Research Methodology Viva Questions

export const frontmatter = { title: "Research Methodology Viva Questions", description: "Common viva voce questions on research methodology with model answers for thesis defense", keywords: ["viva questions", "thesis defense", "oral examination", "research methodology", "academic interview"] };

The viva voce (oral examination) is where you defend your research methodology choices to expert examiners. Unlike written answers, viva responses require you to think on your feet, justify decisions, acknowledge limitations honestly, and demonstrate deep understanding rather than memorized definitions. This guide prepares you for the methodology-focused questions most commonly asked during thesis defenses.

Questions About Your Research Design

"Why did you choose this particular research design?"

How to answer: Connect your design choice directly to your research questions.

Example response: "My research questions sought to examine relationships between workplace flexibility and retention—specifically, whether different flexibility types predict retention differently. This required measuring these variables across a large sample and testing statistical associations, which necessitated a quantitative, cross-sectional survey design. A qualitative approach would have provided depth but not the breadth needed to test these specific hypotheses. An experimental design was impractical because I could not randomly assign employees to flexibility conditions in real organizations."

Key principle: Show that your design was a deliberate, reasoned choice—not the default or the easiest option.

"What are the limitations of a cross-sectional design?"

Honest answer: "The primary limitation is inability to establish causation. My findings show that flexibility is associated with retention intention, but I cannot determine whether flexibility causes higher retention, whether naturally loyal employees select into flexible arrangements, or whether a third variable drives both. A longitudinal design measuring flexibility at Time 1 and retention at Time 2 would strengthen causal inference. I acknowledge this limitation in Chapter 5 and recommend longitudinal follow-up research."

"Why didn't you use a mixed-methods approach?"

Response: Justify your single-method choice without being defensive. "Mixed methods was considered but not adopted for two reasons. First, my research questions were specifically about measuring and testing relationships—quantitative questions that a qualitative component would complement but not answer more directly. Second, practical constraints of a 12-month timeline made adequate depth in both paradigms unfeasible. A superficial qualitative component would have weakened rather than strengthened the study."

Questions About Sampling

"How did you determine your sample size?"

Strong answer: Show the calculation transparently. "I used G*Power software with the following parameters: multiple regression with 5 predictors, anticipated medium effect size f²=0.15 based on Kumar et al.'s (2022) findings, power=0.80, and α=0.05. This yielded a minimum requirement of 92 participants. I targeted 250 to account for the expected 30% non-response rate in email surveys, ultimately obtaining 212 usable responses—comfortably exceeding the minimum."

"Your sample is from only one city. How can you generalize?"

Honest acknowledgment: "You are correct that generalizability beyond Bangalore is limited. I acknowledge this in my limitations section. However, Bangalore represents approximately 35% of India's IT workforce and includes companies of all sizes and ownership types—providing reasonable diversity within this geographic constraint. I recommend replication in other IT hubs (Hyderabad, Pune, Chennai) as future research."

Questions About Instruments

"Why did you use this particular questionnaire?"

Response: "I selected the Minnesota Satisfaction Questionnaire (short form) based on four criteria: (1) it has been extensively validated cross-culturally, (2) prior Indian studies reported good reliability (α = 0.82-0.91), (3) it distinguishes between intrinsic and extrinsic satisfaction dimensions relevant to my framework, and (4) its 20-item length minimized respondent fatigue in a multi-measure survey."

"What was the reliability in YOUR study?"

Always report from YOUR data, not just the original: "In my study, overall α was 0.86, with intrinsic satisfaction at 0.83 and extrinsic satisfaction at 0.79—all above the 0.70 threshold and consistent with prior studies."

Questions About Analysis

"Why did you use regression instead of ANOVA?"

"My research questions involved both categorical predictors (work arrangement type) and continuous predictors (age, experience) predicting a continuous outcome (satisfaction). Regression accommodates both predictor types simultaneously and allows testing mediation, which ANOVA alone cannot. I also ran ANOVA as a supplementary analysis to confirm group differences, reporting results in Appendix D."

"How did you handle missing data?"

"Missing data was minimal (3.8% overall). Little's MCAR test confirmed data was missing completely at random (χ²=34.2, df=28, p=.19). I used listwise deletion for the primary analysis (reducing n from 212 to 204) and verified results were unchanged using multiple imputation as a sensitivity check."

"What about common method bias?"

"Since all variables were measured through the same self-report survey at a single time point, common method variance is a legitimate concern. I addressed this through: (1) Harman's single-factor test—one factor explained only 28% of variance, below the 50% threshold; (2) procedural remedies—assured anonymity, separated predictor and outcome sections within the survey, and used different scale formats; and (3) I acknowledge this as a limitation and recommend future studies use multi-source data."

Difficult Questions

"If you could do this study again, what would you change?"

Be honest and specific—this shows reflective maturity: "Three things: First, I would add a Time 2 measurement (6 months later) to capture actual turnover rather than just intention. Second, I would include objective productivity data alongside self-reports to address common method concerns. Third, I would supplement the survey with 10-15 interviews to understand WHY hybrid work produces the highest satisfaction—the quantitative data shows the pattern but not the mechanism in detail."

"How do you respond to the criticism that your effect size is small?"

"The partial eta-squared of .040 is indeed small by conventional standards. However, I would offer three contextual points: First, in organizational research, effects above .02 are considered practically relevant given the complexity of human behavior. Second, at scale—with millions of IT employees across India—even a small effect translates to meaningful differences in well-being across thousands of workers. Third, the effect was consistent across subgroups and robust to alternative analytical specifications, suggesting it is a genuine, if modest, phenomenon."

Tips for Viva Success

1. Know your thesis intimately — Reread it completely before the viva
2. Anticipate weaknesses — Prepare honest, constructive responses to likely criticisms
3. Think aloud — Examiners want to see your reasoning process, not just conclusions
4. It is okay to say "I don't know" — But add "however, if I were to explore this, I would..."
5. Be confident but not defensive — Accept valid criticisms gracefully
6. Bring your thesis — It is acceptable to reference specific pages during discussion

Conclusion

A successful viva defense demonstrates that you understand not just what you did but WHY you made each choice, what the alternatives were, and what the limitations imply. Prepare by critically examining your own work as an external examiner would—identifying every vulnerability and preparing thoughtful, honest responses.