Perspectives and Practices

Regarding Written
Corrective Feedback in
Swedish Context
A Case Study

Aparna Balachandran

Dept. of Language Education

Course code: USX2AK
Degree project 15 hp, second cycle
Language Education
Autumn 2017
Supervisor: Tore Nilsson
Perspectives and Practices
Regarding Written Corrective
Feedback in Swedish Context

A Case Study

Aparna Balachandran

Abstrakt
Denna studie undersökte uppfattningar om praktiker när det gäller skriftliga återkopplingsmetoder för
två lärare i engelska på ett svensk gymnasium. Projektet som utformades som en fallstudie undersökte
också de kontextuella faktorerna som påverkar lärarnas kunskaper, övertygelser och skriftliga
återkopplingsmetoder genom att samla in data med hjälp av Think-Aloud protokoll sessioner,
halvstrukturerade intervjuer och analys av kommenterade studenttexter. Resultaten tyder på att det finns
skillnader i lärarnas uppfattningar och faktiska praxis när det gäller att lära sig att skriva och
tillhandahålla skriftlig återkoppling på grund av de kontextuella faktorerna. Resultaten visar också att
lärarnas akademiska identitet, erfarenhet och institutionella krav tillsammans med andra faktorer som
betyg, tid, studentmotivering och klassrummstorlek har påverkat lärarens beslutsprocess.

Keywords
Written Corrective Feedback, Teacher cognition, Teachers as decision makers, Contextual factors

1
Contents
1. Introduction 3
1.1. Aim and research questions ....................................................................... 4
2. Background 4
2.1. Theoretical background ............................................................................. 4
2.1.1 Teacher cognition ............................................................................... 5
2.1.2 Written corrective feedback: Objectives and effectiveness ........................ 6
2.1.3 Recommended written corrective feedback practices ............................... 7
2.1.4 Written corrective feedback in instructional context ................................. 8
3. Method 8
3.1. Think-aloud protocol ................................................................................. 9
3.2. Interview .............................................................................................. 10
3.3. Participants and setting ........................................................................... 10
3.4. Data collection: Think-aloud Protocol and interview ..................................... 10
3.5. Data analysis ......................................................................................... 11
3.6. Validity and reliability ............................................................................. 12
3.7. Ethical considerations ............................................................................. 12
4. Results and analysis 13
4.1. Approaches to teaching writing ................................................................ 13
4.2. Teachers’ WCF philosophies and practices .................................................. 14
4.2.1. Approaches to WCF .......................................................................... 14
4.2.2. Type of errors to correct.................................................................... 15
4.2.3. Type of feedback .............................................................................. 17
4.2.4. Challenges ...................................................................................... 18
5. Discussion 19
5.1. Teachers’ stated beliefs and practices about WCF ........................................ 20
5.2. Contextual factors determining teachers’ knowledge, beliefs and practices ..... 21
5.3. Conclusion ............................................................................................ 22
References……………………………………………………………………………………24
Appendix 1………………..…………………………………………………………………26
Appendix 2………………..…………………………………………………………………27
Appendix 3………………..…………………………………………………………………29

2
1. Introduction
Written corrective feedback (WCF) is considered to have a pivotal role in second language (L2) writing
and therefore, feedback and error correction strategies have been studied extensively from different
theoretical perspectives. According to Hyland and Hyland (2006), research in the field of second
language learning is constantly evolving and recent theories of language learning have resulted in
significant changes in pedagogical practices which in turn have established the role of various feedback
techniques in writing pedagogy. As a result, product oriented summative feedback has been
supplemented by a more process oriented formative feedback, involving a variety of feedback strategies
at different levels by combining teacher written feedback and peer or collaborative feedback. This shift
from product oriented writing to process oriented writing from the 1970’s onwards has underlined the
significance of feedback in L2 writing and various theories of language learning such as the dialogic
nature of writing, socio-cultural theories of learning promoting peer and collaborative feedback, genre
theories about the nature of writing and theories about process oriented writing highlighted the potential
of corrective feedback in generating opportunities for learning and student motivation. (Hyland &
Hyland, 2006)
The existing body of research on written corrective feedback focuses mainly on its impact on student
learning. There is an ongoing debate on the effectiveness of various types of corrective feedback
(Chandler, 2003; Ferris, 2004; Truscott, 1996) but researchers have agreed upon the fact that there is no
conclusive evidence regarding the effectiveness of written corrective feedback in different contexts. It
should be noted that feedback provided by teachers is the most commonly practiced form of corrective
feedback in classroom contexts and hence, attention should be given to the philosophies and practices
of language teachers regarding corrective feedback. Even though error correction and feedback are
closely associated with classroom practices, the majority of the existing literature on teacher cognition
regarding corrective feedback is based on data elicited from outside the classroom contexts (Borg, 2015).
Storch also points out that most of the data for understanding teacher cognition and feedback research
is generated in laboratory-like conditions resulting in the lack of ecological validity and pedagogical
relevance (Storch, 2010). Researchers have therefore emphasized the need for a context specific
approach based on the actual practices within various pedagogical contexts. As a result, teachers’ beliefs
and feedback practices have gained importance in language learning research in recent years (Lee,
2014). Hence, it is essential to include teachers’ voices in feedback research to address the issues and
challenges in providing written corrective feedback.
Hyland & Hyland point out that despite the extensive research on corrective feedback conducted for the
past three decades, many questions regarding feedback practices remain unanswered leaving teachers
unsure about its full potential (Hyland & Hyland, 2006, p. 83). Several reviews of the existing literature
on corrective feedback also have identified a sizable gap in the research in inquiring what the teachers
actually do while responding to student writing and why do they do it that way (Ferris, 2014, p. 7). So,
in order to understand the needs of language instructors and learners, and to bridge the gap between
research and practice, written corrective feedback research should be shifted to naturalistic pedagogic
contexts. Moreover, Lee observes that the majority of feedback research has been carried out in first
language (L1) and English as Second Language (ESL) college contexts in English-dominant countries
(Lee, 2014, p. 2) and therefore, there is a growing need for the inclusion of the under-represented
scenarios such as English as Foreign Language (EFL) and elementary and secondary school education
contexts across the world. This project, therefore, intends to investigate the actual classroom practices

3
of high school teachers in providing written corrective feedback in the Swedish context in order to
understand the current issues regarding written corrective feedback.

1.1. Aim and research questions

This project aims at analyzing the beliefs and the actual written corrective feedback practices of two
teachers in a high school in an urban area. Based on data obtained from qualitative research methods
such as Think-Aloud Protocol (TAP) and semi-structured interview, the study intends to investigate
teachers’ knowledge and beliefs regarding written corrective feedback and the role of contextual factors
in shaping language teachers’ cognition and written corrective feedback practices. The study is designed
after the research questions:
• What are teachers’ stated beliefs about written corrective feedback?
• What are the contextual factors determining teachers’ knowledge, beliefs and practices?

2. Background
2.1. Theoretical background
This project analyses written corrective feedback practices in the Swedish context from the perspective
of teacher cognition. As mentioned earlier, the majority of the research in corrective feedback practices
is focused on the effectiveness of corrective feedback in enhancing student learning based on the data
elicited from outside the pedagogical context (Chandler, 2003; Ellis, 2009a; Ellis, 2009b; Ferris, 2014;
Borg, 2003; Truscott, 1996). Hence, incorporating teacher cognition as a variable in written corrective
feedback research will be substantially beneficial in better understanding the issues and challenges faced
by language instructors in providing corrective feedback (Mori, 2011).
Barnard and Burns state that “the management, motivation and sustainability of learning can be
understood only by exploring what teachers believe and do in their specific working contexts” (Barnard
& Burns, 2012, p. 2) because teachers as decision makers are responsible for putting into practice the
principles and procedures mandated by various theories and methodologies. Kumaravadivelu also
emphasizes the challenges faced by teachers as practitioners of methods in realizing these principles and
practices in its purest form in the actual classroom contexts because the theoretical assumptions are not
always derived from realistic classroom experiences (Kumaravadivelu, 1994, p. 29). This implies that it
is essential to consider how teacher cognition works in language learning in order to bridge the gap
between theory and practice. According to Mori (2011), any additional information from the point of
practitioners will be useful for researchers to understand the complex cognitive process behind teachers’
corrective feedback practice, which is not represented extensively in existing feedback research. Such
an understanding about the ways in which teachers respond to student writing will enable methodologists
and curriculum developers to update and adapt the existing concepts and procedures in language
teaching. Hence, it is important to take teachers beliefs, experiences and knowledge into account in order
to implement new concepts and principles effectively in classrooms (Mori, 2011; Barnard & Burns,
2012).

4
 2.1.1 Teacher cognition
A conceptual change in the study of teaching had marked its beginning in the late 1960’s as a result of
the developments and innovations in cognitive psychology. This resulted in a growing interest in the
impact of thinking on behavior, the recognition of teachers’ role as active thinking individuals and the
recognition of the limitations of the existing approach in conceptualizing a generalizable model of
teacher effectiveness based on that (Borg, 2015). The implication of this conceptual change in education
research, according to Borg, was a growing interest in the unobservable cognitive aspects behind
teachers’ classroom practices to deal with the growing challenges and demands in instructional settings
(Borg, 2015, p. 7). Teacher cognition research therefore, takes its beginning in cognitive psychology
with a strong interest in the ways in which knowledge and beliefs influence human actions.
In his seminal work on teacher cognition, Simon Borg defines teacher cognition as the ‘unobservable
cognitive dimension of teaching’ which constitutes what teachers think, believe and know (Borg 2003,
p. 81). According to Borg (2003), teacher cognition research ventures to answer some key questions
such as what constitutes teacher cognition, how does it develop and how does it interact with teachers’
learning and practices in the classroom. In an effort to answer these questions and to clarify the
conceptual ambiguities associated with teacher cognition research, Borg reviewed 64 studies on teacher
cognition, published between 1976 and 2002. He arrived at the conclusion that there are four major
factors shaping teacher cognition at different levels and in various degrees and they are teachers’
previous education, professional education and coursework, contextual factors and classroom practices.
While teachers’ early cognitions are shaped by their schooling followed by teacher training programs
and professional coursework, contextual factors and actual classroom practices modify their existing
cognitions. For instance, Borg points out that newly graduated teachers tend to follow recent pedagogical
theories and methods that they have learnt during their professional training while experienced teachers
create and internalize specific instructional practices according to the situational needs. Hence,
contextual factors such as curriculum, instructional needs, student motivation and time have a significant
role in teachers’ decision making and professional development. (Borg, 2003)
Similarly, Dona M Kagan (1992) also suggests that since teaching as an activity situated in person,
teachers’ professional development is extremely subjective depending on various factors among which
the uncertainties associated with classroom instruction plays a significant role. This implies that teaching
is a form of problem solving in which teachers create and internalize an elaborate system of knowledge
and beliefs to adapt their practice to the situational needs (Kagan, 1992). The teachers’ role as decision
makers is also emphasized as one of the characteristic features of the post-method era, which advocated
a shift from the conventional single method approach in teaching to a context sensitive language
education (Kumaravadivelu 1994). This implies that teachers as practitioners have to construct
classroom oriented theories of practice out of their theoretical knowledge, classroom experience and
experimentation (Kumaravadivelu, 1994). This process of constructing teacher autonomy is a part of the
‘unobservable cognitive dimension of teaching’ otherwise known as teacher cognition (Borg, 2003).
The body of research on teacher cognition and its implication on education for the past two decades
recognizes teachers as “active, thinking decision-makers who make instructional choices by drawing on
complex, practically-oriented, personalised, and context-sensitive networks of knowledge, thoughts, and
beliefs” (Borg, 2003, p. 81). An implication of this assumption is that teacher cognition is highly
personalised by teachers’ previous learning experience, professional training, institutional and
contextual factors such as curriculum, textbooks and classroom practices. This means that teachers
generate context specific classroom oriented instructional practices out of their knowledge, experience,
assumptions and beliefs whthin the academic and administrative constraints. This applies to every aspect

5
of language instruction of which written corrective feedback has a pivotal role in writing pedagogy. In
the classroom context, while responding to student writing, teachers make critical decisions on various
aspects such as the choice of errors to respond, the appropriate corrective feedback strategies and the
timing of corrective feedback (Ellis, 2009b). Written corrective feedback as response to student/ peer
writing is an interactive and contextual real-world practice. It is therefore important to consider how
teachers react to student writing and the factors determining teachers decision making in providing
corrective feedback in specific classroom contexts.

2.1.2 Written corrective feedback: Objectives and effectiveness

Hattie and Timperley give a broad definition of feedback as “information provided by an agent (e.g.,
teacher, peer, book, parent, self, experience) regarding aspects of one’s performance or understanding”
(Hattie & Timperley, 2007, p. 81). Feedback can be both positive and negative. Positive feedback is a
confirmation from the teacher to validate a correct linguistic utterance by the student in order to
encourage learning. Negative feedback on the other hand, is an oral or written comment from the teacher
following an incorrect response by the student. Corrective feedback (CF) is therefore a form of negative
feedback as it involves error treatment. It can be both oral and written. One distiction between oral
corrective feedback and written corrective feedback (WCF)is that oral CF is mostly provided as part of
the instruction within the classroom. WCF, on the other hand, is usually written responses provided by
the teacher after reviewing student writings, which is not an immediate response as oral feedback.
Hence, while providing WCF, teachers may have to address a number of issues in individual student
writing ranging from mechanical errors to content problems using different strategies and methods. By
doing so, teachers help students to “notice the mismatch between their interlanguage and target
language” (Chandler, 2003, p. 293) which facilitates second language acquisition. WCF is therefore an
important tool in drawing students’ attention to linguistic form.
However, the effectiveness of WCF is a topic of debate and there is no conclusive evidence regarding
this. The debate on WCF started with Truscott’s (1996) case against the effectiveness of WCF in which
he argued that WCF feedback is not only ineffective, but also damaging (Truscott, 1996). Based on
Krashen’s assumptions, comprehensible input is sufficient for second language learning and if that
condition is fulfilled, the learner will automatically acquire the necessary grammatical knowledge via
Language Acquisition Device (Mitchell, Myles, & Marsden, 2013, p. 160). Error correction, on the other
hand, makes students defensive and hence they avoid using complex linguistic constructions which in
turn affects language learning. Drawing on Krashen’s observation, Truscott argued that WCF generates
only a superficial form of knowledge which is called pseudolearning (Truscott, 1996, p. 345) which has
a negative effect on learning. Hence, Truscott’s objections against WCF are that it can affect student
motivation as it leads to stress and anxiety leading to simplified writing, and that error correction is a
time consuming process that affects productive learning (Truscott, 1996). Pienemann & Keßler (2012)
have also made a similar observation that it is impossible to teach a grammatical form through correction
or by any means if the learner has not attained the right level of proficiency or in other words, the
developmental readiness to acquire language (Pienemann & Keßler, 2012).
It should be noted that researchers have different opinions regarding the nature of L1 and L2 acquisition.
While some researchers argue that it is sufficient to provide the same type of instructions to L2 learners
as L1 students, others argue that L2 acquisition is significantly different from L1 acquisition. Therefore,
research in the field of L2 acquisition should address the L2 learners’ linguistic gap in comparison with
native speakers (Ferris, 2003, p. 16). Hence, Swain (2005) challenges Krashen’s (1987) argument that
exposure to rich comprehensible input is the only requirement for L2 acquisition based on the empirical

6
evidence from French immersion classrooms in Canada (Swain, 2005). She emphasized that output as
an active communication process promotes learning by encouraging the learners to test their acquired
linguistic knowledge in authentic contexts. By doing so, learners pay attention to the structure of the
target language and the gaps in their L2 knowledge and production. Hence, inviting learners’ attention
to specific linguistic features in the communicative process helps them to attain proficiency in L2 setting
(Abbuhl, Gass, & Mackey, 2012). Long (1988) also holds the view that some attention to linguistic form
is necessary in L2 instruction (Long, 1988). A according to Bitchner and Storch, a growing body of
research in recent years have reported potential advantages of WCF in L2 development and many of
these studies are focused on the effects of WCF on the accuracy of L2 learners’ written output (Bitchener
& Storch, 2016). One of the advantages of L2 writing and WCF in L2 development, according to
Bitchner and Storch, is that the written output is permanent unlike oral output and hence can be revisited
by the learners, which provides an opportunity to develop metacognitive awareness about language
learning. Furthermore, in written output, learners have additional time to think and therefore plan their
linguistic utterances in writing (Bitchener & Storch, 2016).

2.1.3 Recommended written corrective feedback practices

Studies on the effectiveness of WCF have systematically investigated the reasons for the failure to
design effective corrective feedback strategies focusing on the impact of different variable such as the
type of feedback, timing of feedback and other external contextual factors in L2 writing context.
Summarizing the research of composition scholars, Ferris (2003), points out that teacher commentary
on student writing is criticised as it is often used as a means to justify grades with excessive focus on
errors and with a tendency to appropriate student writing. In this context, teacher commentary can
demotivate students and make them comply with the teachers’ wishes rather than focusing on the actual
problems in writing. To avoid this, teachers should replace vague directives with text specific comments
to communicate effectively with students. Moreover, teacher commentary should be treated as an
ongoing conversation between the teacher and the learner in a larger classroom context. (Ferris, 2003)
Researchers have also examined the potential of different types WCF in second language learning based
on the focus and form of the corrective feedback and also the types of errors teachers should address in
teaching writing. According to Ellis, Loewen and Erlam,
Corrective feedback takes the form of responses to learner utterances that contain an error. The
responses can consist (a) an indication that an error has been committed, (b) provision of the correct
target language form, (c) metalinguistic information about the nature of the error, or any combination
of these (Ellis, Loewen, & Erlam, 2006, p. 340)

