5. PhD revisited: English writing instruction in Norwegian upper secondary
school1The chapter presents the overall results of a PhD study
(Horverak, 2016) from the University of Agder, focussing specifically
on its practical implications for the teaching of English in Norway. This
is an article-based thesis, with four published articles (Horverak,
2015a, 2015b, 2016a, 2016b). The PhD thesis in its entirety – with
theoretical, methodological and empirical details – can be found
– a linguistic and genre-pedagogical perspective
- Side: 98-117
- DOI: https://doi.org/10.18261/978-82-15-03074-6-2019-06
- Publisert på Idunn: 2019-06-18
- Publisert: 2019-06-18
- Creative Commons (CC BY 4.0)
This chapter reports a PhD study (Horverak, 2016) that investigated English writing instruction practices and the effects of a genre-pedagogical approach defined as scaffolding writing instruction. The results suggest that this approach supports the students to improve their argumentative writing. The chapter discusses various aspects of writing instruction and feedback on writing, and concludes by advocating the need for a coherent model for teaching writing in English teacher programmes.Keywords: scaffolding writing instruction, genre-pedagogy, argumentative writing, feedback
From the early 19th century to the 1960s, second language instruction in Europe was dominated by the grammar-translation method, and grammar instruction of written language was in focus. This changed when new approaches emerged in the 1960s, and later with the audio-lingual method focusing on listening and talking, the natural approach based on Krashen’s ideas about the importance of meaningful input (Krashen, 1988) and the communicative method focused on communicative competence (Hymes, 1972). These approaches focused on learning to speak the target language, rather than producing written texts, and could be said to present a reaction to the neglect of communicative situations and the strong focus on grammar in the grammar-translation method. At the same time, it could be argued that the strong focus on communicative situations led to a neglect of grammatical knowledge needed to produce formal, well-written texts (Lehmann, 1999), an important aspect of learning English. Organising argumentative texts has generally been considered a challenge on various levels in the educational system, both internationally (Andrews, 1995; Beard, 2000) and in Norway (Berge, Evensen, Hertzberg, & Vagle, 2005).
Due to deteriorating PISA results (Kjærnsli, Lie, Olsen, Roe, & Turmo, 2004), there has been an increased focus on basic skills in the curricula in Norwegian schools (Norwegian Directorate for Education and Training [UDIR], 2006), and writing skills is one of these. This need to improve writing skills is what inspired my PhD work, where I investigated writing instruction practices in English and tried out a genre-pedagogical approach to teaching writing. This approach is partly based on systemic functional linguistics (SFL), a theory that focuses on language choices and language in context (Halliday & Matthiessen, 2014). This is relevant according to the purposes of learning English in school; we need to learn English to communicate in various situations (UDIR, 2013a).
The research questions of my PhD study were as follows:
How is English writing instruction carried out in upper secondary schools in Norway?
What effects does applying systemic functional linguistics through a genre-pedagogy approach to teaching writing have on students’ writing skills?
The study investigated how English teachers conducted writing instruction and how feedback was used in the writing process. Furthermore, students’ perceptions of these practices were investigated. Finally, a genre-pedagogical approach to teaching writing was implemented in several groups to investigate whether this could support students in developing their writing skills.
In Halliday’s SFL, grammar is presented as a network of resources that may be applied when receiving or producing text (Halliday & Mathiessen, 2014). The network of possibilities outlined is stratified into 1) context, 2) semantics, and 3) lexicogrammar, with each of the strata providing a set of variables or options for the language user (see Table 5.1).
|Stratification||Instantiation||Explanation and examples|
|1) Context||Register variables||Type of action/topic |
Social relationship between sender and receiver
Role of the language, written or spoken channel
|2) Semantics||Meanings||Representations of experiences of the world|
Interaction, expressing certainty, necessity and intensity
Organising information, and creating coherence
|3) Lexico-grammar||The grammatical system and words|
Table 5.1 shows that we make choices from the resources in the linguistic system according to what type of register is expected in the context. If it is a formal context, we choose a formal language; in a more informal context, we may use more informal words such as “kids” and “teens”. The context comprises the register variables of a) field (type of action/topic), b) tenor (social relationship), and c) mode (the role of the language, written or spoken).
