Educational data mining approach for engineering graduate attributes analysis

Download
  1. (PDF, 756 KB)
  2. Get@NRC: Educational data mining approach for engineering graduate attributes analysis (Opens in a new window)
DOIResolve DOI: http://doi.org/10.24908/pceea.v0i0.6456
AuthorSearch for: ; Search for: ; Search for:
TypeArticle
Proceedings titleProceedings of the Canadian Engineering Education Association (CEEA) Conference, June 19-22, 2016, Halifax, NS, Canada
Series titleProceedings of the Canadian Engineering Education Association
ConferenceCanadian Engineering Education Association (CEEA) Conference, June 19-22, 2016, Halifax, NS, Canada
ISSN2371-5243
Article number138
Pages# of pages: 7
AbstractCurriculum improvement and graduate attributes assessments have become recently a serious issue for many Canadian engineering schools. Collecting assessment data concerning graduate attributes and the students’ learning is an important step of curriculum evaluation and the continuous improvement process. To be successful, this improvement process needs appropriate methods and tools for data analysis. Recent developments in the field of Psychometrics and Educational Data Mining (EDM) provide multidimensional item response models able to take into account student and curriculum attributes as parameters. The primary intent of these new models is to predict student successes based on students past performance and the assessment map underlying the tests they completed. We demonstrate in this paper that these models can also be used to analyze the assessment map. In the psychometric and Educational Data mining literature, assessment maps are usually represented as a parameter that associates items to competencies in a matrix called Q-matrix. This concept draws its origins from the Rule-Space Model that was introduced in the eighties to statistically classify student item responses into a set of ideal response patterns associated to different cognitive skills. A method based on the Additive Factor Model has been successfully implemented to analyse the Q-matrix corresponding to the assessment maps used in the graduate assessment process. The results of 17 volunteering anonymous students completing 36 courses at the Université de Moncton between winter 2010 and fall 2015 semesters was analysed with our method. Results obtained provided interesting and useful information regarding the assessment map and the overall assessment process that are presented and discussed in this paper.
Publication date
PublisherCanadian Engineering Education Association
LanguageEnglish
AffiliationInformation and Communication Technologies; National Research Council Canada
Peer reviewedYes
NPARC number23002242
Export citationExport as RIS
Report a correctionReport a correction
Record identifier27763f32-9b61-439f-be58-3063892297cc
Record created2017-09-13
Record modified2017-09-13
Bookmark and share
  • Share this page with Facebook (Opens in a new window)
  • Share this page with Twitter (Opens in a new window)
  • Share this page with Google+ (Opens in a new window)
  • Share this page with Delicious (Opens in a new window)
Date modified: