Written for university students in the UK and internationally. Conventions and terms vary between countries and institutions, so your own course handbook and your tutor’s guidance are always the final authority.
One of the earliest and most important decisions in any research project is also one of the most commonly rushed: should your study be qualitative, quantitative, or a mix of both? Students often pick the method that feels most familiar and then quietly bend their research question to fit it. The better order is the reverse. Decide what you are really trying to find out first, and the question will usually point you toward the approach that belongs to it. This article explains the difference in plain terms and gives you a way to choose with confidence.
What each approach is actually for
Quantitative research measures. It is built to answer questions about how many, how much, how often, and whether a relationship exists between things, and it answers them with numbers you can analyse statistically. Surveys with closed questions, experiments, and the analysis of existing datasets are typical quantitative methods. Qualitative research, by contrast, explores. It is built to answer questions about why, how, and what something means to the people involved, and it answers them with words rather than numbers. Interviews, focus groups, open-ended questions and the close analysis of documents are typical qualitative methods. Neither is superior; they answer different kinds of question.
Let the question guide the method
The cleanest way to decide is to write your research question down and look at what it is really asking. “Does a training programme increase employee productivity?” leans quantitative, because it asks you to measure productivity and test whether it changes. “Why do employees resist a new training programme?” leans qualitative, because it asks for people’s reasons, experiences and interpretations in their own words. The wording is a useful clue: verbs like measure, compare, test and predict point toward quantitative work, while verbs like explore, understand, describe and interpret point toward qualitative work. Treat these verbs as clues rather than fixed rules, though — quantitative studies can investigate explanations, and qualitative studies can describe patterns. What settles the method is the kind of evidence your question needs, not the verb it happens to use.
The training example also carries a useful warning about cause. Measuring that productivity rose after a programme does not, by itself, show the programme caused the rise, because many things change at the same time. Establishing cause needs a design built for it — such as a controlled comparison between a group that received the training and a comparable group that did not — rather than a simple before-and-after measurement. Being honest in your write-up about what your design can and cannot show is itself a sign of careful research.
When mixed methods make sense
Some questions genuinely contain both elements. You might want to measure whether productivity rose and understand why employees felt the way they did about the change. A mixed-methods design can be powerful here, but its value comes from integrating the two strands, not simply running a survey and some interviews side by side: the interviews should help explain the numbers, and both should be drawn together in your discussion. Without that integration you really have two half-studies. It is also demanding — a mixed-methods project asks you to do two kinds of research well, which takes more time, more skill and more words than many undergraduate projects allow — so choose it because your question truly needs both, not to avoid committing to one.
Count the practical cost before you commit
Method choice is not only about the question; it is also about your resources. Quantitative work often calls for a larger sample, but how large depends on the design, the precision you need, the size of the effect you expect and the analysis you plan to run — there is no single number that automatically makes results meaningful. Qualitative work typically needs far fewer participants, but it demands many more hours of careful transcription, coding and interpretation, and those hours are easy to underestimate. Be honest with yourself about your timeline, your access to participants and your own skills before you lock in a design, because the wrong choice here is expensive to reverse later.
Sampling: who you study and why
Whichever approach you choose, you will have to justify your sample. Quantitative studies often aim for a sample large and well-drawn enough to generalise from, which is why random or systematic sampling is common. Size alone does not make a sample representative, however — a large but poorly drawn sample can still be badly biased. What matters is how participants were selected and who may have been left out, so describe your sampling method and its limitations rather than leaning on numbers. Qualitative studies usually use purposive sampling, deliberately choosing participants who can speak to the question in depth, and they do not claim to generalise in the same statistical way. Explaining clearly who you studied, how you chose them and why that choice fits your question is one of the simplest ways to make your methodology convincing to an examiner.
Match your analysis to your design
Finally, plan the analysis before you collect anything. Quantitative data points you toward descriptive statistics and then the right test for your question, such as a comparison of means or a measure of correlation. Qualitative data points you toward a structured process such as thematic analysis, where you code the material and build themes from it. Knowing which analysis you will use in advance protects you from the most demoralising situation in research: collecting a pile of data you then discover you cannot properly analyse.
Think about quality: validity and trustworthiness
Examiners expect you to show that your results can be believed, and the language for this differs between the two traditions. In quantitative work you talk about validity, whether you measured what you intended to measure, and reliability, whether the measurement would give consistent results if it were repeated. In qualitative work you talk about trustworthiness, built from ideas such as credibility, transferability and the careful, transparent handling of your own influence on the data. You do not need to turn your methodology into a textbook, but showing that you have thought about how to make your findings credible, and about what could threaten that credibility, is one of the clearest signs of a mature researcher.
A quick worked example
Imagine you are interested in how students use a new library app. “What percentage of students use the app each week, and does usage differ by year of study?” is quantitative: you would run a survey with closed questions and compare the numbers. “How do students experience the app, and why do some abandon it after a week?” is qualitative: you would interview a small, purposively chosen group and build themes from what they say. “How widely is the app used, and why do patterns of use look the way they do?” suits a mixed-methods design, because it needs both the measurement and the explanation. Notice how the same broad topic produces three different designs depending on exactly what you ask.
Whatever you decide, write your reasoning into the proposal and the methodology chapter rather than leaving it implicit. Choose the approach according to the evidence your question requires, the study design and the practical constraints you face; the wording of the question alone does not decide the method. Examiners are far more convinced by a student who can explain why a design was chosen, and what its limitations are, than by one who presents a method as if it were the only option available.
Further reading: standard research-methods texts, such as Creswell and Creswell’s Research Design, set out these distinctions in more detail.
Getting a second opinion on your design
Choosing and justifying a methodology is one of the stages where early feedback saves the most trouble, because mistakes made here follow you all the way to submission. Students who want a second opinion can consult FIZBS for support in pressure-testing a research design and its analysis plan before any data is collected, with the scope agreed beforehand. Check your university’s rules on external assistance, and keep the research your own. Choose the method that fits the question, not the one that fits your comfort zone, and the rest of the project becomes far more straightforward.

Leave a Reply