How to Write Research Methodology and Discussion Section: A Brief Overview

A strong research methodology and discussion section hinges on one core move: connect every design choice and every result to “why it matters,” not just what was done or found. Methodology must clearly define the study’s population and sampling plan—target population, sample, sample size, and sampling technique—then document where the questionnaire came from, how it was distributed, how data were collected, and which analysis methods were used. The discussion section then starts by restating the study objectives so readers can track the purpose, followed by results that are compared with prior research, with the critical requirement that significance (or lack of it) is explained rather than merely reported.

Sampling is treated as a central methodological dilemma, especially in Pakistan where probability sampling is often treated as a requirement for PhD-level work. The transcript challenges that assumption, arguing that probability sampling can be extremely difficult to implement in practice, and that convenience sampling is commonly used in published papers—provided it can be justified. Still, if probability sampling is feasible and can be defended, it can be used. The key is not blind adherence to a preferred technique, but transparent justification tied to the study’s setting and feasibility.

Questionnaire handling is another non-negotiable detail. The methodology should specify the questionnaire’s sources: who developed it, how many items it contains, whether sample items or the full instrument appears in the appendix, and whether any secondary data sources were used. It should also describe the data collection process—such as whether questionnaires were distributed online during COVID-19 or collected in person—and name the analysis techniques used. For theses, the transcript emphasizes adding the advantages of the chosen analysis approach, not only listing the method. It also highlights a common comparison in quantitative research: covariance-based structural equation modeling (CB-SEM) versus variance-based structural equation modeling (PLS-SEM), often labeled as covariance-based SEM (CBM) and variance-based SEM (VBM) in writing.

In the discussion, results must be tied to theory and to practical meaning. The transcript gives examples of how to move from “significant relationship” to “why significant”: internal marketing can improve customer satisfaction by first improving employee satisfaction, which then supports company performance—so the “why” chain matters. When culture moderates the internal marketing practice–employee satisfaction relationship in tourism and hospitality, the implication is that cultural congruence should be considered in both research and practice to strengthen that relationship. Similarly, findings should be interpreted in light of theory; for instance, CSR initiatives can strengthen employee organizational identification and support organizational performance through the social identity perspective.

Finally, the conclusion must align with objectives and include limitations, future research directions, and—most importantly—strong implications. Weak implications can lead to rejection in high-quality journals. The transcript urges authors to articulate theoretical, practical, policy, social, and organizational implications with focus and strength, showing exactly how the study benefits individuals, management, organizations, and society.