in terms of the language and ability to reflect and report, cannot be avoided and is a feature of the 68 S. Mercer ELIA 6, 2005-6, pp. 63-91 individual differences that exist amongst the learners. Naturally, a larger sample size will ensure that there is more likelihood of receiving rich, comprehensive diary reports. In fact, with a large sample size and given the range and variation among the entries it ensures that “even a limited type of triangulation is achieved if the data come from more than two individual diaries” (Murphy-O’Dwyer, 1985, p. 101). A further impact of the use of journals is that keeping a diary may affect the routines of the learner and consequently also their experience of the class (Willig, 2001). It is possible that reflecting on the class or simply knowing that they will have to write about it later may affect their behaviour and attitude towards the class. Furthermore, through their reflection, learners may change or alter their thoughts, beliefs or feelings (Ewald, 2004; HarriAugstein & Thomas, 1991). As McDonough and McDonough (1997, p. 122) comment “the act of writing itself is a way of structuring, formulating and reacting to that experience, which is then available for reflection and analysis”. From a more positive perspective, Bailey (1983) sees diary keeping as both therapeutic and beneficial for the language development of the learner. No research method is without its flaws, but learner journals seem to be an excellent method for accessing learner beliefs and emotions. Thus, they were considered to be the best tool for collecting the longitudinal, learner internal data in situ necessary for the purpose of this study. It is essential in any study to acknowledge the potential problems associated with a particular method and consequently it was decided to research the journals on a meta level. The following table summarises some of the key arguments surrounding the use of diaries as a research tool. Advantages Disadvantages Suitable for internal factors Incompleteness Using journals to investigate the … 69 ELIA 6, 2005-6, pp. 63-91 Long-term, longitudinal studies Demands student commitment & cooperation Insight to learner perspective Intrusion into learners’ private sphere Situated within context Not all learners react positively to journal keeping Holistic – complexity & interrelated nature revealed Possible lack of focus in responses Honest – less risk of ‘reactivity’ Not all learners able to reflect deeply Allows for wide sample Cannot probe responses Some benefit for the learner in terms of self-awareness, language use. Distortion due to keeping of journal and act of reflecting Fig. 1: Summary table of some of the key arguments surrounding the use of journals as a research tool 3. The Study This exploratory study was carried out with the aim of assessing the suitability of journals as a research tool and in order to answer two research questions: • Which emotions play a significant role in the language learning classroom from the perspective of the learner? • What factors appear to affect the learners’ emotional experience of the language classroom? It should be noted that the wording of the second question suggests some kind of directionality and causal effects, however, this is not the 70 S. Mercer ELIA 6, 2005-6, pp. 63-91 intention. The study was interested in exploring which factors appeared to be affecting which emotions without attempting to suggest directionality or causality. The study was carried out amongst 73 advanced, tertiary level students of English studying in one of the initial language courses at the University of Graz, Austria. The students were from three different classes, each class had a different teacher but the same course content and syllabus. I also taught one of the classes. Students were asked to keep a weekly diary throughout the semester which lasted approximately 15 weeks. To allow students to have time to do so, they were permitted to do one less portfolio task as ‘compensation’ for the time spent on the weekly journals. However, due to curriculum and institutional demands, it was compulsory for all learners to keep the journal. It has been suggested (McDonough & McDonough, 1999, p. 127) that some who do not wish to keep the journal may provide unhelpful entries as a form of resistance to having to keep the diary. It was felt, however, that it would be better to include entries from all types of students, not just volunteers, in order to gain a more complete picture. As will be seen in the results section, although there were some less helpful entries none of the entries appear to be deliberately unhelpful or uncooperative. Following the practical advice offered by Allwright & Bailey (1991, p. 190-193) the students were given a sheet of guidelines to reduce the potential amount of redundancy by helping the learners focus on relevant content. Nevertheless, care was taken to not be too prescriptive as this would have perhaps lost valuable data and would maybe have been less motivating as a task for the learners. Learners were also asked to provide some basic background information, which would not compromise their anonymity, to help gain a picture of their particular situation. Students were allocated numbers to connect the first half of their journals and the second half, but the journals were otherwise anonymous, the writer’s identity was protected and confidentiality was guaranteed. This is crucial in terms of ethics and in order to encourage honest, full responses. Using journals to investigate the … 71 ELIA 6, 2005-6, pp. 63-91 A key decision was to collect the diaries at two points during the course; the first part half-way through the semester and the second at the end (for practical reasons it was not possible to collect entries every