In the field of education, there is a wide range of literature that addresses learning styles and the way they should be used in the classroom. In that regard, it can be seen that the constant progress in the educational field implies the occurrence of new methods of learning as well as new findings regarding the implementation of these styles in the classroom.
The issue of accommodating the learning styles and the instructions is also in focus, as the individualization of the teaching methods can be seen to take a separate direction in taking the students through the curriculum. Defining learning styles, it can be said that the wide range of definitions mostly ranges from the preferred sensory modalities to descriptions of personality characteristics, and cognitive information processing. (Smith & Renzulli, 1984, p. 44)
Taking the perspective of the wide array of learning styles definitions and the results of studying the correlation between learning styles, teaching instructions and the outcomes on students’ performance, an evaluation of the research methodologies implemented in each study might help to evaluate each study and accordingly the value of their evidence and findings. In that regard, this paper presents an evaluation of studies presented in four articles based on the methodologies and the design of each study.
The evaluation would be addressed to answer the question “Should instruction be matched to student learning style?”, as well as recommending the design and the method suitable to address such question, based on the results of the evaluation. The main evaluation criteria used in this paper are based on the consistency of the study as well as the appropriateness of research methods.
Research Problem
It should be noted that despite the fact that the studies presented had a common subject, the formulation of the research problem and the accordingly the conclusions made differed from study to study, as narrowing the questions to specific setting might change the statement of the study. Accordingly, the research questions and hypotheses formulations can differ according to the type of the study chosen for the research, i.e. quantitative, qualitative or mixed.
Generally, the research statement can be considered as the signpost which “establishes the central direction of the study” (Creswell, 2003, p. 105), after which the focus of the study can be narrowed to “specific questions to be answered or predictions (i.e. hypotheses) to be tested (Creswell, 2003)
In Adkins and Brown-Syed (2002), the main question posed in the study was related to the identification of the learning styles by the students and the accommodations necessary to make to exhibit styles and creating learning activities in practice. The explicit statement of the question was “whether information-oriented students are more likely than a control group to exhibit sequential and visual learning styles” (Adkins & Brown-Syed, 2002, p. 3). The study is of a descriptive type, and although the choice of using a control group might imply the establishment of a cause and effect relationship, the nature of the question posed implies a descriptive type of study.
The question complied with the chosen type of study, i.e. quantitative, where the authors used a hypothesis, i.e. “a prediction the researcher holds about the relationship among variables” (Creswell, 2003, p. 108), where the authors shaped and focused the purpose of the study. Additionally, the hypothesis can be stated as directional, as it specifies the direction that the researcher expected to emerge in the study. The hypothesis can be represented as “Library and Information Science (LIS) students are more likely to be oriented toward visual information and technology-based learning.
In Felder and Spurlin (2005), the questions posed in the study complied with the qualitative type, where the questions asked were in the form of associated subquestions, although the research implemented a mixed study approach. Due to the study being focused on Index of Learning Styles (ILS), “an instrument designed to assess preferences on the four dimensions of the Felder-Silverman learning style model” (Felder & Spurlin, 2005, p. 103), choosing a common ground for the evaluation of the studies led to selecting the closer question in context from the list of the subquestion stated in the study. The question is “What conclusions regarding the reliability and the validity may be inferred from the research studies, which were conducted using the ILS.” The reliability and the validity of the questionnaire was part of the study, where the authors’ assessment of the test-retest reliability, internal reliability and construct validity was a part of the study, as the scores obtained through the qualitative part of the study did not include such measurements.
A closer study question was formulated in McNeal and Dwyer (1999), in terms of studying the relation between the learning style, the instructions and the performance outcome. The study’s problem was stated as investigating “the relationship that exists between students’ expressed learning style and method of instruction… in terms of performance on specific criterion measures” (George & Francis, 1999, p. 337). Thus the study’s question was stated as “Whether congruence between learning style and instruction affect the performance of students.”
Finally, the problem established in Yildirim et al. (2008), investigates the differences in influence between the leadership style of the teachers and the learning style of the students, in terms of the influence on students’ performance outcomes. Similarly, the selection of a quantitative type of study implies the formulation of a question and a hypothesis, which nevertheless, was not explicitly stated in the study. Thus, the question that could be extracted based on the methodology implemented in the study was “Which factor affected students’ performance more: leadership of teachers or learning style of the student.” Accordingly, it can be seen that problems statement in this study did not include the instructions matching the learning style.
Literature Review
Basically, the literature review in a study is the section where the researcher refers to the efforts which already have been done in the topic, including the results of previous studies, and additionally, provides a background of the ideas used in other works. In that regard, the assessment of the literature review in a study can be based on the relevancy of the covered readings, clarity of the material, credibility of the literature and what most important, the review should indicate to the gaps in knowledge, which the conducted study should fill in.
