Using Association Rule Mining to Uncover Rarely Occurring Relationships in Two University Online STEM Courses: A Comparative Analysis

Item request has been placed! ×
Item request cannot be made. ×
loading   Processing Request
Read Online Read More Add to Saved list
  • Author(s): Valdiviejas, Hannah; Bosch, Nigel
  • Language:
    English
  • Source:
    Grantee Submission. 2020Paper presented at the International Conference on Educational Data Mining (13th, 2020).
  • Publication Date:
    2020
  • Document Type:
    Speeches/Meeting Papers
    Reports - Research
  • Additional Information
    • Peer Reviewed:
      Y
    • Source:
      5
    • Sponsoring Agency:
      Institute of Education Sciences (ED)
    • Contract Number:
      R305A180211
    • Education Level:
      Higher Education
      Postsecondary Education
    • Subject Terms:
      Metacognition; STEM Education; College Students; Online Courses; Educational Technology; At Risk Students; Disproportionate Representation; Minority Group Students; Student Characteristics; Research Methodology; Student Behavior; Computer Mediated Communication; Grades (Scholastic); Correlation
    • Abstract:
      Metacognition is a valuable tool for learning, particularly in online settings, due to its role in self-regulation. Being metacognitive is especially crucial for students who face exceptional difficulties in academic settings because it grants them the ability to identify gaps in their knowledge and seek help during difficult courses. Here we investigate metacognition for one such group of students: college students traditionally underrepresented in STEM (UR-STEM) in the context of two online university-level STEM courses. Using an automatic detection tool for metacognitive language, we first analyzed text from discussion forums of the two courses; one as a prototype and another as a replication study. We then used association rule mining to uncover fine-grained relationships in the online educational context between underrepresented STEM student status, online behavior, and self-regulated learning. In some cases, we inverted association rules to find associations for underrepresented minoritized students. Implications of the results for teaching and learning STEM content in the online space are discussed. Finally, we discuss the issue of using association rule mining to analyze commonly occurring patterns amongst an uncommon smaller subset of the data (specifically, underrepresented groups of students). [This paper was published in: "Proceedings of the 13th International Conference on Educational Data Mining (EDM 2020)" (pp. 686-690).]
    • Abstract:
      As Provided
    • IES Funded:
      Yes
    • Publication Date:
      2020
    • Accession Number:
      ED607169