Based on this definition, Ellis (2009a), has further summarised the discussions regarding the choice of
errors to correct and the choice of corrective feedback strategy. Regarding the type of errors to correct,
researchers suggest teachers should pay attention to global errors rather than local errors in student
writings. Global errors are the errors that affect the comprehensibility of the text while local errors are
the minor errors affecting single elements in a sentence. Ellis points out that there are two issues
regarding the choice of errors and they are whether to correct all the errors or to select specific errors in
student text. The former approach is extensive and unfocused while the latter is intensive and focused.
Ellis highlights that methodologists often advise focused WCF as the best practice and emphasize the
need to focus attention on a few types of errors rather than commenting on all the errors in student
writing. Similarly, indirect feedback where the error is located and indicated without providing
correction is advocated by researchers as it encourages self-correction (Ellis, 2009a; Ellis, 2009b; Ferris,
2014). According to Ferris (2014) indirect feedback is perceived to be beneficial in long-term student
development when compared to direct feedback and similarly, Ellis (2009b) highlights the advantages

7
of indirect feedback over direct feedback as it facilitates guided learning and problem solving. He also
reviews a number of studies, comparing the effectiveness of using metalinguistic clues or error codes to
indicate an error, and concludes that using error codes or brief grammatical descriptions in the margin
promotes noticing and hence enhances accuracy. The use of electronic resources in providing WCF is
also mentioned by Ellis (2009a) in which he observes that providing a brief comment on each error and
with links to resources showing the correct form can encourage self-correction and metalinguistic
awareness in students. (Ellis, 2009a; Ellis, 2009b)

2.1.4 Written corrective feedback in instructional context

In an instructional context, WCF can have multiple functions such as to inform students about their
expectations and evaluation of the assignment and also used as ‘feed forward’ for students to improve
and enhance their performance in future writing. According to Ferris (2003) written feedback aims at
informing students whether or not their text conveys the intended meaning effectively, helping them
understand the potential questions and concerns from the perspective of an audience to encourage self-
evaluation and finally provide students with a motive for revision (Ferris, 2003) . However, a number
of factors can influence teachers’ beliefs and practices regarding corrective feedback. According to Diab
(2005), the form and focus of teachers’ commentary on student writing are significantly varying
depending on the genre, the setting, limitations such as time and the size of the class and other factors
such as the specific time during the semester and also the requirements of the students. Similarly,
contextual factors such as teachers’ academic identity, institutional and curricular demands and teachers’
own personalities can have an impact on their beliefs and practices (Diab, 2005; Li & Barnard, 2011).
In a case study conducted in a high school in Hong Kong, Amy Tsui (1996) describes the contextual
factors creating problems for the teacher and the teacher’s decision making in the instructional context
to resolve the problem. In an attempt to resolve students’ anxiety and demotivation to write, the teacher
was asked to introduce process writing in her classroom which turned out to be highly effective
according to the teacher and the students. However, the teacher had to revert to product writing as it
required much longer time to complete each task and the students ended up writing fewer texts in
comparison with other classes. Another problem the teacher experienced was that the students started to
slip in terms of grammatical accuracy which would disadvantage students in examinations and
assessments. Consequently, the teacher was not encouraged to continue with process writing by the
institution. According to Tsui, this is an example of how curricular constraints override teachers’
attempts to execute methodological innovations in teaching. However, the case study reveals that the
teacher, out of her dissatisfaction with product approach, adapted her practice by modifying process
approach by integrating process and product-oriented approaches. Tsui points out that this demonstrates
how teachers create their own personal knowledge system out of situational needs. (Tsui, 1996)

3. Method
This project is designed as a case study involving two participants for a number of reasons. The aim of
this project is to understand how the knowledge, beliefs and experiences of teachers shape their written
corrective feedback practices in classroom contexts. Hence, this project resorts to case study as a
methodology as it is effective to analyse the WCF practices of teachers in detail. Moreover, it is effective

8
and practical to conduct case studies as it involves a limited number of participants and sites. Small
samples can however, provide data that is otherwise not represented in research involving large sample
size. Since case studies can provide insights into the complex and dynamic aspects of language learning
and human interaction, this method is particularly useful in studying the perspectives and practices of
teachers about corrective feedback. (Duff, 2012)
This case study is guided by the conceptual framework based on teacher cognition research, theories
and previous research and assumptions on corrective feedback. Since this project aims at having an in-
depth understanding about the feedback practices of teachers in a specific context, qualitative research
methodology is used to elicit data. Qualitative research method is considered as an effective practice in
studying participants in their naturalistic setting and hence it suits the purpose of this study (Richards,
2009). According to Mackey and Gass, “the term qualitative research can be taken to refer to research
that is based on descriptive data that does not make (regular) use of statistical procedures” (Mackey &
Gass, 2005, p. 162). Qualitative research aims at the careful and detailed description of data elicited
from studying individuals and events in their natural setting, without altering the contextual factors.
Since this project intends to take the contextual factors affecting teachers’ decision making into account
qualitative research method will be effective to get an insight into the phenomenon at a micro level. In
other words, qualitative research will be helpful in getting an emic perspective, an understanding of the
concept from the perspective of the participants, which is the objective of this study (Mackey & Gass,
2005, pp. 163-164). Therefore, this research involves data triangulation by using multiple data types to
obtain an objective understanding of the phenomena from the perspective of the participants in their
natural context. The methods used to elicit data for this study are Think-aloud protocol, semi-structured
interview and written text samples from students.

3.1. Think-aloud protocol

Originally used in psychological research, Think-aloud protocol (TAP) is considered as an effective
method for understanding cognitive processes involved in context-free tasks (Barnard & Burns, 2012,
p. 129). TAP is a type of verbal report in which participants are required to speak aloud their thought
while they perform a given task. Verbal reports are categorized based on the temporal frame and the
level of details in recording. TAPs are concurrent verbal reports as it is collected while the participants
perform the task. Concurrent verbal reports are further categorized as non-metalinguistic, in which the
participants verbalize their thoughts per se, and metalinguistic in, which the participants are required to
verbalize additional details such as explanations and justifications for the actions involved in performing
a task. Non-metalinguistic concurrent verbal reporting is used in this project because this project is
interested in analyzing how teachers think while they evaluate student writing. While evaluating a
student text, the teacher has to make decisions based on a number of factors such as the ways in which
he or she has prepared the students before carrying out the task, the rubric and the knowledge
requirements, the knowledge level of the student, her own knowledge and experiences. Based on that
the teacher has to choose appropriate feedback strategies considering the type of errors, feedback timing
and even the effectiveness of different types of feedback. This involves complex and dynamic cognitive
procedures which are otherwise unobservable. TAP is therefore, a suitable method to observe the
cognitive aspects of teachers’ decision-making process by encouraging them to verbalize what they
actually think while they carry out the task. In that way, the strategies used by the teachers to treat
different types of errors can be observed. (Bowles, 2010)

9
 3.2. Interview
Interview, on the other hand, is one of the most commonly used instrument for data elicitation that has
an important role in case studies (Friedman, 2011). Interview as a data elicitation method helps to
investigate the problem at hand from the participants perspective. The purpose of the interview questions
and the aim of the interviewer is to stimulate participants reflections on a specific topic. The interviewer
shouldn’t therefore, dominate the interview, but provide supportive feedback to encourage the
participants to discuss the topic in detail. It is therefore, an effective tool in case studies as it helps the
researcher to investigate different perspectives of the participants. Hence, this project employs semi-
structured interview, which falls between unstructured and structured interviews, as a data elicitation
method. In a semi-structured interview, the interviewer prepares an interview protocol with a set of
questions focusing on the topic and the research questions. Unlike structured interviews, semi-structured
interviews use open ended questions, which can be further developed according to the participants
response and the needs of the research. Friedman identifies some generalized guidelines for semi-
structured interviews and the need to avoid closed ended questions is one of them. Similarly, ‘leading
questions’ imposing a point of view or leading to specific types of responses should be avoided.
Interview questions should be comprehensible, and they should not be complex. The interviewer should
be focused on one thing at a time rather than asking multiple questions at once. (Friedman, 2011)

3.3. Participants and setting

Two teachers in English from a high school in Stockholm participated in this study. Teacher 1 has both
English as her first subject and Spanish as her additional subject and she is working with students who
attend Social Science programme. Teacher 2 has English as his first subject and German, Leadership
and Organization Economics as his additional subjects. He belongs to the Economics department.
Teacher 1 has less than ten years of experience in teaching while teacher 2 has been working since 1999
and he has been participating in various continual teacher education programmes during his career as a
teacher. Both teachers are from Sweden and Swedish is their native language. This case study focuses
on the WCF practices of the teachers in teaching their first subject, English in their classrooms.

3.4. Data collection: Think-aloud Protocol and

interview
The teachers were asked to carry out a writing task in their respective classes. A writing task from a
previous National Test in English was used for this purpose for authenticity and the task was to write to
a weekly magazine discussing different kinds of temptations people today are faced with. The teachers
were asked to prepare the students as they usually do, and the students wrote the task as an in-class essay
during a one-hour lesson. The essays were then collected, and two random texts were chosen for each
teacher to evaluate as part of the TAP sessions at a suitable time and place. The interviews were
scheduled two days after the respective TAP sessions. Both the TAP and the interview were carried out
entirely in English.
Before evaluation, the teachers were provided with an informed consent form (Appendix 1), and
information on the procedure of TAP session as follows:
You are expected to verbalize whatever comes to your mind while evaluating the text. You should keep
talking. You may ask question while doing the task, but I may not answer them. I will not tell you when

10
 to stop the task. You can stop whenever you feel the task is complete or if you feel uncomfortable with
the task (Appendix 3).

