What are qualitative and quantitative measurements?
Quantitative measurement represents a characteristic with numbers; qualitative evidence describes meaning, experience or context. A waiting time is quantitative. A person’s explanation of why accessing the service was difficult is qualitative. Both can be collected systematically, and both can contain error or bias.
The method should follow the question. “How long did people wait?” requires a defined time measure. “What made the wait difficult?” calls for experience and context. Neither answer substitutes for the other.
Examples across common workflows
| Workflow | Quantitative measure | Qualitative evidence |
|---|---|---|
| Training | Assessment score or completion rate. | What helped or prevented application. |
| Customer service | Resolution time or repeat-contact rate. | The unresolved part of the experience. |
| Employee development | Participation or a defined survey score. | Examples of support and barriers. |
| Partner reporting | Delivery volume or expenditure. | Explanation of delays or changed conditions. |
| Program evaluation | A defined outcome at follow-up. | How people describe the change and other influences. |
Define the measure before collecting it
For a number, specify the unit, population, period, calculation and missing-data rule. For qualitative evidence, specify the question, sampling approach, collection context and analysis method. A vague measure does not become useful because it appears in a dashboard.
For example, “engagement” could mean attendance, a survey scale or a description of involvement. Name the construct and choose an approach that fits it. Do not switch meanings between collection and reporting.
A worked example: score and explanation
In an illustrative service survey, 40 of 50 respondents report that their issue was resolved. That is 80% of respondents. Ten describe unresolved issues, including repeated requests for the same documents and difficulty reaching the right team.
The percentage describes the observed result. The comments suggest practical areas to investigate. They do not prove which process caused the result, nor do they represent everyone who did not respond. Follow up or examine service records before choosing an intervention.
Can qualitative evidence become numbers?
You can code responses into categories and count them. Keep the original text and coding rules so the count can be checked. State whether a response can receive more than one code and whether the denominator is people, responses or mentions.
A count of “communication” mentions is not automatically a measure of communication quality. Some people write more than others, and an open question may not invite everyone to discuss the same issue. Describe what the count represents.
Combine evidence in one review
| Finding | Explanation | Next question |
|---|---|---|
| A score improved. | Comments describe better access. | Did access improve across groups or only some? |
| A score stayed level. | Some people improved while others struggled. | Is the average hiding different experiences? |
| Many records are missing. | Follow-up contact was difficult. | Who is absent, and how does that limit the conclusion? |
Keep disagreement visible. If interviews and survey results point in different directions, investigate sampling, timing and question wording rather than forcing them into one positive story. See mixed method surveys.
Check reliability and interpretation
- Use a suitable validated instrument when the decision requires one.
- Pilot questions for clarity and accessibility.
- Train collectors and document changes in administration.
- Check coding consistency and review ambiguous examples.
- Keep missing data, contradictory evidence and uncertainty visible.
AI can help organize text or identify candidate themes. Review source passages, exceptions and any proposed scoring before use. A fluent summary is not evidence that the underlying interpretation is correct.
Keep the record useful for the next question
Preserve definitions, dates, sources and relevant permissions with the evidence. Where the purpose supports it, connect scores and explanations to the same person, partner or event. This makes later interpretation easier without requiring every study to identify individuals.
For practical reporting, continue to survey report examples.
How Sopact reduces coding and reporting work
A measure and its accompanying account should stay connected through review. Otherwise every new reporting question creates another matching exercise.
A workflow with repeated manual work
- Define from an initial sampleRead material and agree on the codebook.
- Apply it across the datasetCode responses and check the result.
- Revise a definitionReturn to affected material and recode it.
- Reconnect the numbersReconcile coded results with ratings and context, then rebuild the view.
The Sopact workflow
- Your team owns the definitionsDecide what each code means and improve it as you learn.
- Apply coding across the eligible dataAutomate application; people review quality and exceptions.
- Reprocess after a definition changesReapply the revised definition across the configured scope instead of recoding each response by hand.
- Ask across coded text and numbersKeep the response, rating and relevant record context connected; inspect the evidence behind the result.
This compares workflow patterns, not a claim that every research tool requires manual coding or separate files. Some already automate parts of this work; compare the complete cycle.
For this codebook-based workflow, the main saving is repeated application and reconnection—not the removal of human judgment. A changed definition can be reapplied across the configured data while reviewers concentrate on quality, exceptions and interpretation. Coded text stays connected to the relevant ratings and context.
Count the recurring work in ownership cost. Include setup, coding, recoding after revisions, source reconciliation, review and reporting, plus your actual platform and processing expenses. A worked scenario of four cycles of 4,000 responses illustrates 272 fewer annual staff hours; it is an assumption-based example, not a customer benchmark. Existing automation, review needs and implementation effort can substantially change the result.
Adjust the workload assumptions and compare total effort →
A reliable assistant should calculate from the selected records and let a reviewer open the supporting evidence. Check the data scope, definition, denominator and access permissions. Reproducible arithmetic does not make every AI interpretation correct.
Watch: Why Qualitative Analysis Stays Small — And How to Scale It
See why revising a codebook creates repeat work, and how connected coding and quantitative analysis change that workload.
Frequently asked questions
Is quantitative evidence always objective?
No. Question wording, sampling, measurement and interpretation can bias numerical results.
Is qualitative evidence only opinion?
No. It can be systematically collected and analyzed to understand experience, meaning and processes.
Can I calculate an average for any rating scale?
Not automatically. Consider the scale’s properties and purpose; distributions or medians may be more appropriate for some ordinal measures.
Do I always need both types?
No. Use the evidence needed for the question. Combining them is useful when each addresses a different part of the problem.
