A mixed-method survey carries closed and open-ended questions on one record. A worked questionnaire example, the four formats, and how to design one.
A mixed-method survey is a single instrument that collects both quantitative and qualitative data — a rating and the open-ended reason behind it — from the same respondent, so the number and its explanation stay linked on one record. It is the questionnaire form of mixed-methods research: closed questions for scale, open questions for meaning, joined at the participant rather than merged in a slide. One instrument, one record, both kinds of evidence.
The reason most quant-plus-qual efforts fail is that they run as two surveys: a rating scale in one tool, an interview or open-ended form in another, reconciled by hand at the end. The connection that makes the data valuable — this person's score and this person's reason — is exactly what gets lost when the two live apart.
Key takeaways
The promise of mixed methods is a joint display: a table where a quantitative result sits beside the qualitative theme that explains it, for the same people. That display is trivial when both came from one instrument and one record, and nearly impossible when they came from two tools that never shared an identifier.
Sopact calls the architecture One Record, One Contact ID: the rating and the open-ended reason are captured on the same form, bound to the same participant, so the score and the story never separate. The three mixed-methods designs — explanatory, exploratory, convergent — and worked studies are on the mixed-methods research examples page, and the analysis on mixed-methods data analysis.
This page is the instrument node: the questionnaire that carries both kinds of question. The broader case for using both methods at all is on the qualitative and quantitative methods page, and the instrument architecture beneath it on survey design.
You design a mixed-method survey by pairing every rating with the open-ended question that explains it, capturing both on one instrument bound to a persistent identifier, and theming the open-ends against a codebook on arrival so the qualitative side is analyzed rather than filed. The pairing is the design; everything else supports it.
The failure mode is a survey that asks ten ratings and one generic “any other comments” box at the end. That box is not mixed method; it is an afterthought nobody reads. The table pairs each design choice with what a real mixed-method instrument does instead.
A mixed-method survey pairs each rating with a reason, on one record, read together. Read the last column: it is the difference between two datasets and one joint finding.
| Design choice | Two separate surveys | A mixed-method survey |
|---|---|---|
| Pairing | Ratings and open-ends in different tools | Each rating paired with its explaining question |
| Identity | Reconciled by name at the end | One Contact ID from first response |
| Open-ends | A generic comment box, unread | Themed against a codebook on arrival |
| Output | A bar chart and a pull quote | A joint display: result beside its reason |
| Timing | Merged in the final report | Joined at collection, read as it lands |
Every row collapses to one requirement: the number and the reason have to share a record. That is One Record, One Contact ID — the architecture that turns a rating and an open-end into a joint finding instead of two files someone staples together.
When the quantitative and qualitative sides are joined only at reporting time, the join is a manual merge that ages immediately. Theming the open-ends on arrival means the joint display exists continuously — the score and the reason are linked the moment both are collected. That is the premise of the Loop, Sopact's method for continuous impact intelligence: collect clean at the source, analyze the moment data arrives, improve while you can still act.
The Loop is also what makes a mixed-method finding defensible. The quantitative result and the qualitative theme both trace back to the same respondent, so a claim about why a number moved resolves to the person who said it. That standard has its own chapter in traceability and transparency.
One method, three moves that never stop
Then the cycle runs again, a little sharper each wave. Read the method: the Loop methodology →
The fastest way to feel the difference is to pair one rating with its reason on one record. Each prompt below pastes into Sopact Sense's Assistant, or reasons through with your team; the arrow above each links the Academy walkthrough that shows the expected output and the tips.
Academy walkthrough → Pair every rating with its reason
For this program, design a mixed-method survey where each rating is paired with the open-ended question that explains it: [PASTE PROGRAM + OUTCOMES]. List the rating questions, the paired open-ended prompts, the persistent identifier, and the segments to capture at intake. Flag any rating with no explaining question. Return the instrument.
Academy walkthrough → Theme the reasons on arrival
Theme these open-ended responses against the codebook and join each to the respondent's rating: [PASTE CODEBOOK + RESPONSES with respondent_id + rating]. Return respondent_id, rating, assigned theme, and flag where the theme and the rating disagree. Then show the theme distribution within each rating band.
Academy walkthrough → Define both sides once
Turn this mixed-method instrument into a data dictionary covering both the quantitative and qualitative fields: [PASTE INSTRUMENT]. For each field give the name, definition, answer type, and allowed values or codebook reference. Flag any field two waves would treat differently. Return a table.
Academy walkthrough → Build the codebook for the open side
Draft the codebook the open-ended half of this survey will be themed against: [PASTE OPEN-ENDED QUESTIONS + FRAMEWORK]. Give 6-10 codes with definitions and include/exclude rules, and map each code to the rating it most often explains. Return the codebook.
Each walkthrough is short and practical: what to do, the prompt to run, the output to expect, and the tips that keep it reliable.
Watch: unifying the qualitative and quantitative workflow at the source, on one participant record.
A mixed-method survey is a single instrument that collects both a rating and the open-ended reason behind it from the same respondent, so the number and its explanation stay linked on one record. It is the questionnaire form of mixed-methods research. Sopact keeps both on One Record, One Contact ID, so a quantitative result and the voice that explains it display together rather than in two disconnected files.
Pair every rating with the open-ended question that explains it, capture both on one instrument bound to a persistent identifier, and theme the open-ends against a codebook on arrival. A generic comment box at the end is not mixed method. Sopact designs the pairing into the instrument, so the join exists at collection rather than being reconstructed at analysis.
A workforce program asks participants to rate their job-search confidence and, on the same form, to explain what changed it; the rating shows how much and the open-end shows why, for the same person. A sequential design might follow a survey with interviews of specific respondents. Sopact's examples keep the score and the reason on one record so the joint display is automatic.
A mixed-method survey is the instrument — one questionnaire carrying both kinds of question; mixed-methods research is the broader design that may combine surveys, interviews, and documents in explanatory, exploratory, or convergent sequence, covered on the mixed-methods research examples page. Sopact treats the survey as the collection node feeding the wider design.
A sequential design collects one type of data, then the other, to build on it: explanatory sequential runs the survey first and interviews to explain the results, while exploratory sequential interviews first and surveys to test at scale. Both need the two waves linked by a persistent identifier. Sopact's One Contact ID keeps the sequence joined at the participant across every stage.
Compute the closed questions, theme the open-ends against a fixed codebook, then join each theme to the rating it explains for the same respondent — producing a joint display rather than a bar chart beside a pull quote. Sopact themes on arrival and keeps both on one record, so the analysis is a query; the deeper analysis workflow is on the mixed-methods data analysis page.
Because they run as two surveys in two tools, so the score and the reason never share an identifier and are reconciled by hand at the end — losing exactly the connection that makes mixed methods valuable. Sopact's One Record, One Contact ID captures both on one instrument, so the join is built in rather than reconstructed.
Next: see the three designs worked out on the mixed-methods research examples page, or the case for both methods on the qualitative and quantitative methods page.
