Generative AI and Academic Integrity in Universities
Notice: This is a sample project for study and reference. Submitting it as your own work violates most universities' academic integrity policies.
Abstract
About This Research Topic
Ask any lecturer what has changed most about student writing in the last two years, and the answer usually circles back to the same tool: ChatGPT. Within a single academic session, generative AI moved from a novelty to a near-default study companion for many undergraduates, sitting alongside textbooks and lecture notes rather than replacing them outright — at least in theory. In practice, the line between assistance and substitution has become difficult to police, and universities across Nigeria are only beginning to catch up. A recent study of undergraduate students in a public university in Rivers State puts numbers to what many educators have suspected: heavier generative AI use tracks with lower self-reported critical thinking scores, and it is a meaningful predictor of academically dishonest practices. If you are a student, supervisor, or researcher trying to make sense of your own findings on this topic, it can help to see how a similar chapter one is structured — you can browse the sample research library for comparable methodology and framing before you finalise your own. This article walks through that Rivers State study in full: what it measured, what it found, and what it means for how Nigerian universities write policy for the AI era.
Main Abstract
Generative AI tools — ChatGPT, Gemini, and Copilot chief among them — have moved from the margins to the mainstream of undergraduate academic life in a remarkably short window. This study set out to measure that shift among undergraduates at a public university in Rivers State, focusing specifically on two outcomes educators worry about most: independent critical thinking and academic integrity. Guided by five objectives, five research questions, and three hypotheses, the researcher surveyed 300 students drawn by stratified random sampling from the faculties of Education, Social Sciences, and Engineering, using a five-point Likert-scale questionnaire; 268 responses were usable. The results, analysed with SPSS version 26 using both descriptive statistics and inferential tests (Pearson correlation and simple linear regression), showed a statistically significant negative relationship between the extent of generative AI use and students' self-reported critical thinking (r = -0.41, p < 0.05). Generative AI use also predicted academic integrity concerns, accounting for roughly 29% of the variance in self-reported dishonest academic practices. Beyond the numbers, the study found that institutional policy on generative AI is thin and inconsistently enforced, and that students themselves are uneasy about AI hallucination, unfair advantage in grading, and unequal access to paid AI tools. The overall conclusion is a balanced one: generative AI has genuine academic value, but using it without disclosure or institutional guardrails puts both independent thinking and academic honesty at risk. The study recommends clearer institutional policy, assessment redesign that is harder to outsource to AI, and deliberate AI-literacy and ethics teaching within the curriculum.
Chapter One Preview
Background to the Study
Generative AI — large language model systems capable of producing human-like text, code, and other content from a simple prompt — became a mainstream classroom presence almost overnight following the public release of ChatGPT in late 2022. Within months, university students worldwide were using such tools for essay drafting, literature summarising, brainstorming, and step-by-step problem solving. The appeal is easy to understand: generative AI can explain a difficult concept on demand, suggest a structure for an argument, or turn a vague idea into a workable outline in seconds.
That convenience is exactly what worries many educators. Critical thinking — the ability to analyse, evaluate, and synthesise information into a reasoned, independent judgment — is one of the outcomes higher education is supposed to build. If students routinely offload that reasoning work to an AI system rather than doing it themselves, the underlying skill may simply not develop the way it would through effortful practice. UNESCO's global guidance on generative AI in education makes a similar point at the policy level, calling for a human-centred approach that keeps students actively reasoning rather than passively consuming AI output.
The second concern is more direct: academic integrity. Generative AI can produce a complete, plausible-sounding essay or assignment response in the time it takes to type a prompt, and many institutions still lack a clear position on when that constitutes dishonesty. This matters even for students who never intend to cheat outright, since inconsistent expectations and detection tools can produce results that look suspicious for reasons that have nothing to do with copying — a pattern explored further in this site's piece on why a project can score high for plagiarism even when nothing was copied. In Nigerian public universities, including the one studied here, policy and curriculum design have generally lagged well behind the pace at which students have adopted these tools — which is precisely the gap this study was designed to measure.
Statement of the Problem
Generative AI use among Nigerian undergraduates — including those in Rivers State — has grown far faster than the institutional response to it. Lecturers report, anecdotally, a decline in independent analytical writing and a corresponding rise in suspected AI-generated submissions, but hard evidence to confirm or quantify that impression has been limited. Meanwhile, most Nigerian universities, this one included, still have no clearly communicated policy on what counts as acceptable generative AI use in coursework. That vacuum leaves students and lecturers to work out the rules informally, opens the door to unfair advantage for students who use AI undisclosed, and puts pressure on academic standards more broadly. This study was designed to replace anecdote with evidence — measuring the real relationship between generative AI use, critical thinking, and academic integrity, and mapping how aware students actually are of whatever institutional guidance exists.
Aim and Objectives of the Study
The study aims to examine generative AI's impact on critical thinking and academic integrity among undergraduate students in a selected public university in Rivers State. Specifically, it set out to:
● Examine the extent of generative AI use among undergraduate students in their academic work
● Determine the effect of generative AI use on students' critical thinking skills
● Assess the effect of generative AI use on academic integrity practices among students
● Evaluate students' awareness of institutional policies on generative AI use
● Identify the challenges and concerns associated with generative AI use in academic work
Research Questions
● What is the extent of generative AI use among undergraduate students in their academic work?
● What is the effect of generative AI use on students' critical thinking skills?
