The field of learning styles is wide and affected by several inputs, thus leading to different views and concepts. Various learning style models are in literature, each is proposing different descriptions, approach and classifications of learning types. Coffield et al.(2004b) identified 71 models of learning styles and categorized 13 of them as major models with respect to their theoretical importance in the field, their wide spread use, and their influence on other learning style models. Furthermore, a lot of researches have been carried out in the last decades with respect to different aspects of learning style models.
As stated by Coffield et al. (2004b), about 2000 articles have been written related to the Myers-Briggs Type Indicator (Briggs Myers, 1962) between 1985 and 1995 and more than 1000 publications have been written about the Kolb learning style model (Kolb, 1984) as well as the Dunn and Dunn learning style model (Dunn and Dunn, 1974). (Akkoyunlu & Soylu, 2008) stated that numerous studies have investigated the impact of learning styles in community college courses (Jones, Reichard & Mokhtari, 2003, Terry, 2001). Few studies to date have evaluated the students' perceptions in learning styles and blended learning environment (Lemire, 2002; Raschick, Maypole & Day, 1998; Terrell & Dringus, 1999; Simpson & Du, 2004; Richmond & Liu 2005). The studies about learning styles mostly focus on the success of learners in traditional learning environments, attitudes towards learning environments or the rate of involvement in the learning environment (Akkoyunlu & Soylu, 2008). The researcher didn't find in the literature any investigation correlating the learning styles with the Maritime Education and Training field.
To date, no single definition of the term learning style has been identified; it has been defined as follows:
Honey and Mumford (1992) defined learning styles as "a description of the attitudes and behaviors which determine an individual's preferred way of learning". Felder (1996) defined learning styles as "characteristic strengths and preferences in the ways they [learners] take in and process information". James and Gardner (1995) defined learning styles more precisely by saying that learning style is the "complex manner in which, and conditions under which, learners most efficiently and most effectively perceive, process, store, and recall what they are attempting to learn" (Graf & Kinshuk, 2005). People learn in different ways as the tendency to adopt a particular strategy in learning (Akkoyunlu & Soylu, 2008). Most students have a preferred learning style but some may adapt their learning styles according to tasks (Pask, 1976). Learning style may also be defined as personal qualities that influence a student's ability to acquire information, to interact with peers and the teachers, and otherwise participate in learning experiences (Grasha, 1996). Learning styles are the traits that refer to how individuals are approaching their learning tasks and processing information (Kemp, Morrison & Ross, 1998). Jensen (2003) defined it as a preferred way of thinking, processing, and understanding information.
Counting on the ideas and aspects of the meaning of learning styles, other terms such as learning strategy and cognitive style are often used in a similar context or even interchangeable to the term learning style. In the following paragraphs, definitions of the terms learning strategies and cognitive styles are introduced and the difference to learning styles is described.
Learning strategies can be seen as short term methods that students apply in a particular situation. These strategies can change with the time, teacher, subject, and situation. When learning strategies are frequently used by students, learning styles can be derived from these strategies (Pask, 1976b). Based on Pask's work, Entwistle, Hanley, and Hounsell (1979) define a learning strategy as "the way a student chooses to tackle a specific learning task in the light of its perceived demands" and learning style "as a broader characterization of a student's preferred way of tackling learning tasks generally". Furthermore, they argued that distinct learning styles underlie learning strategies.
According to Jonassen and Grabowski (1993), learning styles can also be seen as applied cognitive styles in the domain of learning, removed one more level from pure processing ability. As evidence of this removal, learning styles are usually based on self reported learning preferences. For measuring them, instruments are used that ask learners about their preferences. In contrast, cognitive styles are identified by task-relevant measures, which test the actual ability or skill. The next subsection introduces several commonly used learning style models.
Subsequently, the implications of learning styles for education as well as criticism and challenges of the field of learning styles are discussed.
2.1 Common Models of Learning Styles
As stated before, a high number of learning style models exists in literature. Coffield et al. (2004b) classified learning style models into 5 families which are based on some overarching ideas behind the models, attempting to reflect the views of the main theorists of learning styles. Coffield, Moseley, Hall, & Eccjestone (2004) in the LSRC (Learning & Skills Research Centre) reference report mentioned that the first family relies on the idea that learning styles and preferences are largely constitutionally based including the four modalities: visual, auditory, kinesthetic, and tactile. The second family deals with the idea that learning styles reflect deep-seated features of the cognitive structure, including patterns of abilities. A third category refers to learning styles as one component of a relatively stable personality type. In the fourth family, learning styles are seen as flexibly stable learning preferences. The last category moves on from learning styles to learning approaches, strategies, orientations and conceptions of learning.
Table 2.1: Summary of described learning style models
This section describes 10 commonly used learning style models. The selection of these models is based on Coffield's review (Coffield et al., 2004a), including the theoretical importance in the field, their widespread use, and their influence on other learning style models (Popescu, 2010). Additionally, the applicability of the learning style models in technology enhanced learning was considered as important criterion, including the application of learning style models in already existing systems as well as their potential to be used in a system. Since this thesis focuses on learning styles rather than on cognitive styles, models that measure the cognitive abilities and skills rather than self-reported learning preferences were excluded. Therefore, no models of the second family were described, where learning styles are seen as features of the cognitive structure. Table 2.1 shows the selected learning style models grouped according to the classification by Coffield et al. (2004b) and ordered according to the dependencies of the models among each other.
