Assessment & Evaluation in Higher Education Vol. 31, No. 1, February 2006, pp. 71–90 Complex skills and academic writing: a review of evidence about the types of learning required to meet core assessment criteria James Elandera*, Katherine Harringtonb, Lin Nortonc, Hannah Robinsonc and Pete Reddyd aThames Valley University, UK; bLondon Metropolitan University, UK; cLiverpool Hope University College, UK; dAston University, Birmingham, UK. Psychology & JamesElander Evaluation 0 100000February Ltd in Higher Education 31 Taylor and (print)/1469-297X (online) 2005 &Article 2006 Original Francis 0260-2938 DepartmentLondon 10.1080/02602930500262379 Metropolitan UniversityCalcutta House, Old Castle StreetLondonE1 7NTUKj.firstname.lastname@example.org CAEH_A_126220.sgm AssessmentFrancis Assessment criteria are increasingly incorporated into teaching, making it important to clarify the pedagogic status of the qualities to which they refer. We reviewed theory and evidence about the extent to which four core criteria for student writing—critical thinking, use of language, structuring, and argument—refer to the outcomes of three types of learning: generic skills learning, a deep approach to learning, and complex learning. The analysis showed that all four of the core criteria describe to some extent properties of text resulting from using skills, but none qualify fully as descriptions of the outcomes of applying generic skills. Most also describe certain aspects of the outcomes of taking a deep approach to learning. Critical thinking and argument correspond most closely to the outcomes of complex learning. At lower levels of performance, use of language and structuring describe the outcomes of applying transferable skills. At higher levels of performance, they describe the outcomes of taking a deep approach to learning. We propose that the type of learn- ing required to meet the core criteria is most usefully and accurately conceptualized as the learning of complex skills, and that this provides a conceptual framework for maximizing the benefits of using assessment criteria as part of teaching. Introduction Assessment criteria, sometimes called marking criteria, play an increasingly impor- tant role in higher education. One reason for this is the priority attached to improv- ing the reliability and validity of marking. Another is the trend towards transparency and explicitness in all aspects of student assessment. Potentially the most significant *Corresponding author. Psychology Group, Thames Valley University, St Mary’s Road, Ealing, London W5 5RF, UK. Email: email@example.com ISSN 0260-2938 (print)/ISSN 1469-297X (online)/06/010071–20 © 2006 Taylor & Francis DOI: 10.1080/02602930500262379 72 J. Elander et al. reason, however, is the increasing use of assessment criteria in teaching, usually in the form of marking exercises where students themselves actively engage with the criteria. Essays and related written work provide opportunities for students to demonstrate some of the most demanding learning outcomes, but students are often more confused about what constitutes a good essay than they are about the criteria for other types of assignment. One survey of first year students taking a study skills module, for example, showed that essay writing was the most commonly requested topic for advice and guidance (Elander, 2003a). Confusion among students is understandable considering that professional academics also often struggle to specify what constitutes a good essay; Sadler (1989), for example, identified over 50 published criteria for written composition. Studies reported as many as 10 (Branthwaite et al., 1980) or 18 criteria (Norton, 1990) that tutors considered important, and a recent report listed 12 potential criteria for essays (Andrews, 2003). Across disciplines and institutions, however, there are a few core criteria that have a central role in the shared perception of what is important in good student writing. Four such core criteria, identified from an analysis of published assessment criteria in psychology, business studies and geog- raphy, are critical thinking, use of language, structuring, and argument (Elander et al., 2004). Table 1 gives examples of each of those core criteria. The purpose of using assessment criteria in teaching has been to improve students’ understanding of the criteria, thereby improving their performance in assessments. The rationale has usually been to provide opportunities for students to develop tacit rather than explicit knowledge about the meaning of the criteria and how they are applied. This, it is argued, should lead students to develop understandings of assess- ment that correspond more closely to those of their tutors, whose ability to assess students’ work is usually the result of shared practical experience rather than internal- isation of explicit rules or standards. ‘It follows that inviting students into the shared experience of marking and moderating should also enable more effective knowledge transfer of assessment processes and standards’ (Rust et al., 2003, p. 152). The case for this type of teaching was made previously by Sadler (1989) in the context of formative assessment. ‘Providing guided but direct and authentic evaluative experi- ence for students enables them to develop their evaluative knowledge, thereby bring- ing them within the guild of people who are able to determine quality using multiple criteria’ (p. 135). One such initiative was part of a first year geography module in which students discussed the criteria and undertook marking exercises. Student feedback was gener- ally positive but there was no attempt to estimate the effects on student performance (Pain & Mowl, 1996). In another, marking exercises using assessment criteria were included in a first year psychology skills module, along with other study skills exer- cises including IT, note-taking and library use. Student feedback was generally positive, but there was no clear improvement in student achievement (Elander, 2003a). In the largest, most systematic, and most rigorously evaluated programme of this kind, workshops with marking exercises improved student performance in a first year business module, with signs that the benefits might even endure over time and Core assessment criteria 73 Table 1. Core assessment criteria for student writing Core criterion Example Critical thinking/critical evaluation Does the author present material in a critical manner? (Pain & Mowl, 1996). Clear application of theory through critical analysis/critical thought of the topic area (O’Donovan et al., 2000). Evaluation includes conceptual/ methodological critique and an appreciation of alternative perspectives and current controversies (Elander, 2002). Use of language/writing style Is it generally clear, readable and well presented? Does it make the reader want to read it? Correct use of spelling and grammar? (Pain & Mowl, 1996). Language fluent, grammar and spelling accurate (Price & Rust, 1999). Material and arguments presented clearly and coherently (Elander, 2002). Structuring Does the essay have a clear, logical and well-defined structure? (e.g. is there an introduction, middle and conclusion?) (Pain & Mowl, 1996). Good essay structure; sections obvious (Oates, 2002). Clear structure, material organised well (Elander, 2002). Does the conclusion draw together the various important points made in the main body of the essay? (Pain & Mowl, 1996). Extremely well organised answers whose structure reflects the development of argument (Elander, 2003b). Argument Does the author sustain a well-reasoned and supported argument? (Pain & Mowl, 1996). Logical argument clearly present throughout (Oates, 2002). Good development shown in summary of arguments based in theory/literature (O’Donovan et al., 2000). Note: For uniformity’s sake most examples are criteria for satisfactory or good performance. transfer to some extent to other modules (Rust et al., 2003). A self-assessment initia- tive for Masters level students was based on a similar model of extending communi- ties of practice about assessment to include students (Elwood & Klenowski, 2002). Initiatives like those described above are likely to continue to be developed in the future and their success will depend on the type of learning that is identified as required to demonstrate the criteria concerned. Note that assessment criteria them- selves refer to qualities of the outcomes of student learning: they are ‘properties of the text’, consistent with the definition of an assessment criterion as ‘a distinguishing property or characteristic of any thing, by which its quality can be judged or esti- mated, or by which a decision or classification can be made’ (Sadler, 1987, p. 194). Types of learning, on the other hand, are the activities in which students engage as they go about producing the work to which the assessment criteria will be applied. 