By Micah Asamba, Director, Keynea Research and Analytics.
Academic research in Kenya, particularly at the Master’s and PhD levels, sits at a complex intersection of ambition, institutional expectation, and structural constraint. Over the past decade, postgraduate enrolment has expanded across universities, yet the systems that support research have not scaled proportionately. The result is a landscape where many capable students struggle to convert ideas into rigorous, timely, and publishable work.
The Structural Bottlenecks
Several persistent constraints shape postgraduate research outcomes.
First, supervisory load remains high. In many institutions, a limited number of experienced supervisors oversee large cohorts. This reduces the frequency and depth of feedback. Students often receive delayed or fragmented guidance, which affects momentum and clarity.
Second, there is a skills gap in research design and analysis. Many students enter postgraduate programmes with strong theoretical grounding but limited exposure to applied methods. Challenges emerge in operationalizing variables, selecting appropriate analytical techniques, and interpreting results with precision.
Third, access to resources is uneven. Subscription databases, statistical software, and specialized tools are not consistently available. Where access exists, training is often insufficient. This creates a dependence on secondary summaries rather than direct engagement with primary literature.
Fourth, formatting and institutional requirements introduce another layer of complexity. Universities maintain strict guidelines on structure, referencing, and presentation. Students spend considerable time reconciling content with formatting standards, often at the expense of deeper analytical work.
Finally, there is the issue of time pressure and competing responsibilities. Many postgraduate students are working professionals. Balancing employment, family obligations, and research leads to fragmented effort and prolonged completion timelines.
The Entry of Artificial Intelligence into Research
Artificial intelligence has entered this environment as both an opportunity and a source of concern. Tools such as large language models can assist with:
- Structuring research proposals
- Summarizing literature efficiently
- Refining language and clarity
- Supporting data interpretation frameworks
- Enhancing productivity in drafting and revision
For a system constrained by time, access, and supervision, these tools can reduce friction. They can help a student move from conceptual uncertainty to a structured draft more quickly. They can also improve the technical quality of writing, especially for those who struggle with academic expression.
However, the introduction of AI does not remove the underlying demands of research. It shifts them.
The Risk Landscape
The primary risk is over-reliance without understanding. When AI is used to generate content without critical engagement, the result is often superficially coherent but conceptually weak work. Arguments may lack grounding in context. Citations may be inaccurate or unverifiable. Methodological sections may appear structured but fail under scrutiny.
A second risk is loss of intellectual ownership. Research is not merely a product. It is a process of inquiry, reflection, and interpretation. If this process is outsourced entirely to automated systems, the researcher loses the opportunity to develop analytical depth.
A third concern is integrity and originality. Universities increasingly scrutinize work for authenticity. The presence of AI-generated patterns, unsupported claims, or inconsistent voice can raise questions about authorship.
Responsible Use of AI in Research
The question is not whether AI should be used. It is how it should be used.
A responsible framework involves several principles:
1. AI as an assistant, not an author
AI should support tasks such as structuring ideas, refining language, or organizing content. The core arguments, interpretations, and analytical decisions must remain with the researcher.
2. Verification of all outputs
Any information generated by AI must be checked against credible sources. Citations should be traced to original publications. Data points must be validated.
3. Contextual grounding
Research in Kenya requires sensitivity to local realities. AI models are trained on global datasets and may not reflect local nuances. The researcher must ensure that all analysis is grounded in the relevant context.
4. Transparency in process
Where AI tools significantly support drafting or analysis, this should be acknowledged where appropriate. Transparency strengthens credibility.
Retaining Authenticity and Originality
Authenticity in research is not defined by the absence of tools. It is defined by the presence of independent thought.
To retain originality:
- Engage deeply with primary literature rather than relying on summaries
- Develop a clear conceptual framework before drafting
- Write initial drafts in your own words before refining
- Use AI to improve expression, not to replace reasoning
- Maintain a consistent voice throughout the document
Originality also emerges from contextual insight. Kenyan research gains strength when it reflects local data, institutional realities, and policy environments. No automated system can substitute for lived context and field engagement.
The Way Forward
Improving postgraduate research outcomes in Kenya requires both systemic adjustment and individual discipline.
At the institutional level, there is need for:
- Strengthened methodological training
- Reduced supervisory overload
- Improved access to research tools and databases
- Clearer alignment between expectations and support
At the individual level, researchers must:
- Take ownership of their work
- Build methodological competence
- Use tools strategically rather than dependently
- Prioritize clarity, evidence, and structure
Artificial intelligence will continue to shape the research landscape. Its value will depend on how it is integrated into scholarly practice. When used with discipline, it can enhance productivity and quality. When used without control, it can undermine both.
The goal is not to resist change, but to retain the core principles of research: rigor, originality, and intellectual honesty.
Micah Asamba
Director, Keynea Research and Analytics
Supporting structured, credible, and high-quality academic research in Kenya.