● What is the effect of generative AI use on academic integrity practices among students?
● What is the level of students' awareness of institutional policies on generative AI use?
● What challenges and concerns are associated with generative AI use in academic work?
Significance of the Study
Empirical Nigerian research on generative AI in higher education is still thin on the ground, so this study adds directly to a literature that badly needs local evidence rather than borrowed assumptions from Western contexts. For university management and policymakers, the findings offer a concrete starting point for drafting generative AI policy that reflects how students in Nigerian public universities are actually using these tools, rather than guessing. Lecturers stand to benefit too: understanding the real extent and pattern of generative AI use can inform assessment redesign that is harder to complete with AI alone, while still leaving room for its legitimate study benefits. Students working through their own research — particularly at the proposal or chapter-one stage — often find it useful to get outside feedback on structure and methodology before submission; one-on-one research coaching can strengthen that original work without becoming a substitute for it. Finally, the study serves as a reference point for other researchers examining generative AI, critical thinking, or academic integrity in higher education.
Scope of the Study
The study is delimited to undergraduate students in three purposively selected faculties — Education, Social Sciences, and Engineering — within one public university in Rivers State. It covers four dimensions: the extent of generative AI use, its effect on critical thinking skills, its effect on academic integrity practices, and students' awareness of institutional policy. The period under review runs from 2023 to 2026, capturing the window of rapid generative AI adoption in higher education.
Operational Definition of Terms
Generative Artificial Intelligence (Generative AI): a class of AI systems capable of producing human-like text, code, or other content in response to user prompts, such as ChatGPT, Gemini, and Copilot.
Critical Thinking: the capacity to analyse, evaluate, and synthesise information in order to form well-reasoned judgments and independent arguments.
Academic Integrity: adherence to ethical principles of honesty, originality, and fairness in academic work, including proper acknowledgement of sources and avoidance of unauthorised assistance — a definition consistent with the six fundamental values set out by the International Center for Academic Integrity.
Academic Dishonesty: any act that violates the principles of academic integrity, including submitting AI-generated content as one's own original work without disclosure.
Cognitive Offloading: the practice of using an external tool, such as generative AI, to reduce the mental effort otherwise required to complete a task.
Institutional Policy Awareness: the extent to which students are informed of, and understand, their institution's rules governing acceptable generative AI use in academic work.
Conclusion
The picture this study paints is not a simple ban-it-or-embrace-it choice. Generative AI clearly offers real academic value — on-demand explanation, brainstorming support, and faster first drafts — but the data show that unregulated, undisclosed use comes with a measurable cost to independent critical thinking and to academic honesty. What Nigerian universities do next matters more than the technology itself: clear policy, assessment that is genuinely AI-resilient, and deliberate teaching of AI literacy and ethics are the levers most likely to keep the benefits while containing the risks. For students and researchers working on similar topics, getting the structure of chapter one right — objectives, hypotheses, scope, and definitions clearly stated — is half the battle, and it's worth studying a few well-built examples before you start writing your own.
Frequently Asked Questions
1. Does using ChatGPT for assignments count as academic dishonesty?
It depends on your institution's policy and on disclosure. Using generative AI to brainstorm or check understanding is generally different from submitting AI-generated text as your own original work without acknowledgement — the latter is the practice most institutions and integrity bodies treat as dishonest.
2. Does generative AI actually reduce critical thinking skills?
This study found a statistically significant negative relationship between the extent of generative AI use and students' self-reported critical thinking (r = -0.41, p < 0.05), suggesting heavier, less critical use is associated with weaker self-reported reasoning skills — though self-reported measures have their own limits.
3. How much of academic dishonesty can generative AI use explain?
In this sample, generative AI use accounted for roughly 29% of the variance in self-reported academically dishonest practices — a meaningful share, though it leaves most of the variation explained by other factors.
4. Do Nigerian universities have clear policies on generative AI use?
Not yet, in most cases. This study found generally low policy clarity and enforcement at the institution examined, a gap that mirrors the wider Nigerian higher education landscape.
5. What faculties were covered in this study?
Education, Social Sciences, and Engineering, purposively selected from a public university in Rivers State, with 300 students sampled and 268 valid responses analysed.
6. What is AI hallucination, and why does it worry students?
Hallucination refers to generative AI producing confident-sounding but inaccurate or fabricated information. Students in this study flagged it as a real concern, since relying on hallucinated content can quietly damage the accuracy of academic work.
7. Is generative AI use the same as plagiarism?
Not exactly. Plagiarism is presenting someone else's work as your own; undisclosed AI-generated content raises a related but distinct integrity question, since the 'someone' is a model rather than a person — but many institutions still treat undisclosed AI use as a form of dishonesty.
8. Can generative AI be used ethically in research and coursework?
Yes — for brainstorming, explaining concepts, checking grammar, or summarising your own notes, with clear disclosure where your institution requires it. The concern in this study is undisclosed substitution, not assistance itself.
9. What can universities do to make assessments more AI-resilient?
This study recommends redesigning assessments to rely more on in-class reasoning, oral defence, and process documentation — methods that are harder to complete purely through AI output — alongside clearer policy and AI-literacy teaching.
10. Where can I see how a study like this is structured for my own project?
You can review comparable Education-department studies, including their objectives, hypotheses, and methodology, in the sample research library for reference and structural guidance.
Purchase to unlock the full material.