74 J. Elander et al. Confusion between the two is easy because the same forms of words are often employed in descriptions of the activities of students and the properties of the work they produce. For example, a student may engage in critical thinking, and subse- quently produce an essay that may (or may not) have certain textual properties that the marker recognizes as a demonstration of the outcome of critical thinking. The same goes for use of language, structuring, and argument, which can all be used to refer to activities undertaken by students and the properties of the work they submit for assessment. Notwithstanding the distinction between the meanings of terms that refer to prop- erties of the text (assessment criteria) and those that refer to activities of students (types of learning), it is important to know about the types of learning that can most helpfully be associated with core assessment criteria. That is, although there is no identity between assessment criteria and types of learning, any given criterion may be a property of text that is more likely to result from one type of learning than another. Certain types of learning may therefore be more likely than others to increase students’ ability to produce work that demonstrates the relevant criteria, which has important implications for the way assessment criteria are used in teaching. For example, certain assessment criteria may describe properties of work that are the result of students learning skills that will transfer from one context to another. Or they may be properties of work that are the result of a deep approach to learning that reflects engagement with a specific area of knowledge. Or they may be properties of work that are the result of complex learning, where subject knowledge is combined with personal qualities and social practices. Each of these possibilities has different implications for how to achieve the best results from using assessment criteria in teaching. Assessment criteria that describe properties of work resulting from learning skills could contribute to changing the way skills are regarded in higher education. The debate about skills in higher education has included critical analysis of skills provision in universities (Kemp & Seagraves, 1995), analysis of the conceptual problems inher- ent in the notion of transferability (Hinchliffe, 2002), and analysis of the problem of separating ‘thinking skills’ from subject knowledge (Bridges, 1993). An analysis of the extent to which transferable skills can be appropriately assessed by core assessment criteria for student writing would inform those debates, and would also be informa- tive about the extent to which teaching based on assessment criteria should be skills- oriented. Assessment criteria that describe properties of work resulting from a deep approach to learning would have different implications, for whereas skills are amenable to train- ing, deep approaches to learning are associated with motivational factors and active student engagement in the discipline. Since a deep approach to learning is desirable, and since assessment criteria codify desirable qualities of students’ work, it is perhaps natural to assume that the criteria represent the expected outcomes of a deep approach to learning. This is especially true for essays, which are often assumed to be the ideal form of assessment of the outcomes of taking a deep approach to learning. One study showed that students were more likely to adopt a deep approach for essay assignments than for multiple-choice examinations, and that adopting a deep rather Core assessment criteria 75 than a surface approach was associated with better essays (Scouller, 1998). However, the extent to which a deep approach to learning can be promoted by teaching innovations is uncertain, and interventions using assessment criteria might lead instead to students merely simulating the outcomes of a deep approach. Deep approaches to learning were increased by an intervention among very able and highly motivated students (Biggs & Rihn, 1984), but interventions in other student groups have resulted in increases in surface, or instrumental, approaches to learning (e.g., Maguire et al., 2001). Complex learning has been described as the ‘construction of new knowledge’ (King, 1997) and is a recent alternative conceptualization of desirable outcomes of education, especially in relation to employability. Complex learning aims at the ‘inte- gration of knowledge, skills and attitudes; the coordination of qualitatively different constituent skills; and the transfer of what is learnt to daily life or work settings’ (van Merrienboer et al., 2003, p. 5). It is distinct from learning generic skills and taking deep approaches to learning in that academic learning is associated with psychologi- cal characteristics similar to those underpinning ‘graduate identity’ (Holmes, 2001) and ‘social practices’ (Knight & Yorke, 2003). Analysis of the ways and extent to which assessment criteria represent the properties of work resulting from complex learning could help to promote employability through teaching innovations linked directly to subject learning, in line with the increasingly close links between pedagogy and employability (Tynjälä et al., 2003). For example, identifying possible links between the qualities represented in core assessment criteria and complex learning could help develop interventions to improve students’ meta-awareness of what assess- ment criteria represent, and promote more autonomous approaches to learning that may be more likely to transfer to other contexts, including employment settings. Generic skills, deep approaches to learning, and complex learning therefore repre- sent broad categories or types of learning that differ in a number of ways. Table 2 summarizes the distinctions between them, as discussed above, in terms of whether they can be separated from subject knowledge, whether they can be improved by specific training (training in study methods, for example, rather than the training provided by higher education more generally), and whether they are related to personal development and employability. This is a very approximate classification that is based partly on evidence from research, but also on the ways that the terms Table 2. Features of three types of learning Type of learning Generic skills Deep learning Complex learning Separable from subject knowledge Yes No No Improvable by specific training Yes Maybe Maybe Related to personal development No Maybe Yes Related to employability Maybe Maybe Yes 76 J. Elander et al. ‘skills’, ‘deep approaches to learning’ and ‘complex learning’ have been defined, and on the assumptions that are often made about them. Identifying the type of learning that is most likely to lead to the successful demon- stration of core assessment criteria would provide a first step in the more systematic organization of teaching based on assessment criteria, and this is the main aim of the present paper. Taking each criterion in turn, we consider how they have been concep- tualized, the extent to which the qualities they represent are capable of being sepa- rated from subject knowledge, the extent to which the activities required to demonstrate them can be promoted by specific and non-specific training, and the range of benefits that have been obtained by such training. The analysis is based on a comprehensive review of theory, practice, and research evidence identified by elec- tronic searches in the science, social science, and arts and humanities databases for English language reports published since 1981, using search terms from the following sets: (1) assessment, criteria, marking, marking scheme, grade descriptor, level descriptor; (2) critical thinking, critical evaluation, critical analysis, academic writing, essay writing, writing style, academic voice, academic language, academic English, register, structure, argument; and (3) skills, generic skills, transferable skills, key skills, deep learning, approaches to learning, learning styles, complex learning, social practices, employability, learning objectives, learning aims, learning outcomes. Reports were also identified from the bibliographies of reports found by the electronic searches. Critical thinking/critical evaluation Garside (1996) reviewed definitions of critical thinking and concluded that it is usually defined in terms of a skill component and an attitude component. McPeck (1981, p. 8), for example, suggested that ‘the core meaning of critical thinking is the propensity and skill to engage in an activity with reflective scepticism.’ Four defining features of critical thinking have been suggested: (a) clear, precise, accurate, relevant, logical and consistent thinking; (b) a controlled sense of scepticism or disbelief about claims, assertions and conclusions; (c) taking stock of existing information and iden- tifying holes and weaknesses; and (d) freedom from bias and prejudice (Garside, 1996). Many give critical thinking a central role in learning (e.g., McPeck, 1981; Beyer, 1987), and suggest that essays are well suited to the assessment of critical thought: ‘the answer format for a critical thinking test should permit more than one justifiable answer and good answers should not be predicated on being right, in the sense of true, but on the quality of the justification given for a response’ (Tynjälä, 1998, p. 175). Let us first consider the skills elements of critical thinking, which was often included in the lists of skills that proliferated during the 1980s and 1990s, usually as a thinking or cognitive skill. It was rated eighth in importance out of 20 ‘transferable employment skills’ by graduates who were asked about the importance of skills in their current jobs (Smith et al., 1989). Beyer (1985) identified 10 specific critical thinking skills: (1) distinguishing between verifiable facts and value claims, (2) Core assessment criteria 77 determining the reliability of a source, (3) determining the factual accuracy of a state- ment, (4) distinguishing relevant from irrelevant information, (5) detecting bias, (6) identifying unstated assumptions, (7) identifying ambiguous or equivocal claims or arguments, (8) recognising logical inconsistencies or fallacies in a line of reasoning, (9) distinguishing between warranted and unwarranted claims, and (10) determining the strength of an argument. Defining very specific, discrete critical thinking skills should open the door to skills-based interventions to promote critical thinking. One programme, for example, focused on questioning and showed that students asked more questions, and especially more critically evaluative questions, after training in questioning (Keeley et al., 1998). Garside argued that ‘critical thinking involves a set of skills that are most effectively taught within the context of a subject area. Since it is impossible to think critically about something of which one knows nothing, critical thinking is dependent on a sufficient base of knowledge’ (Garside, 1996, p. 215). That view is supported by eval- uations of interventions to promote critical thinking, which have been more success- ful when integrated in subject teaching than when delivered through study skills programmes. In one qualitative evaluation of nursing education, a study skills programme had some benefits for the development of critical thinking, but more significant gains were obtained by relating critical thinking to students’ clinical prac- tice and academic learning (Girot, 1995). Another study of nursing education showed that discipline-specific teaching rather than more general academic ability had a bigger influence on the development of critical thinking (Miller, 1992). In biol- ogy, critical thinking instruction and subject content were successfully integrated, with improvements in both reasoning skills and content knowledge from pre-course to post-course tests (Chapman, 2001). Recommendations about specific classroom activities and teaching methods to promote critical thinking include active student participation, meaningful interaction, and opportunities for students to challenge and question (Garside, 1996). Psycholog- ical research indicates that cooperative learning should be more effective than competitive learning (Johnson et al., 1981), but in a direct comparison between students taught in lectures and those taught in group discussions, there was no differ- ence in outcome (Garside, 1996). However, critical thinking was measured with a multiple-choice test, which may not be the most appropriate measure. Let us turn now to the attitude or propensity element of critical thinking, for there is evidence that critical thinking is related to learning styles. For example, one study showed that critical thinking measured with the Watson-Glaser test was correlated with a deep processing approach to learning (Gadzella & Masten, 1998). There is also evidence that critical thinking is influenced by non-academic as well as academic expe- riences, making critical thinking resemble complex learning, in which personal char- acteristics influence outcomes. Terenzini et al. (1995), for example, found that both instructional and out of class experiences made unique contributions to gains in critical thinking, over and above pre-college levels of critical thinking and other characteristics. Critical thinking therefore involves elements that resemble skills and can be improved by specific training, but these are not generic skills, and they can be 78 J. Elander et al. improved more effectively by integration with subject teaching. However, critical thinking is not just a matter of knowing more about one’s discipline, for it is affected by learning styles and out of class experiences, suggesting a complex learning process. Bailin et al. (1999b) described critical thinking as comprising a set of intellectual resources including knowledge of concepts, standards and procedures, plus certain habits of mind. Perhaps most importantly, critical thinking includes the application of standards and criteria: ‘critical thinking is not promoted simply through the repe- tition of “skills” of thinking, but rather by developing the relevant knowledge, commitments and strategies and, above all, by coming to understand what criteria and standards are relevant’ (Bailin et al., 1999a, p. 280). Sadler (1989) argued that the key skill required for students to improve their academic performance was evalu- ation: ‘For an important class of learning outcomes, the instructional system must make explicit provision for students themselves to acquire evaluative expertise. It is argued that providing direct and authentic evaluative experience is a necessary (instrumental) condition for the development of evaluative expertise’ (p. 143). Of all the core criteria therefore, critical evaluation has a special status in the context of improving student performance, and familiarising students with the criteria that are applied to their own work, and providing opportunities for them to apply those crite- ria themselves, may be an especially effective method to promote critical and evalua- tive thinking more generally. Use of language Many universities offer generic courses on academic writing, and student guides often treat writing as a generic skill that can be developed independently of what is being written about. In the analysis below we identify three levels of writing: correctness, register, and academic literacy. Correctness involves grammar, punctuation, spelling, ˇ and referencing, and Ivanic (2004) associated the ‘skills discourse’ of writing with [n a or c] assessment practices that focus mainly on accuracy. This is consistent with a view of writing as a teachable generic skill that will transfer to almost every