The register variables influence what type of meanings we express on the semantic stratum (Halliday & Matthiessen, 2014). We have choices concerning a) ideational meaning: what representations of experiences in the world to include, b) interpersonal meaning: how to use language to interact, and c) textual meaning: how to organise the information and create coherence. On the lexicogrammatical level, we make choices concerning, for example a) what type of verbs to use to realise the ideational meaning, b) how to express modality to realise the interpersonal meaning, and c) how to use cohesive ties to realise the textual meaning.
The aspects of language described in SFL are relevant in the English subject, which has a focus on writing coherent texts suited to purpose and situation (UDIR, 2013a). It is also relevant within genre-pedagogical approaches as they focus on identifying what is typical of the language and structure of different types of texts. Genres are seen as constructed by stages, a kind of set pattern (Martin, 2012). A stage represents an element in a text, for example a topic sentence introducing a paragraph. Knowing these stages is an important step in learning how to produce texts of various genres. Different teaching-learning cycles have been developed within genre-pedagogical contexts, and my PhD study applied an adjusted version of Feez’s cycle (1999), as elaborated on in Hyland’s (2004) book on Genre and Second Language Writing. In this cycle, there are five steps included; 1) developing the context, 2) modelling and deconstructing the text, 3) joint construction of the text, 4) independent construction of the text, and 5) linking to related texts. This cycle was applied when analysing the data collected to investigate writing instruction practices and when developing a teaching intervention to investigate the effects of a genre-pedagogical approach.
In a review of research about genre as a tool for developing instruction in L1 and L2 contexts, three different traditions are outlined: a) English for specific purposes, or ESP, b) North American new rhetoric studies, and c) Australian systemic functional linguistics, on which this study is based (Hyon, 1996). Hyon (1996) points out that the understandings of genre in the different traditions have different advantages when applied in the classroom. The advantage of the Australian genre-pedagogy tradition and ESP is that these types of approaches provide students with insight into the linguistic features of texts and guidelines for presenting these. The advantage of the North American new rhetoric approach is that it provides a fuller perspective on context and function of genres.
In the tradition of teaching English as a second language, there has been an increased interest in how functional language descriptions may be used as a resource for making meaning (Schleppegrell, 2013). At Georgetown University in the United States, Heidi Byrnes has implemented a theoretical approach based on the concept of genre and SFL to a foreign language writing programme in German. Many studies conducted in this programme have reported the effectiveness of genre-pedagogical approaches in the context of advanced foreign language learners (Byrnes, Crane, Maxim, & Sprang, 2006; Ryshina-Pankova, 2010). Byrnes argues that there is a need for a long-term curricular trajectory for researchers and educators in contexts with advanced language learning, and she suggests that the construct of genre may provide a theoretical foundation for this (Byrnes, 2012).
Much research on how genres are learnt by more advanced L2 learners in college or university contexts is undertaken in ESP contexts and is focused on professional settings (Gimenez, 2008). These studies generally claim that support is needed to meet writing expectations in different genres in working life. Another researcher, Swales, is well known for his work with English in academic and research settings (1990), and he provides a model that is applicable to teaching more advanced compositions and the teaching of English for Academic Purposes, drawing on linguistic and sociolinguistic theory. Many studies on genre-based approaches have focused on how heightened genre-awareness may help learners to contextualise the genre of their writing (Tardy, 2009). These studies argue that when students’ genre-awareness increases, this has a positive impact on their writing ability. This is a central aspect in the type of genre-based approach applied in the current study, which is conducted in a Norwegian context, an under-represented context in second language writing research.
