Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.

Using Generative AI in Higher Programming Education: An Empirical Evaluation

Open access
Grewe, Dennis, Vedislav, Raluca-Maria, Scherz, W. · 2026 · Proceedings of the 18th International Conference on Computer Supported Education

Abstract

The growing use of generative artificial intelligence (GenAI) tools like ChatGPT presents both opportunities and challenges for programming education, especially for novices. While these tools can assist with code generation, debugging and conceptual understanding, there is a risk that they will promote superficial learning and the uncritical adoption of generated code. This paper presents the development and empirical evaluation of a best-practice guideline designed to support students in the reflective and effective use of generative AI in programming contexts. We conducted an experimental study with 36 first-semester computer science students, comparing a control group using AI without guidance against an experimental group equipped with our structured guide. The results show that while the quantitative performance metrics remained similar between the groups, the students using the guideline demonstrated a significantly higher self-reported understanding of their solutions and exhibited more reflective AI interaction patterns. Observational data revealed that guided users critically evaluated (66.7% vs 55.6%), modified (61.1% vs 22.2%), and rejected (38.9% vs 16.7%) AI-generated suggestions more frequently. A supplementary AI guidance workshop conducted provided convergent validation: students independently identified debugging and conceptual understanding as their primary skill gaps requiring AI support.

Overview

CoChat Summary

## TL;DR

This paper evaluates a best-practice guideline for using generative AI in programming education with 36 first-semester computer science students. The guideline did not change quantitative performance, but it was associated with higher self-reported understanding and more reflective AI use.

## Key findings

  • In an experimental study with 36 first-semester computer science students, the control group used AI without guidance and the experimental group used a structured guide.
  • Quantitative performance metrics remained similar between the two groups.
  • Students using the guideline reported significantly higher self-reported understanding of their solutions and showed more reflective AI interaction patterns.
  • Guided users more often critically evaluated AI suggestions (66.7% vs 55.6%), modified them (61.1% vs 22.2%), and rejected them (38.9% vs 16.7%).

## Methods

  • The paper developed a best-practice guideline for reflective and effective use of generative AI in programming contexts.
  • It conducted an experimental study with 36 first-semester computer science students, comparing a control group using AI without guidance to an experimental group using the guide.
  • The main analysis compared quantitative performance metrics, self-reported understanding, and observed AI interaction patterns; a supplementary AI guidance workshop was also used for convergent validation.

## Limitations

  • The paper does not report any explicit limitations in the provided text.
  • Evident from the text: the study sample was small (36 students) and limited to first-semester computer science students, so generalizability may be constrained.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.5220/0014726700004021

Cites

0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 Valid · 0 Not valid · 0 Not checked · 8 Own results and interpretations

3 more claims

Create a free account to see all 8 claims and every source verdict.

See all claims — free account

Citations

Cites

The publisher hasn’t deposited this paper’s reference list.

Cited by

Still being checked.

Metadata

Authors
Grewe, D., Vedislav, R.-M., & Scherz, W.
Affiliations
  • Esslingen University of Applied Sciences (DE)
  • HTWG Hochschule Konstanz - Technik, Wirtschaft und Gestaltung (DE)
Venue
Proceedings of the 18th International Conference on Computer Supported Education
Published
2026
Pages
1451-1458
OpenAlex
W7161961719
Cited by
0
References
0
FWCI
0.00
Type
conference paper
Language
en
Open access
Licence
cc-by-nc-nd
Locations
  • International Conference on Computer Supported Education (publishedVersion, cc-by-nc-nd)
  • URN-Resolver at the German National Library (German National Library) (submittedVersion, cc-by-nc-nd)
Topics
  • Physical Sciences › Computer Science › Computer Science Applications › Teaching and Learning Programming
  • Social Sciences › Education › Artificial Intelligence in Education
  • Physical Sciences › Computer Science › Information Systems › Software Engineering Research
Keywords
generative AI in education, programming education, ChatGPT, novice programmers, best practice guidelines, code generation, debugging, conceptual understanding

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Using Generative AI in Higher Programming Education: An Empirical Evaluation. https://app.cochat.ai/papers/10.5220/0014726700004021
The paper
Grewe, D., Vedislav, R.-M., & Scherz, W. (2026). Using Generative AI in Higher Programming Education: An Empirical Evaluation. Proceedings of the 18th International Conference on Computer Supported Education. https://doi.org/10.5220/0014726700004021

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.