By chapter three, a lot of students start to relax a bit, thinking the hard part is already behind them after chapters one and two. In reality, this is often the chapter that decides whether your data will actually answer your research questions or not. A weak methodology chapter can force you to redo your whole data collection process later, so it is worth slowing down here instead of rushing through it.
Most Nigerian university formats expect research design, population of the study, sample size and sampling technique, sources of data, instrument for data collection, validity and reliability of the instrument, method of data analysis, and sometimes ethical considerations. Your department might add a section or rename one, but this is the core structure you are working with.
Your research design should match your research questions, not the other way round. If you are measuring relationships between variables using numbers, a quantitative design using a survey or questionnaire is the typical choice. If you are exploring experiences, perceptions, or processes in more depth, a qualitative design using interviews or case studies fits better. Some projects combine both. Whichever one you go with, briefly justify why it suits your specific objectives instead of just naming it and moving on.
Clearly define your population first, meaning the full group your study is really about, before you narrow down to your sample. If your population is quite large, like all small business owners in a state, explain how you arrived at your sample size. This is one of the places supervisors push back the most, asking students to justify their number instead of just picking a round figure that sounds reasonable.
A commonly used approach for unknown or very large populations is a formula like the Taro Yamane formula, which gives you a sample size based on your population estimate and an acceptable margin of error, usually five percent. If your population is small and known, say all the staff in one specific organization, you might be able to study the entire population instead of sampling, which is called a census study.
State clearly whether you used a probability technique, like simple random or stratified sampling, or a non probability technique, like purposive or convenience sampling, and explain why that particular technique fits your study. If your project involves specific subgroups, like comparing responses across different departments or age ranges, stratified sampling is usually a stronger choice than a simple random pick, since it makes sure each subgroup is properly represented.
Describe the tool you used to collect your data, most commonly a structured questionnaire for quantitative studies or an interview guide for qualitative ones. Briefly explain how the instrument is structured, for example how many sections it has and what each section is measuring, and tie each section back to your research objectives so it is clear the instrument was actually designed to answer your questions.
Validity is about whether your instrument actually measures what it claims to measure, often established through expert review by your supervisor or another lecturer in the field. Reliability is about consistency, often measured statistically using something like Cronbach's Alpha for questionnaires with multiple items measuring the same concept. If you ran a pilot test before your main data collection, mention it here, it strengthens your reliability claim quite a lot.
State clearly which statistical tools or software you used, like SPSS, and which specific techniques you applied, such as descriptive statistics, correlation, regression, or a t-test, and briefly connect each technique to the specific objective or hypothesis it was used to address. If you are not sure which test actually fits your data, our guide on choosing between t-test, ANOVA, correlation, and regression breaks it down in plain terms.
Choosing a research design that does not match the research questions
Picking a sample size with no formula or justification behind it
Describing the instrument without tying it back to your objectives
Skipping validity and reliability, or mentioning them without explaining how they were actually established
Naming a statistical test without explaining why it fits your data
A properly justified methodology chapter makes your chapter four data analysis far easier to defend later on. If you want expert eyes on your research design, sample size, or your full methodology chapter, my team at ProjectPal works with students across Nigerian universities on exactly this. Message us on WhatsApp with your topic and department to get started.
Send your topic and department on WhatsApp and get support built around your actual project.