setting where writ- ing is used. Register is much more a matter of writing style, including the length and complex- ity of sentences, the way in which values are expressed, the ways in which writers can distance themselves from other people’s terms, and the use of active and passive sentences. ‘Our choice of register when we write displays our attitude towards our reader and towards the subject matter we are writing about. … One of the main char- acteristics of the register appropriate for academic writing is that it does not resemble the register of conventional speech’ (Fabb & Durant, 1993, pp. 72, 74–75). Though not divorced from acts of understanding and knowledge creation, the concept of register emphasises the role of conventions in the construction of knowledge and ˇ resembles to some extent the ‘genre discourse’ of writing (Ivanic , 2004). [ o a rn c] Biber et al. (2002) described a multidimensional analysis of ‘register variation’ in the written and spoken language to which students were exposed, and exemplars from published academic writing can be used to identify elements of the academic register Core assessment criteria 79 that students might be encouraged to adopt. Studies of published academic writing have focused on sentence length and structure (Harnon, 1992), use of audience engagement by addressing readers directly in the text (Hyland, 2001), use of antici- patory (it) clauses (Hewings & Hewings, 2002), directives (Hyland, 2002), and inclu- sion in the text of a voice intended to be attributable to the reader (Thompson, 2001). Studies of student writing have also helped to understand the significance of register. For example, one of the factors that differentiated students at different levels was assertiveness versus cautiousness in writing, which included confidence and the feel- ing of being in control when writing and discussing essay plans with tutors and other students, as opposed to being pessimistic, unenterprising and externally constrained (Branthwaite et al., 1980). Cautiousness is nevertheless an important part of academic writing because of the need to balance different views and avoid absolute judgements, leading to constructions such as ‘it can be suggested that’ or ‘it could be argued that’. In one analysis, for example, ‘frequent use of a “bold” or “very bold” style in students’ essays by no means guaranteed a high grade’ (Francis et al., 2002, p. 174). Academic literacy subsumes all the other assessment criteria for essays and proceeds from the assumption that understanding and creating knowledge in a disci- pline takes place through language, not independently from it, so that writing is part of the formulation as well as the presentation of thoughts (Lea & Street, 1998; Warren, 2003). ‘The student has to learn to speak our language, to speak as we do, to try on the peculiar ways of knowing, selecting, evaluating, reporting, concluding and arguing that define the discourse of the community’ (Bartholomae, 1985, p. 134). Lavelle (1993) identified five student writing processes that span the three levels of writing we have identified. The ‘low self-efficacy’ process was characterized by low confidence in writing abilities and little concern for surface aspects of composition, grammar and punctuation. The ‘spontaneous-impulsive’ process involved an off-the- cuff impromptu approach to writing. The ‘procedural’ process was a methodological approach aiming to meet requirements. The ‘elaborationist’ process involved personal engagement in writing. The ‘reflective-revisionist’ process involved active reworking of writing and the emergence of meaning. Low self-efficacy, procedural, and spontaneous-impulsive writing processes were associated with surface approaches to learning, and reflective-revisionist and elaborative writing processes were associated with deep approaches to learning (Lavelle & Zuercher, 2001). However, a deep approach to learning does not appear to be the only type of learning involved in using language well in essays. Adherence to grammatical and other conventions is probably rightly treated as a transferable skill, and at least some aspects of register are often treated as skills: ‘… academic register is a convention you learn to adopt so that your essays “sound right”’ (Fabb & Durant, 1993, p. 74). Whitehead (2002) concluded that the development of an academic writing style is a skill that students must be willing to learn. There is also evidence that workshops focusing on writing can be effective, at least in changing the writing processes adopted by students. For example, workshops for graduate students led to reductions in 80 J. Elander et al. procedural and spontaneous-impulsive processes and increases in elaborationist processes (Biggs et al., 1999). At lower levels of ability and performance, therefore, where the emphasis is mainly on correctness, use of language is a criterion for judging the outcome of using generic skills, and can be differentiated quite clearly from other core criteria. At higher levels, however, where the emphasis is on register and academic literacy, it subsumes other core criteria, and describes the outcomes of a complex process in which subject learning is central, qualifying therefore as a criterion for judging properties of work that results from taking a deep approach to learning. Structuring Peck and Coyle (1999) distinguished between form-driven and content-driven approaches to writing. They associated form-driven approaches with aspects of struc- turing that resemble generic skills, and offered advice about structuring techniques that can be learned, such as ‘how to build an essay, including how to shape a para- graph and a sentence’: ‘One of the most useful rules in writing an essay … is the rule of three, which is that the core of an essay should be divided into the three stages of setting the issue up, pushing the issue along, and then seeing where we arrive. This kind of division allows one to start to impose a shape on the raw material of the essay. … this same structure … will work for virtually any essay set on any subject at univer- sity’ (pp. 97–100). That advice would address many of the structuring criteria shown in Table 1, espe- cially those proposed by Pain and Mowl (1996), but structure goes beyond generic skills, as demonstrated by the advice offered by authors with content-driven approaches to student writing. These approaches highlight the dependence of struc- ture on content, consistent with research showing that higher grade student essays had structures that were more closely integrated with content (Prosser & Webb, 1994). For example, Pirie (1985) has argued that ‘each paragraph must be recogniz- able as a logical next step in a coherently developing argument that directly answers the set question … the most effective order will almost always emerge through thought about the particular problems which have occurred to you during your research on each essay’s specific topic’ (p. 58). Similarly, Creme and Lea (1997) proposed that: ‘By “structure” we mean both the way a piece of writing is organised and—more importantly—what work it is doing: its function in the assignment. We are particularly interested in how the structure constructs the relationships between different ideas’ (pp. 84–88). The content-driven perspective aligns structuring with a deep approach to learning, consistent with the rationale for the SOLO taxonomy (Structure of Observed Learn- ing Outcomes), which is often used to assess the structural complexity of essays (Biggs & Collis, 1982; see also Atherton, 2005). ‘SOLO … is based on the assump- tion that the learning quality is reflected in the level of complexity with which the learning outcome is structured, regardless of whether the item learnt is a skill, a concept, or a problem’ (Biggs, 1988, p. 197). That assumption is supported by stud- ies showing that more complex essay structures are related to deeper approaches to Core assessment criteria 81 learning (e.g., Biggs, 1987). Campbell et al. (1998) found that essays with more complex conceptual structures were more likely to have been written by students who reported active, reconstructive note-taking, and who reported building arguments rather than