My PhD study included a mixed-method design (Teddlie & Tashakkori, 2009), divided into three phases, as illustrated in Figure 5.1 below. Phase 1 included triangulation of the qualitative approaches of interviews, classroom observations, and collection of teaching material. In Phase 2, the English writing instruction (EWI) questionnaire was developed (Horverak & Haugen, 2016) and distributed to randomly selected upper secondary schools. Phase 3 included a quasi-experiment with a teaching intervention. It is defined as a quasi-experiment as the students were not randomly selected or assigned to conditions, and the study also lacked a control group (Shadish, Cook, & Campbell, 2002).
Using a mixed-methods approach to investigate how English writing instruction is carried out in Norwegian upper secondary schools enabled me to analyse data both at a local level (Phases 1 and 3) and a national level (Phase 2). The qualitative data from Phase 1 of the study, which focused on teachers’ perceptions of writing instruction practices, provided a basis for developing a questionnaire and collecting quantitative data focused on students’ perceptions in Phase 2. The qualitative study also provided a basis for the quasi-experiment investigating how linguistic theory applied through genre-pedagogy may contribute to support students in developing their writing skills in Phase 3.
In Phase 1, the informants were contacted via acquaintances in the southern region of Norway. Nine teachers were interviewed individually and six other teachers participated in a focus group interview, totalling 15 participating teachers. The teachers in this study came from 7 different upper secondary schools in total, and they varied in terms of age, gender, educational background and teaching experience. In Phase 2, the student sample was collected through a systematic sampling procedure (Langdridge & Hagger-Johnson, 2013). Schools were chosen at a fixed interval from a comprehensive list of upper secondary schools in Norway. Fifteen schools participated in the survey, and this resulted in 522 student respondents when respondents with missing values were excluded. In Phase 3, four English teachers with one class each of upper secondary school students participated in the experiment. This resulted in 83 student participants, about 20 from each class. Most of them had Norwegian as a first language (91.6%). Their final grades in written English from lower secondary school ranged from 3 to 6, with the majority getting the grades 4 and 5. All the participating groups were general studies students, which may explain that the participating students generally had rather good grades in English from lower secondary school.
In Phase 1, the interview guide used included questions concerning the structuring of texts, adjusting writing to purpose and situation and the use of various feedback strategies. The material from the interviews was complemented with notes from 13 observations of writing instruction lessons, three of which included different types of feedback situations. In addition, teaching material used to teach the five-paragraph essay and connectors, and material used for teacher, self- and peer assessment was collected.
In Phase 2, I developed and used the English writing instruction questionnaire (EWI), which included two main parts with different categories: 1) EWIT – English writing instruction questionnaire – teaching, and 2) EWIF – English writing instruction questionnaire – feedback. This questionnaire includes items about the teaching of narrative and argumentative writing, items about the students’ self-confidence concerning narrative and argumentative writing, and items concerning various feedback strategies, such as teachers’ follow up of feedback, students’ follow up of feedback and self- and peer assessment. In the first main part, the five categories included were: 1) narrative texts, 2) self-confidence and narrative texts, 3) argumentative texts, 4) self-confidence and argumentative texts, and 5) formality level. In the second part, the four categories included were: 1) teacher’s follow-up of feedback, 2) working to improve, 3) self-assessment, and 4) peer assessment. In addition, some background variables concerning gender, first language, and grades obtained were included, as well as questions concerning what types of texts the students had written and frequency of writing.
In Phase 3, the data consisted of pre- and post-tests. The students were asked to write a text where they discussed American values and social issues in the American society, and include relevant sources attached to the task. There was an overall scoring from 1 (lowest) to 6 (highest) of the tests in accordance with criteria used in examination evaluation guidelines (UDIR, 2013b). Each of the three main categories of structure, language, and content, as well as various items in all categories, were also scored. The evaluation form was divided into three main categories: structure, language and content. Each of these main categories consisted of 6 or 7 items describing what is included in each category. The category “structure” included the items: introduction, paragraph division, topic sentence, coherence of arguments, conclusion and cohesive links such as connectors and pronouns. The category “language” included the items: spelling, grammar, sentence complexity, vocabulary, formality level and modality. The category “content” included the items: appropriateness of the answer in relation to the task, clarity of the topic, relevance, thoroughness, discussion, use of sources and literature list. This evaluation form was based on criteria set in the official censor guidelines (UDIR, 2013b). Three raters evaluated each test, and a one-way random intraclass correlation was computed to check for inter-rater reliability. Furthermore, background variables were mapped; gender, first language, grades, and self-confidence level.