Two texts were chosen for each teacher and the first task in the TAP sessions were carried out as a trial.
Permission from the teachers were sought to be present during the TAP sessions to make field notes.
The TAP sessions were started with greetings followed by an informal briefing of the procedure as
mentioned above. I repeated the information in the consent form orally for them and their permission
was sought before recording the sessions. The participants were asked if they have any further questions
before they started the evaluation. Once they started the TAP session, I sat down against the participants
in order to make field notes and to observe their facial expressions and body language while doing the
task. I remained silent throughout the sessions and did not interfere during the sessions other than
gesturing them to speak whenever necessary. However, I reviewed the pilot session of the TAP with my
participants and suggested improvements in the process before they started the actual task. The time,
date, venue and the name of the participants were marked in the field notes.
The follow up interviews, spanning approximately 30 minutes, were carried out two days after the TAP
sessions. The interview questions (Appendix 2) were formed after listening to the Think-Aloud
recordings and the questions were open ended. Such information as justifications and explanations for
their decision making was elicited through interview questions and discussions formulated considering
the verbal reports and the marked student texts. Hence, the semi-structured interview incorporates
stimulated recall to certain extent as it is conducted after the TAP sessions and the questions were
prepared in reference to that, seeking clarifications for the strategies used by the participants in providing
WCF. For instance, the question number (8), Do you attempt to correct all the mistakes in the text?
(Appendix 2) has sub questions seeking for clarifications on the teachers’ practices regarding selecting
specific errors in student texts. In the interview protocol, the interview ends with a discussion with the
participants by asking for further clarifications on the marked student texts. However, the interview
questions were designed after the data obtained from the TAP and starting with the direct written
questions in the interview protocol, the interviewer already initiates discussions on teachers’ feedback
practices during the interview itself. Furthermore, this project focuses also on the typology of written
corrective feedback, based on Ellis’ classification (Ellis, 2009a), and hence the questions from number
5 – 9 are formulated after this, without directly referring to the terminology provided by researchers.
This is to make sure that the questions are not very complicated with theoretical jargons so that the
participants feel comfortable to discuss their reflections on their use of different strategies in classroom.
Following the interview, the participants were asked to respond to a single preference survey question
in which they were required to mark how important it is for them to address different error types, such
as mechanical errors, lexical errors, grammatical errors, organizational problems and content problems
in student essays (Appendix 2). The TAP sessions and interviews were audiotaped using a mobile phone
voice recorder application to avoid filtering of information. Field notes were taken to supplement any
missing data Finally, the marked student essays were collected from teachers as a support material for
analyzing the TAP sessions and the interviews. Moreover, the marked student essays have written
comments from the teachers, which is an important source of data.

3.5. Data analysis

A thematic content analysis of the collected data was done after transcribing the recorded interviews
and TAP sessions. Content analysis, according to Friedman, is an approach to analyse qualitative data

11
by finding patterns using systematic coding procedures (Friedman, 2011, p. 191). In order to understand
teachers stated beliefs and practices regarding WCF, the data was categorized based on the type of errors
(Global/Local), Type of feedback strategies used based on Ellis’ (2009a) classification and the
challenges faced by teachers in teaching writing and providing feedback. Data from both interview and
TAP is thus analysed and categorized to find patterns. The results obtained were developed and refined
to look for the contextual factors determining teachers WCF practices.

3.6. Validity and reliability

Bowles (2010) points out that the challenges to validity can vary according to the type of verbal reports
and the potential challenge to the validity of concurrent reports is reactivity because “it is not known
whether the act of verbalizing while completing a task is reactive, acting as an additional task and
altering cognitive processes rather than providing a true reflection of thought” (Bowles, 2010, p. 14). In
other words, the act of verbalising the thoughts while performing the task imposes an additional
cognitive processing on the participants which can affect the way in which they perform the task.
However, non-metalinguistic verbalisations are largely non-reactive because the participants are not
required to provide additional information during verbalization of their cognitive process (Bowles, 2010,
pp. 14-15). Hence, according to Bowles (2010), non-metalinguistic verbal reports are fairly accurate
even though they might slow down processing as the participants might take additional time to verbalize
their thoughts. In order to minimize reactivity and to ensure validity, this project has employed non-
metalinguistic TAP and additional information and explanations were gathered using semi-structured
interviews, conducted two days after TAP sessions. Interviews were purposefully scheduled two days
after TAP in order to avoid giving the participants explicit details that might affect the accuracy of verbal
reports. Moreover, the semi-structured interview was designed to supplement the data obtained from
non-metalinguistic TAP sessions, stimulating the participants to explain and justify their decisions
during task performance. Furthermore, this project has followed the guidelines provided by Bowles
(2010) to ensure validity of TAP sessions by providing uniform verbalization instructions to both
participants. Instructions included a short description of the project, the rationale for thinking aloud and
general instructions about how to think aloud (Appendix 3). The participants were given opportunity to
perform a warm-up task and to ask for clarification before they started the actual task. The participants
were reminded to continue thinking aloud throughout the process using gestures without disturbing the
cognitive process. (Bowles, 2010, pp. 120-121)
Finally, this project has used multiple data types which can be compared for validation of results. The
use of multiple data types generates both etic and emic perspectives which is effective in terms of
validity but can be logistically challenging (Duff, 2012). Triangulation of data in which multiple
research methods and multiple sources of data are used in a single investigation is recommended as it
helps in ensuring reliability and validity of research (Mackey & Gass, 2005).

3.7. Ethical considerations

The participants in this study were contacted via email along with a letter of information describing the
aim and the design of the research. Since the data collection methods involved recording the TAP
sessions and interviews using a voice recorder application, this project has followed the guidelines
provided by Swedish Research Council by assuring anonymity of the participants (Swedish Research
Council, 2017, p. 15). A consent form (Appendix 1), assuring that the materials collected will only be

12
used for this project and that any use of the information contained in the material would not reveal the
participants name and identity. Moreover, the consent form informed the participants that both the TAP
and interview would be recorded and disguised extracts from the material would be quoted in the thesis.
The participants were also informed about their right to withdraw from this study at any time without
providing any specific reason to do so. In addition to that, the information in the consent form is repeated
orally and the consent for recording the tasks were sought before each TAP sessions and interview.

4. Results and analysis

The results of this research are presented and analysed in this chapter with reference to the research
questions. The findings and their interpretation are presented under the thematic categories discussed
in data analysis section.

4.1. Approaches to teaching writing

When asked about their approach to teaching writing, both teachers emphasized the significance of pre-
writing activities in providing sufficient input in their classroom. The pre-writing activities consisted of
brainstorming for ideas and discussions on various perspectives, themes and focus of the writing task.
Both teachers reported that they always tried to provide examples, which are often advanced texts so
that the students will be familiar with various textual patterns. Both teachers mentioned that they
prepared their students to be familiar with the genre or the type of the essay they are going to write, as
shown in Extracts 1 and 2.
I try to show them essays where there is structure in it so that they not only hear it from me, but they
can also actually see a good example for what structure is …. … try to exemplify many times, as often
as I can… (Extract 1: Interview, teacher 1)
According to teacher 2,
So, whenever they select the subject, they need to focus on having lots of possible angles to that subject,
in order for them to write a convincing essay… I try to... in different ways, exemplify that… I have
sample text. Very often I can have quite an advanced one… (Extract 2: Interview, teacher 2)

Teacher 1 reported that she prefers to focus on the overall structure of the essay and language as a whole
rather than providing explicit grammar instructions during the prewriting activities or while providing
WCF.
“(When) we talk about writing, we talk less about grammar and more about structure and language as
a whole. More that you have to be formal and have a big variation in your language…, should have a
more advanced language than what you have when you speak” (Extract 3: Interview, teacher 1).

Teacher 2 also suggested the importance of cohesion in writing and he reported that he prepared his
students to write a ‘convincing’ essay.
So… I work with building the subject, building the perspective, building the angles... for the students.
‘How can I address this? What’s my point? What is a convincing introduction? Why would I want to
continue reading that…?’ (Extract 4: Interview, teacher 2)

Therefore, pre-writing activities in class included discussions on how to write thesis statements, topic
sentences and convincing introduction. However, it is evident from the interview that the teachers

13
mentioned some of the linguistic features of writing tasks such as, how the language of writing differ
from everyday spoken language, the use of passive form in formal writing and the strategies to achieve
cohesion in writing. For instance, teacher 1 made the contradicting statement (see Extract 3), that she
chose to talk about grammar in particular during pre-writing and when she provides feedback because
of the knowledge requirement dictated by the curriculum in which fluency is an important aspect.
We talk more about that (linguistic aspects), we talk about grammar because of the knowledge
requirement...that, it needs to have fluency which you kind of lose if you don’t have correct grammar,
but that’s like the only thing within the knowledge requirement which actually hints that you need to
grade the grammar. (Extract 5: Interview, teacher 1)

According to her, the text loses fluency if the students make grammatical errors. For her, fluency is the
only thing in the knowledge requirement that suggests the need to grade grammar according to the
Swedish curriculum.