building information in their essay construction. Not all of the evidence is positive, however: Lavelle (1997) found no significant relationships between the structural complexity of student essays and measures of students’ composition processes. However, even content-driven approaches acknowledge that there are common or repeated elements or principles to structuring essays: ‘whatever piece of academic writing you are attempting, whatever subject or course you are doing, you will be putting together all the components into a structured whole’ (Creme & Lea, 1997, p. 39). Structuring an essay can be thought of as a skill, therefore, in that it will improve with practice, but its relationship with content means that it should be treated as a complex rather than a generic skill, and is probably best taught in disci- pline-based settings. The SOLO taxonomy can be used to help with teaching as well as assessment of student writing (e.g., Boulton-Lewis, 1995), and there is evidence that students who have been taught in ways that combine subject knowledge with complex skills produce essays with more complex structures. Students taught using writing tasks in which they had to transform knowledge by applying it and criticizing it, for example, produced essays with more complex structures than those who attended lectures on the same topics (Tynjälä, 1998). There are therefore aspects of structuring that qualify as generic skills, and struc- turing can be improved by specific training. At higher levels of performance, however, structuring is more closely linked to content and has been shown to be related to deep approaches to learning. This does not preclude certain complex transferable skills playing a role. One study examined essay structuring in relation to students’ under- standing of the assessment criteria, and found that students with a better understand- ing of ‘organisation’, ‘synthesis’, and ‘critical evaluation’ produced essays with more complex structures (Campbell et al., 1998). This suggests that instruction based on core assessment criteria could help improve students’ ability to structure their writing. Argument Argument is arguably the defining feature of the essay: ‘Your essay is your argument; everything else makes sense because of it’ (Bonnett, 2001, pp. 50–51). What exactly constitutes a good argument is something of a moot point, however, and the concept of argument may contribute significantly to the ‘connoisseur’ model of student assessment: ‘I can recognize a good piece of student writing when I see it. I know when it is well structured and has a well-developed argument but it is difficult to say exactly what I am looking for, let alone describe a good argument more fully’ (a lecturer quoted in Creme & Lea, 1997, pp. 36–37). Students may also recognize the importance of argument but have difficulty in articulating what arguments consist of. In one study where students were interviewed about essay writing, 44% mentioned the importance of presenting their own views or 82 J. Elander et al. opinions, making this the most frequently mentioned factor (Read et al., 2001). Argu- ment has to go further than presenting one’s own view, of course. Branthwaite et al. (1980) found that students were much more likely than lecturers to emphasize the need for ‘original’ thought in essays, and students who believed in presenting their own opinions obtained lower grades for their essays than those who did not. A common view is that argument is the antithesis of a transferable skill, being rather the process by which personal and academic fulfilment is achieved: ‘Argument is not simply a “transferable skill”. It isn’t something to be ranked alongside computer literacy and time management. The ability to argue … is the core attribute of all forms of advanced level education. But it’s more than that too. For argument goes to the heart of who we are and what we want to do with our lives’ (Bonnett, 2001, p. 3). Some descriptions of argument resemble complex learning in that they relate academic knowledge to personal transformation and enablement: ‘The ability to engage in argument is what makes learning exciting. To feel comfortable with debate changes your relationship with education and just about everything else. It transforms you from a passive and bored receptacle of another’s wisdom into a participant; into someone who is neither scared by, nor indifferent to, the society around them but actively involved in its interpretation and transformation’ (Bonnett, 2001, p. 1). More analytical approaches to understanding argument are possible, however: ‘an argument can be divided into several components of which the main ones are the claim and the grounds. The claim can either be presented in the form of a conclusion, or it may (or may not) include a conclusion. As both the claim and the conclusion indicate the opinion of the writer, the purpose of the grounds is to provide evidence for them’ (Marttunen & Laurinen, 2001, p. 139). Having defined argument in that more systematic way, it is possible to identify skills associated with each element. For example, analytical skills have been associated with identifying the components of an argument; evaluative skills have been associated with determining whether the claims are valid, whether the grounds support the claim, and whether the conclusions are balanced; and constructive skills have been associated with presenting the arguments (Marttunen, 1992). Several factors have also been associated with the ability to argue, including intelligence (Perkins, 1985), age (McCann, 1989), and level of education (King et al., 1990). The identification of specific skills, or elements, involved in argument should open the door to interventions to improve students’ ability to develop arguments in their essays, and there is some evidence about what form those interventions should take. Perhaps the first question is the extent to which argument can be separated from subject discipline. Andrews (1997) suggested that there are generic elements of argu- ment at undergraduate level that can be separated from discipline-specific elements. Regardless of the discipline, all academic arguing involves negotiating a new position or defending an existing one in relation to others. Argument is, ‘an arrangement of linguistic, visual and/or physical propositions in engagement with one or more other points of reference in order to change or assert a position’ (Andrews, 1997, p. 267). Mitchell and Riddle (2000) developed a generic undergraduate argument module, Core assessment criteria 83 and suggested that ‘quality of argument needs to find a place on institutional agendas in the same way that, for instance, key skills have done’ (p. 13). Andrews (1997, p. 266) also stressed, however, that disciplinary constraints play a major part in shaping the nature of argument. Different ways of seeing and arguing pertain in different disciplinary fields. One limitation of generic aspects of argument is that they tend to be expressed in very abstract, academic language, as some of the definitions given in this section illustrate, which is not useful for helping students understand what is involved. Also, in generic argument teaching, the topics and source materials must refer to general knowledge because students do not have a shared body of disciplinary knowledge: ‘recourse to a general topic is limited in that it bypasses the problem of how to help students improve within specialised disciplinary fields whose arguments deploy particular theoretical perspectives, types of evidence and authority … we felt that the greatest benefit to students would come from guidance and practice within their subject disciplines’ (Mitchell & Riddle, 2000, p. 18). There is some evidence that argument can be improved within disciplines. In one study, education students produced written texts on topics encountered in the course books and lectures, then used email to express their opinions, present the grounds for those opinions, criticize the views of other students, and defend their own opinions if criticized by others. Those students performed better than a control group when the commentaries they wrote on specimen arguments were assessed for