Identifying Course Trajectories of High Achieving Engineering Students through Data Analytics

Omaima Almatrafi, Aditya Johri, Huzefa Rangwala, Jaime Lester2016 · DOI

Abstract: Dr. Johri studies the use of information and communication technologies (ICT) for learning and knowledge sharing. This research explores how data mining can reveal the specific academic paths that lead to student success or struggle in engineering programs.

Introduction

In higher education, Grade Point Average (GPA) is frequently utilized as the primary metric for predicting student retention. However, GPA is a lagging indicator, identifying struggles only after they have occurred. This study proposes that a more granular understanding of "academic paths"—the specific sequence and combination of courses—is necessary to proactively support students. By analyzing the trajectories of high- and low-achieving students, researchers can identify patterns that correlate with success.

Methodology

The study analyzed 13,950 records from 630 students at a large public university between 2009 and 2014, focusing on Computer Science (CS), Information Technology (INFT), and Civil Engineering (CEIE) majors. Two primary analytical methods were used:

  1. Frequent Pattern Mining (Apriori Algorithm): Identifying sets of courses frequently taken together in a single semester.
  2. Network Visualization: Creating graphs where nodes represent courses and edges represent transitions between semesters or concurrent enrollment.

Key Findings: The Path to Success vs. Stagnation

The "Heavy Lift" Trap

A significant finding was that low-achieving students (GPA < 3.0) often fall into a "heavy lift" trap. They frequently enroll in combinations of difficult courses simultaneously that high-achievers avoid. In Computer Science, for instance, low performers often attempted CS101, CS112, and MATH113 in the same term—a workload that proved systematically overwhelming.

Figure 4: Network visualization of course trajectories for low-performing Computer Science students, showing complex interconnections and potential 'heavy lift' combinations.

Bottlenecks and Strategic Postponement

High achievers tend to follow a tighter, more connected sequence, whereas low achievers often postpone "consequence courses" (prerequisites). This creates a domino effect. In the IT track, high achievers typically complete IT206 by their 4th semester, while low achievers often delay it until the 6th or 7th, significantly pushing back graduation.

Figure 6: Course trajectory network for low-performing Information Technology students, illustrating delayed progression and the impact of bottleneck courses.

The Cohort Advantage

In majors like CEIE, high achievers moved through the program as a cohesive cohort, evidenced by highly connected network graphs in their final years. Conversely, low achievers showed "enormous repetitions," indicating they were retaking classes and losing the support of a consistent peer group.

Conclusion

This research demonstrates that student success is not merely a matter of individual ability but is heavily influenced by the sequence of academic choices. Network visualizations provide a powerful tool for academic advisors to "flag" problematic pathways early. Future work will focus on identifying specific "bottleneck" courses more formally and designing prerequisite structures that prevent students from falling into high-risk trajectories.

Summary updated

Abstract

Dr. Johri studies the use of information and communication technologies (ICT) for learning and knowledge sharing
BibTeX
@misc{almatrafi2016identifying,
  author = {Omaima Almatrafi and Aditya Johri and Huzefa Rangwala and Jaime Lester},
  title = {{Identifying Course Trajectories of High Achieving Engineering Students through Data Analytics}},
  year = {2016},
  doi = {10.18260/p.25519},
}
Figure