Rural Area Learners ’ Writing Self-efficacy Development : A Qualitative and Quantitative Approaches

Ilyana Jalaluddin Universiti Putra Malaysia E-Mail: ilyana@upm.edu.my Abstract: This case study aimed to evaluate the rural area learners’ writing self-efficacy using two different approaches which are qualitative and quantitative approach. It involved three form four students and a teacher for six months. In this study, the learners’ writing self-efficacy was investigated using a combination of quantitative and qualitative (classroom observation). Classroom observation was based on three characteristics: persistence in accomplishing language tasks, self-awareness of English proficiency, and willingness to engage in language activities. These characteristics which were concept of self-efficacy. This study did not focus on the score of WSE scales alone but also placed a heavy emphasis on the perceptions and actions of the form four students and teacher. Altogether, 15 non-participant classroom observations, 10 interviews with each of the three students and 10 teacher’s interviews were carried out to understand the participants’ selfefficacy phenomena in their learning to write. Findings showed that the combination of different method of collecting data for writing self-efficacy was a feasible way in explaining rural learners’ writing self-efficacy development.


INTRODUCTION
Writing is not an easy task as it is a highly complex and demanding task that requires a number of skills to be performed.It is a complex cognitive activity as it involves attention at multiple levels: thematic, paragraph, sentence, grammatical and lexical (Tillema, 2012).Pajares, Britner, & Valiante, (2000); Pajares, Hartley, & Valiante, (2001) noted that "writers, in contrast to readers, produce/create texts rather than simply consume them and, writers often have minimal environment/curricular input".For example, "when given a topic to write about, the ideas and text generated require a knowledge base on which the individual can draw" (MacArthur, Graham, & Fitzgerald, 2008).In addition, "the complexity of the task, the solitary nature of the activity, with no immediate feedback and the effort needed to persist in the task are other aspects of writing that can adversely affect AL-TA'LIM JOURNAL, 24 (3), 2017, (255-265) (Print ISSN 1410-7546 Online ISSN 2355-7893) Available online at http://journal.tarbiyahiainib.ac.id/index.php/attalimwriting" (Bizzaro, 2004;Canagarajah, 2011;DeFord, 1981;Glasswell & Parr, 2009;Root, Steinberg, & Huber, 2011).Although writing is teachable, the transformation of thought into written communication is a difficult activity that requires many other levels of complementary skills.Scott & Vitale (2003) identified that learners' writing problems range "from lower level mechanical problems such as spelling, capitalization, and punctuation, to higher order cognitive and metacognitive problems such as planning and revision".The writer also suffers from the disadvantage of not getting immediate feedback from the reader and sometimes not getting feedback at all (Harmer, 2006(Harmer, , 2007)).In spite of the difficulties, Hyland (2007) found that it is possible to teach the necessary skills via process approach to learners so that they are able to express their ideas competently.Sari & Nufus (2016) pointed out that students will not learn writing if they are not trains to reflect their own writing and teachers should provide authentic opportunities for practice that will lead students to develop their confidence and interact with the rest of the society.Bruning & Horn (2000); Heidari, Izadi, & Ahmadian, (2012); Kyles & Olafson, (2008) further elaborated that "nurturing learners' positive beliefs about writing, fostering authentic writing goals and contexts, providing learners with a supportive context for writing, and creating a positive emotional classroom environment are he conditions that determine learners' motivation to write" (Barkaoui, 2007;Ivanič, 2004).

Social Cognitive Theory: Fundamental of the Study
Social cognitive theory is used as the theoretical framework to pursue the study on the development of the learners' self-efficacy.Overall, Bandura's social cognitive model emphasizes effective learning as involving three elements: the person (internal), the behavior, and the environment.This is because "how people interpret the results of their own behavior informs and alters their environments and the personal factors they possess which, in turn, inform and alter subsequent behavior" (Pajares et al., 2001).This is the foundation of Bandura's conception of reciprocal determinism, the view that (a) personal factors in the form of cognition, affect, and biological events, (b) behavior, and (c) environmental influences, create interactions that result in a triadic reciprocality.In the model of triadic reciprocality, the behavior, personal factors, and environmental events all operate as interacting determinants of one another.This model is illustrated in Figure 1 below: Briefly, the theory portrays that the three factors which are environment, personal factors and behavior are all constantly influence one another.In relation to the link between personal factors and behavior, learners' selfefficacy beliefs can influence choice of tasks, effort, persistence, and achievement (Schunk & Pajares, 2004, 2009).This implies that when the learners have high self-efficacy in writing, they are probably more optimistic and confident in completing their writing tasks.Conversely, learners' behaviors can also alter efficacy beliefs.For example, as they work on their writing tasks, they notice their progress and capabilities in writing.This goal progress and accomplishment will convey to the learners that they are capable of performing well.As a consequence, it enhances self-efficacy for continued writing.As noted by Zimmerman (2000), "learners' academic accomplishments can often be better predicted by their selfefficacy beliefs than by their previous attainments, knowledge, or skills".

The Connection of Self-Efficacy in Writing
Linnenbrink & Pintrich (2003) defined self-efficacy as "people's judgments of their capabilities to organize and execute courses of action required to attain designated types of performances".In writing skills, self-efficacy is a judgment of the confidence that one has in one's abilities to perform written English task successfully including correctly punctuating writing and creating grammatically correct samples of writing (Hashemnejad, Zoghi, & Amini, 2014).Spicer (2004) previously outlined three dimensions which govern selfefficacy.The first dimension refers to the level of task difficulty.According to Spicer (2004), "the magnitude of one's self-efficacy beliefs will differ upon how difficult he/she perceives a task to be; a task may be perceived as easy resulting in high self-efficacy, whereas a task thought to be difficult may lower self-efficacy".The second dimension is the area or domain to which one 's self-efficacy beliefs are applied.Spicer (2004) termed the second dimension as generality.For example, through generality, learners may have high self-efficacy for writing when they believe that writing is a necessary component of their study.Thus, they will work hard, have high perseverance and will be successful at it.A learner with low writing selfefficacy may feel that whenever he is confronted by writing, he would not be able to complete the required task successfully.The final dimension of one's self-efficacy is strength (Chemers, Hu, & Garcia, 2001;Linnenbrink & Pintrich, 2003;Maddux & Kleiman, 2012).According to Spicer (2004), strongly held self-efficacy beliefs are less likely to be challenged than are weaker beliefs.Thus, a learner with low writing self-efficacy may readily change his self-efficacy beliefs when encountering difficulties, even if he had previously been experiencing some success.Strength is a dimension which must be considered when measuring self-efficacy (Bandura, 1997).Strength, as used by Bandura (1997), describes "how strong a person's sense of self-efficacy is".People who have strong beliefs in their capabilities approach difficult tasks as challenges to be mastered rather than as threats to be avoided" (Bandura, 1997).Such an affirmative orientation fosters interest and engrossing involvement in activities.In sum, one of the best ways of knowing whether one is capable of some performance is by actually attempting it.Repeated success at an activity results in high self-efficacy, while failures will lower self-efficacy, unless lack of effort or adverse circumstances are involved (Bassi, Steca, Delle Fave, & Caprara, 2007;Palmer, 2006).Once a strong sense of efficacy (or inefficacy) is established, it perhaps generalizes to similar tasks and situations.This is because according to (Bandura, 1997), sometimes, a learner does not have to directly perform a task to gain efficacy information, but by watching others succeed on a task can raise his/her own sense of efficacy, especially if the person perceives himself/herself to be similar to those observed.A learner may think that "if he can do it, why can't I?".By the same token, observing others who are similar to us fail despite high effort lowers our efficacy (Bandura, 1997).Thus, it is important for teachers to encourage interaction and cooperation among learners in class, so that they can see how others work and at the same time emulate the way their friends work.

The Importance of Analyzing Writing Self-Efficacy
Bandura's theory of perceived selfefficacy overall predicts that a child's selfperception of writing self-efficacy will affect his/her subsequent writing growth.It means that an individual who holds positive writer self-perceptions will probably pursue opportunities to write, expand more effort during writing engagement, and demonstrate greater persistence in seeking writing competence.
Understanding why a learner perceives inability to perform or achieve will give evidence to understanding how to correct writing problems.In this study, it may help to contribute to understanding the importance of teacher's assistance that might affect the learners' writing self-efficacy and writing skills in English.Therefore, it is hoped that a connection can be established by having a detailed analysis of how teacher's assistance affect the rural learners' writing skills and writing self-efficacy.It is also hoped that information obtained from the writing selfefficacy scale can be useful for monitoring individual learners.The scale may be able to assist teachers to identify learners whose selfefficacy are either initially below the norm or who do not respond positively as a result of writing instructions.In addition, by having writing self-efficacy analysis, it might enable the researcher to describe what possible learners' writing self-perceptions that the teacher is unaware of.Furthermore, the information on teacher's assistance and writing self-efficacy will be useful to teachers in helping them to select suitable instructional approaches and learning materials for the learners.This will help the teachers to modify current classroom learning environment to address areas of need in the learners' writing self-efficacy and become more aware of the indirect cues that they send to learners regarding their writing performance.According to Tanyer (2015), enhancing writing selfefficacy may play a positive role in enhancing motivation and persistence in target writing task as it can minimize the amount of stress and anxiety, and encourage attempts of objective setting related to writing.Thus, it is useful to ask learners how difficult they find the tasks are or whether they believe that they are able to complete the tasks.

METHOD
Both quantitative and qualitative approaches are adopted in order to obtain a more holistic view of the research issues.
Meanwhile the case study method is employed in order to facilitate a more in-depth probe into the matter.According to Gardner (2009:557), a mixed method develops a holistic picture and analysis of the phenomenon being studied with an emphasis on "thick" rather than "selective" description".It helps to "clarify and explain relationship found to exist between variables" (Gardner 2009: 558).The objective of this study is to investigate the level of rural area learners' writing self-efficacy.Learners' writing self-efficacy was evaluated using the writing self-efficacy scale adapted from Bottomley, Henk and Melnick (1998).Both provided the researcher with the descriptive statistics such as mean and overall scores which described the level of the learners' writing selfefficacy.The other reason for using this approach is that the element of subjectivity or bias interjected by the qualitative approach can be reduced to some extent.In order to complement and support the questionnaire findings, the qualitative approach was chosen.
Here, this study also places a heavy emphasis on the perceptions and actions of Form Four learners and teacher through classroom observations, and interviews with the learners and teacher.By having classroom observation and interviews, it allowed the researcher to assess the participants' unobservable (such as feelings) and observable behavior which were used in the language tasks.Classroom observation was used in this study to further explain the phenomenon of writing self-efficacy which was obtained from the WSE questionnaire earlier.Meanwhile, the interviews with the learners and teacher were used to describe the rationale of support given, and the learners' writing self-efficacy level.Indirectly, the combination of the qualitative and quantitative approaches here can help to confirm or cross validate relationship discovered between the variables, as "when quantitative and qualitative methods can be compared to see if they converged on a single interpretation of a phenomenon" (Gardner 2009:558).

Data Collection
At the beginning of the study, one class was asked to conduct a self-appraisal for their writing self-efficacy.The instrument was adopted and modified from the writing selfefficacy scale used by (Bottomley, Henk, & Melnick, 1997).The 37 items on the writing self-efficacy scale measure how confident the students feel about their writing abilities; the aspects of writing for self-evaluation on the scale include ideas and content, organization, paragraph formatting, voice and tone, word choice, sentence fluency and conventions.Moreover, the instrument measures the students' confidence level on the Likert scale of strongly agree to strongly disagree, as (Pajares et al., 2001) emphasized that "since neither a Likert-type scale nor a 0-100 scale is more difficult or longer than the other, using any of the format that adds predictive utility and correspond to the outcome being measured are especially warranted".Based on the result obtained, three learners were chosen as participants namely Farah (the highest selfefficacy writer), Haslina (the average selfefficacy writer) and Aishah (the lowest self-efficacy writer).These learners were then observed in the classroom.Learners' behavior which was associated with the self-efficacy in writing was observed twice a week for six months.This study adopted non-participant classroom observation for analyzing selfefficacy to understand the participants' selfefficacy phenomena in their learning to write.As Wiersma & Jurs (2005) point out that analysis in a qualitative research is a process that allows an accurate description and interpretation of the phenomenon.A detail report which includes the learners' action and dialogues was recorded and transcribed to show an interpretive description of their behavior which was associated with writing self-efficacy.According to Wang & Pape (2007), evidence of the learners' self-efficacy beliefs can be obtained based on three characteristics: persistence in accomplishing language tasks, self-awareness of English proficiency, and willingness to engage in language activities.These characteristics which were used by Wang & Pape (2007) were based on Bandura's concept of self-efficacy.The descriptions for each category can be summarized as in Table 1 below: Table 1: Wang and Pape (2007) description in observing learners' behaviour associated with self-efficacy in writing.