The teaching intervention in the Phase 3 experiment focused on argumentative writing, or essay writing, and was inspired by genre-pedagogy, or more specifically, on the teaching-learning cycle as presented by Hyland (2004). Compared with Hyland’s (2004) cycle, there was more emphasis on teacher instruction than on co-construction with students in the model applied in the intervention. The following steps were included:
Setting the context: The students reflected over what text types exist and purposes of writing.
Modelling, revealing key features of genre: The students were presented with a model text and were instructed in how to construct an essay with introduction, main part and conclusion. The model text was deconstructed by identifying the different stages of the text, e.g. topic sentences, supporting details, and critical reflections.
Writing preparation and practice: The students worked with relevant texts on the topic of social issues and values in the USA, such as rap lyrics and political speeches. They learnt about using sources and referring to sources, and they practised finding relevant quotes and arguments from texts.
Grammar instruction: The students were taught and worked with exercises in the following grammatical topics: a) cohesive ties: conjunctions, subjunctions, conjuncts and pronouns, b) modality, c) formal and informal language, d) vocabulary: work with synonyms to create coherence.
Independent writing with support from feedback: The students were given the pre-test with feedback and worked to improve it.
Comparing to other genres and contexts: The students investigated differences between formal and informal texts.
Based on these steps, teaching material was developed and distributed to the participating teachers, who applied this in their English classes.
In all phases of the study, genre-pedagogy constitutes a framework for the analysis. This means that the analysis was driven by theoretical or analytic interests, defined as a deductive or a theoretical thematic analysis (Braun & Clarke, 2006). In Phase 1, the data material was categorised according to various steps of Feez’s (1999) teaching-learning cycle, as elaborated on in Hyland (2004). I chose this model as a framework for the analysis as it reflects ideas presented in official guidelines relevant for this context. There is a focus on adjusting writing to context in the English subject curriculum, and using feedback in the process of learning is central in the programme Assessment for Learning that has been implemented in Norwegian schools. Hence, the three steps of the teaching-learning cycle that influenced the choice of three main themes in the analysis were 1) setting the context, 2) modelling, revealing key features of genre, and 3) independent writing with support from feedback. The data was analysed, identifying elements relevant for these three themes. The subthemes under each main theme were developed through a coding process carried out in Nvivo.2Nvivo is software widely used within qualitative research to organise and analyse research data.
In Phase 2, the analysis presented the total responses on each score in percentages for all the categories in the survey, which were 1) narrative texts, 2) self-confidence and narrative texts, 3) argumentative texts, 4) self-confidence and argumentative texts, 5) formality level, 6) teacher’s follow-up of feedback, 7) working to improve, 8) self-assessment, and 9) peer assessment. In addition, medians were reported to give indications about central tendencies. Medians were reported instead of means as the data were based on ordinal scales and therefore non-parametric.
In Phase 3, the teaching intervention was applied at the beginning of the first semester for first year upper secondary school students. The students had a pre-test prior to receiving any English teaching on this level, and after the intervention, they had a post-test. There was no control group, so in order to measure improvement, the gain from pre- to post-test was measured. Paired t-tests were used to check whether the students had improved significantly from pre- to post-test. Cohen’s d was calculated to measure the size of the effect of the teaching intervention. In addition, a multiple regression analysis was performed to see whether the background factors, i.e. gender, first language, grades or self-confidence-level may explain some of the variance in the results. The quantitative analysis was complemented by examples from students’ texts that reflected the development from pre- to post-test.