argumentation skills (Marttunen, 1992). In another study, the argument and counter-argument in the email messages of education students in tutor-led and student-led email discus- sion groups were assessed. The level of argument improved over time and was higher in messages that included counter-arguments targeted at others’ standpoints. Students in the student-led groups produced more and better counter-argument than those in tutor-led groups (Marttunen, 1997). A third study of this type suggested that different learning environments may promote improvements in different aspects of argument: students who took an argumentation course delivered face-to-face improved at putting forward counter-arguments, whereas those who took the same course delivered by email improved at identifying and judging grounds. The overall impact of the course was not impressive, however, and outcomes were not demon- strably better than in a control group who were not taught to argue (Marttunen & Laurinen, 2001). In another study, essays written by education students who had attended constructionist, writing-to-learn discussion groups were compared with those written by students attending lectures on the same topics. The essays were anal- ysed by looking at the numbers of sentences belonging to epistemic categories, which showed that the constructionist group had developed more elements of argument, with fewer descriptions and more generalizations, classifications and comparisons (Tynjälä, 1998). Argument therefore appears on first consideration to be a vague concept that is related very much to personal development and self-efficacy, so that it could be treated as a form of complex learning. On closer analysis, however, it comprises a number of more specific complex skills that can be improved by training in the context of subject learning. 84 J. Elander et al. Discussion Skills are involved to some extent in the demonstration of all the criteria we examined: the activities required improve with experience, can be applied to a range of material, and have been shown to respond to specific instruction. However, none of the criteria describe the outcomes of truly generic or transferable skills, for in every case the skills involved can be separated only in a limited way from subject knowledge. For example, use of language and structuring are criteria for judging outcomes that require trans- ferable skills only at lower levels of performance; at higher levels, meeting the criteria requires a deep approach to learning because integration with subject knowledge is needed. Critical thinking is a criterion for judging outcomes of learning that are more separable from subject knowledge, and the activities involved resemble complex learning because they are related to personal development, but demonstrating critical thinking also involves a range of complex cognitive skills. The activities involved in meeting the criteria for argument also resemble complex learning, especially when the nature of argument is considered superficially, but argument, like critical thinking, may most usefully be construed as a criterion for judging the demonstration of complex cognitive skills. Each type of learning we have considered has different implications for the use of assessment criteria in teaching situations. For example, treating the criteria as descriptions of the outcomes of using generic skills would imply working with the criteria one by one, away from the disciplinary context, and targeting novice and poorly performing students. Treating the criteria as descriptions of the outcomes of taking a deep approach to learning would imply closer links with subject knowledge, more recognition of the inter-relationships among core criteria, and delivery to students across a wider range of ability and experience. Treating them as descriptions of the outcomes of complex learning would imply linkage with personal development and performance outside the academic setting. In fact, most initiatives to use assess- ment criteria in teaching have been discipline-based and delivered to first year students (Pain & Mowl, 1996; Elander, 2003a; Rust et al., 2003). Our analysis suggests that core assessment criteria for essays could underpin a wider range of learning support, including interventions for more experienced and advanced students. We propose that the type of learning required to demonstrate the core criteria for written work is the learning of ‘complex skills’, and that the concept of complex skills can help to inform the design and delivery of teaching interventions that use assess- ment criteria. We propose complex skills because the abilities required to demon- strate core criteria cannot be considered in isolation from one another, are closely intertwined with subject knowledge, and, as they develop, help to promote a deep approach to learning and personal attributes like self-efficacy that are associated with complex learning. The reason for identifying core criteria with the manifestations of skills is that they describe the outcomes of things that students do in the process of understanding and producing knowledge. They are descriptions of abilities that are needed in order to understand and produce knowledge, and those abilities can be Core assessment criteria 85 acquired, practised, refined, and even go rusty without use. The reason for identifying the relevant skills as ‘complex’ rather than ‘generic’ is that they embody disciplinary practices of knowledge creation, and that understanding and acquiring each skill involves understanding and acquiring all of the others. Complex means firstly, ‘consisting of or comprehending various parts united or connected together’, and secondly, ‘complicated, involved, intricate; not easily anal- ysed or disentangled’ (Simpson & Weiner, 1989). Core assessment criteria for essays are complex in the sense that they are connected together and not easily analysed or disentangled. Assessment criteria tend to overlap conceptually with one another, and, ‘many are operationally correlated together, so that whenever an attempt is made to change a piece of writing according to one dimension, other properties are inevitably affected at the same time’ (Sadler, 1989, p. 131). In an essay, therefore, the qualities referred to in the assessment criteria work together to produce the kind of work expected. Analytically, the skills required may be identifiable as separate, but under- standing, using, and learning each skill happens in relation to understanding, using, and learning all of the others. This is not to say that specific criteria cannot provide an individual focus in teaching sessions. However, not only does an understanding of each criterion come into focus only through an understanding of all the others, but also, learning to employ each of the skills required to demonstrate the criteria is facil- itated by learning to employ all of the others. The skills required to meet core assessment criteria are complex also in the way they are related to subject knowledge. At the lower end of the performance range, criteria like using language and structuring can be specified, and the skills and abilities required to meet those criteria can be promoted, with little reference to a specific discipline. As those skills become more highly developed, however, they become much more closely linked with the content of the writing, and more specific to disci- plinary knowledge. The concept of complex skills therefore puts the knowledge back into skills. The concept is similar to academic literacy, which was defined as, ‘the complex of linguistic, conceptual and skills resources for analysing, constructing and communicating knowledge in the subject area’ (Warren, 2003, p. 109), and which also makes very strong links between subject knowledge and writing skills. Complex skills and academic literacy are both concerned with the construction of disciplinary knowledge, but complex skills can be more explicit about the nature of the skills required, especially when those skills can be mapped onto assessment criteria. The concept of complex skills is consistent with the development of study skills interventions that emphasize integration with disciplinary content rather