2.1.1 Personality Types as defined by Myers-Briggs
Myers-Briggs Type Indicator (MBTI) (Briggs Myers, 1962) is a personality test and is not focused specifically on learning. Nevertheless, the personality of a learner influences his/her way of learning and therefore, MBTI includes important aspects for learning. Besides, other learning style models are based on considerations of MBTI. Based on Jung's theory of psychological types (Jung, 1923), the MBTI distinguishes a person's type according to four dichotomies: extroversion/introversion, sensing/intuition, thinking/feeling, and judging/perceiving. All possible combinations can occur, which result in a total number of 16 types (Graf & Kinshuk, 2005).
The extrovert and introvert dimension refers to the orientation of a person. The preferred focus of people with an extrovert attitude is on the surroundings such as other people and things, whereas an introvert's preferred focus is on his/her own thoughts and ideas. Sensing and intuition deal with the way people prefer to perceive data. While sensing people prefer to perceive data from their five senses, intuitive people use their intuition and prefer to perceive data from the unconscious. The judgment based on the perceived data can be distinguished between thinking and feeling. Thinking means that the judgment is based on logical connections such as "true or false" and "if-then" while feeling refers to "more-less" and "better-worse" evaluations. However, judgment and decisions are in both cases based on rational considerations. The last dichotomy describes whether a person is more extroverted in his/her stronger judgment function (thinking or feeling) or in the perceiving function (sensing or intuition). Judging people prefer step by step approaches and structure as well as coming to a quick closure. Perceiving people have a preference for keeping all options open and tend to be more flexible and spontaneous (Graf, 2007).
The preferences on the four dimensions interact with each other rather than being independent, and for a complete description of a person's type, the combination of all four preferences needs to be considered.
The standard version of the MBTI is a 93 items Form 'M' (Myers and McCaulley, 1998). The previous version is a 126 items Form 'G' (Myers and McCaulley, 1985), also an abbreviate version of 50 items. The instruments include a series of forced choice questions, related to the four bipolar scales, and the personality type calculation is based on the answers.
2.1.2 Pask's Serialist/Holist/Versatilist Model
During the development of the conversation theory (Pask, 1972, 1976a, 1976b), Pask studied patterns of conversations between individuals to identify various styles of learning and thinking. A critical method according to the conversation theory is the "teachback" approach, where students teach their peers. Different patterns for designing, planning, and organizing of thought as well as for selecting and representing information were investigated, resulting in the identification of three types of learners (Pask, 1976b). Serialist students use a serial learning strategy. They tend to concentrate more narrowly on details and procedures before conceptualising an overall picture. They typically work from the bottom up, learn step-by-step in a linear sequence and concentrate on well defined and sequentially ordered chunks of information. According to Pask, serial learners tend to ignore relevant connections between topics, which can be seen as their learning deficit. In contrast, holists use a holistic learning strategy. They tend to concentrate on building broad descriptions and use a top-down approach. They focus on several aspects of the subject at the same time and use complex links to relate
Multi leveled information. While they are good in building interconnections between theoretical, practical, and personal aspects of a topic, holistic learners do not focus on enough details, which can be seen as their learning deficit. Versatile learners employ both, serial and holistic learning strategies. They engage in global and detailed approaches and succeed in achieving a full and deep understanding. Therefore, versatile learners are proficient at learning from most or all modes of instruction (Graf, 2007).
Pask developed some tests such as the Spy Ring History Test (Pask and Scott, 1973) and the Clobbits Test (Pask, 1975) as measure for serial, holistic and versatile thinking. Some years later, Entwistle (1981; 1998) and Ford (1985) developed self-report inventories for identifying a preference for serial, holistic, and versatile learning styles.
The Study Preference Questionnaire developed by Ford (1985) provided students with pairs of two statements (one on the left side and one on the right side) and asked them to indicate their degree of agreement with either statements, or to indicate no preference, using a 5 point scale. Entwistle's learning style model (described in the next section) is based on Pask's work. With respect to his model, Entwistle designed inventories to tap into a number of dimensions of study attitudes and behaviours, including also the serial/holistic/versatile dimension (Entwistle, 1981, 1998).