4.2. Teachers’ WCF philosophies and practices

This section presents a detailed analysis of the teachers’ stated beliefs and practices regarding WCF
according to the aspects identified in section 3.5. The analysis focuses on four such aspects, which are
approaches to WCF, types of errors to correct, type of feedback and the challenges faced by teachers in
providing WCF. The results are presented and analysed in the respective subsections.

4.2.1. Approaches to WCF

Teacher 1 reported that she provides feedback, often oral feedback, on a regular basis to her students.
She mentioned that she makes her students write two long essays every semester, which are assessed
and graded by the teacher. Detailed feedback for those essays are given by the teacher orally and in
writing. She talks to her students individually, justifying the grade and providing suggestions for the
next time.
I try to talk to everyone whenever I hand their essays back… And when I give those back I always try
to talk to them, like ‘you have got this grade, and this is why...look at this...maybe you could improve
this...think about this next time’... (Extract 6: Interview, teacher 1)

She mentioned that grading is also a motive for her to make students revise their written assignments. If
she does not have enough material to grade a student, or if the students’ performance is not satisfactory,
she will make the students revise and improve the essay before grading.
I kind of test every knowledge requirement many times during the year. But if I can see, may be in
March, or like beginning of April that I don’t have enough to grade someone…, to give them a grade
in June, I will make them rewrite. (Extract 7: Interview, teacher 1)

Teacher 1 also reported that she prefers to talk to them rather than provide WCF and that she does not
make students rewrite their essays in normal cases. She chooses specific students to have individual
writing conferences based on their performance and knowledge level.
Teacher 2 reported that he has an elaborate way of responding to student writing and he uses technology
significantly in providing WCF. He makes his students share their written assignment on Google
Documents so that he can watch them write and can provide comments instantaneously.
I give them feedback… in writing… in Google. They also can comment or disagree with me, in
writing… like ‘I don’t agree. I think I actually pointed this out…’ Sometimes we have an online
discussion on certain aspect… (Extract 8: Interview, teacher 2)

14
All the student writings will be thus saved in a folder as a portfolio for each student, which both the
teacher and the students can revisit to assess development. He makes sure that the students revise the
texts based on his feedback by asking them to change the colour of the text once they have rewritten the
specific sections or have resolved the issues in the texts. Also, he reported that he meets his students
individually at least twice during the course to discuss their performance and to suggest improvements
based on the portfolio. One of the advantages of responding to student texts digitally, according to him,
is that the whole process becomes a communicative activity providing the teacher and the students with
an opportunity to respond in real time. Teacher 2 believes that WCF is very efficient and effective, if
combined with digital technology. Moreover, it helps him document students’ progress over a period of
time and the permanency of such a procedure helps students to revisit and reflect on their performance.

4.2.2. Type of errors to correct

Teacher 1 believes that the most important aspect of a written text is structure, both the structural
framework and ‘the inner structure’. By inner structure, she means the organisation of thoughts and
ideas in an essay. She justifies this by quoting the phrase ‘relativt strukturerat’ from the criteria for grade
E in curriculum (Skolverket, 2011, p. 61)
Immediate attention would be, if the texts lack structure completely, on both inner structure and the
actual structure. It says…’relativt strukturerat’ on the E level of the knowledge requirement and if you
can’t find that in a text, it’s definitely an F. No matter how good the text is. (Extract 9: Interview,
teacher 1)

Therefore, teacher 1 strongly believes that WCF should only focus on the global errors such as errors
that affect comprehensibility. Local errors, according to her, do not actually affect the flow of the text
and hence are not a concern in writing.
If it is a linguistic error, I do nothing. Because, I think that doesn’t matter... It doesn’t stop the essay
from having fluency, it doesn’t stop the essay from being easy to understand and follow. It doesn’t stop
the structure and nothing like that… And someone who has English as their mother tongue wouldn’t
have a problem understanding the essay. (Extract 10: Interview, teacher 1)

She mentioned that she does not correct all the errors in student texts and if she finds a linguistic error
in student writing, she normally leaves it untreated. According to her, a couple of errors in the text does
not affect the comprehensibility of the essay. Neither does she provide the students with the correct
linguistic form if she locates an error in a text, unless the students repeat the same errors, such as subject-
verb agreement, in their writing. However, she gives the students suggestions to improve problems at
content and organizational level.
On the other hand, teacher 2 believes that both the overall structure and the details matter in writing and
even if grammar is not mentioned explicitly in the curriculum, it is an integral part of writing. According
to him,
The broad strokes of the adaptation to topic or genre and all those things are really important… But I
think also the, the grammar and vocab is very important, for the overall impression. If you have a great
essay in terms of structure and adaptation, …the fluency is lacking for the reader if there are lot of
disturbing errors”. (Extract 11: Interview, teacher 2)

Therefore, he emphasized that teachers should pay attention to errors concerning the overall content,
organization and the structure of the essay and also, to the minor errors like grammar and vocabulary.
However, the teacher believes that correcting everything in student texts is not a good practice.
Therefore, he does not correct all the mistakes in student texts unless it is an exceptionally good essay.
He said that he, normally focuses on providing positive feedback and always look for the trend in
selecting specific errors to correct.

15
During TAP, Teacher 1 focused mainly on variation and fluency in the text and made explicit comments
on the organization of the essay, as shown in Table 1. She commented also on the clarity of expression
in the introductory paragraph and made remarks on the student’s choice of words well adapted to the
subject.
...short but efficient...hmm... introduction… about temptations and it makes...hmm... the subject pretty
clear… (Extract 12: TAP, teacher 1)
It’s well balanced... hmm...if you see to…to...hmm...the introduction and the main body and the
conclusion(s)..., or...the conclusion. The conclusion is really good (Extract 13: TAP, teacher 1)

When the data obtained from TAP was colour coded, it was observed that teacher 1 focused mostly on
organization and structure of the essay. She made fifteen comments on structure, organization, variation
and fluency while six comments were on formal issues such as prepositions, spelling mistakes and
subject-verb agreement. Moreover, most of the negative comments were on the formal issues in the text.
Teacher 2, as shown in Table 1, made explicit and specific comments on both organization and form
equally. The teacher expressed frustration when the student made repeated grammatical errors. For
instance,
Hmm… sub-verb…has…sees…too bad she has so many verbs, that’s a bit disturbing. (Extract 14:
TAP, teacher 2)

Organization, content and form were given equal importance in his comments and he made altogether
sixteen comments on the organization of the essay with special attention to thesis statement, topic
sentence and cohesion. However, the teacher does not make explicit comments on fluency and variation.
Teacher 2 commented on nineteen formal errors on grammar, such as genitive problem, subject-verb
agreement and spelling during TAP.
The two participants, therefore have different preferences when it comes to the type of errors that should
be given immediate attention. When the marked student texts were examined and compared with the
interview and TAP data, it became evident that teacher 1 has left many local errors untreated. Those
errors were not located in the text while she commented extensively on the structural framework of the
essay such as introduction, body and conclusion, along with fluency and cohesion. Her comments on
language form were mostly focused on the implicit grammatical practices such as syntactic fluency and
hence she encouraged variation in student writing. However, her statement that she occasionally
corrected the errors repeated frequently in the text, was not consistent with her practice. Even though
Table 1: An overview of the type of issues targeted by the teachers during TAP
Type of issues Teacher 1 Teacher 2
Content 4 4
Organisation 3 8
Fluency 3 0
Repetition 3 0
Style 2 4
Grammar 1 11
Mechanics 3 4
Vocabulary 2 4

16
the teacher commented on the informal use of language during TAP, she did not mark or provide any
written comments on the student’s use of contractions throughout the essay.
Teacher 2 on the other hand, holds the view that both global and local errors should be addressed while
providing WCF. According to him, grammar and vocabulary is as important as content and organization
of the text. Analysis of the TAP data and the marked student texts reveals that the teacher’s approach
conforms with his beliefs. He has made explicit comments on grammatical, lexical and mechanical
errors in student writing along with content and organisational problems. However, the marked student
text revealed that the teacher has commented on majority of the linguistic errors in the written texts on
contrary to his statement that he does not correct all the mistakes in student texts.

