The mixed method designs and how to choose the sequence, plus the join most designs under-specify: both strands on the same units.
Mixed method design is mostly a decision about sequence and integration: a convergent design collects quantitative and qualitative data at the same time and compares them; an explanatory sequential design collects numbers first, then uses qualitative work to explain them; an exploratory sequential design collects qualitative first, then builds and tests a quantitative measure. Every design ends in the same requirement — the two strands have to be joined on the same units — and that join is what most designs specify least. Sequence is the visible choice; integration is the hard one.
The gap teams fall into is designing the sequence carefully and the integration not at all. They decide whether interviews come before or after the survey, then assume the joining will work itself out, and it does not: the strands end up in separate tools with no shared key, and the design’s integration step becomes an unplanned manual scramble.
Key takeaways
A mixed method design has two decisions, and teams reliably make one. The first, sequence, is visible and gets attention: convergent, explanatory sequential, exploratory sequential, or embedded. The second, integration, is where the design either works or falls apart, and it gets waved at with phrases like “the strands will be triangulated.” But triangulation is not a plan; it is a hope unless the design specifies how the two strands connect on the same units, with what key, in what tool.
Specifying the join means deciding, at design time, that the strands share a unit key. Sopact calls the result the Connected Record: quantitative and qualitative both landing on one participant, so whatever the sequence, integration is built into collection rather than reconstructed after. The design’s hardest step is handled by the data model, not left to end-stage effort, the foundation the whole mixed methods research practice needs.
Executing mixed designs moved through three eras. First, two studies run separately and compared in the write-up, with the sequence honored and the join improvised. Then parallel tools with a planned but painful manual join, which often degraded to asserting agreement. The current era shares a unit key across both strands from the start, so the join specified in the design actually happens in the data.
The one test that separates the eras: does your design name the key that connects a participant’s quantitative and qualitative data, and does that key exist at collection? A design that specifies sequence but not the shared key has planned the visible half and left the integration to chance. If the join has no key, the sequence is the only part of the design that will survive contact with the data.
Sequence should follow what you are trying to do. Use explanatory sequential when you have or expect quantitative results and need to understand why — the numbers raise the questions the qualitative work answers. Use exploratory sequential when the territory is unfamiliar and you need qualitative work to define what is worth measuring before you build the instrument. Use convergent when you want to corroborate, comparing what the numbers and the words say about the same moment. Embedded designs nest one strand inside a larger study of the other.
Each sequence implies a different integration point, but all of them require the same shared key, so choosing sequence is the interesting decision and securing the join is the non-negotiable one. Getting both right is what separates a design that delivers integrated findings from one that produces two datasets and a hopeful discussion, the same standard the mixed methods research tools are evaluated against.
Choose the sequence by the question — explanatory sequential to explain numbers, exploratory sequential to explore before measuring, convergent to corroborate — and in every case, secure the join by giving both strands a shared unit key at collection so integration is designed in, not improvised. The design decision is really two: the visible sequence and the essential join.
The output is a design that produces integrated findings: whatever the sequence, a participant’s numbers and words connected on one record, so the integration the design promised actually happens. Because Sopact lands both strands on the Connected Record, the join is guaranteed by the data model, and the sequence can be chosen freely to fit the question, as the worked mixed methods research examples show.
The designs differ by sequence, but every one ends in the same join — both strands on the same units. Securing that key at collection is what most designs under-specify.
| Design | Sequence | Use it to |
|---|---|---|
| Convergent | Both strands at once | Corroborate numbers and words on the same moment |
| Explanatory sequential | Quantitative, then qualitative | Explain the numbers you have |
| Exploratory sequential | Qualitative, then quantitative | Explore before you build a measure |
| Embedded | One strand nested in the other | Support a larger study with a secondary strand |
The practice these designs serve is mixed methods research; the tooling that secures the join is mixed methods research tools.
Longitudinal and mixed-methods designs are usually treated as after-the-fact analysis: collect everything, then, months later, try to stitch it together. The value of reading data is highest while collection is still open, when a wave can be chased and a confusing number can be explained. That is the premise of the Loop, Sopact’s method for continuous intelligence: collect clean at the source, analyze the moment data arrives, improve while there is still time to act.
The Loop is also what makes a longitudinal or mixed-methods claim defensible: every figure traces back to the response it came from, on the same unit across waves and methods, the standard detailed in Loop traceability.
One method, three moves that never stop
Then the cycle runs again, a little sharper each time. Read the method: the Loop methodology →
The fastest way to strengthen a design is to secure the join in your own data. Export your quantitative and qualitative data with IDs, then paste the prompts below into Sopact Sense’s Assistant, or reason through them with your team. The arrow above each links the Academy walkthrough with the expected output and tips.
Academy walkthrough → Analyze longitudinal data
Here are several waves of data from the same participants on the same IDs: [ATTACH]. Track each participant across waves, show the trajectory of the key measures, flag anyone who dropped out, and surface the open-ended comments that explain the biggest movements.
Academy walkthrough → Analyze pre, mid, and post data
Here are pre and post responses from the same units on the same IDs: [ATTACH]. Report change per unit as real pairs against each baseline, flag anyone who did not move or regressed, and quote the answer that explains each flag.
Academy walkthrough → Connect quant and qual data
Here are our quantitative measures and the open-ended comments on the same IDs: [ATTACH]. Show which themes in the comments explain the weakest numbers, quote a comment for each, and tell me which cases to look at more closely.
Academy walkthrough → How to build a data dictionary
Here are the measures I collect across waves and methods: [PASTE]. Build a data dictionary entry for each — exact wording, scale, wave schedule, and what would invalidate a comparison — so wave two and method two stay comparable to wave one.
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: keeping the same unit connected across waves and methods on one record.
Convergent (both strands at once, compared), explanatory sequential (quantitative then qualitative to explain it), exploratory sequential (qualitative then quantitative to build a measure), and embedded (one strand nested in the other). All end in the same join on the same units. Sopact keeps both strands on the Connected Record so that join is designed in.
Explanatory sequential collects numbers first and uses qualitative work to explain them; exploratory sequential collects qualitative first and uses it to build and test a quantitative measure. The choice follows the question. Sopact keeps both strands joined regardless of sequence, so integration is secured either way.
The integration — connecting the two strands on the same units — which most designs specify least while carefully planning the sequence. Sopact handles it in the data model by landing both strands on one participant record, so the join is guaranteed rather than improvised.
By the question: explanatory sequential to explain results you have, exploratory sequential to explore unfamiliar territory before measuring, convergent to corroborate. Sopact secures the join in every case, so the sequence can be chosen freely to fit the question rather than to make integration easier.
To give both the quantitative and qualitative strands a shared unit key at collection, so a participant’s numbers and words connect automatically. Without it, integration becomes a manual scramble. Sopact stamps that shared key by keeping both on the Connected Record from the start.
Triangulation is a goal — corroborating findings across strands — not a design or a plan. It only works if the strands are joined on the same units. Sopact makes triangulation real by keeping the number and the reason on one record rather than leaving them in separate tools.
It lands both the quantitative and qualitative strands on the Connected Record with a shared unit key from collection, so whatever sequence the design chooses, integration is built in. The join — the design’s hardest step — is handled by the data model rather than left to end-stage effort.
Next: see the practice on mixed methods research, or the tooling on mixed methods research tools.