2.1.3 Entwistle's Deep, Surface and Strategic Learning
The research conducted by Entwistle and his colleagues (Entwistle, 1981, 1998; Entwistle, McCune, and Walker, 2001) deals with the involvement of students' intentions, goals and motivation in their learning approach. Entwistle argued that the students' orientations to and conceptions of learning lead to and are affected by the student's typical approaches to learning. The model is based on research by Pask (1976b), Marton (1976), and Biggs (1979) and distinguishes between three approaches for learning and studying (Entwistle, McCune, and Walker, 2001): learners applying a deep learning approach are intrinsically motivated and have the intention to understand the ideas for themselves. They learn by relating ideas to previous knowledge and experiences, looking for patterns and underlying principles, and checking evidence and relating it to conclusions. They examine logic and arguments cautiously and critically, develop an understanding of the topic, and become actively interested in the course content. In contrast, learners who apply a surface learning approach are extrinsically motivated and aim merely at meeting the requirements of the course. They treat the course content as unrelated bits of knowledge, try to identify those elements of a course that are likely to be assessed and focus on memorizing these details. They carry out procedures routinely and find difficulty in making sense of new ideas presented (Coffield et al., 2004). Popescu (2010) see little value or meaning in either courses or tasks set, study without reflecting on either purpose or strategy, and feel undue pressure and worry about their work. In the strategic learning approach, students combine the deep and surface approach in order to achieve the best possible outcome in terms of marks. Students who adopt the strategic approach put consistent effort into studying, manage time and effort effectively, find the right conditions and materials for studying, and monitor the effectiveness of ways of studying. They are alert to assessment requirements and criteria and gear work to the perceived preferences of teachers.
For measuring the adopted approach of learning and studying of students, several versions of a questionnaire have been evolved such as the Approaches to Studying Inventory (ASI) (Ramsden and Entwistle, 1981), the Course Perception Questionnaire (CPQ) (Ramsden and Entwistle, 1981), the Revised Approaches to Studying Inventory (RASI) (Entwistle and Tait, 1995), the Approaches and Study Skills Inventory for Students (ASSIST) (Entwistle and Tait, 1996), and the Approaches to Learning and Studying Inventory (ALSI) (Tyler and Entwistle, 2003). Since Entwistle's model is based on Pask's serial and holistic learning strategy, this concept is also included in the questionnaires. For example, in the ASSIST, the currently most often used instrument, the serial and holistic learning strategy is included as subcategory of the deep learning approach (Graf, 2007).
2.1.4 Grasha-Riechmann Learning Style Model
The Grasha-Riechmann learning style model (Grasha and Riechmann, 1975; Riechmann and Grasha, 1974) focuses on the students' social interaction with their teachers and fellow students in the classroom environment. Grasha and Riechmann identified three bipolar dimensions in order to understand the students' behaviour with respect to their social interaction: the participant/avoidant, collaborative/competitive, and dependent/independent dimension. The participant/avoidant dimension indicates how much a student wishes to become involved in the classroom environment. Students who adopt a participant style desire to learn the course content and enjoy attending the class. They take responsibility for their own learning and enjoy participating in the learning activities. In contrast, students who adopt an avoidant style do not like to learn and do not enjoy attending the class. They also do not take responsibility for their learning and avoid taking part in the course activities.
The collaborative/competitive dimension measures the motivation behind a student's interactions with others. Collaborative learners are characterised as learners who are cooperative, enjoy working with others, and see the classroom as a place for learning and interacting with others. On the other hand, competitive learners see their fellow students as competitors. They have the motivation to do better than others, enjoy competing, and see the classroom as a win-lose situation. The dependent/independent dimension measures attitudes toward teachers and how much the students desire freedom and control in the learning environment. Dependent students see the teacher as the source of information and structure. They want to be told what to do by authorities and learn only what is required. Independent learners are characterised as confident and curious learners. They prefer to think for themselves and work on their own.
For measuring the preference of students with respect to the six learning styles, a 90- item self-report inventory called Student Learning Styles Scale (SLSS) (Grasha and Riechmann, 1975) was developed. The questionnaire is created in particular for college and high school students. It is divided in six subcategories, each for one learning style.
Each subcategory consists of 15 questions. Students are asked to rate their agreement or disagreement to these questions on a 5-point Likert scale. Considering the issue that the styles may change from class to class for each student, two different forms are designed, one that assesses a general class, and the second that relates to a specific course.
2.1.5 Dunn and Dunn Learning Style Model
The Dunn and Dunn learning style model (Dunn and Dunn, 1974; Dunn and Griggs, 2003) was originally proposed in 1974 and then refined and extended over the years. The model distinguishes between adults and children and includes five variables where each variable consists of several factors. The environmental variable includes sound, temperature, light, and seating/furniture design. The sociological variable incorporates factors dealing with the preference for learning alone, in a pair, in a small group, as part of a team, with an authority, or in varied approaches (as opposite to in patterns). For children, additionally the motivation from parents/teachers is included as factor. The emotional variable consists of the factors motivation, conformity/responsibility, persistence, and need for structure. The physical variable is comprised of factors regarding perception/modality preferences (visual, auditory, tactile/kinaesthetic external, kinaesthetic internal), food and drink intake, time of day and mobility. The psychological variable was added later to the model and includes factors referring to global/analytic preferences, right or left hemisphericity, and impulsive/reflective preferences.
For detecting the learning style preferences according to the Dunn and Dunn learning style model, different versions of questionnaires were developed. The Learning Styles Inventory (Dunn, Dunn, and Price, 1996) was developed for children and exists in three versions (kindergarten to grade 2, grade 3 and 4, grade 5-12). This inventory consists of 104 questions which employ a 3-choice or 5-choice Likert scale. The Building Excellence Inventory (Rundle and Dunn, 2000) is the current version for adults. It includes 118 questions and employs a 5-point Likert scale. As a result, a high or low preference for each factor is identified.
2.1.6 Gregorc's Mind Styles Model
Gregorc's mind style model (Gregorc, 1982a; Gregorc, 1982b; Gregorc, 1985) is based on two dimensions dealing with the preferences for perception and ordering. Regarding perception, people can prefer an abstract or concrete way of perception, or some combination of both. Abstract perception refers to the ability to process information through reason and intuition, often invisible to our physical senses. In contrast, concrete perception emphasises the physical senses and refers to the ability to process information through these senses. The ordering dimension deals with the way a learner is arranging, prioritising, and using information in either a sequential or random order, or in a combination of both. While a sequential style pertains to use a linear, step-by-step organisational scheme, a random order style refers to the use of a network-like format which relates data to each other in a variety of ways. The perceptual and ordering preferences can be combined into four basic mediation channels which lead to four types of learners.
The concrete sequential learners prefer to use their five senses for processing information and are considered as orderly, logical, and sequential. These learners look for authority and guidance in a learning environment and prefer to extract information from hands-on experiences. The concrete random learners are characterised by the need to experiment with ideas and concepts and will employ trial-and-error in learning. They like to explore the learning environment, are considered as insightful, can easily move from facts to theory, and do not like authoritative interventions.
The abstract sequential learners have their strengths in the area of decoding written, verbal, and image symbols. They prefer rational and sequential presentations and are good in synthesising ideas and producing new concepts or outcomes to new conclusions. They will defer to authority and has a low tolerance for distractions.
The abstract random learners are characterised by a keen awareness of human behaviour and an ability to evaluate and interpret atmosphere and mood. They prefer an unstructured learning environment and collaborations with others, are good in seeing relationships, tend to be reflective and need time to process data before reacting to it. A more detailed description about the characteristics and preferences of the four types of learners is provided by Gregorc (1982a; 1982b). The Gregorc Style Delineator (Gregorc, 1982b; Gregorc, 1985) is a self-report instrument to detect learners' preferences for the two dimensions and therefore their preferred channels. The instrument presents the students with 40 words arranged in 10 columns of four items each. The learners are then asked to rank the four words relative to how they fit to themselves (1 for being least and 4 for being most like themselves). Scores for each of the four learner types can range from 10 to 40, calculated by summing up the ranks of the respective words for each channel.
2.1.7 Kolb's Learning Style Model
The learning style theory by Kolb (1984) is based on the Experiential Learning Theory, which models the learning process and incorporates the important role of experience in this process. Following this theory, learning is conceived as a four-stage cycle. Concrete experience is the basis for observations and reflections.
These observations are used to form abstract concepts and generalizations, which again act as basis for testing implementations of concepts in new situations. Testing implementations results in concrete experience, which closes the learning cycle (Graf & Kinshuk, 2005).
According to this theory, learners need four abilities for effective learning: a) Concrete Experience abilities, b) Reflective Observation abilities, c) Abstract Conceptualization abilities, and d) Active Experimentation abilities. On closer examination, there are two polar opposite dimensions: concrete/abstract and active/reflective. Kolb (1981) described that "as a result of our hereditary equipment, our particular past life experience, and the demands of our present environment, most of us develop learning styles that emphasize some learning abilities over others". Based on this assumption, Kolb identified four statistically prevalent types of learning styles.
Convergers' dominant abilities are abstract conceptualization and active experimentation. Therefore, their strengths lie in the practical applications of ideas. The name "Convergers" is based on Hudson's theory of thinking styles (Hudson, 1966), where convergent thinkers are people who are good in gathering information and facts and putting them together to find a single correct answer to a specific problem.
In contrast, Divergers excel in the opposite poles of the two dimensions, namely concrete experimentation and reflective observation. They are good in viewing concrete situations in many different perspectives and in organizing relationships to a meaningful shape. According to Hudson, a dominant strength of Divergers is to generate ideas and therefore, Divergers tend to be more creative.
Assimilators excel in abstract conceptualization and reflective observation. Their strength lies in creating theoretical models. They are good in inductive reasoning and in assimilating disparate observations into an integrated explanation.
Accommodators are the contrary of Assimilators. Their dominant abilities are concrete experience and active experimentation. Their strengths lie in doing things actively, carrying out plans and experiments, and becoming involved in new experiences.
They are also characterized as risk-takers and as people who excel in situations that call for adaptation to specific immediate circumstances. For identifying learning styles based on Kolb's learning style model, the Learning Style Inventory (LSI) was developed (Kolb, 1976) and revised several times (Graf & Kinshuk, 2005). The current LSI version (Kolb and Kolb, 2005) uses a compulsory grading technique to assess a candidate's learning preferences (Concrete Experience, Reflective Observation, Abstract Conceptualization and Active Experimentation). Individuals are asked to complete 12 sentences about their preferred way of learning. Each sentence has four endings and the individuals are asked to rank the endings according to what best describes how they learn (4 = most like you; 1 = least like you). The results of the LSI indicate the individuals' preferences for the four modes. Furthermore, their score for the active/reflective and concrete/abstract dimensions can be derived from the preferred modes, which again lead to the preferred type of learning style.
