Why These Questions Are Almost Guaranteed
If your thesis is quantitative, expect to be asked how you ensured validity and reliability — these are among the most predictable questions in a Malaysian quantitative viva. Examiners ask because the credibility of every statistical finding rests on whether the instrument measured what it claimed to measure and measured it consistently. Many candidates lose ground here not because their study was weak but because they confuse the two concepts under pressure, or recite textbook definitions without connecting them to their own instrument. Preparing a clear, study-specific account of both, in plain language, is one of the higher-value pieces of viva preparation you can do.
Explaining Reliability in Your Study
Reliability is consistency — whether your instrument produces stable results. Explain it through what you actually did rather than in the abstract. The most common evidence is internal consistency reported as Cronbach’s alpha: “Each construct in the questionnaire returned a Cronbach’s alpha above 0.70, indicating acceptable internal consistency.” If you ran a pilot study, mention that you tested and refined the instrument before the main data collection. If you used test-retest or inter-rater reliability, describe the procedure and the result. The key is to attach a number and a method to the concept — examiners want to hear that reliability was something you measured and reported, not a quality you simply assumed your instrument possessed.
Explaining Validity Without Confusing the Types
Validity is whether the instrument measures what it is supposed to measure, and the difficulty for many candidates is keeping the types distinct. Address them one at a time. Content validity is about whether the items cover the construct fully, usually established through expert review: “Three subject experts reviewed the items for relevance before piloting.” Construct validity concerns whether the instrument behaves as the theory predicts, often supported through factor analysis. Criterion validity, where relevant, is about whether your measure relates to an external benchmark. Name only the types you genuinely addressed; claiming a form of validity you did not test invites a follow-up question you cannot answer. A confident answer connects each type to a specific action you took: who reviewed the items, what analysis you ran, and what the result was. Be ready, too, for the follow-up that asks about threats — what could have undermined validity or reliability in your design, and how you mitigated it — because examiners often probe whether you understand the limits of your own safeguards rather than just the textbook definitions. Rehearsing validity and reliability as separate, evidence-backed answers before the viva keeps the two concepts from collapsing into one another when you are put on the spot.