4.2.3. Type of feedback

When asked about the type of feedback they provide, teacher 1 reported that she provides very little
written feedback on student texts and when she does, she provides indirect feedback by locating the
error by marking it in the text. She reported that she uses error codes to locate repeated errors, but never
provides the students with the correct linguistic form. The teacher mentioned in the interview that she
often uses metalinguistic clues to the nature of the error by using error codes. She reported that she
provides elaborate comments on repeated errors but does not comment on every error in the text.
if it’s something that keeps coming back…Yes. But I don’t underline or circle or make comment to
every mistake”. (Extract 15: Interview, teacher 1)

Even though she commented on errors such as subject-verb agreement issues during TAP, she did not
provide any kind of metalinguistic clue to the errors. On the other hand, she located the error by
underlining it within the text. When asked about this, she mentioned that she preferred oral feedback to
WCF and hence the errors were located for her to identify the issues in the text to discuss later with the
student. However, this was in contradiction to the statements during interview.
Teacher 2 provided metalinguistic corrective feedback combined with electronic corrective feedback by
using error codes and a hyperlink to a concordance file that explains and exemplify the correct usage.
He keeps a separate document with explanations and examples for grammatical structure, vocabulary,
phrases, organization, thesis statement, topic sentence and genre. He uses error codes while providing
feedback with a hyperlink directing to this document which enables the students to refer to the linguistic
forms with specific examples.
I have a link pasted, link to another Google Document that I use for improving English, which is focused
primarily on grammar and vocab issues. I have a set number of abbreviations. So, they can always
through the course access this document. So, there you have both grammar issues but also thesis
statement and topic sentences. Both the broad strokes and the details…, everything from the spelling
to interpunction to… or punctuation and… pronouns or word order or word formation and so on...
(Extract 16: Interview, teacher 2)

He said that he normally focuses on providing positive feedback and always looks for the trend in
selecting specific errors to correct. He also mentioned that he highlights the texts in different colours
suggesting different levels of improvement needed so that the students will not be discouraged to see so
many errors marked in their text. It should be noted that he works in the department of Economics and
he teaches two other subjects, Leadership and Organization and German. Hence, he works with cover
letters and job applications in his classroom and he mentioned that he is very strict with the linguistic
features in such instructional contexts. He has a more extensive procedure when it comes to teaching
writing cover letters and job applications and he usually prepares the students beforehand to write
multiple drafts and to expect more comments on each draft.

17
 When they write job applications and cover letters…, I am very strict with that. I have prepared
them…that you are gonna be seeing a lot of corrections probably…so be ready for that. And then they
work really hard to get better and then they will get another feedback… like still some red, but… it
takes a lot of time to do that. (Extract 17: Interview, teacher 2)

The Data from TAP and the evaluated student texts were in confirmation with his claims. Also, his
statement that he focuses on combining positive feedback with constructive WCF was in alignment with
his practice. Teacher 2 expressed appreciation while reading the text such as “very promising”, “she has
really improved”, “good”, “she has actually picked up” and “it’s good”. However, he provided corrective
feedback on a number of grammatical, lexical and mechanical errors in the text by commenting on
subject-verb agreement, verb form, punctuation, genitives and spelling, even though he reported that he
only looked for recurring errors in such cases. Teacher 2 concluded his comments on student texts as
follows:
If she improves…hmm… thesis statement in the introduction…and…works harder on subject verb
agreement and…hmm…just more idiomatic phrases…hmm…she can reach even higher, but this will
be a C. hmm…ya…she has improved. (Extract 18: TAP, teacher 2)

During the TAP sessions, both teachers provided positive comments on student texts. Teacher 1’s
comments on the text were mostly positive; out of 300 words in the transcribed data, more than 200
words were positive. According to teacher 1,
And for comments to the students I would say... that even though a lot of the text is...hmm... it has a
variation, you need to...perhaps find another word for ‘temptation’ because it keeps coming back
throughout the text many many times and it...it's...hmm...when you've read it a couple of times it's the
only...word...that you actually see… And look at for some...some...easy, silly, spelling mistakes and….
Otherwise well done! (Extract 19: TAP, teacher 1)

She also mentioned the positive aspects of the text in the beginning, using words like “good”, “well-
adapted”, “nice” and “positive”.

4.2.4. Challenges
According to teacher 1, the most challenging aspect in teaching writing is time. She reported that even
though writing multiple drafts helps students to improve fluency and accuracy, it is impossible to do
that every time the students perform writing tasks because of limited time and the size of the class.
I wish that… I did that every time… I try to do it as many times as possible… To have them write
something and I will collect it and… give them comments and give it back and… keep doing
that…because it takes forever and… I have 33 students in my classes. (Extract 20: Interview, teacher
1)

Therefore, she resorts to peer feedback as a means to overcome these constraints and also limits the
revision process to larger essays that they write during the course.
According to teacher 2, teaching students how to plan an essay with a thesis statement, convincing
introduction and logical paragraph divisions with topic sentences (Extract 4) is the challenging part in
teaching writing, as the students may not always understand these concepts.
I stress the importance of that because it helps them… First, it’s like a…. a bit challenging, like they
don’t understand what is it and… what’s the point…? (Extract 21: Interview, teacher 2)

So, he provides advanced texts as examples to his students in order to make them notice how these
aspects are used in ‘good writing’. He mentioned that he employs a genre-based approach and process
writing approach in in his class.

18
Both the teachers observed that time and workload are the most challenging aspects in providing WCF.
Teacher 2 referred to institutional and contextual factors such as convincing the students why they
received a certain grade. Both the teachers pointed out that they have 33 students in their classrooms
and keeping up with deadlines and providing feedback to each student after completing every written
task is time consuming. Moreover, making students understand what the problem is and how to resolve
it is a demanding process. Teacher 1 pointed out that WCF may not be always worth the time invested
by the teacher because the students are too young to process it in an effective and responsible manner.
I am not sure that they read, but I write…I have to say...I am not sure...if they read, I am not sure they
understand what the problems are. (Extract 22: Interview, teacher 1)

It should be noted that the participants differ in their belief about the effectiveness of WCF which is
reflected in their classroom practices. Teacher 1, for instance preferred to talk to her students and provide
oral feedback because she believed that WCF is not an effective strategy with young learners. She
believed that even if she writes comments on student writing, they may not read her comments or if at
all they take the initiative to read her comments, they may not understand the problem and how to resolve
that problem. However, teacher 2 believes that WCF is very effective if it is provided with positive
feedback as well. According to him, WCF is necessary in the Swedish context because even if students
are very fluent in speaking due to exposure to English language, they lack ‘precision’, or in other words,
accuracy in writing.
Swedes tend to have too good self-confidence about their English skills…because they understand a
lot, speak fluently very often. But they are not so precise… (Extract 23: Interview, teacher 2)

During the interview, he mentioned that from his experience he has learnt to adapt according to the
needs of the students and to communicate his philosophies efficiently with his students. For instance,
the teacher mentioned,
So, I have learnt more now…they need to feel that I am there…seeing what they do…they don’t
necessarily need to have my comment all over… (Extract 24: Interview, teacher 2)

While describing his approach to WCF during the interview, he reported that out of his experience, he
has successfully learnt to prepare his students to take responsibility in their learning. Furthermore, both
teachers mentioned that they attend teacher conferences and teacher improvement programs on a regular
basis and hence they were aware and updated about the theories and methodologies in language teaching.
However, as practitioners they have to negotiate with their professional environment to realize their
beliefs and knowledge regarding language teaching.

5. Discussion
The aim of this research is to analyse the philosophy and actual practices of teachers in providing WCF
in their classrooms and the contextual factors behind them. In this chapter, the results of the study are
discussed to address the research questions in relation with background literature. The two research
questions in consideration here are
• What are teachers’ stated beliefs and practices about WCF?
• What are the contextual factors determining teachers’ knowledge, beliefs and practices?

19
The conclusion section which follows, includes the discussions on the strengths and weaknesses of this
study, the didactic implications and future research.