2.1.8 Honey and Mumford's Learning Style Model
The learning style model by Honey and Mumford (1982) is based on Kolb's Experiential Learning Theory and is developed further on the four types of Kolb's learning style model (Kolb, 1984). The active/reflective and concrete/abstract dimensions are strongly involved in the defined types as well. Furthermore, Honey and Mumford stated that "the similarities between his model [Kolb's model] and ours are greater than the differences" (Honey and Mumford, 1992).
As described by Graf (2007), Honey and Mumford's learning style model the types are called: Activist (similar to Accommodator), Theorist (similar to Assimilator), Pragmatist (similar to Converger), and Reflector (similar to Diverger). Activists involve themselves fully in new experiences, are enthusiastic about anything new, and learn best by doing something actively.
Theorists excel in adapting and integrating observations into theories. They need models, concepts, and facts in order to engage in the learning process. Pragmatists are interested in real world applications of the learned material. They like to try out and experiment on ideas, theories, and techniques to see if they work in practice. Reflectors are people who like to observe other people and their experiences from many different perspectives and reflect about them thoroughly before coming to a conclusion. For Reflectors, learning occurs mainly by observing and analyzing the observed experiences. The Learning Style Questionnaire (LSQ), a self-report inventory for identifying learning styles based on the Honey and Mumford learning style model, as well as its manual was initially developed in 1982 (Honey and Mumford, 1982), revised in 1992 (Honey and Mumford, 1992) and then replaced in 2000 (Honey and Mumford, 2000) and again revised in 2006 (Honey and Mumford, 2006). Currently, two versions of the LSQ exist, one with 80 items and one with 40 items.
2.1.9 Herrmann "Whole Brain" Model
The Herrmann "Whole Brain" model (Herrmann, 1989) is based on the split brain research that was carried out by Roger Sperry (1964), separating the brain in the left and right cerebral hemispheres. In addition, the Herrmann "Whole Brain" model considers, following MacLean (1952), the hypothesized functions of the brain's limbic system.
Accordingly, individuals are modeled with respect to how they process information using either a cerebral mode, by thinking about the problem, or a limbic mode, which is a more active approach based on experimentation.
The Herrmann "Whole Brain" model distinguishes between four modes or quadrants. Learners who have a primary preference for quadrant A (left hemisphere, cerebral) prefer logical, analytical, mathematical, technical thinking and can be considered as quantitative, factual, and critical. Learners with a primary preference for quadrant B (left hemisphere, limbic) tend to be sequential and organized, like details, structure and plans and have a structured, organizational and controlled thinking style. Learners with a primary preference for the quadrant C (right hemisphere, limbic) are characterized as emotional, interpersonal, sensory, kinesthetic, and musical. Learners who have a primary preference for quadrant D (right hemisphere, cerebral) tend to be visual, holistic, and innovative and prefer conceptual, synthesizing, and imaginative thinking. For identifying the preferred quadrant, the Herrmann Brain Dominance Instrument (HBDI) was developed (Herrmann, 1989). The HBDI is a self-report inventory, containing 120 questions. As a result of the HBDI, a brain dominance profile is calculated, which shows the primary, secondary and tertiary preferences.
2.1.10 Felder-Silverman Learning Style Model
In Felder-Silverman learning style model (FSLSM) (Felder & Silverman, 1988), learners are described by values on four dimensions. These dimensions are based on major dimensions in the field of learning styles and can be viewed independently from each other. They show how learners prefer to process (active/reflective), perceive (sensing/intuitive), receive (verbal/visual), and understand (sequential/global) information. While these dimensions are not new in the field of learning styles, the way in which they describe a learning style of a student can be seen as new. While most learning style models, which include two or more dimensions, derive statistically prevalent learner types from these dimensions, such as the models by Myers-Briggs (Briggs Myers, 1962), Gregorc (1982a), Kolb (1984), and Honey and Mumford (1982), Felder and Silverman describe the learning styles by using scales from +11 to -11 for each dimension (including only odd values). Therefore, the learning style of each learner is characterized by four values between +11 and -11, one for each dimension(Graf, Lin, & Kinshuk, 2005). These scales facilitate describing the learning style preferences in more detail, whereas building learner types does not allow distinguishing between the strength of the preference. Additionally, the usage of scales allows expressing balanced preferences, indicating that a learner does not have a specific preference for one of the two poles of a dimension. Furthermore, Felder and Silverman consider the resulting preferences as tendencies, meaning that even a high preference learner for a particular learning style can act differently sometimes.