week). As it was feared that some students would not in reality keep the journals consistently and regularly, it was hoped that the danger of all the entries being written at the end of the course in one sitting could be reduced by having at least two collection points. The first collection date also provided the researcher with a brief chance to check some of the entries and, if necessary, make alternative suggestions for the second part, which in fact was not needed. This half-way point was also used to send a note to all the participants thanking them for their contributions for the first part, restressing their confidentiality and highlighting their helpfulness for the researcher and, hopefully, themselves. In doing so it was hoped that their motivation would be maintained through the final part of the diary keeping process. Crucially, the journals collected from the first half underwent a preliminary analysis. This informed the design of the final questionnaire which was distributed to all participants at the end of the course after the second part of the journals had been collected. The questionnaire had two purposes; firstly to validate the data collected from the journals and secondly to provide meta-feedback on the use of journals from the learner perspective. The questionnaire was composed of a mixture of closed and open-ended items and was distributed in the final class which meant that all participants completed it. Finally, some volunteers were interviewed at later dates after the end of the classes to clarify issues from the preliminary analysis of the journals and questionnaires and also to further validate data from other research tools. As one of the teachers, I also kept a journal of the class in order to compare the teacher/researcher perspective with the learner perspective of the class and importantly also to experience first hand the keeping of a weekly journal. The diaries were analysed for salient, frequent and widely distributed trends/factors (Allwright & Bailey 1991, p. 193) using content analysis. The open-ended sections of the questionnaires have been analysed in the same 78 S. Mercer ELIA 6, 2005-6, pp. 63-91 “I don’t know if this journal had an effect on my language learning but it was a nice experience to feel that lecturer is interested in my feelings.” 4.3. From the Researcher Perspective Keeping a journal is not a habit I have ever had nor enjoyed and this project was no different. To be perfectly honest I disliked having to keep the journal; even though it was my own research and for my own benefit, it still became a chore. In fact, there were three occasions during the term when I forgot about the journal completely until several days later by which point specific details were distant memories. It became clear that the duration of the journal keeping was significant and, like my students, my entries were more thorough and reliable at the outset than towards the end of the fifteen week period. Furthermore, it felt awkward to sit and write about what had happened and it was inevitable that I began to analyse events immediately. The heightened awareness of the class and preoccupation with thinking about what I would write later certainly affected a change in my behaviour and thinking. One of my entries reads, for example, “(the class)… is rather dry – unlike so many of the classes I have ever had, ironic that this is the one we write about.” And then I continue later in the same entry whilst reflecting on why the class is so quiet, “I also wonder if I am a little more tense than normal as somehow I feel they are commenting on me and I know they are writing journals. I am so preoccupied with the emotional climate that I fear I am losing my naturalness as a teacher”. Interestingly as the course progressed the journals became less of a preoccupation whilst in class, but I never completely forgot about their presence or their potential content. However, it may be that as teacher-researcher I experienced the journal keeping differently to the students. It is possible that my knowledge of the argument that journals may distort the experience led me to write my entries differently as opposed to the journal keeping itself. Yet, if this were the case, although implying that the journal keeping itself is not at fault, it would highlight the fact that as a researcher my journal entries may be Using journals to investigate the … 79 ELIA 6, 2005-6, pp. 63-91 distorted by my knowledge of research factors and other studies and issues in the area. Either way it illustrates the fact that keeping a diary as a researcher is not without its problems. Bailey (1983) explains that there is need for some caution when respondents have knowledge of the topic and an agenda in their research. As Long (1983) points out although the researcher may be able to offer rich insights into the language learning experience, their dual role implies a distortion of reality and means that their attention is also drawn to the diary as a task and away from the “classroom processes per se” (ibid: 25), which I certainly felt to be true in my situation. There is the danger with ‘expert’ diaries (McDonough & McDonough, 1999, p. 129) that their writers are too aware of issues involved and possible content of entries. Although they may be able to provide deeper introspection, I would suggest that this in itself is a distortion of reality. They may highlight issues that ‘non-researchers’ would have been unaware of or which they would not have drawn attention to, hence presenting an inaccurate emphasis. For these reasons, it is wise to be cautious when it comes to many of the diary studies which involve a ‘researcher-turned learner’ (Bailey, 1983), in which the writer’s ‘expertise’ may distort the entries (McDonough & McDonough, 1999). Furthermore, ensuring that the diarist and the person who analyses the entries are different people would possibly afford more objectivity and less retrospective interpretation. To conclude, my personal experience of keeping the journal was, on the whole, rather negative. However, I must conclude that on re-visiting my journal entries, despite my dislike of the task and misgivings about the content, I did, in fact, find that there was much more of use and relevance to my study than I had at first been aware of. Their usefulness only really became apparent as I began to analyse them in more detail and compare perspectives in relation to those of the learners. 