Categories
Description 1. Persistence in accomplishing writing tasks Learners persisted longer in the activities that they believed they could do well on, but gave up easily or avoided performing tasks that they could not do very well.This proved to be a clue that their persistence in accomplishing the writing tasks might be an indicator of their self-beliefs 2. Self-awareness of English proficiency Learners' awareness of their English proficiency also known as perceived competence.When learners show lack of competence in performing writing tasks, they are categorised as low efficacy writers.For example, a learner who is unable to retell a story as he has limited vocabulary (unable to explain in English).This can be further confirmed through an interview asking him to give reasons for not being able to retell the story.

Willingness to engage in language activities
Learners show different levels of willingness to participate in writing tasks.A classroom observation can provide evidence for a relationship between learners' self-efficacy and their willingness to participate For example, an observation showed that a learner was quite a passive participant in the classroom.When asked for reason during an interview, the learner admits that he could not answer and was concerned with his classmate's responses to his answers.
The researcher used the categories above when coding the data for the learners' behavior associated with self-efficacy in writing.This was later triangulated with the questionnaire and interviews.Each of the participants was interviewed after each observation by the researcher to understand the sources of their writing self-efficacy and why they acted or responded to a writing task in such a way.By asking questions, it was hoped that the researcher could elicit the learners' self-efficacy and the reasons behind the action .