The active/reflective dimension is equivalent to the respective dimension in Kolb's model (1984). Active learners learn best by working actively with the learning material, by applying the material, and by trying things out. Moreover, they tend to be more interested in communicating with others and prefer to learn by working in groups where they can discuss about the learned material. In contrast, reflective learners prefer to think about and reflect on the material. Concerning communication, their preference is to work alone or in a small group together with one good friend. The sensing/intuitive dimension is taken from the Myers-Briggs Type Indicator
(Briggs Myers, 1962) and has also similarities to the sensing/intuitive dimension in Kolb's model (Kolb, 1984). Learners with a sensing learning style like to learn facts and concrete learning material, using their sensory experiences of particular instances as a primary source. They like to solve problems with standard approaches and also tend to be more patient with details (Graf, Kinshuk, & Liu, 2009). Furthermore, sensing learners are considered realistic and sensible; they tend to be more practical than intuitive learners and tend to relate the learned material to the real world. In contrast, intuitive learners prefer to learn abstract learning material, such as theories and their underlying meanings, with general principles rather than concrete instances being a preferred source of information. They like to discover possibilities and relationships and tend to be more innovative and creative than sensing learners. Therefore, they score better in open-ended tests than in tests with a single answer to a problem. This dimension varies from the active/reflective dimension in a significant way, as the sensing/intuitive dimension deals with the preferred source of information while the active/reflective dimension covers the process of transforming the perceived information into knowledge, whilst the third dimension, were visual/verbal dimension deals with the preferred input mode. The dimension differentiates learners who remember best what they have seen (e.g., pictures, diagrams, flow-charts), from learners who get more out of textual representations, regardless of the fact whether they are written or spoken.
Moreover the fourth dimension, learners are distinguished between a sequential and global way of understanding. This dimension is based on the learning style model by Pask (1976b), where sequential learners refer to serial learners and global learners refer to holistic learners. Sequential learners learn in small incremental steps and therefore have a linear learning progress. They tend to follow logical stepwise paths in finding solutions.
In contrast, global learners use a holistic thinking process and learn in large leaps. They tend to absorb learning material almost randomly without seeing connections but after they have learned enough material they suddenly get the whole picture. Then they are able to solve complex problems and put things together in novel ways; however, they have difficulties in explaining how they did it. Because the whole picture is important for global learners, they tend to be more interested in overviews and in a broad knowledge, whereas sequential learners are more interested in details (Graf, Viola, Kinshuk, & Leo, 2006).
For identifying learning styles based on the FSLSM, Felder and Soloman developed the Index of Learning Styles (ILS) (Felder and Soloman, 1997), a 44-item questionnaire. As mentioned earlier, each learner has a personal preference for each dimension. These preferences are expressed with values between +11 to -11 per dimension, with steps +/-2. This range comes from the 11 questions that are posed for each dimension (Graf, Viola, Leo, & Kinshuk, 2007).
2.2 Implications of Learning Styles in Education
Educational researchers and theorists consider learning styles as an important factor in the learning process and agree that incorporating them in education has potential to make learning easier for students. Furthermore, Felder argued that learners with a strong preference for a specific learning style might face difficulties in learning if their learning style is not supported by the teaching environment (Felder and Silverman, 1988; Felder and Soloman, 1997). Thus, from a theoretical aspect, it can be argued that incorporating the learning styles of students makes learning easier for them and increases their learning efficiency. On the contrary, learners who are not supported by the learning environment may experience problems in the learning process.
Graf (2007) mentioned that learning styles can be considered in different ways in education. A first step is to make learners aware of their learning styles and show them their individual strengths and weaknesses. The knowledge about their learning styles helps students to understand why learning is sometimes difficult for them and is the basis for developing their weaknesses. Furthermore, students can be supported by matching the teaching style with the learning styles of the students. Due to the nature of learning styles, providing students with learning material and activities that fit their preferred ways of learning seems to have high potential to make learning easier for them. However, the matching approach aims at a short-term goal, namely to make learning as easy as possible at the time students are learning. Looking at long-term goals, educational theorists such as Messick (1976), Kolb (1984) and Grasha (1984) suggested that learners should also train their not-preferred skills and preferences. Messick argued that when learners acquire more educational experience, they are required to adapt to a variety of instructional methods and styles. The ability to adapt to different instructional styles will prepare them with important life skills. For example, providing visual forms of instruction to verbal learners will force them to develop their visual skills. For Grasha, the mismatching approach is relevant in order to make learning interesting and challenging for students and Kolb argued that the educational objectives for mismatching are personal growth and creativity. However, in Gregorc's model, learning styles are seen as stable, and therefore he argued that a mismatched approach can harm students (Gregorc, 2002). Felder advises against the unintentional, permanent mismatch of teaching styles and learning styles, where teachers are unaware of their own learning styles and may, as a result, teach only according to this style, thus favoring certain students and disadvantages others (Felder, 1993).
Summarizing these aspects, conclusion can be drawn that the mismatching approach should be applied intentionally and depending on the applied learning style model as well as on the learner's needs. In an environment, where students get their individual learning material and activities, the matching and the mismatching approaches can be applied in a regulated method, depending on specific circumstances such as the current learning goal, the experience of the learner in a specific subject, their motivation and so on. A less intensive approach for teachers is to support their learners by including learning material and activities in their courses that address different learning styles rather than teaching in a way that contain only one learning style. For example, if the learning material consists mainly of abstract material, teachers can include some concrete examples to support a sensing/concrete learning style or if the teacher is mainly lecturing in the course, he/she can consider adding some group work activities in order to support active learner. By adopting different learning styles, some activities match with the student's strength and some other with their weakness. Accordingly, the composition is not controlled since the course is the same for all students.
2.3 Learning Styles Critique
The field of learning style is multifaceted and although lot of researches had been conducted, important questions are still unanswered and debatable issues are under debate. The challenge is to clarify these debatable issues, answer the open questions and provide a clear understanding.
Currently, plenty of learning style models exists, each integrating some aspects of learning, and some of them overlapping with each other. Such amount of learning style models leads to criticism and the question on how to integrate all different dimensions of learning styles in education, from a practical view, which learning style model is most appropriate and shall be used with the cadets training onboard the training vessel. Furthermore, the similarities and relationship between these different learning style models and dimensions are mostly not elaborated.
Accordingly, a challenge of the field of learning styles is to carry out research that involves all learning style models and dimensions, fetch clarity in its relationships to each other as well as to other relevant factors of learning (e.g., cognitive styles and cognitive abilities), evaluate them in order to discover major learning style models/dimensions, and build up a holistic model that integrates all relevant aspects of learning styles.
Furthermore, debatable issues such as the question whether learning styles are stable or not over time, subject and environment should be clarified. Depending on the basic ideas behind the learning style models, theorists made different point of views for the degree of stability within their learning style models. The extreme theorists in this aspect state that learning styles similar to learning strategies, thus as flexible and changeable from context to context and even from task to task. Some theorists see learning styles as "flexibly stable", arguing that previous learning experiences and other environmental factors form the learning styles of students. Others link learning styles strongly to cognitive styles and abilities and argue that they are stable over a long period of time or even see them as God-given and not changeable.
However, based on the incorporation of particular dimensions in different models with different ideas about the stability, controversial issues occur. For example, the serial and holistic learning style by Pask (1976b) is related to the sequential and random style by Gregorc (1982a). However, Pask considers the dimension as relatively flexible while Gregorc claims that the learning styles are not changeable. Therefore, future research is needed in order to shed light on the stability of specific dimensions as well as learning style models.
Another issue of criticism deals with the implications of learning styles in education. While the effectiveness of the matching approach seems to be insightful and is one of the very popular recommendations supported by educational theories, inconsistent results are obtained by studies dealing with studying the reflection on achievement when providing matched and mismatched instructions for learners with different learning styles.
Yet the overall impression is that even if the concept of learning style were acceptable, the prospect of matching is unrealistic and largely unsupported by research (Doyle and Rutherford, 1984; see also Candy, 1987 and Curry, 1983). In a useful review of the assimilation of cognitive style into adult education, Joughin (1992) criticizes the assumption that matching will enhance learning as simplistic, ignoring "both the potential value of creatively mismatching teacher and learner and the equivocal outcomes of research on matching itself (p.7), a view shared by Ruble and Stout (1993) in particular reference to LSI.
Currently, no unchallenged and hard evidence exist that learning style matching approach has a significant positive effect on the students' achievement (Coffield et al., 2004b). As Jonassen and Grabowski (1993) summarized, several reasons for such inconsistent results are known in the field of aptitude-treatment interaction (ATI) research. Limitations might include "small samples size, abbreviated treatments, specialized aptitude constructs or standardized tests, and a lack of conceptual or theoretical linkage between aptitudes and the information-processing requirements of the treatment" (Jonassen and Grabowski, 1993, p. 28). This conclusion shows that more research is required to get a clear result about the effect of specific learning styles and other factors on achievement.
However, the main criticism regarding the matching approach is that it is simply "unrealistic, given the demands of flexibility it would make on teachers and trainers" (Reynolds, 1997, p. 121). In traditional learning, teachers would have to routinely change their teaching style to accommodate the different learning styles in a class. Therefore, the feasibility of the matching approach is depending on the number of students and on the adopted learning style model. Pask (1976b), for example, distinguishes between three learning styles, Honey and Mumford (1982) propose four types of learners, the Myers- Briggs Type Indicator (Briggs Myers, 1962) includes 16 different types and in the Felder- Silverman learning style model (Felder and Silverman, 1988), learners can have up to 625 (=54) different learning styles when arranging each of the four dimensions into five groups (e.g., strong active, moderate active, balanced, moderate reflective, strong reflective). Therefore, teachers might not have the capacity to provide each learner with an individual combination of learning material and activities as soon as the number of students and the number of different learning styles increase. However, in MET ISSP as technology enhanced learning, changing the teaching styles for each student and therefore tailoring courses to the individual needs of students is possible, even for a high number of different learning styles and almost independent on the number of students. Lot of research is done in the area of adaptive educational systems, and recently more and more research deals with personal characteristics of learners, such as learning styles (Sabine, 2007). In Chapter XXX, a description on adaptive educational systems incorporating learning styles is provided and in Chapter XXX, an approach for the proposed Maritime Self Study Program associated with adaptive educational systems in order to provide adaptive maritime courses for the deck cadets with respect to the Felder-Silverman learning style model is introduced.
