Learning Styles And Pedagogy In Post-16 Learning: A .

LSRC referencepage 1Section 1A systematic review of learning-styles modelsIntroductionA complex research fieldHow can we teach students if we do not know howthey learn? How can we improve the performanceof our employees if we do not know how we ourselveslearn or how to enhance their learning? Are thelearning difficulties of so many students/employeesbetter understood as the teaching problems oftutors/workplace training managers? How can wepretend any longer that we are serious about creatinga learning society if we have no satisfactory responseto the questions: what model of learning do we operatewith and how do we use it to improve our practiceand that of our students/staff/organisation? Theseare just some of the issues raised by those researcherswho for the last 40–50 years have been studying thelearning styles of individuals.Yet beneath the apparently unproblematic appealof learning styles lies a host of conceptual and empiricalproblems. To begin with, the learning styles field is notunified, but instead is divided into three linked areasof activity: theoretical, pedagogical and commercial.There is a strong intuitive appeal in the idea thatteachers and course designers should pay closerattention to students’ learning styles – by diagnosingthem, by encouraging students to reflect on themand by designing teaching and learning interventionsaround them. Further evidence for the idea that wehave individual learning styles appears to be offeredwhen teachers notice that students vary enormouslyin the speed and manner with which they pick up newinformation and ideas, and the confidence with whichthey process and use them. Another impetus to interestin post-16 learning styles is given by a governmentpolicy that aims to develop the necessary attitudesand skills for lifelong learning, particularly in relationto ‘learning to learn’. These are widely assumed bypolicy-makers and practitioners to be well delineated,generic and transferable.The logic of lifelong learning suggests that studentswill become more motivated to learn by knowing moreabout their own strengths and weaknesses as learners.In turn, if teachers can respond to individuals’ strengthsand weaknesses, then retention and achievementrates in formal programmes are likely to rise and‘learning to learn’ skills may provide a foundation forlifelong learning. Perhaps a more instrumental impetusis provided by pressures on resources in many post-16institutions. For example, if students become moreindependent in their learning as a result of knowingtheir strengths and weaknesses, then negative effectsfrom lower levels of contact between lecturers andstudents will be counterbalanced if students developmore effective learning strategies which they can useoutside formal contact time.The first area is a growing body of theoretical andempirical research on learning styles in the UK, theUS and Western Europe that began in the early yearsof the 20th century and is still producing ideas andan ever-proliferating number of instruments. Our reviewhas identified 71 models of learning styles and we havecategorised 13 of these as major models, using criteriaoutlined below. The remaining 58 (listed in Appendix 1)are not critically analysed in this report. Many consistof rather minor adaptations of one of the leading modelsand therefore lack influence on the field as a whole;a large number represent the outcomes of doctoraltheses. Some offer new constructs 1 (or new labelsfor existing constructs) as the basis for a claim to havedeveloped a new model. Others have been used onlyon very small or homogeneous populations, and yetothers have had a brief vogue but have long falleninto obscurity. It is important to note that the fieldof learning styles research as a whole is characterisedby a very large number of small-scale applicationsof particular models to small samples of studentsin specific contexts. This has proved especiallyproblematic for our review of evidence of the impactof learning styles on teaching and learning, since thereare very few robust studies which offer, for example,reliable and valid evidence and clear implications forpractice based on empirical findings.The second area is a vast body of research intoteaching and learning which draws researchers fromdiverse specialisms, mainly from different branchesof psychology, but also from sociology, businessstudies, management and education. Researchersworking in the field of learning styles across or withinthese disciplines tend to interpret evidence andtheories in their own terms. Evidence about learningis guided by contrasting and disputed theories frompsychology, sociology, education and policy studies,and valued in different ways from different perspectives.Education is also influenced strongly by politicalideologies and social values that create preferencesas to which type of theory is given greatest weight.The problem is compounded by the way in whichacademic researchers develop their reputations byestablishing individual territories and specialisms,which are then stoutly defended against those froma different perspective. This form of intellectual trenchwarfare, while common throughout academia, is nota particular feature of the learning styles literature,where the leading theorists and developers ofinstruments tend to ignore, rather than engage with,each other. The result is fragmentation, with littlecumulative knowledge and cooperative research.1Bold italic text indicates the first usage in the text of a term in the glossary(Appendix 3).