5. Preliminary Results of Learners’ Emotional Experiences To date approximately only a third (n = 25) of the journals have been analysed with a view to answering the two research questions introduced at the outset of this paper: 80 S. Mercer ELIA 6, 2005-6, pp. 63-91 • Which emotions play a significant role in the language learning classroom from the perspective of the learner? • What factors appear to affect the learners’ emotional experience of the language classroom? At this stage, the data can only suggest possible themes and issues to be followed up in more detail as the analysis progresses. It should be noted that the results presented here are not intended to be conclusive, but merely to serve as a brief survey of some of the preliminary issues present in the journals. They may provide food for thought for other researchers. 5.1. General Issues and Themes When examining the journals distinct learner profiles immediately emerge. Individual learners clearly have topics and themes that reoccur throughout their journals and which are of particular importance to them personally. In some cases, it is possible to identify learning style preferences, personality traits and, in other cases, specific issues that are of central importance, such as pronunciation, relationships to peers in class, the task itself, grades, etc. One learner, for example, is fixated on accuracy and it is a dominant theme throughout every single one of his/her entries; she says “As I’m a real perfectionist… I’ll have to get aware of the fact that mistakes are not exclusively bad but help you to improve your language skills.” An interesting aspect of the journals is that learner attitudes, beliefs and motivations appear in some cases to change within the individual over time, which is an important reflection of the dynamic nature of these factors. For example, “What was good today was that I was not afraid of using the language with speaking with my colleagues. As I am repeating the course I am familiar with the vocabulary which made it easier for me to say something” Then a week later the entry, “I feel not very comfortable when speaking in English because I have to think so much about what I would like to say.” One journal says “I like working in groups” and then a few weeks later “Group work is getting boring now”. There are many examples of this Using journals to investigate the … 81 ELIA 6, 2005-6, pp. 63-91 dynamism which would support arguments in favour of longitudinal studies in this area. Another factor that is evident from this initial analysis are the contradictions within each group. It becomes clear that each learner has their own set of beliefs and frame of reference against which the classroom experience is evaluated. This shows how some learners enjoy certain task types and others dislike them or equally how some students find a task useful whilst another finds it unhelpful. This would correspond to findings by Koch and Terrell (1991) who note that there is variation in the ways in which students respond to certain learning activities and tasks. This appears to be due to individual learning styles and changes that take place within the individual and hence “instructional techniques should not be thought of as intrinsically ‘good’ or ‘bad’” (Ibid: p. 124). From a teacher perspective it becomes clear that the group is composed of individuals and the teacher needs to be aware of this variety and attempt to ensure that there are tasks to suit all interests and learning preferences. In terms of research it should serve as caution against generalisations about groups or tasks, etc. Finally as the journals were analysed it became increasingly clear how highly interrelated the affective concepts were and how difficult, in fact virtually impossible, it is to separate them for analytical purposes. It points to the need for a holistic approach to studying these factors and influences on them and questions the sense of separating and compartmentalising these concepts when the reality is much more complex than research would like it to be. 5.2. Specific Emotions and Influencing Factors Nevertheless, it is necessary for research to attempt to extract individual factors from the complexity of the data in order to make some comprehensible sense of what affective factors appear to be present in the journals and in relation to which other factors. It should be repeated that this study does not attempt to show causality but wishes to explore patterns and possible relationships rather than suggesting any directionality or causality. 