Research reliability and validity
There were certain challenges with the reliability and the validity of the findings in this study. The main limitation in Phases 1 and 2 was that most of the collected data was self-reported, and in Phase 1, there was a limited sample. As the development of the student questionnaire was based on a genre-pedagogical framework and an interview study carried out in a Norwegian context, there were perhaps elements relevant in the context of second language writing instruction that were left out. However, the combination of qualitative and quantitative data strengthened the validity of the findings. Furthermore, there was a large, random sample in Phase 2. The main limitation in Phase 3 was the lack of a control group. Still, the inclusion of both pre- and post-tests made it possible to investigate changes to see if these reflected topics that were included in the teaching intervention.
My PhD study investigated how English writing instruction is carried out in Norwegian upper secondary schools (Phases 1 and 2) and tested whether applying linguistic theory through a genre-pedagogical approach to teaching writing has a positive effect on students’ writing competence (Phase 3). In Phase 1, the data revealed that the teachers generally focused on teaching the students how to write argumentative texts, or 5-paragraph essays, using model texts and focusing on adjusting language and structure to genre and context in line with a genre-pedagogy approach to the teaching of writing. Many of the teachers gave feedback during the writing process, but not all had the capacity to do so, even though they saw this as the ideal. The findings of the survey conducted in Phase 2 supported these findings, and revealed that many students are uncertain about their writing competence. The main findings in Phase 3 was that the students benefitted from a genre-pedagogical approach to teaching writing regardless of gender, first language and level.
Teachers’ writing instruction practices
In Phase 1, I investigated teachers’ perceptions of how writing instruction was carried out, including how they worked with feedback. Data from interviews were triangulated with data from observations and teaching material. I found that the participating teachers generally focused on teaching students how to write argumentative texts, and the students were given the 5-paragraph essay format as a practical template for an argumentative text to prepare them for the exam and for higher education. Model texts were used to demonstrate how argumentative texts are structured. The teachers also included some instruction on using connectors and adjusting language to the formality level required in the context.
There were, however, different opinions about how detailed the instruction should be. The teaching material collected shows that some teachers provided very detailed instructions about the different elements that were to be included in each paragraph by deconstructing example texts. Others were sceptical about supplying too detailed instructions as they feared this would restrict the students’ creativity. As pointed out by one of the teachers: “If you just use a template imitatively, then it is never totally wrong, but it might not be brilliant either”. In opposition to this, another teacher argued: “The problem is that they cannot be creative and good at this if they do not know what they are to relate to […] they have to know the basic structure first”.
In terms of feedback, the teachers reported that the typical issues they commented on were how to structure argumentative texts, how to use sources and give thorough arguments and how to adjust the language to the right formality level. When it came to feedback strategies, many of the informants in this study reported that they used a type of process-orientated strategy where the students received comments on drafts, with a focus on how to adjust the text to genre requirements, both in terms of language and structure. The observations revealed that students worked well with revising when this approach was applied. As one of the teachers argued: “Students become very motivated by receiving help in a process where it is still possible to get better results”. Another teacher reflected: “Once I have started working like this, I don’t see that there is any other way of doing it”. Many teachers also reported that the students were given a grade on a revised version, and those who did not said that they would have done so if they had sufficient capacity. The teachers also used self-assessment strategies of various types, but they expressed that they could develop better practices for this. The teachers were more sceptical about peer assessment as they worried about how students would feel if they were asked to show others their written texts. The findings from the teachers’ interviews and observations were followed up in a student survey.
