Education for Health

: 2018  |  Volume : 31  |  Issue : 2  |  Page : 136--137

Investigating postgraduate physician's learning style trend using multivariate statistical analysis

Abdelhakim Abdelhadi1, Samer Altalafha2,  
1 Department of Engineering Management, Faculty of Engineering, Prince Sultan University, Kingdom of Saudi Arabia
2 Department of Emergency Medicine, Royal Commission Health Service Program, Jubail Industrial City, Kingdom of Saudi Arabia

Correspondence Address:
Abdelhakim Abdelhadi
Department of Engineering Management, Faculty of Engineering, Prince Sultan University
Kingdom of Saudi Arabia

How to cite this article:
Abdelhadi A, Altalafha S. Investigating postgraduate physician's learning style trend using multivariate statistical analysis.Educ Health 2018;31:136-137

How to cite this URL:
Abdelhadi A, Altalafha S. Investigating postgraduate physician's learning style trend using multivariate statistical analysis. Educ Health [serial online] 2018 [cited 2022 May 26 ];31:136-137
Available from:

Full Text

Dear Editor,

Linking learning style to teaching strategy can be important in improving students' knowledge of subject matter. The Felder and Silverman Index of Learning Style[1] (ILS) consists of 44 questions to assess preferences on four sets of responses. ILS is available at the World Wide Web free of charge. Results range on scale of 1–11, with 11 indicating a high preference of active/reflective, sensing/intuitive, visual/verbal, and sequential/global learning style. Each learning style requires its own specific teaching approach.

By realizing that many problems are similar and by grouping similar problems, a single solution can be found to a set of problems.[2] This study looked at enhancing the teaching methodologies used by medical professionals by utilizing the concept of an engineering philosophy called Group Technology; identifying similarities among attributes of sets of objects and clustering them into cells.

Pairwise similarity coefficients between physicians and their learning style can be found using ILS output questionnaires. These similarities are then organized into a matrix. This matrix is used as an input to one of the clustering algorithms, such as complete-linkage clustering (CLINK), where groups are formed by merging nearest neighbors on the basis of the greatest similarities between them.[3] It works as follows:

Start with M clusters and an M × M symmetric matrix of similarities, D = {dik}Find the maximum similarity in D = {dik}Merge the corresponding objects U and V to get the cluster [UV]The similarity between [UV] and any other cluster Q is computed by

dUVQ = max {dUQ,dVQ}

The values dUQ and dVQ are the similarity between clusters U and Q and clusters V and Q, respectively.

The following is an example:

The study participants were 18 physicians selected from hospitals located in Saudi Arabia. Participants were given the link to answer the questions of ILS. Participant number one answers were as follows: active: 0, reflective: 5, sensening: 1, intuitive: 0, visual: 5, verbal: 0, and sequential: 0, global: 5. Participant number two answers: active: 0, reflective: 4, sensening: 1, intuitive: 0, visual: 8, verbal: 0, and sequential: 0, global: 2, and so on. Results were tabulated in a matrix.

Minitab software is used to develop clusters of physicians in a study based on their learning styles using complete linkage clustering. Results are presented in [Figure 1]. Participants 7, 8, 9, 18, 16, and 17 share the same learning style at about 33% similarity level; they have strong preference toward sensing and visual and moderate preference toward being sequential. On the other hand, participants 3, 4, 13, 10, and 12 belong to moderate sensing and visual and well-balanced toward being global at a similarity level about 60%.{Figure 1}

An instructor can use a specific teaching approach to convey the materials to the learning physicians.[4] Felder and Silverman suggested the following techniques to be used as teaching approaches to the sensing learners: Provide concrete information such as actual facts and data and use a problem-solving approach. The sample size of this study was limited to 18 physicians which opens the door to expand the study to much larger sample sizes for additional information on the technique.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Felder RM, Silverman LK. Learning and teaching styles in engineering education. Eng Educ 1988;78:674-81. Available from: http// [Last accessed on 2017 Jan 15].
2Abdelhadi A, Alwan LC, Yue X. Managing storeroom operations using cluster-based preventative maintenance. J Q Maint Eng 2015;21:154-70.
3Johnson R, Wichern D. Applied Multivariate Statistical Analysis. 6th ed. Upper Saddle River, NJ: Pearson Education; 2007.
4Kalaca S, Gulpinar M. A Turkish study of medical student learning styles. Educ Health (Abingdon) 2011;24:459.