The third area consists of a large commercial industrypromoting particular inventories and instruments.Certain models have become extremely influentialand popular: in the US, for example, the Dunn, Dunnand Price Learning Styles Inventory (LSI) is usedin a large number of elementary schools, while in theUK, both Kolb’s Learning Style Inventory (LSI) andHoney and Mumford’s Learning Styles Questionnaire(LSQ) are widely known and used. The commercialgains for creators of successful learning stylesinstruments are so large that critical engagementwith the theoretical and empirical bases of their claimstends to be unwelcome.Many teachers use the most well-known instrumentswith explicit acknowledgement of the source anda clear idea of why they have chosen a particular model.However, it is also common, particularly on in-servicetraining, management or professional developmentcourses, for participants to analyse their learning stylesusing an unnamed questionnaire with no accompanyingexplanation or rationale. In many ways, the use ofdifferent inventories of learning styles has acquired anunexamined life of its own, where the notion of learningstyles itself and the various means to measure itare accepted without question. Mainstream use hastoo often become separated from the research field.More problematically, it has also become isolated fromdeeper questions about whether a particular inventoryhas a sufficient theoretical basis to warrant eitherthe research industry which has grown around it,or the pedagogical uses to which it is currently put.A final aspect of complexity is that researchersproduce their models and instruments for differentpurposes. Some aim to contribute to theoryabout learning styles and do not design theirinstrument for use in mainstream practice. By contrast,others develop an instrument to be used widely bypractitioners in diverse contexts. This differenceaffects the type of claims made for the instrumentand the type of research studies that evaluate it.These three areas of research and activity andtheir potential and pitfalls, militate against the typeof integrative review that we have carried out forthe Learning and Skills Research Centre (LSRC).We have found the field to be much more extensive,opaque, contradictory and controversial thanwe thought at the start of the research process.Evaluating different models of learning styles andtheir implications for pedagogy requires an appreciationof this complexity and controversy. It also requiressome understanding of ideas about learning andmeasurement that have preoccupied researchers ineducation, psychology and neuroscience for decades.The extensive nature of the field surprised us: weunderestimated the volume of research which has beencarried out on all aspects of learning styles over the last30 years, although most of it refers to higher educationand professional learning rather than work in furthereducation (FE) colleges. Three examples illustratethis point. First, in 2000, David Kolb and his wife Aliceproduced a bibliography of research conducted since1971 on his experiential learning theory and LearningStyle Inventory (LSI) : it contains 1004 entries. Second,the website for the Dunn and Dunn Learning StylesQuestionnaire (LSQ) has a bibliography with 1140entries. Lastly, it has been estimated that 2000 articleshave been written about the Myers-Briggs Type Indicator(MBTI) between 1985 and 1995 (see our evaluationslater in this report for more detail).The enormous size of these literatures presents veryparticular problems for practitioners, policy-makersand researchers who are not specialists in this field.It is extremely unlikely that any of these groups will everread the original papers and so they are dependent onreviews like this one, which have to discard the weakestpapers, to summarise the large numbers of high-qualityresearch papers, to simplify complex statisticalarguments and to impose some order on a field whichis marked by debate and constructive critique as wellas by disunity, dissension and conceptual confusion.The principal tasks for the reviewers are to maintainacademic rigour throughout the processes of selection,condensation, simplification and interpretation, whilealso writing in a style accessible to a broad audience.In these respects, the field of learning styles is similarto many other areas in the social sciences whereboth the measurement problems and the implicationsfor practice are complex.Competing ideas about learningConflicting assumptions about learning underpinmainstream ideas about learning and the best-knownmodels of learning styles. For example, some theoriesdiscussed in this report derive from research intobrain functioning, where claims are made that specificneural activity related to learning can be identifiedin different areas of the brain. Other influential ideasderive from established psychological theories, suchas personality traits, intellectual abilities and fixedtraits which are said to form learning styles. From thislatter perspective, it is claimed that learning styles canbe defined accurately and then measured reliably andvalidly through psychological tests in order to predictbehaviour and achievement. Claims about learningstyles from the perspective of fixed traits lead to labelsand descriptors of styles as the basis for strong claimsabout the generalisability of learning styles. Thesecan take on unexpected predictive or controversialcharacteristics. For example, the belief that stylesare fixed has led to propositions that marriage partnersshould have compatible learning styles, that peoplefrom socially disadvantaged groups tend to havea particular style or, as Gregorc (1985) believes, thatstyles are God-given and that to work against one’spersonal style will lead to ill-health (see Section 3.1for evaluation of his Style Delineator).