Students’ perceptions of how writing instruction is carried out
According to the results of the survey, most students had written argumentative texts in English in upper secondary school, but only about half of the group reported having written narrative texts. This was in line with the findings of the teacher interviews. How often students had written exercises as 1) tests at school, 2) home assignments, or 3) homework, varied from several times a month to never. A majority of the students reported a frequency of at least once every semester in all three categories. The main findings based on the data from the first part of the questionnaire, were that a majority of the students expressed uncertainty about whether they had been taught narrative and argumentative writing, and about formality levels of language (Table 5.2).
|Taught formality level||6||5||14||19||26||18||11||5|
Note. As decimal numbers are rounded off, the added sum of the rows does not necessarily equal 100%. NW = Narrative writing, AW = Argumentative writing, “Taught NW” and “Self-confidence NW” include four items each, “Taught AW” and “Self-confidence AW” include six items each, “Taught formality level” includes four items.
The scores were lower on the questions concerning whether they had been taught narrative writing than in the other two categories of whether they had been taught argumentative writing and formality level of language. A majority of the students also expressed little confidence about being able to write both argumentative and narrative texts. However, they were somewhat more confident that they could write argumentative texts. According to the students’ responses, feedback strategies varied in English (Table 5.3).
The majority of students reported that they never or seldom received new evaluations on revised texts, even though a high percentage reported that the teacher made them work on revisions. The majority also answered that they agree that they worked with improving their texts (Table 5.4).
The majority of the students reported that they agreed that they used self-assessment strategies (Table 5.4). Half of the group reported that they had participated in peer assessment. The findings of this study revealed that feedback strategies were not fully exploited in English teaching in upper secondary schools in Norway. This confirms the findings in Phase 1 of this study, which investigated teachers’ practices. In Phase 3 of the study, a teaching intervention was developed based on both genre-pedagogical theory and the findings of the previous phases of this study.
The effects of a genre-pedagogical approach to teaching writing
The teaching intervention in the quasi-experiment included a focus on adjusting language to context, using an appropriate text structure and applying various feedback strategies in the writing process. The main findings in this phase of the study was that the students improved significantly from pre- to post-test in all the three main categories of structure, language and content (Table 5.5). Their paragraphs were better structured, and they used sources better to give thorough argumentation.
|M(SD)||M(SD)||M (SD)||95% CI.||t||E.S|
|Structure||3.13(0.81)||3.74(0.88)||0.61(0.70)*||(0.46 - 0.76)||7.97||.72|
|Language||3.53(0.78)||3.90(0.81)||0.37(0.58)*||(0.24 - 0.50)||5.83||.46|
|Content||2.87(0.74)||3.66(0.86)||0.80(0.70)*||(0.64 - 0.95)||10.41||.99|
|Total||3.10(0.80)||3.75(0.88)||0.65(0.70)*||(0.49 - 0.80)||8.47||.77|
Note. Paired sample t-tests, df = 82, *p< .001 (Two-tailed), effect sizes are calculated as Cohen’s d, scale: 1 – 6.
In the category of language, the students improved most in the use of modal expressions and correct formality level, which were both topics that were focused on in the teaching intervention. Multiple regression analysis revealed that students improved regardless of gender, first language and level (Table 5.6).
|Grade, lower secondary||09(.46)|
Note. First language is coded 1 = Norwegian, 2 = others, including English Grade in lower secondary = grade on written English. β = standardised regression coefficient, *p< .05 (Two-tailed).
In table 5.6, we see that the four variables included in this analysis predicted 12.4% of the variance in the scores (R squared = .124). Hence, the type of genre-pedagogy applied in the current study supported different types of students in improving their writing skills. Students improved regardless of gender, first language and previous grade in English.
In the qualitative analysis in this study, examples from students’ texts illustrated how students improved. One of the students started the conclusion in her pre-test essay with “My opinion about this is that it is terrible!” In the post-test, the same student started the conclusion with “To sum up we see that America has many different social issues they need to work on.” The student changed from an informal, emotional style to a more formal, neutral statement starting with the connector “To sum up”, indicating a conclusion. A good deal of students used connectors signalizing a conclusion, either “To sum up”, or “To conclude”, or connectors signalizing a contradiction, such as “However”. The examples from student texts complemented the findings of the quantitative analysis in this study.