RESULT AND DISCUSSION
Firstly, findings for the three learners' writing scores indicate that one can outperform others in certain areas even though the three participants have different level of writing selfefficacy.This contradicts previous findings such as Boscolo (2008); Ergul (2004); Hidi (2001); Mahyuddin et al., (2006); Mascle (2013); Pajares et al., (2000) which point out that self-efficacy and academic performances including writing outcomes are related.In general, the previous results revealed that writing self-efficacy makes an independent contribution to the prediction of writing outcomes and mediates between previous and subsequent achievement in writing.
Nevertheless, in this study, Farah was very confident in writing as revealed in the interview and the WSE questionnaire scores.She revealed her active attitude and even ambitious in the class discussions and written assignments.Haslina's writing self-efficacy score was also higher than Aishah, and she did represent an average writing self-efficacy level.She was occasionally ambitious in class discussions, persistent in writing revisions and showed strong interest in the class activities.Besides, her self-efficacy questionnaire total score was the second highest score among the participants.Meanwhile, Aishah had the lowest writing self-efficacy scores and did not persist longer in the writing task.She also did not show interest in participating in the class discussions.In addition, her writing selfefficacy score was the lowest among the three participants and also among her classmates.
Nevertheless, their results for writing self-efficacy did not correlate with their writing performance as predicted earlier in this study and previous research.In this study, Farah's writing achievement was almost the same as Aishah in terms of the directed writing.Similarly, Haslina showed mixed achievements where she managed to outperform Farah and Aishah in the directed writing tasks.Meanwhile, Aishah, though had a very low writing self-efficacy still could outperform Farah and Haslina in the continuous writing.These inconsistent findings may in part be explained by the context-sensitivity of selfefficacy beliefs.It is likely that different sources influence self-efficacy differently in different academic areas and at different academic levels.This justifies and supports the argument put forward by Schunk & Pajares (2009) that self-efficacy is actually subject and situation specific which means that a learner might judge his/her competence high in mathematics for example, but within mathematics, the learner might feel efficacious about algebra but not geometry.This suggests that within an academic area, high self-efficacy does not imply that learners feel highly confident in all academic areas.Similarly, in this study, though Farah showed high self-efficacy in writing, it did not necessary imply to all types of writing.Thus, the findings are varying where certain learners could outperform others in certain area even though one reported a higher self-efficacy level from another.
Secondly, though the writing performance did not predict the writers' writing scores, the participants' ability perception as from the WSE Scale findings however did correlate with their behavior in the writing lesson during the classroom observation.This supports Pajares, Johnson, & Usher (2007) view that the manner in which the learners engage text is mediated by the interpretations learners make about the skills they possess.For example, Farah had high writing self-efficacy scores, and her behavior in the classroom within the six months portrayed correspondingly.Farah showed a high interest to participate in writing activities and persisted in any situation.Similarly, Haslina's and Aishah's behavior was consistent with their writing self-efficacy scores.Haslina was active depending on the writing activities and persisted longer depending on certain factors.This behavior was consistent with her writing self-efficacy scores where she depicted strong belief only in certain aspects.Aishah was the lowest scorer in terms of the writing self-efficacy scale, and respectively showed a very passive behavior in the writing classroom whenever the teacher approached her.Obviously, the order of the participants' writing self-efficacy scores corresponded to their behavior throughout the six-month classroom observation.Specifically, looking at the five aspects from the Writing self-efficacy (WSE) scale, the three participants' behavior also corresponded to each aspect.Overall, the three participants' writing self-efficacy scores can be summarized as below: Based on Table 2 above, the General Progress (GP) dealt with the overall aspects of writing.Meanwhile, the Specific Progress (SP) focused more on specific aspects of writing such as vocabulary, topic sentence, and coherence, part of speech, organization and content descriptions.Farah rated her skills generally as average, while her SP as high.Consistently, during the class discussion, she showed her confident in her skills by trying to provide answers regarding any language errors even though the teacher did not ask her to do so.In the group work as well, she tried to check her group's writing and rectify any language or content errors.Haslina rated both her GP and SP as average.Consistent with the classroom observation data, she did not portray a constant active contributor and only tried to rectify language or content errors when the teacher scolded her or when her group was unable to improve their writing.It portrayed her as if on the verge of "can do it" and "cannot do it".On the other hand, Aishah clearly indicated a low perspective with regards to the SP and GP.This was also consistent with her behavior in the classroom where she did not want to try and help her group, and just admitted did not know how to write it.She also avoided from being asked by the teacher when the class was trying to rectify the language errors in the writing.
The five basic categories above also imply that the writers' self-perception judgments do not operate independently, but rather overlap and influence one another.For example, the general and specific perceptions of writing progress (the GP and SP) were based on some extent upon the learners' sense of how their performance compares with their classmates (OC), the kind of positive social feedback (SF) they received about their writing, and their internal comfort while engaging in the composing activities (PS).In the way, these interactions support the notion that literacy learning is both complex and social.
As the socio-cultural theory puts forward that sociocultural context influences the development of participants' learning process, while the socio-cognitive theory argues that learners need self-efficacy to complete tasks.With these two theories as the basis, this study portrays that learners' socio cultural context influenced their writing self-efficacy and at the same time writing self-efficacy also influenced their socio cultural context in which it altered the teacher's approach.
The consistency between the writing self-efficacy scores from questionnaire and the classroom observation data also suggests that it is possible to combine findings from both methods in order to see the learners' writing self-efficacy in details.From previous years until now, self-efficacy has become a welldefined concept that is supported by a growing body of research.Linnenbrick and Pintrich (2012) provide a general framework for conceptualizing self-efficacy in terms of behavioral, cognitive and motivational engagement with implications for classroom practices.Pajares and Schunk (2001) examine quantitatively how self-efficacy relates specifically to reading and writing using questionnaires.Meanwhile, Wang and Pape (2007) look at learners' behavior in general to describe the self-efficacy development.Finally, Pajares and Schunk (2001) provides practical instructional recommendations based upon each theorists work that teachers can employ to raise self-efficacy in reading and writing.In this study, the concept of writing self-efficacy is explored from multiple perspectives.Through the classroom observation and writing selfefficacy questionnaire, this study managed to provide an insight in conceptualizing selfefficacy in terms of behavioral, cognitive, and motivational engagement with implications for classroom practices.Assessing the learners' self-efficacy beliefs through observation plus writing self-efficacy questionnaire is thus warranted because these beliefs are not always self-evidenced.Capable individuals often hold deep insecurities that they will not readily admit.But through observation, their behavior are managed to be elaborated in detail and thus give an insight to the research of writing selfefficacy field.
In other words, the participants' effort, persistence and willingness as evidenced through the observation managed to be linked to their perceptions obtained through survey methods.With two combined, it can support one another in writing the self-efficacy research field.

