UX Glossary · Q
Quantitative research
Also: Quant
Research producing numbers: how many, how often, how long, measured at a scale that supports inference.
Quantitative research measures. Analytics, surveys with closed questions, A/B tests, task success rates, time on task, standardised instruments such as the System Usability Scale.
Its counterpart, qualitative research, explains. The two answer different questions and neither substitutes for the other.
What it is good at
Scale and confidence. It tells you how widespread something is, whether a difference is likely to be real, and whether a change moved a number. It is how you decide which of five problems to fix first, because it tells you which one affects most people.
What it cannot do
Tell you why. A drop-off rate identifies a step; it does not identify the confusion. And a number is only as good as the question that produced it: survey questions are notoriously easy to write badly, and asking people to predict their own future behaviour produces confident, unreliable answers.
Sample size honesty
Small products often cannot run valid quantitative studies, and pretending otherwise is common. A difference measured on 40 sessions is usually noise. Saying “we do not have the traffic to answer this quantitatively” is a legitimate finding.
In practice
The reliable pattern is quant to find where and how much, qual to find why, quant again to confirm the fix moved the number.