82 S. Mercer ELIA 6, 2005-6, pp. 63-91 The most frequently occurring affective factor mentioned is anxiety. It occurs, not unsurprisingly, in relation to making mistakes and tests or exams. It was also often mentioned in the context of learners comparing themselves with others in the class. In fact, comparing oneself to others appeared often in the journals and was strongly related to feelings about oneself and one’s ability. “I was more nervous but then I noticed that I’m not worse in talking than the others”. Although this entry is clearly about anxiety, it also contains a statement of self-efficacy, namely a judgement of one’s ability to do something and this self belief seems to be based on a comparison of one’s ability to others. With this single example alone the inter-related nature of affective factors and the complexities involved in their analysis should be evident. An interesting aspect of the anxiety factor was that the factors with which it co-occurred were relatively constant among all the learners, whereas other factors, such as motivation and interest showed much greater variation in terms of influential co-occurring factors. Another key factor, which occurred frequently, was self-confidence (it should be noted that the terms self-confidence, self-esteem, self-worth, self-concept are often used interchangeably and their meanings vary but a full discussion of their definitions is beyond the scope of this paper, cf. Pajares, 1996). In this study, self-confidence is understood as a more global, general feeling in contrast to the more domain-specific self-efficacy. In the journals, this global feeling of confidence or “feeling comfortable” cooccurred often in relation to a comparison with others but also in relation to a metacognitive knowledge of strategies, e.g. “I got more confident when I’m talking in English because of my goal-setting which worked well.” Self-efficacy, as domain-specific perceived feelings of competence, is naturally linked to more global feelings of self-confidence and they are clearly interrelated. In line with studies in educational psychology, the journal entries clearly show two different types of confidence, the one more general and the one task or situation specific. I took statements about the ease or difficulty of a task or activity as indicators of learner’s self-efficacy beliefs. The self-efficacy references often co-occurred in relation to task, metacognitive knowledge of strategies and also references to previous experiences. Using journals to investigate the … 83 ELIA 6, 2005-6, pp. 63-91 “I feel comfortable now I know different ways of giving opinion.” “It is often a problem for me to express exactly what I actually want to say. I sort of have a concept in my head but can’t find the appropriate words”. “I am familiar with the vocabulary which made it easier for me to say something.” As expected, motivation was a frequently cited factor as well. It is important to note that motivation can be positive or negative (demotivation). Feedback from tests or homework was often cited in relation to motivation but naturally this can be positively or negatively motivating, e.g. “(the test)… went quite well for me so I was willing to work” and then the same student comments after a test where she felt she had been marked too harshly and the grade seemed unfair “I didn’t want to say anything because to work in class doesn’t count anyway... I just didn’t want to be encouraged – what for?” Also a test or upcoming exam was seen by learners as a motivation to study, often because of their stated fear of failing, “I am more motivated because I don’t want to fail a second time”. Motivation was referred to both as a global concept by learners “I am really motivated to learn languages, any language really” but also in more specific immediate context-specific terms, such as related to task or topic. This leads me to make the provisional distinction at this stage in my analysis between long-term, seemingly more stable motivation and shortterm, more domain-specific motivation which also seems more dynamic in nature. Such short-term motivation was frequently cited in relation to mood and this will be discussed below in more detail. Finally, the effect of out-ofclass events was centrally inter-linked with mood and motivation. This close relationship between mood, out-of-class events and motivation occurred with notable frequency but at this stage in the analysis it is not possible to provide any detailed interpretation of it. The significance of out-of-class events appears indisputable from the journals that have been analysed so far. This naturally raises questions about 84 S. Mercer ELIA 6, 2005-6, pp. 63-91 research which isolates the learner within the ‘vacuum’ of the language classroom, particularly in the field of affect. As Delamont and Hamilton (1984, p. 21) caution “while it is possible, for research purposes, to regard the classroom as a social unit in its own right, it is only with considerable difficulty that it can be regarded as self-contained”. It becomes clear that many studies make a false dichotomy between life in and out of the class. Ideally, studies, which attempt to understand learner motivations and affective reactions in class, need to understand the learner as a whole. This means attempting to understand the interplay of factors that affect a learner in their lives and learning out of class too. In addition, many students mentioned experiences with their English out of class which affected their motivation and feelings about the course and class. A study needs to accommodate the influence of out of classroom encounters and learning opportunities as well. Limiting language learning to the classroom setting alone is an oversimplification of a very complex, interconnected reality which sees learners learning language and interacting with the language both in and out of class. Finally, the key affective factor of mood was often mentioned in all the journals. Reeve (2005, p. 316) distinguishes between moods and emotions explaining that moods are more enduring than emotions and that people are always feeling something in terms of a mood, whereas an emotion may be rarer and tends to be more specific. He claims that a positive or negative mood can have a significant effect on cognition, behaviour and motivation. It was evident from