than trans- ferability. Hattie et al. (1996), for example, concluded from a meta-analysis that study skills interventions should take place in context, using tasks from the target domain, and aim to promote learner activity and metacognitive awareness. Complex skills are also consistent with the concept of ‘cognitive apprenticeships’ (Collins et al., 1989). Hattie et al. (1996) even classified study skills interventions according to the SOLO taxonomy. Unistructural programmes involved direct teaching of mostly mnemonic devices, whereas relational programmes were delivered in relation to particular content and were used for ‘near transfer’. Interventions using assessment criteria and 86 J. Elander et al. adopting a complex skills approach would be expected to qualify as relational programmes in Hattie et al.’s taxonomy. One possible risk associated with raising the profile of assessment criteria, and especially with characterizing the criteria as describing the successful execution of skills, is that what will be promoted is strategic or instrumental approaches rather than deep approaches to learning. In a study of university students’ goals, those with ‘performance goals’ and ‘multiple goals’ took the evaluation criteria more into account when deciding what learning strategies to use, compared with those with ‘learning goals’ (Valle et al., 2003). It is not surprising that assessment criteria should be related to performance goals, but performance and learning goals are not necessarily mutually exclusive. One of the aims of using assessment criteria to support learning should be to extend the benefits of understanding assessment criteria to students with learning goals, and to encourage those with performance goals to use the assessment criteria in ways that facilitate learning. Norton (2004) proposed that assessment criteria be operationalized as ‘learning criteria’, shifting the emphasis from outcome to process in a way that is intended to promote a deep approach to learning and avoid encouraging students to adopt a strategic approach. In conclusion, assessment criteria have provided the focus for innovative efforts to support students’ learning and achievement, which are likely to be further developed in the future. The impact of those interventions will depend on educators arriving at a clearer understanding in pedagogic terms of the meaning of assessment criteria and the types of learning in which students must engage to produce written work that meets the criteria. Our analysis suggests that the type of learning required to demon- strate core criteria for essays and other academic writing is the learning of complex skills that are transferable from one task to another within disciplines and are amena- ble to improvement with practice and instruction. This provides a pedagogic frame- work for future initiatives to promote student learning and achievement by using assessment criteria in teaching. The concept of complex skills also enables teaching about assessment criteria to support subject learning and knowledge acquisition, by integrating the activities required to meet the criteria with those required to learn the subject matter of the discipline concerned. Acknowledgements We are grateful to a referee for helpful comments on a previous draft. Work on the article was supported by HEFCE funding through an FDTL4 grant for Assess- ment Plus, a consortium project to develop methods and materials to support student learning using assessment criteria. For details, see http://www.assessment- plus.net. Notes on contributors James Elander is Professor of Psychology at Thames Valley University. Katherine Harrington is Director of the Centre for Scientific Literacy at London Metropolitan University. Core assessment criteria 87 Lin Norton is Professor of Pedagogical Research at Liverpool Hope University College. Hannah Robinson was a Research Assistant at Liverpool Hope University College at the time this work was conducted. Pete Reddy is a Learning and Teaching Fellow at Aston University. References Allen, R. (Ed.) (2002) The Penguin English dictionary (London, Penguin). Andrews, R. (1997) Reconceiving argument, Educational Review, 49(3), 259–269. Andrews, R. (2003) The end of the essay? Teaching in Higher Education, 8(1), 117–128. Atherton, J. S. (2005) Learning and teaching: solo taxonomy [On-line]: UK. Available online at: http://www.learning and teaching.info/learning/solo.htm (accessed 26 September 2005). Bailin, S., Case, R., Coombs, J. R. & Daniels, L. B. (1999a) Common misconceptions of critical thinking, Journal of Curriculum Studies, 31, 269–283. Bailin, S., Case R., Coombs, J. R. & Daniels, L. B. (1999b) Conceptualising critical thinking, Journal of Curriculum Studies, 31, 285–302. Bartholomae, D. (1985) Inventing the university, in: M. Rose (Ed.) When a writer can’t write (New York and London, Guilford). Beyer, B. (1985) Critical thinking: what is it? Social Education, 49, 270–276. Beyer, B. (1987) Practical strategies for the teaching of thinking (Boston, MA, Allyn and Bacon). Biber, D., Conrad, S., Reppen, R., Byrd, P. & Helt, M. (2002) Speaking and writing in the univer- sity: a multidimensional comparison, TESOL Quarterly, 36(1), 9–48. Biggs, J. (1987) Process and outcome in essay writing, Research and Development in Higher Educa- tion, 9, 114–125. Biggs, J. (1988) Approaches to learning and to essay writing, in: R. R. Schmeck (Ed.) Learning strategies and learning styles (New York and London, Plenum Press). Biggs, J. & Collis, K. (1982) Evaluating the quality of learning: the SOLO taxonomy (New York, Academic Press). Biggs, J., Lai, P., Tang, C. & Lavelle, E. (1999) Teaching writing to ESL graduate students: a model and an illustration, British Journal of Educational Psychology, 69, 293–306. Biggs, J. B. & Rihn, B. (1984) The effects of intervention on deep and surface learning approaches to learning, in: J. Kirby (Ed.) Cognitive strategies and educational performance (New York, Academic Press). Bonnett, A. (2001) How to argue: a student’s guide (Harlow, Pearson Education). Boulton-Lewis, G. M. (1995) The SOLO taxonomy as a means of shaping and assessing learning in higher education, Higher Education Research and Development, 14(2), 143–154. Branthwaite, A., Trueman, M. & Hartley, J. (1980) Writing essays: the actions and strategies of students, in: J. Hartley (Ed.) The psychology of written communication (London, Nichols and Cogan Page). Bridges, D. (1993) Transferable skills—a philosophical perspective, Studies in Higher Education, 18, 43–51. Campbell, J., Smith, D. & Brooker, R. (1998) From conception to performance: how undergradu- ate students conceptualise and construct essays, Higher Education, 36, 449–469. Chapman, B. S. (2001) Emphasising concepts and reasoning skills in introductory college molecu- lar cell biology, International Journal of Science Education, 23, 1157–1176. Collins, A., Brown, J. S. & Newman, S. E. (1989) Cognitive apprenticeship: teaching the craft of reading, writing, and mathematics, in: L. B. Resnick (Ed.) Knowing, learning and instruction: essays in honour of Robert Glaser (Hillsdale, NJ, Lawrence Erlbaum Associates), 453–493. Creme, P. & Lea, M. R. (1997) Writing at university: a guide for students (Buckingham, Open University Press). 