Qualitative vs quantitative research differ in the answer each one gives. Qualitative research collects words and observations to explain why people think or act as they do. Quantitative research collects numbers to measure how many people do something, or how much.
In practice, many workplace projects need both: interviews to find the reasons, then a survey to see how common each reason is.
The two methods differ in the data they collect, how the study is run and what you can conclude. Nielsen Norman Group (2017) sets out the split for usability testing: qualitative research answers “why,” and quantitative research answers “how many and how much.” The table follows that comparison.
| Qualitative research | Quantitative research | |
|---|---|---|
| Question answered | Why something happens | How many and how much |
| Data collected | Observations, interview notes and open-ended answers | Metrics such as completion rates, task times and satisfaction ratings |
| Typical methods | Interviews, focus groups, observation, open-ended questions | Closed-question surveys, analytics, benchmark tests |
| Study setup | Flexible: you can rewrite a question between sessions | Tightly controlled: every session runs the same way |
| Main strength | Finds the main problems directly and shows the reasons | Statistical confidence, benchmarking and return-on-investment figures |
| Main limit | Findings rest on the researcher’s interpretation, and a few people may not represent your whole audience | A number does not say why, and a metric needs a reference point to be read |
Group size often differs too, but it does not define the method. In usability testing, NN/g describes qualitative studies with 5 to 8 participants and quantitative studies with often more than 30. A separate NN/g video (2024) recommends 40 participants for many quantitative UX studies as a balance of precision, risk and practicality. Those numbers come from usability testing. For an HR survey, market research or an academic study, the sample you need depends on the purpose, the design and how precise the result has to be.
The clearest way to see the difference is to run one question through both methods. The example below is hypothetical: an HR team sees that few employees use a new recognition program and wants to know what to fix.
| Qualitative step | Quantitative step | |
|---|---|---|
| Question | Why do employees not use the recognition program? | How many employees give each reason? |
| Method | Interviews of about 30 to 45 minutes with a few employees who rarely use it | A short survey sent to all employees |
| Sample questions | “Tell me about the last time you wanted to thank a colleague. What did you do?” | “Which of these best describes why you have not used the program?” with the reasons from the interviews as options, plus “Other” |
| What you get | A list of reasons in employees’ own words, for example not knowing the program exists or finding the rewards hard to use | The share of employees who give each reason, which can be split by team or office |
| Decision it supports | What the possible fixes are | Which fix to make first |
The order matters. The survey can only offer answer options someone has already thought of, so the interviews supply them. The reverse order also works: if a usage metric shows a drop, the number alone does not explain it, and NN/g notes that researchers often add qualitative sessions to understand the specific issues. Either way, the second round needs its own participants and rewards, so plan for both before you start.
Start from the decision you need to make, not from the method. Write the question in one sentence, then match it to the method that can answer it.
| Question you need answered | Method | Why |
|---|---|---|
| Why do new customers cancel in their first month? | Qualitative | The answer is a set of reasons that people give in their own words |
| What share of customers finish onboarding? | Quantitative | The answer is a share that you can track over time |
| Which of two page designs is faster to use? | Quantitative | The answer compares one metric across the same task |
| What would make our rewards program more useful? | Qualitative, then quantitative | Interviews find the options, and a survey measures which ones people want |
Three checks keep the choice honest:
It can be either, or both. A survey is a way to collect data, so the question type decides what you get. Rating scales, yes-or-no answers and multiple-choice questions produce numbers.
Open-ended questions such as “What would make this program more useful?” produce words. A survey with both gives you a measurement and some of the reasons behind it. Read the written answers theme by theme, as you would interview notes, rather than counting them as votes.
Reward cost is the number of participants multiplied by the reward for each one. There is no single required rate. Massachusetts Digital Service’s compensation guidelines recommend the tiers below for participants recruited organically or through a panel and describe them as typical for industry and government.
Treat them as a reference point: your organization’s policy, the audience and the country set the final amount. For a wider look at reward types, see Top 6 Survey Incentives: How to Reward Survey Participants.
| Study type | Session length | Recommended reward |
|---|---|---|
| Moderated interview or session | 10 to 30 minutes | $50 |
| Moderated interview or session | 31 to 60 minutes | $100 |
| Moderated interview or session | 61 to 90 minutes | $150 |
| Unmoderated test | 15 minutes or less | $25 |
| Unmoderated test | 16 to 30 minutes | $50 |
| Survey | Under 5 minutes | No reward |
| Survey | 5 to 15 minutes | $10 |
| Survey | 16 to 20 minutes | $15-$20 |
Here is a worked example using those tiers. The participant counts are examples: 8 interviews, the 40 participants NN/g suggests for many quantitative UX studies, and a survey of 100.
| Plan | Participants | Session | Reward each | Total |
|---|---|---|---|---|
| Qualitative interviews | 8 | 45 minutes, moderated | $100 | $800 |
| Quantitative usability test | 40 | 15 minutes, unmoderated | $25 | $1,000 |
| Quantitative survey | 100 | 10 minutes | $10 | $1,000 |
| Mixed: interviews, then survey | 8, then 100 | As above | $100, then $10 | $1,800 |
Qualitative sessions pay more per person, while quantitative studies pay more people. Each extra 100 survey respondents at the $10 tier adds $1,000. Rewards are one line in a research budget: recruiting, tools, and analysis time are separate costs.
The same guide also sets its own policies, such as not paying for screening surveys or missed sessions, capping paid responses on high-volume surveys, and paying more for hard-to-reach people (up to $250 per hour for highly specialized participants such as doctors). Treat these as one organization’s choices: set your own rules before you recruit, and check legal requirements for participants outside the U.S.
When you send each reward also matters, for example right after an interview or after a completed survey. Key To Boosting Survey Response Rates covers other ways to raise completion, and Sending Survey Incentives to Participants in Canada and Around the World and The easiest way to pay research participants digitally compare reward options for participants in different countries.
Once the budget is set, Giftronaut sends the rewards. The Giftronaut survey incentives page lists the full feature set.
Top Line Marketing, a consultancy that rewards survey, review and referral participants internationally, reported saving $500 and 5 hours per week with Giftronaut (case study).
Reasons call for qualitative research, measurements call for quantitative research, and many projects use both in sequence. Size each study for its purpose rather than a fixed rule, then price it: participants multiplied by the reward for the session length.
Look at the data it collects. Words, observations and themes make a study qualitative. Numbers, counts and statistics make it quantitative, whatever the number of participants.
Interview notes, open-ended answers and observations are qualitative data. Ratings, task times, completion rates and the share of people choosing an answer are quantitative data.
Quantitative research gives statistical confidence and lets you benchmark and calculate return on investment. A number does not say why people had trouble, so researchers often add qualitative sessions to find the reasons.
Often, but not always. Start qualitative when you do not yet know the possible answers, because a survey needs answer options. Start quantitative when a metric already shows a problem and you then need interviews to explain it.
It depends on the study’s purpose and design. In usability testing, Nielsen Norman Group describes qualitative studies with 5 to 8 participants and recommends 40 for many quantitative studies. Surveys, market research and academic studies are sized for the precision they need.
There is no single required rate. As one public reference, Massachusetts Digital Service recommends $100 for a 31 to 60 minute interview and $10 for a 5 to 15 minute survey in its own guidelines. Adjust for your audience, country and organization’s policy.
Yes. Giftronaut reaches recipients in 225 countries, and a Choice Card lets participants choose their currency at redemption. Check legal requirements for each country before you recruit.