# Qualitative and Quantitative Research

A distinction between evidence used to understand patterns and mechanisms in depth and evidence measured to estimate magnitude, frequency, or difference.

- Type: framework
- Domain: Research & behavior
- Level: Foundational
- Last reviewed: 2026-09-25
- Original: [Soroush’s Product Wiki](<https://wiki.soroush.uk/topic/qualitative-and-quantitative-research/>)

## What it means

Qualitative research directly examines experiences, behaviors, language, and context to interpret how or why something happens. Quantitative research uses predefined measurements and numerical analysis to estimate how many, how much, how often, or how strongly. Either can study attitudes or behavior, and combining them requires a reason rather than a ritual.

## Why it matters

Product teams need both explanation and magnitude. Qualitative evidence can reveal mechanisms and unknowns that metrics miss, while quantitative evidence can test prevalence or change beyond a few observed cases.

## When to use it

Use qualitative approaches to explore unfamiliar situations, diagnose problems, or understand meaning. Use quantitative approaches when constructs can be measured and the decision depends on magnitude or comparison. Sequence or combine them when one mode can frame or explain the other.

## How it works

1. Define the decision and whether it requires explanation, measurement, or both.
2. Choose a design, sample, and data collection method appropriate to that need.
3. Analyse qualitative material transparently or quantify predefined measures with suitable statistical reasoning.
4. Compare conclusions only at the level each sample and method support.
5. Integrate complementary evidence while reporting disagreements and uncertainty.

## A practical example

A streaming team uses interviews and session observation to learn why household switching is confusing, then measures task success across a larger sample after redesign. The first study identifies mechanisms; the second estimates improvement.

## Where it breaks

Small qualitative samples do not estimate population prevalence, while numerical precision does not guarantee valid constructs or explain causes. Mixed methods add cost and can produce incompatible evidence that requires judgment.

## Common mistakes

- Calling open-text survey responses qualitative depth without contextual probing.
- Converting counts from a small purposive sample into population percentages.
- Assuming quantitative evidence is objective regardless of measurement and sampling choices.

## Visualization

[View the visualization on the website](<https://wiki.soroush.uk/topic/qualitative-and-quantitative-research/#visualization-qualitative-and-quantitative-research>)

## Sources

1. [When to Use Which User-Experience Research Methods](<https://www.nngroup.com/articles/which-ux-research-methods/>) — Nielsen Norman Group · Professional guidance
2. [Quantitative User-Research Methodologies: An Overview](<https://www.nngroup.com/articles/quantitative-user-research-methods/>) — Nielsen Norman Group · Professional guidance

## Related topics

- [Triangulation](<https://wiki.soroush.uk/topic/triangulation/>)
- [Survey Design](<https://wiki.soroush.uk/topic/survey-design/>)
- [Thematic Analysis](<https://wiki.soroush.uk/topic/thematic-analysis/>)
