The methodology chapter has one job: give another researcher enough information to replicate the study and get the same result. Everything in it — design choice, instrument, sample, analysis tool — needs a justification, not just a name.
Sign in to mark this article as read and track your progress.
Unit 1 established that genuine research must be valid and verifiable. The methodology chapter is where that criterion is either met or failed. A methodology section that names a design without justifying it, names a tool without naming its version, or describes a sample without explaining how it was selected fails the verifiability test — not because the underlying work was bad, but because the report doesn't give enough information to check.
The guiding question for every sentence in a methodology chapter is: could an independent researcher, reading only this chapter, repeat this study and expect to arrive at comparable results?
Research design. Name the design chosen — from the taxonomy in Unit 2 — and justify the choice. "A quasi-experimental design was chosen because random assignment to the treatment group was not feasible in the live SOC environment, and the closest available comparison group was the adjacent shift team operating on the unmodified ruleset." The justification is not optional. A design named but not justified leaves the examiner wondering whether the researcher considered alternatives.
Data collection instrument and procedure. Describe the instrument in enough detail for replication. For a questionnaire, include the full item list in an appendix and reference it here. For a sandbox experiment, name the sandbox software and version, the operating system image, the execution timeout, and the network isolation setting. For a code analysis tool, name the tool, version, and the ruleset applied. A statement like "the malware samples were executed and their behaviour was recorded" fails the test; "the samples were executed in Cuckoo Sandbox 2.0.7 on a Windows 10 22H2 guest image with network isolation active, for a 120-second observation window" passes it.
Sample. State the population, the sampling frame, the sample size, and the sampling technique, and justify the choice of technique using the reasoning from Unit 3. State any exclusion criteria: "samples smaller than 4 KB were excluded because files below that threshold in this repository are overwhelmingly non-executable headers rather than active payloads."
Analysis method. Name the statistical test or analytical procedure and explain why it suits the data. State the software and function used. Refer to Unit 3's guidance on test selection: if the outcome is continuous and two matched conditions are compared, a paired t-test is appropriate; if it is categorical, a chi-square test. "Accuracy was compared across three classifiers using a one-way ANOVA (scipy.stats.f_oneway, version 1.11.2), with significance level α = 0.05 set in advance."
A methodology chapter that sounds like every choice was optimal is less credible, not more. Real research involves trade-offs: a convenience sample because a probability sample was inaccessible; a two-week observation window rather than a six-month one because of time constraints; a sandbox rather than a live network because live execution of active malware would have required approvals that fell outside the project timeline.
Naming these trade-offs is not an admission of failure. It is what a limitations section is for, and a methodology chapter that anticipates its own limitations section is one that understands the difference between a study's design and the ideal experiment. Examiners routinely reward honest acknowledgement of trade-offs over a methodology section that implies the work was perfectly designed.
A methodology section reads: "The study used Python to analyse the data. Fifty malware samples were collected from VirusTotal. A t-test was run and the results are shown in Section 4." Identify every missing piece of information that the "valid and verifiable" standard requires, and rewrite the analysis sentence to the level of detail this standard demands.