LSRC referenceEven if we dismiss these extreme examples, thenotion of styles tends to imply something fixed andstable over time. However, different theorists makedifferent claims for the degree of stability within theirmodel of styles. Some theories represent learningstyles as ‘flexibly stable’, arguing that previous learningexperiences and other environmental factors maycreate preferences, approaches or strategies ratherthan styles, or that styles may vary from contextto context or even from task to task. Nevertheless,supporters of this view still argue that it is possibleto create valid and reasonably reliable measures andfor these to have diagnostic and predictive use forenhancing students’ learning. By contrast, othertheorists eschew all notions of individual traitsand argue that it is more productive to look at thecontext-specific and situated nature of learning andthe idea of learning biographies rather than stylesor approaches.Competing ideas about learning have led toa proliferation of terms and concepts, many of whichare used interchangeably in learning styles research.For example, terms used in this introduction include‘learning styles’, ‘learning strategies’ and ‘approachesto learning’. In addition, we have referred to ‘models’,‘instruments’ and ‘inventories’. Our investigation hasrevealed other terms in constant use: ‘cognitive styles’,‘conative styles’, and ‘cognitive structures’; ‘thinkingstyles’, ‘teaching styles’, ‘motivational styles’, ‘learningorientations’ and ‘learning conditions’. Sometimesthese terms are used precisely, in order to maintaindistinctions between theories; at other times, they areused very loosely and interchangeably. Some theoristsoffer clear definitions of their key concepts at theoutset, but forget to maintain the limitations theyhave placed on their language in later papers. Ratherthan attempting to offer yet another set of definitionsof each concept, this report aims to define these termsas clearly as possible within particular families of ideasabout learning in order to show how they are used bydifferent learning styles theorists.Implications for defining and measuringlearning stylesIt is possible to explain the main dimensions thatunderpin different approaches to learning styles andthis report does so in later sections. Nevertheless,the competing theories and techniques of measuringthem, and the effectiveness of such measures areso varied and contested that simple choices aboutthe most suitable are difficult to substantiate. Differentideas about learning styles create distinct approachesto identifying the specific attitudes and skills thatcharacterise styles and different measures designedto generalise between learning contexts and typesof learner.Section 1page 2/3Evaluating the claims for various models requiresan understanding of the psychometric vocabularythat underpins particular constructs and measuresof reliability and validity. For example, there arevarious dimensions to validity: including whetherthe various test items appear to capture what they setout to measure (face validity) and whether the rangeof behaviours can be seen to have an impact on taskperformance (predictive validity). In addition, a numberof other types of validity are important, includingecological validity, catalytic validity and constructvalidity. In addition, there is the frequently overlookedissue of effect size.The notion of reliability is also important because someof the most popular models extrapolate from evidenceof reliability to strong assertions of generalisability,namely that learners can transfer their styles to othercontexts or that measures will produce similar resultswith other types of student. We provide a summaryof measurement concepts in a glossary in Appendix 3.Finally, the technical vocabulary needed to understandand interpret the various claims about learningstyles also requires an appreciation that for someresearchers, a reliable and valid measure of learningstyles has not yet been developed; and for some,that the perfect learning style instrument is a fantasy.From the latter perspective, observation and interviewsmay be more likely than instruments to capture someof the broad learning strategies that learners adopt.Those who reject the idea of measurable learning stylesconsider it more useful to focus on learners’ previousexperiences and motivation.Implications for pedagogyA number of options for pedagogy flow from thedifferent perspectives outlined in this introduction.For example, supporters of fixed traits and abilitiesargue that a valid and reliable measure is a soundbasis for diagnosing individuals’ learning needsand then designing specific interventions to addressthem, both at the level of individual self-awarenessand teacher activity. This, however, might lead tolabelling and the implicit belief that traits cannot bealtered. It may also promote a narrow view of ‘matching’teaching and learning styles that could be limitingrather than liberating.In order to counter such problems, some theoristspromote the idea that learners should developa repertoire of styles, so that an awareness of theirown preferences and abilities should not bar themfrom working to acquire those styles which theydo not yet possess. In particular, as students movefrom didactic forms of instruction to settings witha mixture of lectures, seminars and problem-basedlearning, it may become possible for them to usea range of approaches. This can lead to a plan forteachers to develop these styles through differentteaching and learning activities, or it can lead towhat might be seen as a type of ‘pedagogic sheep dip’,where teaching strategies aim explicitly to touchupon all styles at some point in a formal programme.