CONCLUSION AND RECOMMENDATION
The three participants' motivational, cognitive and behavioral engagement discussed above portrayed that the new component which was teacher's assistance was actually influencing the learners' writing self-efficacy.In addition, it also showed that the combination of different method of collecting data for writing self-efficacy was a feasible way in explaining rural learners' writing self-efficacy development after teacher's assistance was provided.Here, cognitive engagement could be obtained through the rural learners' written works and their responses to teachers' question during the activities.Their written works and responses during discussion managed to describe how far they managed to internalize and apply what they discussed in her writing.Both could be the evidence of their thinking process in producing a composition.Meanwhile, their motivational engagement was evident through their verbal and written responses.Verbal responses were gained through the interview where they expressed their feeling and opinions regarding their ability in carrying out the activities and assistance provided.In addition, responses were obtained from the writing self-efficacy questionnaire which described their perception of their own ability in writing.Finally, behavioral engagement was observed through the non-participant classroom observation which was carried out in six months.
Their behavior portrayed their persistence in completing the task, willingness to participate in the writing activities, and their awareness of their proficiency that was how frequent they sought help from peers and teacher upon knowing their own weaknesses.These three components can be summarized as in Table 3 below: -Learners' perception of their own ability and skills in writing -Learners' confidence in completing written task after assistance was given.
Table 3 overall shows the combination of different methods and results for the rural learners' writing self-efficacy.Indirectly, it portrays the possibilities in carrying out the research for writing self-efficacy via qualitative and quantitative methods.In previous studies, self-efficacy were mostly assessed through the assessment based on self-report measure and not directly observed by the researcher.One of the most frequently used techniques for assessing an individual's self-efficacy is through rating scales.With rating scales, an individual is presented a series of statements (e.g."I am good in writing" or "Overall, I am better than my peers in writing") and then asked to indicate his or her level of agreement of disagreement with each statement.The responses are then tabulated to determine either specific-area self-concept scores or general self-concept scores.
However, a close examination of participants' self-efficacy helped the researcher to understand each learner's writing selfefficacy development in detailed.In this study, self-efficacy was measured via different methods which were classroom observation, interviews and writing self-efficacy questionnaire.It was found that the data derived from the qualitative methods in this study did not only support and explain more of the quantitative data derived from writing selfefficacy questionnaire, but ensuring a triangulation in collecting data.Thus, this contributes to the methodological implication of this study.By implementing three different methods of collecting data for self-efficacy, an insight about how and why interpretations of similar attainments from similar sources attainments resulted in different self-efficacy level.
With this combination of quantitative and qualitative methods in assessing writing self-efficacy, it can be implied that learners cannot accomplish tasks beyond their capabilities simply by believing that they can but self-efficacy also involves the "rules for action".In other words, self-efficacy here becomes the internal rules that the individuals follow as they determine the effort, persistence, and perseverance required to achieve optimally as well as the strategies they will use.In addition, the mixed method approach also allowed the researcher to see the connection between teacher and writing self-efficacy clearer.This is because findings derived from both methods managed to answer the question of under what conditions similar self-efficacy can result in different levels of achievement and performance.Here, it develops better understandings of the conditions under which self-efficacy beliefs operate as causal factors in human functioning, through their influence on choice, effort and persistence.Thus, this implies that quantitative efforts will have to be complemented by qualitative studies aimed at exploring how efficacy beliefs are developed, how learners perceived that these beliefs influence their academic attainments and the academic paths they follow, and how the beliefs influence choices, effort, persistence, perseverance, and resiliency.

Table 2 :
The participants' overall scores for writing selfefficacy scale.

Table 3 :
Summary of methods used to collect data for self-efficacy