88 J. Elander et al. Elander, J. (2002) Developing aspect-specific assessment criteria for essays and examination answers in psychology, Psychology Teaching Review, 10, 31–51. Elander, J. (2003a) A discipline-based undergraduate skills module, Psychology Learning and Teaching, 3, 48–55. Elander, J. (2003b) The BPS Stage 1 qualification in health psychology, Health Psychology Update, 12(4), 40–48. Elander, J., Harrington, K., Norton, L., Robinson, H., Reddy, P. & Stevens, D. (2004) Core assessment criteria for student writing and their implications for supporting student learning, in: C. Rust (Ed.) Improving student learning 11. Theory, research and scholarship (Oxford, Oxford Centre for Staff and Learning Development), 200–212. Elwood, J. & Klenowski, V. (2002) Creating communities of shared practice: the challenges of assessment use in learning and teaching, Assessment and Evaluation in Higher Education, 27, 243–256. Fabb, N. & Durant, A, (1993) How to write essays, dissertations & theses in literary studies (London, Longman). Francis, B., Robson, J. & Read, B. (2002) Gendered patterns of writing and degree award, in: G. Howie & A. Tauchert (Eds) Gender, teaching and research in higher education (Aldershot, Ashgate). Gadzella, B. M. & Masten, W. G. (1998) Relation between measures of critical thinking and learning styles, Psychological Reports, 83, 1248–1250. Garside, C. (1996) Look who’s talking: a comparison of lecture and group discussion teaching strategies in developing critical thinking skills, Communication Education, 45, 212–227. Girot, E. A. (1995) Preparing the practitioner for advanced study—the development of critical thinking, Journal of Advanced Nursing, 21, 387–394. Harnon, J. E. (1992) An analysis of 50 citation superstars from the scientific literature, Journal of Technical Writing and Communication, 22(1), 17–37. Hattie, J., Biggs, J. & Purdie, N. (1996) Effects of learning skills interventions on student learning: a meta-analysis, Review of Educational Research, 66, 99–136. Hewings, M. & Hewings, A. (2002) ‘It is interesting to note that …’: a comparative study of antic- ipatory ‘it’ in student and published writing, English for Specific Purposes, 21, 367–383. Hinchliffe, G. (2002) Situating skills, Journal of Philosophy of Education, 36, 187–205. Holmes, L. (2001) Reconsidering graduate employability: the ‘graduate identity’ approach, Quality in Higher Education, 7, 111–119. Hyland, K. (2001) Bringing in the reader—addressee features in academic articles, Written Communication, 18, 549–574. Hyland, K. (2002) Directives: argument and engagement in academic writing, Applied Linguistics, 23(2), 215–239. ˇ Ivanic, R. (2004) Discourses of writing and learning to write, Language and Education, 18, 220–245. o c a r [ n ] Johnson, D. W., Maruyana, G., Johnson, R., Nelson, D. & Skon, L. (1981) Effects of cooperation, competitive and individualistic goal structures on achievement: a meta-analysis, Psychological Bulletin, 89, 47–62. Keeley, S. M., Ali, R. & Gebing, T. (1998) Beyond the sponge model: encouraging students’ questioning skills in abnormal psychology, Teaching of Psychology, 25, 270–274. Kemp, I. J. & Seagraves, L. (1995) Transferable skills—can higher education deliver? Studies in Higher Education, 20, 315–328. King, A. (1997) ASK to THINK—TEL WHY: a model of transactive peer tutoring for scaffolding higher level complex learning, Educational Psychologist, 32, 221–235. Knight, P. T. & Yorke, M. (2003) Employability and good learning in higher education, Teaching in Higher Education, 8(1), 3–16. Lavelle, E. (1993) Development and validation of an inventory to assess processes in college composition, British Journal of Educational Psychology, 63, 489–499. Lavelle, E. (1997) Writing style and the narrative essay, British Journal of Educational Psychology, 67, 475–482. Core assessment criteria 89 Lavelle, E. & Zuercher, N. (2001) The writing approaches of university students, Higher Educa- tion, 42, 373–391. Lea, M. R. & Street, B. V. (1998) Student writing in higher education: an academic literacies approach, Studies in Higher Education, 23, 157–172. Maguire, S., Evans, S. E. & Dyas, L. (2001) Approaches to learning: a study of first-year geogra- phy undergraduates, Journal of Geography in Higher Education, 25(1), 95–107. Marttunen, M. (1992) Commenting on written arguments as part of argumentation skills— comparison between students engaged in traditional versus on-line study, Scandinavian Journal of Educational Research, 36, 289–302. Marttunen, M. (1997) Electronic mail as a pedagogical delivery system: an analysis of the learning of argumentation, Research in Higher Education, 38, 345–363. Marttunen, M. & Laurinen, L. (2001) Learning of argumentation skills in networked and face-to- face environments, Instructional Science, 29, 127–153. McCann, T. M. (1989) Student argumentative writing knowledge and ability at three grade levels, Research in the Teaching of English, 23, 62–76. McPeck, J. E. (1981) Critical thinking and education (New York, St Martin’s Press). Miller, M. A. (1992) Outcomes evaluation—measuring critical thinking, Journal of Advanced Nursing, 17, 1401–1402. Mitchell, S. & Riddle, M. (2000) Improving the quality of argument in higher education: final report (London, School of Lifelong Learning and Education, Middlesex University). Norton, L. S. (1990) Essay writing: what really counts? Higher Education, 20, 411–442. Norton, L. (2004) Using assessment criteria as learning criteria: a case study in psychology, Assess- ment and Evaluation in Higher Education, 29, 687–702. Oates, J. (2002) Essay TMA scoring/feedback criteria, Personal communication. O’Donovan, B., Price, M. & Rust, C. (2000) The student experience of criterion-referenced assessment (through the introduction of a common criteria assessment grid), Innovations in Education and Teaching International, 38, 74–85. Pain, R. & Mowl, G. (1996) Improving geography essay writing using innovative assessment, Journal of Geography in Higher Education, 20(1), 19–31. Peck, J. & Coyle, M. (1999) The student’s guide to writing: grammar, punctuation and spelling (Houndmills, Basingstoke and London, Macmillan). Perkins, D. N. (1985) Postprimary education has little impact on informal reasoning, Journal of Educational Psychology, 77, 562–571. Pirie, D. B. (1985) How to write critical essays: a guide for students of literature (London, Methuen). Price, M. & Rust, C. (1999) Business assessment criteria grid, Personal communication. Prosser, M. & Webb, C. (1994) Relating the process of undergraduate essay writing to the finished product, Studies in Higher Education, 19, 125–138. Read, B., Francis, B. & Robson, J. (2001) ‘Playing safe’: undergraduate essay writing and the presentation of the student ‘voice’, British Journal of Sociology of Education, 22, 387–399. Rust, C., Price, M. & O’Donovan, B. (2003) Improving students’ learning by developing their understanding of assessment criteria and processes, Assessment and Evaluation in Higher Education, 28, 147–164. Sadler, D. R. (1987) Specifying and promulgating achievement standards, Oxford Review of Educa- tion, 13, 191–209. Sadler, D. R. (1989) Formative assessment and the design of instructional systems, Instructional Science, 18, 119–144. Scouller, K. (1998) The influence of assessment method on students’ learning approaches: multi- ple choice question examination versus assignment essay, Higher Education, 35, 453–472. Simpson, J. A. & Weiner, E. S. C. (Eds) (1989) The Oxford English dictionary (2nd edn) (Oxford, Clarendon Press). Smith, D., Wolstencroft, T. & Southern, J. (1989) Transferable personal skills and the job demands on graduates, Journal of European Industrial Training, 13(8), 25–31. 90 J. Elander et al. Terenzini, P. T., Springer, L., Pascarella, E. T. & Nora, A. (1995) Influences affecting the devel- opment of students’ critical thinking skills, Research in Higher Education, 36(1), 23–39. Thompson, G. (2001) Interaction in academic writing: learning to argue with the reader, Applied Linguistics, 22(1), 58–78. Tynjälä, P. (1998) Traditional studying for examination versus constructivist learning tasks: do learning outcomes differ? Studies in Higher Education, 23, 173–189. Tynjälä, P., Valimaa, J. & Sarja, A. (2003) Pedagogical perspectives on the relationships between higher education and working life, Higher Education, 46, 147–166. Valle, A., Cabanach, R. G., Nunez, J. C., Gonzalez-Pienda, J., Rodriguez, S. & Pineiro, I. (2003) Multiple goals, motivation and academic learning, British Journal of Educational Psychology, 73, 71–87. van Merriënboer, J. J. G., Kirschner, P. A. & Kester, L. (2003) Taking the load off a learner’s mind: instructional design for complex learning, Educational Psychologist, 38, 5–13. Warren, D. (2003) A discipline-based approach to developing academic literacy, in: D. Gosling & V. D’Andrea (Eds) International conference on the scholarship of teaching and learning: proceedings 2001–2002 (London, Educational Development Centre, City University), 109–117. Whitehead, D. (2002) The academic writing experiences of a group of student nurses: a phenome- nological study, Journal of Advanced Nursing, 38, 498–506.