In McNeal and Dwyer (1999), the literature review can be characterized through the reliance on mostly one source, where the focus on Kolb’s Learning Study Inventory was reflected in the review. In that regard, the lack of the extensive reading was seen through the reference to works accomplished on this topic by “most of the researches” (George & Francis, 1999, p. 337). In that regard, although the study merely utilized the previous work, i.e. the utilization and assessment of Kolb’s inventory, narrowing the area of research to nursing programs, it did not refer to other works to justify the need for such study.
Similarly, mostly Kolb’s work was cited in Felder and Spurlin (2005). However, as the study included a summary of previous works’ results and definition, the author referred to other researches, although they were spread throughout the paper. Nevertheless, the researcher did not distinguish between the research, theory, own results and opinion in the paper. On the other hand, in Yildirim et al. (2008), the literature review was considered in the beginning of the study setting the background of the knowledge used in the paper.
Additionally the review pointed out to the gaps in knowledge, e.g. “Given the lack of evidence in favour of matching teaching and learning styles” (Yildirim, Acar, Bull, & Sevinc, 2008, p. 74). Although the review can also be characterized by unselective referencing, i.e. using a long string of references for single information, indicating that the author was not selective in the works to cite, specifically for general information, e.g. “It is argued that students’ learning styles differ (Kolb, 1976, 1981; Marton & Saljo, 1976; Richardson, 1990)” (Yildirim, et al., 2008, p. 73).
In Adkins and Brown-Syed (2002), the literature review sets the background to the study’s problem statement, in terms of referring to the suggestions of the literature in that regard. However, the most of the reference of the work is taking place through the discussion section, where the results are compared to the earlier suggestions, without critiquing the earlier information or justifying the reasons for the new study.
The Research
As quantitative studies examine the relationship between variables in a measurable way, it can be seen that the selection of quantitative approach is mostly justified. In educational research related to methods, styles, and instructions, mostly the independent variable will be associated with performance, which accordingly is a measurable entity. In that regard, it can be seen that all of the studies implemented quantitative designs, except for the Felder and Spurlin (2005) study, in which a mixed methods was used. In the quantitative studies, or in quantitative parts of the studies, the problem was identifying the factors influencing the outcome, i.e. the dependent variable.
Accordingly, the dependent variable across the studies can be identified as follows:
- In Adkins and Brown-Syed (2002), the dependent variable is the score of xxx derived or measured by the Learning Styles Inventory.
- In Yildirim et al. (2008), the dependent variable is the academic achievement in the course.
- In Felder and Spurlin (2005), a mixed method was used, where the qualitative part was used to identify the variables through an analysis of the literature, while the quantitative part was used to assess the relation between the variables. In that regard, the dependent variable can be said to be the preferences of learning styles across students, and the strengths of such preferences.
- In McNeal and Dwyer (1999), the dependent variable measured is the test score of the achievement resulted from combining four individual criterion measures.
Additionally, all of the studies were cross sectional, i.e. the data were collected at one point of time. (Creswell, 2003) In general, the methods of research included in there of the articles can be classified as quantitative experimental and cross-sectional, where only in Felder and Spurlin was in the usage of a mixed method. The exception can be seen Felder and Spurlin (2005), where the study can be characterized as a descriptive type for the quantitative part, while it exploratory for the qualitative part. In that regard, the qualitative part was exploratory. The experiment part of the design implies that the researchers did intervene in the observation by altering one or more factors in the research and studied the effect of doing so.
Resolution
One of the main aspects of assessing a research study is the determination of its validity, i.e. investigating whether the study measuring what it is intended to measure. In that regard, the study should clearly indicate the measured aspects, and accordingly operationally define them if necessary.
Analyzing the internal validity of the studies, the main purpose is to investigate whether causal relation between the variables cannot be explained by a confounding variable which was not considered and controlled by the chosen research design. The confounding or extraneous variable serves as a threat to the internal validity of a study. This makes one difficult to conclude the true effect of an independent variable on the dependent or outcome variable. Accordingly, such criterion can be considered related to the consistency of the study, based on which it will be assessed. The issue of controversy can be seen through the work of Yildirim et al., where the perception of teachers’ leadership style indicated by the students can have several confounding variables such as personal attributes of the teacher, the course difficulty, and the success in other subjects. Similarly, in Adkins and Brown-Syed, the study involved perception in identification of learning styles, where being a cross-sectional study, the instant measurement can be accordingly confounded by several factors, e.g. the personal preferences of the students, the perception of specific learning style, educational history, and etc.
Another factor to consider, in terms of limitations is the instruments used in measurement and the possibility of the interpretation. Instrument decay can be defined as the possibility of changing the nature of the instrument and the scoring procedures, as a result of which a different interpretation of the result might occur. (Wallen & Fraenkel, 2001) In that regard, some of the instruments lacked the necessary description regarding the way the instrument were used , while for the modifications made to existent instruments, mentioned through the introduction or the literature review part, the nature of such modifications was not mentioned, e.g. inventory questionnaire in Adkins and Brown-Syed, and behavioral theories’ evaluation in Yildirim et al. The evidences of reliability can be seen through the usage of Cronbach’s alpha in Yildirim et al., in which the items lowering the Cronbach’s coefficient were excluded. Generally, it can be seen that the researcher used a value of over.60 for the item to be acceptable. Statistical analysis was used to establish the validity of that study. Additionaly, the method of sampling, with the exception of Yildirim et al., implied non-random selection, and random assignment, whereas in Adkins and Brown-Syed, the assignment was also non-random, all of which can be reflected through the generalizability of the findings, being related to the specific setting of the research.
On the other hand, in two other studies, the procedures and the instruments used were identified with various degrees of descriptiveness. This can be evident in the criterion measures, which were used to measure the achievements in McNeal and Dwyer, and the procedures used to measure the validity in Felder and Spurlin.
In terms of the appropriateness of the methods used, three of the articles, i.e. George, H. M., & Francis, D. (1999), “Effect of learning style on consistent and inconsistently designed instruction”, Yildirim et al.(2008), “ Relationships between teachers’ perceived leadership style, students’ learning style, and academic achievement”, and Adkins, D., & Brown-Syed, C. (2002), “Accommodating All Learners: Critical Inquiry and Learning Styles in the LIS Classroom”, examined the relation between the variables, where the independent variables were in some way or another related to the learning styles, instructions, the congruence between them, their perception and identification.
Dependent variable was always related to the performance and achievement, as mostly such indicator is the most important in assessing the effect and the success of the intervention made. In Felder, R. M., & Spurlin, J. (2005), “Applications, Reliability and Validity of the Index of Learning Styles”, the mixed approach was of analytical nature rather than examining the correlation between the variables.
Mixed method research is an “approach to inquiry that combines or associates both qualitative, [exploring and understanding the meaning individuals and groups ascribe to a social or human problem], and quantitative forms” (Creswell, 2003). Thus, the method used can be seen appropriate for the objectives stated in the study, determination of the dimensions of the ILS model (qualitative), testing the reliability and the validity of the studies conducted using ILS (quantitative).
In general, three of the studies, which were related to the examination of the relation between the learning styles and the performance, stated that a positive correlation exists.
Recommendation
Generalizing the studies used in the context of the question posed in this paper, i.e. should the instruction and the learning styles should be matched, certain modifications should be considered, based on the evaluation of the four researches on education. Thus, the problem of matching the learning styles with the instruction can be stated as a directional hypothesis that “matching the instructions to the learning style positively affects the performance of the students”.
Defining the variables, where the independent variables would be a defined scale rate of congruence between the style and the learning style and the dependent is the score of the students, an important step would the inclusion and the definition of any confounding variable s that might affect the process of measurement. In that regard, a quantitative longitudinal study, where the measures would be taken through a defined interval would minimize the interference of the confounding factors.
Additionally, an important aspect to consider is the validity and the reliability of the study, where based on the analysis of the articles, it could be recommended that the instruments of measurement would be tested by an independent education board, or through conducting a preliminary pilot study, after which modifications can be made to the measuring instrument, the sample or the variables included in the study. In terms of generalizability, the sampling techniques can be adjusted in case a random sample cannot be obtained through repeating the repetition of the study.
Conclusion
In the paper four articles in were evaluated, in order to have a representation of the factors to consider when researching the question of matching the instructions and the learning styles of the students. Three of the four articles were closely related to the aspects considered in this paper, and they were fairly consistent in their methods. The recommendations outlined were related to the area of validity, instruments used, and confounding factors. Generally, the methods used were appropriate for the problems stated, and they established a relation between the examined variables. The fourth article, with more distant approach to the theme examined, was nevertheless, exemplary in terms of examining the reliability and the validity of the results, in terms of definition, description and clarity. These considerations were outlined in the recommendations sector on how to develop the study for the main question formulated in the paper.
References
Adkins, D., & Brown-Syed, C. (2002). Accommodating All Learners: Critical Inquiry and Learning Styles in the LIS Classroom. Paper presented at the 68th International Federation of Library Associations and Institutions (IFLA).
Creswell, J. W. (2003). Research design : qualitative, quantitative, and mixed method approaches (2nd ed.). Thousand Oaks, Calif.: Sage Publications.
Felder, R. M., & Spurlin, J. (2005). Applications, Reliability and Validity of the Index of Learning Styles. International Journal of Engineering Education, 21(1), 103-112.
George, H. M., & Francis, D. (1999). Effect of learning style on consistant and inconsistently designed instruction. International Journal of Instructional Media, 26(3), 337.
Smith, L. H., & Renzulli, J. S. (1984). Learning Style Preferences: A Practical Approach for Classroom Teachers Theory into Practice 23(1), 44-50.
Wallen, N. E., & Fraenkel, J. R. (2001). Educational research : a guide to the process (2nd ed.). Mahwah, N.J.: Lawrence Erlbaum Associates.
Yildirim, O., Acar, A. C., Bull, S., & Sevinc, L. (2008). Relationships between teachers’ perceived leadership style, students’ learning style, and academic achievement: a study on high school students. Educational Psychology, 28(1), 73-81.