5.1. Teachers’ stated beliefs and practices

about WCF
The first research question in this study sought to investigate teachers’ stated beliefs and practices about
WCF, which is identified as a part of process oriented formative feedback by Hyland and Hyland (2006).
Therefore, teachers’ approaches to teaching writing were analysed in the previous chapter. The results
from this study demonstrate that the teachers recognise the importance of process writing and the need
to provide abundant comprehensible input in language learning as pointed out by Krashen and Truscott
(Krashen, 1987; Truscott, 1996). Using advanced example texts as a guide, the teachers make their
students familiarise and internalize the use of specific vocabulary, transition words, phrases and passive
constructions so that the students can practice them in their own writing (see Extract 1 & 2, Approaches
to teaching writing). However, disparities in beliefs and actual practice were observed during TAP. Even
though the teachers discussed extensively on familiarising the students with different types of texts
during the prewriting sessions, they did not make any comments on the genre of the essay during the
evaluation process. It should be noted that the given student assignment demanded certain formal
features in writing and the genre of the essay was relevant in the context of evaluation.
When it comes to teaching writing, teacher 1 believed that the most important aspect to teach is the
overall structure of the text. She stated that she talks more about structure and cohesion, during pre-
writing instructions (Extract 3, Approaches to writing). However, she contradicted this statement later
in the interview while describing her actual procedures (Extract 5, Approaches to writing). The data
from TAP also suggested that the teacher looked for implicit grammatical aspects such as syntactic
fluency, grammatical cohesive devices and the adaptation of grammatical structures according to the
specific type of writing. Despite the disparity in her beliefs and practice, teacher 1 is referring to implicit
linguistic features in writing by emphasizing the significance of fluency and variation in student texts.
The importance of this practice is identified and discussed in detail by Frodesen (2014) as it helps writers
to ‘expand their repertoire of linguistic choices’ (Frodesen, 2014).
The teachers’ beliefs about teaching writing are reflected in WCF approach as well. Teacher 1 in this
study believed that providing corrective feedback on global errors is beneficial in improving student
writing rather than focusing on local errors. This is one of the recommended feedback practices by Ellis
(Ellis, 2009b). The teachers believed that they provided focused or intensive feedback by commenting
only on recurring errors in student texts. However, the data from TAP and the sample texts showed that
teacher 2 commented extensively on student texts while the teacher 1 left even recurring errors untreated.
Teacher 1 stated that she preferred oral feedback to WCF for a number of reasons. She believed that her
students are too young to process the comments on linguistic form. In other words, the teacher believed
that error correction is not an effective strategy to teach her students grammatical forms as they may not
have the developmental readiness to process her comments (Extract 22, Challenges). This belief is in
alignment with the observations made by Pienemann and Keßler (Pienemann & Keßler, 2012).
According to teacher 1, WCF is time consuming making it difficult to practice each time the students
perform writing tasks (Extract 20, Challenges & Extract 17, Type of feedback). Researchers like Weigle
have identified this as a disadvantage of WCF, that it is time-consuming and open to misinterpretation
by the learners (Weigle, 2014, p. 233). Moreover, teacher 1 believed that written comments can have a

20
negative effect on student motivation as too many comments in the written text can trigger anxiety. This
is one of the objections raised by Truscott in his case against WCF (Truscott, 1996).
Teacher 2 also agreed to this observation and hence he prepares his students to expect more comments
while writing certain types of texts (Extract 17, Type of feedback). However, he believed that WCF as
a communicative practice, is significantly effective in improving accuracy. According to him some
attention to linguistic form is necessary in Swedish context as the students tend to slip in terms of
accuracy in writing despite being fluent in oral production (Extract 23, Challenges). Therefore, teacher
2 believed that it is essential to incorporate WCF into the communicative context by making it a dialogue
between the teacher and the student, using information and communication technologies (Extract 8,
Teachers’ WCF philosophies and practices). This reflects Ferris’ (2003) observation that teacher
commentary should be treated as an ongoing dialogue between the students and the teacher in the larger
classroom context (Ferris, 2003).
Both teachers acknowledge WCF as a means to invite attention to the gaps in students’ knowledge and
language production by providing them information regarding their current writing skills (Extract 8,
Teachers’ WCF philosophies and practices). The teachers’ approach to feedback strategies varied in the
degree of details provided on student texts. It is evident that the teachers’ choice of feedback strategies
was motivated by their beliefs. For instance, teacher 1 located the errors in text without providing
additional information using error codes or metalinguistic clues. She reported that she located the errors
to understand the pattern of errors so that she can provide focused oral feedback on repeated errors.
Consequently, she left many local errors in the student essay untreated, even though she noticed those
errors and thought about them during verbal reporting sessions. Most of her comments were directed
towards the organizational and content problems and she always had the knowledge criteria as reference
while evaluating the text (Extract 9, 12 &13, Type of errors to correct). Teacher 2, on the other hand
had an elaborate approach to WCF as he believed that accuracy is as significant as fluency (Extract 11,
Type of errors to correct). Unlike teacher 1, he employed a range of WCF strategies such as error coding,
metalinguistic clues and electronic feedback to make teacher commentary effective. The difference in
approach can be attributed to their beliefs about the effectiveness of WCF and their preferences
regarding fluency and accuracy which are determined by their personalised and context sensitive
cognitive dimension of teaching as discussed by (Borg 2003).

5.2. Contextual factors determining teachers’

knowledge, beliefs and practices
Several researchers have recognized teachers as active thinking decision makers, constructing and
internalising personalised theories of knowledge to adapt their practice to satisfy instructional needs in
classroom contexts (Borg, 2003; Kumaravadivelu, 1994; Kagan, 1992). The second research question
addressed the role of contextual factors in shaping the cognitions of the participants in this study. Here,
a number of factors influencing the two teachers’ decision-making process in providing WCF are
identified.
As discussed earlier in the background literature, the difference in WCF approach is determined by
factors such as the teachers’ academic identity, experience, curricular demands, the need to justify the
grades, time, setting and the size of the class (Diab, 2005; Tsui, 1996). In this study, both teachers have
made explicit comments on these factors. Teacher 1 reported that even though she believed that process
writing is significantly effective in teaching writing, time constraints and the size of the class makes it

21
difficult for her to practice it, in her class (Extract 20, Challenges). However, teacher 2, who has more
than fifteen years of experience, has modified his approach to overcome this difficulty, by preparing his
students to work intensively without anxiety (Extract 19, Type of feedback).
Teachers’ academic identity plays a role in motivating their instructional choices. Teacher 1, for instance
works with students who attend Social Science programme while teacher 2, belongs to Economics
department. Therefore, teacher 1 focuses more on communicative activities that help students to develop
necessary skills to understand, analyse and express ideas fluently in an organized manner (Extract 11,
type of errors to correct & Extract 17, Type of feedback). On the other hand, teacher 2 focuses on English
for specific purpose as he makes his students write cover letters and job applications on a regular basis.
The need to focus on linguistic structures used to express the form-meaning relationship as discussed
by Frodesen (2014) is significant in this context (Frodesen, 2014, p. 242).
Finally, both teachers mentioned the need to align their instruction and assessment according to the
curricular needs and to justify the grade that they provide for students (Extract 6 & 7 Teachers’ WCF
philosophies and practices). In order to document and report both formative and summative assessments
of each class consisting of 33 students, the teachers have adjusted and adapted their practices according
to the contextual needs mentioned above. Previous research (Diab, 2005; Li & Barnard, 2011; Tsui,
1996), supports these observations made in this study based on the data from interview and TAP.

5.3. Conclusion
In summary, the study confirms that teachers’ beliefs, knowledge and practices are influenced by their
background and contextual factors as recognised by previous research. Among other factors, grading,
curriculum and institutional requirements determined their practice in classroom contexts. The study
observed that there are discrepancies in the beliefs and actual practices of teachers in teaching writing
and providing WCF due to the contextual demands. The findings in this study have pedagogical
implications for practice as it addresses the challenges and frustrations experienced by teachers as
practitioners. In that way, the study takes forward the discussion to help teachers understand, analyse
and overcome such difficulties. It also emphasizes the need for teachers to evaluate and adapt their own
practices to utilise the complete potential of WCF in language teaching.
Most of the challenges associated with this study was due to the difficulty in finding participants in
limited time. Generalizations were made based on the available data and the case study did not intend
to extent its scope to a larger context. For instance, this study has not addressed aspects such as teachers’
previous educational and professional training. Research in teacher cognition emphasizes that teachers’
experience as students is very significant in understanding their beliefs and practices. Therefore, this
study analyses only a limited number of factors shaping teacher’s philosophies and practices. Moreover,
this project has also some limitations from the methodological perspective. This case study involves
only two participants in a restricted setting, primarily because of practical limitations such as time and
availability of participants. Furthermore, limitations in time made it impossible to carry out a pilot study
and hence, the study encountered problems with data collection methods. However, these limitations
were overcome to certain extent by conducting an in-depth analysis of teachers’ practices and beliefs
using multiple data types. Triangulation of data is the strength of this study as the data from different
sources facilitates cross verification and hence ensure validity.
The follow up of this study therefore should pursue a closer examination of the WCF practices using a
broader sample by overcoming the aforementioned constraints. Future research might extend the study

22
over a longer period of time and observe the effectiveness of WCF practices in improving student
writing. It is essential to consider how and to what extent the students benefit from WCF. A diversity in
participants and location, considering the multicultural classrooms in Sweden, can provide valuable
insights into teacher identity and practice.

23
References
Abbuhl, R., Gass, S. M., & Mackey, A. (2012). Interactionist approach. In S. M. Gass, & A. Mackey,
The Routledge Handbook of Second Language Acquisition (pp. 7-23). London: Routledge.
Barnard, R., & Burns, A. (2012). Researching Language Teacher Cognition and Practice :
International Case Studies. Bristol: Multilingual Matters.
Bitchener, J., & Storch, N. (2016). Written Corrective Feedback for L2 Development. Bristol:
Multilingual Matters.
Borg, S. (2003). Teacher Cognition in Language Teaching: A Review of Research on What Language
Teachers Think, Know, Believe, and Do. Language Teaching, 36(2), 81-109.
Borg, S. (2015). Teacher Cognition and Language Education: Research and Practice. London:
Bloomsbury.
Bowles, M. A. (2010). The Think-Aloud Controversy in Second Language Research. New York:
Routledge.
Chandler, J. (2003). The Efficacy of Various Kinds of Error Feedback for Improvement in the
Accuracy and Fluency of L2 Student Writing. Journal of Second Language Writing, 12, 267-
296.
Diab, R. (2005). Teachers' and Students' Beliefs about Responding to ESL Writing: A Case Study.
TESL Canada Journal , 28 -43.
Duff, P. A. (2012). How to Carry Out Case Study Research. In A. Mackey, & S. M. Gass, Research
Methods in Second Language Acquisition : A Practical Guide (pp. 95-116). Chichester:
Wiley-Blackwell.
Ellis, R. (2009a). A Typology of Written Corrective Feedback Types. ELT Journal, 63(2), 97-107.
Ellis, R. (2009b). Corrective Feedback and Teacher Development. L2 Journal, 1(1), 3-18.
Ellis, R., Loewen, S., & Erlam, R. (2006). Implicit and Explicit Corrective Feedback and the
Acquisition of L2 Grammar. Studies in Second Language Acquisition, 28(2), 339-368.
Ferris, D. R. (2003). Response to Student Writing: Implications for Second Language Students.
London: Routledge.
Ferris, D. R. (2004). The “Grammar Correction” Debate in L2 Writing: Where are we, and where do
we go from here? (and what do we do in the meantime …?). Journal of Second Language
Writing, 13(1), 49-62.
Ferris, D. R. (2014). Responding to Student Writing: Teachers' Philosophies and Practices. Assessing
Writing, 19, 6-23.
Friedman, V. (2011). How to Collect and Analyze Qualitative Data. In A. Mackey, & S. M. Gass,
Research Methods in Second Language Acquisition: A Practical Guide (pp. 180-200).
Chichester: Wiley Blackwell.
Frodesen, J. (2014). Grammar in Second Language Writing. In M. Celce-Murcia, D. M. Brinton, & M.
A. Snow, Teaching English as a second or foreign language (pp. 238-253). Boston: Cengage
Learning.
Hattie, J., & Timperley, H. (2007). The Power of Feedback. Review of Educational Research, 77(1),
81-112.
Hyland, K., & Hyland, F. (2006). Feedback on Second Language Students' Writing. Language
Teaching, 39(2), 83-101.
Kagan, D. M. (1992). Implication of Research on Teacher Belief. Educational Psychologist, 27(1), 65-
90.
Krashen, S. D. (1987). Principles and Practice in Second Language Acquisition. Englewood Cliffs,
NJ: Prentice-Hall International.
Kumaravadivelu, B. (1994). The Postmethod Condition: (E)merging Strategies for Second/Foreign
Language. TESOL Quarterly, 28(1), 27-48.
Lee, I. (2014). Feedback in Writing: Issues and Challenges. Assessing Writing, 1-5.
Li, J., & Barnard, R. (2011). Academic Tutors’ Beliefs about and Practices of Giving Feedback on
Students’ Written Assignments: A New Zealand Case Study. Assessing Writing 16, 137-148.

24
Long, M. H. (1988). Focus on Form in Task-Based Language Teaching. University of Hawai’i
Working Papers in ESL, 16, 35-49.
Mackey, A., & Gass, S. M. (2005). Second Language Research: Methodology and Design. New
Jersey: Lawrence Erlbaum Associates.
Mitchell, R., Myles, F., & Marsden, E. (2013). Second Language Learning Theories. Oxon:
Routledge.
Mori, R. (2011). Teacher Cognition in Corrective Feedback in Japan. System, 39(4), 451-467.
Pienemann, M., & Keßler, J.-U. (2012). Processability Theory. In S. M. Gass, & M. A, The Routledge
Handbook of Second Language Acquisition (pp. 228-246). London: Routledge.
Richards, K. (2009). Trends in Qualitative Research in Language Teaching Since 2000. Language
Teaching, 42(2), 147-180.
Skolverket. (2011). Läroplan, examensmål och gymnasie gemensamma ämnen för gymnasieskola
2011. Stockholm. Retrieved from Skolverket: https://www.skolverket.se/om-
skolverket/publikationer/visa-enskild-
publikation?_xurl_=http%3A%2F%2Fwww5.skolverket.se%2Fwtpub%2Fws%2Fskolbok%2
Fwpubext%2Ftrycksak%2FRecord%3Fk%3D2597
Storch, N. (2010). Critical Feedback on Written Corrective Feedback Research. International Journal
of English Studies, 10 (2), 29-46.
Swain, M. (2005). The Output Hypothesis: Theory and Research. In E. Hinkel, Handbook of Research
in Second Language Teaching and Learning (pp. 471-483). Mahwah, NJ: Lawrence Erlbaum
Associates.
Swedish Research Council. (2017). Good Research Practice. Retrieved from Vetenskapsrådet:
https://publikationer.vr.se/en/product/good-research-
practice/?lang=2.235844289.1114807949.1517250881-447017702.1517250881
Truscott, J. (1996). The Case Against Grammar Correction. Language Learning, 4(2), 327-349.
Tsui, A. B. (1996). Learning How to Teach ESL Writing. In D. Freeman, & J. C. Richards, Teacher
Learning in Language Teaching (pp. 97-119). Cambridge: Cambridge University Press.
Weigle, S. C. (2014). Considerations for Teaching Second Language Writing. In M. Celce-Murcia, B.
Donna, & M. A. Snow, Teaching English as a second or foreign language (pp. 230-245).
Boston: Cengage Learning.

25
Appendix 1
Think-Aloud Protocol Informed Consent Form
I hereby acknowledge my free and willing participation in the Think-Aloud protocol session and the
interview designed in connection with this research project.

I understand that _______________________________________ will take notes and record my

interview and my speech as I “think aloud” while reading through the written texts I have agreed to
evaluate, and that she/he will use these recordings only for research purpose.

I understand that I have the right to withdraw from this study or decline to answer any question if I
feel uncomfortable in any way during the TAP session or the interview.

I also understand (1) that no one except __________________________ will listen to my recording,
(2) that the disguised extracts from the TAP- session and the interview may be quoted in the thesis and
(3) that any use of the information contained in the recording will not reveal my name or other
information that could identify me.

Name of Participant: _______________________________________________

Participant’s Signature: ____________________

Date: ________________

Name, contact information and signature of recorder___________________________________

Date __________________

26
Appendix 2
Interview Protocol
Interview Questions

1. Briefly describe your educational background and experience as a teacher

2. Describe your approach to teaching writing in your classroom (Do you provide writing
prompts? How do you prepare students before writing tasks? Do you make students
write multiple drafts? Do you use peer feedback? Do you use teacher-student writing
conferences?)

3. In your opinion, what kind of errors in a written text should be given immediate
attention?

4. Describe your approach to providing feedback to your students.

5. What are the strategies you use to respond to students writing?

6. Do you usually provide the students with the correct linguistic form?
Yes – Why?
No – Then what do you usually do? Do you underline/circle/locate the error in the text? Or
Do you comment on the margin that there is an error?

7. Do you use error codes? Or Do you use numbers in the text to draw attention to the
grammatical description provided in the margin?

8. Do you attempt to correct all the mistakes in the text?

Yes – Why?
No – Do you select specific errors? How and why?

9. Do you use electronic feedback? (hyperlink)

10. How do you respond to multiple drafts of student texts? To which draft do you usually
provide error correction?

11. To what extend do you rely on the rubric/knowledge requirement by Skolverket? Do

you find it difficult to interpret Skolverket’s document?

12. For you what is the most challenging aspect in providing feedback to your students?

13. In your opinion, how effective is error correction in improving student writing?

14. How do contextual factors such as time, work-load, curriculum and institutional
policies affect your approach?

27
 How important it is for you to address the following issues in student texts?

Type of errors Not at all Slightly important Moderately Very important

important important

Mechanical errors

Lexical errors

Grammatical
errors

Organizational
issues

Content issues

28
Appendix 3
Think-Aloud Instructions
Background of the study
This is a research project designed in connection with the course Självständigt arbete för ämneslärare
i språk inom KPu, HT 17 for the partial completion of the teacher degree programme “Kompletterande
pedagogisk utbildning för ämneslärarexamen - humaniora och samhällsvetenskap” at Stockholm
University. I am interested in understanding the stated beliefs and attitudes of teachers in Swedish
schools and this study is designed to analyze how teachers think when they give feedback to their
students. This project aims at analyzing teachers’ stated beliefs and practices regarding WCF in the
Swedish context by observing and recording the WCF practices of teachers in English and their
knowledge and beliefs on WCF using TAP and interviews.
Instructions
You will be asked to correct two randomly selected student texts and ‘think aloud’ while performing
the task by trying to verbalize what you think when you do the correction which will be recorded with
an audio recorder.
• I want you to say out loud everything you think while you correct the student texts. But you
don’t have to explain or justify your thoughts.
• You will be provided with a warm-up task as a trial. I will record that as well. You may ask
for clarifications before we start the actual task
• You can stop the task at any time if you become uncomfortable
• You may ask questions at any point in the process, but I may not answer them
• I will not tell you when you have completed the task; you must determine this on your own
• I am testing the process, not the participant. So, any difficulties are my fault, not yours

29
Stockholm University
SE-106 91 Stockholm
Telephone: +46(0)8 – 16 20 00
www.su.se

30