Other theorists promote the idea of learning stylesinstruments as a diagnostic assessment tool thatencourages a more self-aware reflection aboutstrengths and weaknesses. For supporters of thisidea, the notion of learning styles offers a way forteachers and students to talk more productively aboutlearning, using a more focused vocabulary to do so.Finally, those who reject the idea of learning stylesmight, nevertheless, see value in creating a moreprecise vocabulary with which to talk about learning,motivation and the idea of metacognition – wherebetter self-awareness may lead to more organisedand effective approaches to teaching and learning.A large number of injunctions and claims for pedagogyemerge from the research literature and we providea full account of these in Section 8, together with anindication of their strengths and weaknesses. However,although many theorists draw logical conclusions aboutpractice from their models of learning styles, thereis a dearth of well-conducted experimental studiesof alternative approaches derived from particularmodels. Moreover, most of the empirical studies havebeen conducted on university students in departmentsof psychology or business studies; and some wouldcriticise these as studies of captive and perhapsatypical subjects presented with contrived tasks.Aims of the projectThe Learning and Skills Development Agency (LSDA)commissioned a number of research projects in post-16learning through a new Learning and Skills ResearchCentre (LSRC) supported by the Learning and SkillsCouncil (LSC) and the Department for Education andSkills (DfES). The University of Newcastle upon Tynecarried out two projects: an evaluation of modelsof learning style inventories and their impact on post-16pedagogy (this report and Coffield et al. 2004) andan evaluation (with the University of Sunderland)of different thinking skills frameworks (Moseley et al.2003). Other projects in the LSRC’s programme includean evaluation by the University of Strathclyde of theimpact of thinking skills on pedagogy (Livingston,Soden and Kirkwood 2003), a report by the universitiesof Surrey and Sheffield on the extent and impactof mixed-age learning in further education (McNair andParry 2003) and a mapping by the University of Leedsof the conceptual terrain in relation to informal learning(Colley, Hodkinson and Malcolm 2003).The evaluation of learning styles inventories wasoriginally a separate project from the evaluation of theimpact of learning styles on post-16 pedagogy. However,the two projects were merged in order to maximise thesynergy between the theoretical research on learningstyles and its practical implications for pedagogy.The aims of the joint project were to carry outan extensive review of research on post-16 learningstyles, to evaluate the main models of learning styles,and to discuss the implications of learning stylesfor post-16 teaching and learning. These broad aimsare addressed through the following research questionsand objectives.Research questionsWe addressed four main questions.1What models of learning styles are influential andpotentially influential?2What empirical evidence is there to support the claimsmade for these models?3What are the broad implications for pedagogyof these models?4What empirical evidence is there that models oflearning styles have an impact on students’ learning?Research objectivesThe objectives that arose from our questionsenabled us to:identify the range of models that are:availableinfluential or potentially influential in researchand practicelocate these models within identifiable ‘families’of ideas about learning stylesevalu

Section 1 Section 2 Section 3 Section 4 Section 5 Section 6 Section 7 Section 8 Section 9 1 9 13 37 47 61 91 119 133 Ackno