An Approach for Non-deterministic and Automatic Detection of Learning Styles with Deep Belief Net

Manga, Maxwell Ndognkon; Ndjodo, Marcel Fouda

doi:10.1007/978-3-030-80126-7_32

Cited by 4 publications

(1 citation statement)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A static approach for learning style detection is carried out by filling out the questionnaires [15][16][17]. It is possible that some of the questionnaires will not be filled in properly, which may lead to inaccurately detecting the learning styles [18][19][20]. The other approaches are based on data-driven methods to build classification models that use the sample data and ILS.…”

Section: Introductionmentioning

confidence: 99%

ILS Validity Analysis for Secondary Grade through Factor Analysis and Internal Consistency Reliability

et al. 2022

View full text Add to dashboard Cite

In differentiated learning, the teacher needs to be aware of the learning styles of students in the classroom to accommodate specific learning preferences, e.g., the Felder–Silverman learning style model. The corresponding instrument, i.e., the Felder–Silverman Index of Learning Style (ILS), was designed to assess learning styles, specifically for engineering students. The ILS has been tested at the middle school level to identify the learning styles; however, validity/reliability had not been established in earlier studies with large samples. The focus of this study was to identify the validity and reliability of an ILS instrument for middle school students (N=450) by investigating the factor structure through factor analysis. This includes internal consistency reliability and constructing validity report of the ILS. An exploratory and confirmatory factor analysis was undertaken to investigate the factor structure to establish validity. As a result of the study, the reliability of the instrument was established. Five-factors emerged through exploratory factor analysis (EFA), which were subjected to confirmatory factor analysis (CFA). The outcome provided five-factors (i.e., Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA), Standardized Root Mean Residual (SRMR), and Goodness of Fit (GFI)), out of which four factors were related to the four dimensions of the Felder–Silverman model, and the fifth factor was related to the association of sensing/intuitive and sequential/global dimensions of the model, which is in agreement with the theoretical construct of ILS. As a result of CFA, ILS entailing 24 items indicates a good fit with five-factor structure. CFI=0.922;TLI=0.927; RMSEA=0.026; SRMR=0.585;GFI=0.911;X2=277;df=42;p=0.60. This study suggests that the ILS for the secondary-grade students needs to be revised with fewer items to improve the reliability, as supported by empirical evidence through the EFA and CFA.

show abstract