Moreover, more research is required on the subject of mismatching teaching styles and learning styles, its effect on learning, and the conditions when such a mismatch is beneficial in terms of either to support learners and make learning more interesting for them or to achieve long-term goals by forcing them to train their weaknesses. Another point of criticism is the method for measuring learning styles. Most learning style models provide a questionnaire, where students are asked about their preferences with respect to the learning style model. These questionnaires raise several problems (Sabine, 2007).
Questionnaires, in general, have to deal with the problem that the given answers might not correspond to the real behavior the questions aim to investigate (Draper, 1996; Paredes and Rodríguez, 2004). The use of questionnaires in general and as an instrument for identifying learning styles is based on several assumptions.
Firstly, the assumption is made that students are motivated to fill out the questionnaire properly and to the best of their knowledge about their preferences.
Secondly, filling out a questionnaire about the preferred way of learning requires that the students are aware of their preferred way of learning. However, Stash, Cristea, and de Bra (2006), for example, identified that the Masters students participating in their study about adaptation to learning styles had only little meta-knowledge on their learning preferences, and Merrill (2002), for example, even argued that most students are unaware of their learning styles. Thirdly, social and psychological aspects such as the students' beliefs about how people should behave can influence their answers on the questionnaire.
Furthermore, using questionnaires for identifying learning styles trigger the assumption that the learning styles are stable for a long period of time. However, as discussed before, the stability of learning styles is still a debatable issue. As soon as learning styles change, the results of the questionnaires will not be valid anymore and students would have to do it again to identify their new learning styles. However, this argues will cause new issues, concerning with investigating how to spot when a learning style changed and how to motivate students to fill out the questionnaire a number of times.
Another issue is the validity and reliability of the questionnaires themselves. According to Coffield et al. (2004b), four criteria have to be fulfilled as a minimum standard for any instrument which is to be used to redesign pedagogy: construct validity, predictive validity, internal consistency reliability, and test-retest reliability. Construct validity means that the instrument actually measures the theoretical construct or trait that it purports to measure. Predictive validity refers to whether the range of behavior can be seen to have an impact on task performance. The internal consistency reliability refers to the homogeneity of the items intended to measure the same quantity that is the extent to which responses to the items are correlated. The test-retest reliability measures the extent to which an individual achieves the same result when performing the questionnaire twice within a specific period (e.g., one month). However, this test is based on the assumption that learning styles are stable, at least during the test period. Most learning style questionnaires are tested according to these criteria. However, instruments often lack one or several of these criteria, researchers achieve inconsistent results or even identify latent dimensions. Coffield et al. (2004b, p. 56) argued that from the 13 major learning style models they have identified and studied, only three of the models "could be said to come close to meet such criteria".
Another point that has to be highlighted, which is focusing on tailoring courses through identifying the learning styles without considering the complex sociopolitical forces in the larger society, 'personal warmth, trust and community'(Giroux, 1981:66), or the different perspectives of feminist and anti racial behavior, Laurillard's conclusion is more convincing. She writes:
"It would therefore be hazardous for an investigation of learning to proceed on the assumption that learning is a process that is independent of external factors, or those students' posses' inherent, invariant styles of learning". (1979:408)
That supported by Curry (1983) who proposes that learning style theories and their supporting instruments can be thought of in three levels, resembling layers of an onion. This model has " cognitive personality style" as relatively stable at the core, an intermediate and less stable layer of "information processing style" (Kolb LSI, for example), and an outer layer called "instructional format preference indicator" allowing for the individual's choice of learning environment. However, in MET ISSP as technology enhanced learning, setting and changing these three levels stated by Curry (1983) is possible.
From all these argumentations, the conclusion can be drained that questionnaires have to deal with several problems and restrictions. People who are using such questionnaires for identifying learning styles should therefore be aware of these problems and restrictions as well as consider the limitations of the questionnaires when interpreting the results. Since the proper identification of learning styles is a crucial issue, challenge is to develop an approach that measures learning styles more accurately and reliably, minimizing the extent to be affected or restricted by other factors. In Chapter XXX, the researcher will introduce an approach to conduct TEL based Maritime Education and Training onboard the training vessel, which aims at overcoming the above mentioned problems and restrictions of questionnaires.
Summarizing this section, it can be concluded that several debates and unsolved problems still exist in the field of learning styles. It seems that we are still far way from a model of learning styles that integrates all relevant aspects of learning styles and provides a clear understanding.
However, the debate and criticism of learning styles show challenges in the field, in addition to the lake of any previous studies about applying learning styles in the MET process. This thesis is an intervention that tackles some of the challenges and introduces new interactive TEL approach which contributes to get closer to solve some of the mentioned problems in the Maritime Education and Training and to be a mile stone in applying the learning styles in a VLE for marine cadets.
One of the most popular learning style inventories and one that is often used in distance learning and for adult research is the Kolb's Learning Style Inventory (Kolb, 1986; Dillie & Mezack, 1991; Dowdall, 1991; Diaz & Cartnal, 1999; Miller, 2005; Liegle & Janicki, 2006).