Society for Educational Studies

Language learning strategies across the curriculum: government policy and school practice.

Vee Harris

Paper presented at the British Educational Research Association Annual Conference, University of Warwick, 6-9 September 2006

Address for Correspondence:

Vee Harris

Department for Educational Studies

GoldsmithsCollege

Lewisham Way

London SE14 6NW

v.

DRAFT ONLY: PLEASE DO NOT QUOTE WITHOUT PERMISSION

Abstract

This paper reports on one aspect of a project funded under the auspices of the Denis Lawton award by the Society for Educational Studies. Recent government concerns to raise pupils' academic achievement have resulted in a plethora of publications for Key Stage 3. These initiatives offer an opportunity to address the balkanised departmental structure typical of secondary schools. One of the aims of the project, carried out with students aged 12-13 years in two London schools, was to explorethe impact on ML performance and motivation ofencouraging students to transfer language learning strategies across English and ML. A further aim wasto investigate the issues raised in the necessary cross-curricular collaboration. Whilst the quantitative data suggests that the project had a positive impact on pupils’ ML performance and motivation, its success in encouraging the transferability of strategies across English and Modern Languages was more limited. Field notes and interviews with teachers and students provide one possible explanation, highlighting the gap between the government’s ‘learning to learn’ agenda and the realities of implementing it.A bewildering array of recommendations hides genuine differences in the aims and objectives in the teaching of the two languages.These can only be resolved by providing teachers with the time to work through them together and by trusting their professional judgement to make the necessary adjustments to the curriculum: both are in short supply in the current educational climate.

Introduction

Since it came to office in 1997, a keystone to New Labour education policy in England has been a determination to raise students' academic achievementwith

improved literacy seen as the lynchpin to success.A year later,the National Literacy Strategy in primary schools was introduced (DfEE 1998). Inspite of considerable debate (Bousted 2001, Hilton 2001) it was extended into Key Stage 3 (the 11-14 age group) through the Key Stage 3 Framework for Teaching English(DfES 2001a)and the Language Across the Curriculum (DfES 2001b)initiative. The Key Stage 3 National Strategy now includes similar frameworks of guidance for other curriculum subjects, setting out the skills and understandings to be taught to pupils over the three year period.

Recalling the Bullock report (1975),Literacy across the Curriculum (2003a) stresses that any literacy learning pupils undertake in their English lessons should be consolidated in other subject lessons and that there should be a common terminology as well as a common pedagogical approach across the curriculum.Following an Ofsted report (2004) that progress in some schools towards developing literacy across the curriculum remained slow and limited, a further publication, Literacy and Learning(2004)was issued, allocatingparticular priorities to be addressed by each subject.

A common theme underlying the various publications is that the teaching of grammar is a key element ofsuccess, with lists of grammatical objectives being a major feature of both the Framework for Teaching English(DfES 2001a) and theFramework for Teaching ML (DfES 2003b).Theheightened emphasis on explicit grammar teaching is likely to increase the distrust between English and ML teachers, since previous studies (Mitchell, Hooper and Brumfit 1994; Pomphrey and Moger 1999) already indicate that one of the factors impeding cross-curricular collaboration is the divergent views they hold on the role of knowledge about grammar.

Alongside the grammatical objectives common to both KS3 Frameworks, there is an alternative context for cross-subject dialogue; one which is not reliant on rote learning of the rules governing a particular language but rather directed towards teaching studentshow to learn any language; on the process rather than the product. Table 1provides two examples of basic language learning strategies common to the two Frameworks (Harris and Grenfell2004).

Table 1: Common language learning strategies

English Framework
Pupils should be taught to: / ML Framework
Pupils should be taught to:

Memorisation

Identify words which pose a particular challenge and learn them by using mnemonics, multi-sensory
re-inforcement and by memorising critical features (p.23) / How to find and memorise the spelling, sound, meaning and main attributes of words (p.45)
Reading
Work out the meaning of unknown words using context, etymology, morphology, compound patterns and other qualities such as onomatopoeia
(p.23) / How to read and understand simple texts using cues in language, layout and context to aid understanding (p.51)

As yet, however, it is not clear if students or even their teachers understand this common ground. Nor is it clear what factors would facilitate or impede the necessary cross-curricular collaboration.

As early as 1994, Whitty et al warned that given prescribed subject-based objectives, teachers may be reluctant to find the time to address additional cross-curricular aims. The plethora of government policies in the last decade is only likely to have exacerbated the situation. Gereluk (2005, p. 8) notes that:

Collaboration requires time and effort amongst staff and a demanding curricular framework may overwhelm an already overworked teacher… The inflexibility of the curriculum may create a situation whereby teachers do not have time to collaborate or see the need to collaborate when every detail has been laid out.

