Assessing Engineering Student Perceptions of Introductory CS Courses in an Indian Context

Utsav Kumar Nareti, Divyansh Gupta, Chandranath Adak, Soumi Chattopadhyay, Emma Riese, Tanujit Chakraborty, Mayank Agarwal, Satendra KumararXiv 2025 · DOI · arXiv

Introduction

Introductory computer science (CS) serves as a critical "gatekeeper" course for nearly all engineering disciplines. Despite its importance, many students find the transition to programming challenging. While pedagogical research in CS is well-established in Western contexts, there is a significant data gap regarding the Indian engineering ecosystem—one of the largest and most competitive in the world. This study investigates the perceptions of over 300 students at an Indian Institute of Technology (IIT) to understand how they navigate assessments, the role of prior experience, and the cultural nuances of help-seeking behavior.

Methodology and Student Demographics

The researchers conducted a longitudinal study at IIT Patna during the Spring 2023 semester, surveying 318 first-year Bachelor of Technology (B.Tech) students. The cohort represented nine distinct academic branches, ranging from computing-focused fields like Computer Science and Engineering (CSE) and Artificial Intelligence (AI) to traditional disciplines such as Mechanical (ME) and Civil Engineering (CE).

Table 1: Student demographics across academic branches, including gender and self-reported prior experience levels.

As shown in the demographic breakdown, there is a notable disparity in prior programming experience (PCE/PAE) across different branches and genders. The study utilized a weekly 25-statement Likert-scale survey to track student sentiment. Given that the resulting data was non-normally distributed (confirmed via Q-Q plots), the authors employed non-parametric statistical methods, including the Mann-Whitney U and Kruskal-Wallis tests, to ensure rigorous analysis.

Analysis of Student Perceptions

The core of the study focused on four key themes: lab assignments, examinations, project work, and the impact of teaching assistants (TAs). The overall sentiment across these categories reveals a high level of engagement and a distinct cultural preference for human-led support.

Figure 1: Likert scale responses for statements S1 through S25, illustrating student agreement and disagreement across various course elements.

Labs as Learning Tools

One of the most striking findings was the perception of lab work. Unlike some Western cohorts who may view labs as a "necessary evil" (S2), Indian students overwhelmingly agreed that lab assignments were primary activities for learning course content (S1). This suggests that in the Indian context, hands-on lab work is viewed as a high-value learning experience rather than a mere administrative hurdle.

The Role of Teaching Assistants

The survey also highlighted a strong preference for seeking help from TAs over online resources (S23 vs. S24). While students in similar Swedish studies often preferred searching the internet, Indian students found TAs more approachable and helpful than lead instructors. However, students also noted significant inconsistencies in how different TAs assessed assignments (S17, S18), pointing to a need for more standardized grading rubrics.

The "Gaussian Myth" and the Digital Divide

A major contribution of this research is the challenge to the "Gaussian Myth"—the assumption that student grades should follow a normal bell curve. The data revealed that performance was non-normally distributed, suggesting that traditional "curving" of grades may be fundamentally inappropriate for CS education.

Furthermore, the study identified a persistent "digital divide." Prior programming experience was the single strongest predictor of success; students with no prior experience averaged a grade of 7.91, compared to 9.13 for those with high experience. This gap was also reflected in gender demographics, where female students reported lower prior technology exposure and achieved lower average GPAs (8.07) than their male counterparts (8.52).

Conclusion

This study underscores that introductory CS courses are not a level playing field. Success is heavily mediated by prior experience and academic branch, creating a divide that instructors must actively work to bridge. While the high value placed on lab work and TA support provides a strong foundation for learning, the lack of grading consistency and the reliance on prior knowledge suggest that more structured support is needed for students entering without a background in computing. Future work will explore how the integration of Large Language Models (LLMs) might further shift these perceptions and impact the digital divide.

Summary updated

BibTeX
@misc{nareti2025assessing,
  author = {Utsav Kumar Nareti and Divyansh Gupta and Chandranath Adak and Soumi Chattopadhyay and Emma Riese and Tanujit Chakraborty and Mayank Agarwal and Satendra Kumar},
  title = {{Assessing Engineering Student Perceptions of Introductory CS Courses in an Indian Context}},
  year = {2025},
  eprint = {2508.06563},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2508.06563},
}
Figure