Discussion: Contributions to the english didactics field
Using a mixed-methods approach to investigate how English writing instruction is carried out in Norwegian upper secondary schools enabled me to analyse data both at a local and a national level. The qualitative data from Phase 1 of the study, which focused on teachers’ perceptions of writing instruction practices, provided a basis for developing a questionnaire and collecting quantitative data focused on students’ perceptions in Phase 2. The qualitative study also provided a basis for the quasi-experiment investigating how linguistic theory applied through genre-pedagogy may contribute towards supporting students in developing their writing skills in Phase 3.
In the following, I will discuss the empirical, theoretical and methodological contributions of the PhD study. The main contributions were perhaps in the empirical and methodological domains, as the study presented new insights about English writing instruction practices in a Norwegian context and methodology that was developed to investigate these. The section on theoretical contributions describes how the theories of genre-pedagogy and systemic functional linguistics were applied and adjusted in the context of teaching argumentative writing in English in upper secondary school.
Empirical contributions: new knowledge of writing instruction practices
First and foremost, my PhD study illustrated how SFL applied through genre-pedagogy may provide a useful framework for English writing instruction in L2 contexts. The use of SFL to describe and support language development has been important in language learning research in recent decades (Byrnes et al., 2006; Christie, 2012; M. Schleppegrell, 2004; M. J. Schleppegrell, 2013). The teaching intervention in my study did not apply the full framework for language studies offered in the SFL tradition, but the focus on modal verbs, formality of language and cohesive links was influenced by SFL. This was included after the deconstruction of text in step two of the teaching learning cycle. In this study, I argued that there is a potential in using meta-language in writing instruction, and applying strategies from the teaching-learning cycle developed in the genre-pedagogy tradition.
The second contribution of my PhD study was knowledge of students’ perception of writing instruction practices and their own writing skills. The students perceived argumentative writing to be more in focus than other types of writing in English. Like the teachers, the students reported differently concerning what practices were applied. Some students agreed that feedback strategies such as teacher, peer and self-assessment were applied during the writing process, which is in line with a genre-pedagogy approach to the teaching of writing (Martin, 2012; Hyland, 2004), whereas some said that they were not. Most of the students expressed an uncertain attitude as to whether they could write argumentative and narrative texts.
Third, my PhD study found evidence that teachers used model texts and writing frames when teaching writing, and that there was a focus on how to structure argumentative texts or five-paragraph essays. Further, the findings revealed that there was a focus on what elements to include in a text in each paragraph, as in the staging approach developed within genre-pedagogy (Martin, 2012). Some teachers also included a focus on how to use connectors to create coherence, how to adjust language to the correct formality level, and how to express modality. These are linguistic elements that are relevant in Halliday’s systemic functional linguistics (Halliday & Matthiessen, 2014). The study also revealed that feedback practices varied, e.g. that some teachers used feedback strategies in the writing process before a final evaluation was given, and some did not. Implementing feedback as part of a writing instruction process is an important element of the teaching-learning cycles developed in the genre-pedagogy tradition (Martin, 2012; Hyland, 2004).
Theoretical contributions: a scaffolding writing instruction approach
A theoretical contribution of my PhD study concerned the application of a linguistic and genre-pedagogical framework to English writing instruction in a Norwegian educational context. I illustrated how the teaching-learning cycle developed within the genre-pedagogy tradition includes elements that are applicable to the Norwegian school system. In line with Byrnes et al. (2006), I argued that the emphasis on the meaning-making properties of language in SFL, e.g. the focus on cohesive links and adjusting language to purpose, makes this theory particularly useful in contexts of language teaching.