The increasing importance attached to performance measures is a further constraint, teachers feeling a: ‘narrowing of focus in their work’(Gewirtz 2002, p.80) andobliged to 'teach to the tests', skewing pupils' integrated knowledge about language (Frater 2000). Finally,the emphasis inmany secondary schools on individual teaching in closed classrooms limits the opportunities for teachers to work with others outside

their subject departments (Hodkinson and Hodkinson 2005).

This paper explores the issues in relation to collaboration between ML and English teachers in the context of a project designed toinvestigate the impact of strategy instruction (SI) directed towards the transferability of learning strategies across the two languages. It does not discuss the findings from the perspective ofcognitive psychology- this will be the subject of a separate paper. Rather the intention is that, although only dealing with two subjects in the curriculum, the lessons learned may be of value in revealing the complexity ofcollaboration across other subject areas.

Language Learning Strategies

The last three decades have seen a considerable growth of interest into how learners process the information and skills involved in learning language. Learning strategies are commonly defined as the skills, tactics and approaches which learners adopt in tackling their language learning. This research has taken place into both L1 learning (mother tongue)and L2learning (second learning learning; either where the learners are living in the host country or where learners' exposure to the new language is in the classroom).

For L1 learning, the roots of strategy research lay in the 1960s and 70s in the development of cognitive psychology through the work of Bruner and others (Bruner et al. 1966; Flavell 1976). This perspective provoked a number of studies contrasting the learning strategies used by 'experts' with those used by 'novices' (see for example Palinscar and Brown's 1984 study on reading). Research into L2 learning also used as its starting point the notion of the so-called ‘Good Language Learner’ (Naiman et al 1978; Stern 1975) and listed the strategies which successful learners adopt. Subsequent studies consistently found that high attaining language learners have a wider repertoire of language learning strategies and use them more frequently than their less successful peers (O'Malley and Chamot 1990).Although the focus of much of the early L2 strategy research was on adult learners of English, more recent studies (Harris et al 2001; Macaro 2001;Grenfell and Harris 1999), have explored the strategies used by UK secondary school pupils to learn a modern language, revealing similar findings.

A major outcome of the research has been the debate over whether learners should be explicitly taught ‘how to learn’. Within L1 contexts, studies suggest that SI results in more effective learning and school achievement (Pressley et al 1995; Adey and Shayer 1994). In L2 learning contexts, McDonough’s review (1999) suggests similarly promising results, although the studies are mainly concerned with older learners learning English in the USA.

Harris and Grenfell (2004) highlight the similarities of L2SI to the pedagogical principles underlying the Key Stage 3 Strategy outlined by Harrison(2002);for example both encourage pupils to 'activate their prior knowledge' and to 'reflect' on the learning process. Both also highlight the importance of extensive practice through collaborative pair and group work so that students learn to apply their knowledge to new learning contexts.

Since the Key Stage 3 National Strategy encourages students to reflect on their language learningin both English and ML classes, and there are learning strategies common to both, then it would be a wasted opportunity not to facilitate the transference of new understandings across the two arenas. Furthermore if it is essential to provide extensive practice in the deployment of strategies, it seems probable that having two contexts rather than one in which to use these approaches can only be beneficial.

No studies have been undertaken with British secondary school students learning a ML to examine the impact of SI directed towards the transferability of learning strategies. Similarly, whilst a number of L2 studies have suggested thedifficulties ML teachers face in implementing learning strategy instruction (Chamot and Keatley 2003; Macaro 2001; Grenfell and Harris 1999; O’Malley and Chamot 1990) none have charted the issues raised when English and ML teachers embark collaboratively on the SI enterprise.

Methodology

Project participants and school context

The investigation was quasi experimental in design using intact class groups. Two schools were involved, referred to as school A and school B. School A is a large, multi-ethnic, mixed,11-18 comprehensive school in a working class area of South East London.School B is a small, mixed, 11-16 school, serving a population of mixed socio-economic status in a London suburb.In each school, two parallel Year 8 classes of 30 studentslearning French (a control and an experimental class) were selected,yielding a total sample of 120students.Although there was a risk of contamination by having the control and experimental class in the same school, it had the advantage of maximising parity of socio-economic background. The experimental classes were taught by the two ML teachers working on the project.The control classes were taught by another experienced teacher in each ML department.Both the control and the experimental classes followed the ML scheme of work in each school but whereas the experimental class was exposed to explicit SI during their French lessons, the control class was not. For reasons outlined in Harris (forthcoming) and Harris and Prescott (2005),the skill areas selected for the SI were reading and listening strategies.Initial lessons focused on familiarising students with the strategies within the ML context. Ten of the subsequentML lessons were specifically designed to encourage students to make the links to strategies they used or could use in English.