the journals that learners felt their moods could strongly affect their reaction and approach to the class. In fact, in the questionnaire, mood was the strongest factor that learners believed influenced their learning. It was mentioned in relation to physiological needs such as being tired, hungry or having a headache. The weather was also frequently cited as affecting their mood and willingness to work. Feedback from peers, the teacher or grades was also mentioned as affecting mood. Finally, one of the key influences on mood was a whole array of out-of-class events, which again highlights the importance of a holistic approach to the study of these factors. Mood also often appeared in relation to short-term motivation as mentioned earlier. Using journals to investigate the … 85 ELIA 6, 2005-6, pp. 63-91 “The weather today was absolutely wonderful that was one reason why everything was easier during our VSE lesson.” “I had breakfast with a friend of mine so I was in a good mood when I came to class.” “I had a quarrel with my boyfriend so I wasn’t too enthusiastic to go to my lessons.” “When I’m in a good mood it’s much easier to learn.” Large parts of these findings, such as the effect of mood, are probably not surprising to teachers who will be aware of the importance of them from their daily teaching experiences. Whilst researchers in language learning have tended to focus on selected affective factors, such as anxiety, some of the less ‘popular’ affective factors, such as mood, have, as yet, received rather scant attention. 6. Conclusions It should be noted that these conclusions are dynamic and represent one stage in an ongoing research process. It is quite possible that as the analysis progresses they may change but the writer of this paper wishes to share the experiences and findings of this study as they stand at present. 86 S. Mercer ELIA 6, 2005-6, pp. 63-91 6.1. The Use of Journals 6.1.1. For the Learner For learners, keeping a journal requires a considerable commitment in terms of time and effort over an extended period of time. It is, therefore, not unnatural that they should also expect to benefit from it in some way. It seems, given the mixed responses, that any possible benefits of keeping a journal will depend on the attitude, personal preferences and learning style of the individual. Following principles of Exploratory Practice, it is particularly important for me that the research benefits both learner and researcher. Building on my experiences with the journals I have begun to explore a more interactive approach where learners write a more structured monthly review and I, as their teacher, respond individually. It is beyond the scope of this paper to discuss this alternative to journals at this stage but it appears to be working well for both the learners and myself. It clearly has a combined research and pedagogical function. From the learner perspective students report that they are able to see the benefits for themselves in tracking their own progress and feelings as well as having the chance to communicate directly with their teacher. More importantly they get an almost immediate response to the review as I write back to individual learners. Clearly this is only possible with a smaller sample size and there is no longer any anonymity. However, the resulting data seems richer, possibly as a result of the relationship that has developed between researcher and learner. A drawback of this approach is that their memory of events may be incomplete and/or distorted given the time lapse since the reviews are only collected once a month. Another alternative to diary studies that has been explored elsewhere in relation to emotions are language learning histories (Oxford, 1996) which are being used to supplement this monthly review method. Using journals to investigate the … 87 ELIA 6, 2005-6, pp. 63-91 6.1.2. For the Researcher Diaries have much to offer to researchers as outlined in the earlier discussion section of this paper. They can be insightful and revealing, especially concerning learner internal factors such as beliefs, feelings, thoughts and emotions. They allow issues to be viewed from the learner perspective in situ and over an extended period of time. However, their quality may vary and ideally there should be a large sample size in order to allow for the problems of variation. The issue of gender-specificity or possible perception of diary keeping as being gender specific remains unresolved at present. Ultimately each researcher will have to select the method most appropriate to their own study, context and their own epistemological position. 6.1.3. For the Teacher As a teacher as well as researcher involved in this study there were many benefits from the teacher perspective. It can serve as a rich, valuable source of feedback on a course and teaching approach. If the journals are collected half-way through the semester, the first half material can be used to inform teaching in the second half of the course. Journals also allow an insight into the learner perspective of the language learning experience and they can sensitise the teacher to learner individuality within a group. 6.2. Learner’s Emotions in Language Learning At this stage of the research conclusions here can only be very tentative, but the journals appear to raise some questions. Given the complex and interrelated nature of many of these concepts a holistic approach to their study would perhaps help lead to a better understanding of their nature. This would also enable the learner’s out-of-class experiences and relationships to be included in the study. In addition, as learners beliefs about themselves and their reactions and motivations appear to change over time, this seems to suggest a need for more longitudinal studies.