Another theoretical contribution of my PhD study was the adjusted model for teaching argumentative writing outlined in the quasi-experimental study, which I defined as scaffolding writing instruction. Compared with Feez’s (1999) teaching-learning cycle as elaborated on by Hyland (2004), the third step, “joint construction”, was replaced by the two steps “writing preparation and practice” and “grammar instruction”. This step included pre-writing exercises such as studying relevant source texts to include in the writing, instruction in how to use and refer to sources, and grammar instruction. I would argue that the teaching-learning cycle model as developed in this study presented a theoretical contribution to linguistics as it applied linguistic theory in practice, as the grammar instruction was based on SFL. It also presented a theoretical contribution to didactics as it applied an adjusted teaching-learning cycle model for teaching argumentative writing.
Methodological contributions and a return to reliability and validity issues
One of the methodological contributions of this study was to present a mixed-methods model for investigating teachers’ and students’ perceptions of writing instruction practices. The first two phases of this study comprised a multilevel mixed design (Teddlie & Tashakkori, 2009), with qualitative data among the teachers and quantitative data among the students. When complementing the interview study with the survey investigating writing instruction practices from another perspective, there was some consensus between the findings of the qualitative study and the quantitative study. Both studies showed that there is an emphasis on argumentative writing in upper secondary school, and both showed that some teachers give feedback to students during the writing process while others do not.
A second methodological contribution of this study was the development of the English writing instruction (EWI) questionnaire. This questionnaire was developed through a validation process where the survey was tested in several English groups. A factor analysis was conducted on the results to confirm the categories in the survey, and Cronbach’s alpha was calculated to check for internal validity (Horverak & Haugen, 2016). This questionnaire can be adapted and applied in other contexts for writing instruction, and is therefore a methodological contribution.
Implications for teaching English
In my PhD study, I urged central educational authorities to consider the need to prioritise writing in English in schools. With the establishment of the National Writing Centre in Norway there has been increased focus on writing in Norwegian schools, but there has not been much focus on writing in English. There is a need for resources to further educate current English teachers in strategies for writing instruction and to give room for teachers to apply teaching strategies that have been proven to be useful, such as formative assessment strategies. There is also a need to include a more coherent model for writing instruction in the English teacher education, and this study suggests implementing the genre-pedagogical approach scaffolding writing instruction presented here in English teacher training programmes and in further education of English teachers.
Still, I think it is important that teachers keep an open attitude to creativity and to writing texts in different ways. We do not want students and teachers that follow templates rigidly without engaging in what they write. I believe that if teachers have a high competence in how to teach writing, it is possible to combine a scaffolding writing instruction approach with an open attitude to students’ individual voices and creativity. If teachers provide students with a variety of model texts, so they can see that there are different ways of solving a writing exercise, this may facilitate creativity and prevent rigidity in English writing instruction.
Suggestions for future research
There is a need to follow up the findings of this study in future research, for example to investigate writing instruction practices from a teacher’s perspective on a larger scale, and to investigate more in-depth how students perceive these practices. Furthermore, there is a need for research on how English writing instruction is carried out at lower levels, as this is scarce in a Norwegian context – for example what type of texts are written in lower secondary school to prepare for further education, and what type of strategies for teaching writing are applied. There is also a need to investigate what feedback practices are applied in lower secondary school, and what prevents teachers from applying more varied feedback strategies in the teaching practices.
Future studies should also investigate whether SFL applied through genre-pedagogy may result in significant improvement when compared with other approaches. In this study there was no control group and the only type of students that participated were first year general studies students. The effects of this type of approach in other contexts also need to be studied. Finally, to establish the efficiency of a genre-pedagogy approach in the teaching of writing based on SFL, there is a need to investigate how students exposed to this type of treatment improve compared to students exposed to other types of treatments.
|1||The chapter presents the overall results of a PhD study (Horverak, 2016) from the University of Agder, focussing specifically on its practical implications for the teaching of English in Norway. This is an article-based thesis, with four published articles (Horverak, 2015a, 2015b, 2016a, 2016b). The PhD thesis in its entirety – with theoretical, methodological and empirical details – can be found here http://hdl.handle.net/11250/2458188|
|2||Nvivo is software widely used within qualitative research to organise and analyse research data.|