A major constraint of the project was that funding did not allow for the English teachers to be released from their timetables to work alongside their ML colleagues, who werereleased for between 3-6 hours a week. Hence although explicit reference to the transferability of the strategies was included in some English lessons, this was done on ad hoc basis. Time did not permit the necessary discussion and joint planning that would have allowed the lessons to run concurrently, mutually reinforcing each strategy taught. However, following the success of the main phase of the project, School A agreed to fund the English teacher to be released for one hour a week over the course of the following year.The intention was to improve the opportunities for cross-curricular collaboration. Unfortunately the English teacher secured a new post after one term and there was no suitable replacement. Although brief, the opportunity in the final phase for a more focused, systematic and complementary implementation of the SI provides a clearer understanding of the factors helping or hindering cross-curricular collaboration.

Research methods

A range of quantitative and qualitative research methods were used; those relevant to this paper are described below.

Quantitative

• Pre and post intervention reading and listening tests in Frenchwere conducted to examine the impact of SI on ML performance.

• A pre and post intervention attitude questionnaire was completed by both the experimental and the control classes. Scored on a 5 point Likert scale from 1 (‘strongly disagree’) to 5 (‘strongly agree’), it included two questions designed to investigate students’ perceptions of the transferability of strategies across ML and English: ‘learning French is different from all the other subjects in school’ and‘I learned English OK so there is no reason why I should not learn another language’. Although open to interpretation, the intention was that by the end of the project, students in the experimental classes would be more likely than those in the control classes to see the similarities between learning ML and learning other subjects in the curriculum and to feel confident in their ability to acquire a new language.

• A short ‘extra’ attitude post-intervention questionnaire was completed by the experimental classes. This asked, for example, whether the SI had helped them to see the links between the two languages and if they preferred ML activities which made the links to English or those focusing just on French.

Results from the questionnaires were subjected to statistical analysis using SPSS version 12XX.

Qualitative

24 students of varying attainment levels in the experimental and control classes were selected as case studies. The selection wasbased on a combination of their ML test score the preceding year and theirCognitive Abilities Test (CAT)verbal score, since the latter is considered the most reliable predictor of ML GCSE performance (NferNelson 2003). Pre and post intervention semi-structured interviews were conducted based on a card prompted task devised for the pilot study (Harris 2004) to discover the extent to which students transferred strategies across English and ML. Students were presented with 16 cards, each of which had a learning strategy written on it and a picture indicating the skill area. Their task was to assign each card to one of four brightly coloured plastic containers labelled:

• I only use it for learning English;
• I only use it for learning ML;
• I use it for learning any language (whether English or ML);
• I do not use it.

The interviews were tape-recorded and transcribed. Analysis employed a ‘grounded theory’ approach; extensive re-reading of the transcripts of both experimental and control classes; categorization; identification of significant similarities or differences in shifts in students’ perception of the transferability of strategies and of underlying themes.

Prior to the start of the project, semi-structured interviews with two English teachers were conducted in School A (the larger school) and with the English teacher, who is also the KS3 co-ordinator, in school B. The card prompted task used with the students served the same purpose with the teachers. Additional questions were designed to explore possible issues raised in the implementation of the strategy instruction. Field notes recorded further informal conversations with the English teachers that took place over the course of the project. Semi-structured interviews with two English PGCE tutors at my own institution were conducted, again using the card format, at the start of the year along with informal discussions subsequently. A semi-structured interview took place with the English teacher involved in the final phase of the project just before she left the school, with a view to sharing her perceptions of the difficulties and benefits of the more systematic approach to cross-curricular SI. Interviews with the teachers and tutors were tape-recorded, transcribed and analysed with a view to identifying the most salient points.

Although a review of government documentation was carried out at the beginning of the project, the final phase, where the ML and English teacher were both released, led to further scrutiny. Detailed field notes of half of the lessons in the two schools were completed either by myself or theteachers involved in the project. 7 ML and two English lessons were video-recorded.

Findings

The main phase of the project

Both the quantitative and the qualitative data suggest that whilst the project had a positive impact on pupils’ ML performance and motivation (Harris forthcoming; Harris 2006), its success was more limited in terms of encouraging the transferability of strategies across English and Modern Languages. One possible explanation appears to lie in the divergence of aims and objectives between the teaching of the two languages;a gap exacerbated by government policies and the lack of adequate release time for the necessary cross-curricular dialogue between the teachers.