From one question
to a checkable finding.
Follow the questionnaire, inspect the dashboard, test an alternative explanation and reconcile the downloads. Every response is generated for demonstration; these are not customer findings.
What was asked?
A fictional workshop follow-up asks about using a feedback technique, confidence, skills to practise, recommendation, team communication, training format, workshop priorities, practice conversations and written suggestions. It demonstrates single answers, rating and recommendation scales, multiple selections, a matrix, image choices, ranking, a number and open text.
The images in the sample’s training-format question are illustrative interface pictures. This example does not test image-click or passage-marking questions. It demonstrates report reconciliation rather than every question type.
The clearest pattern: use differs by department
| Department | Used the technique | Answering people | Percentage |
|---|---|---|---|
| Operations | 18 | 25 | 72.0% |
| Sales | 12 | 25 | 48.0% |
| All responses | 30 | 50 | 60.0% |
The descriptive gap is 24.0 percentage points. In the dashboard, clear filters and inspect the first question, then compare Operations and Sales. For the full group comparison, leave Role set to All.
Test another explanation before acting
Different roles, opportunities to practise, manager support or collection timing could explain a real difference. In this generated fixture each department has five managers and 20 individual contributors. Filtering to managers gives 80.0% versus 60.0% (4/5 versus 3/5); among individual contributors it is 70.0% versus 45.0% (14/20 versus 9/20). The role mix alone does not remove the constructed gap. Those five-person groups are too small for a dependable general conclusion.
For real research, ask whether the technique was relevant and whether respondents had an opportunity to use it. Compare equivalent jobs and exposure periods before claiming a workshop caused a difference.
Three useful cross-checks
- Single choice: 30 Used it, 15 Not yet and 5 No opportunity sum to 50; 60.0%, 30.0% and 10.0% sum to 100.0%.
- Confidence: ten people at each point from 1 to 5 produce a mean of 3.0 out of 5. This is an original-scale mean, not a benchmark score.
- Multiple selection: each person selects two skills, giving 100 selections from 50 people. Percentages use people and sum to 200.0%.
Every question has 50 answers; there are no missing answers or exclusions in this fixture. The sample is deterministic, not a random sample. It has no real invitation denominator, fieldwork dates or population inference.
Download the same example in your preferred format
The report files are generated through Survey Studio’s report exporters from the same fixture. The public Excel copy uses one-decimal display formatting while retaining the underlying numeric precision. The package contains fictional row-level data, the codebook and import scripts. No private survey data is included.
The analysis package includes CSV, Excel, JSON, a codebook and scripts. Run the SPSS syntax to create a native SAV file in SPSS; native SAV, DTA and Parquet are not direct website outputs. See the format and import guide.
Turn findings into a testable next step
In a real study, discuss barriers with the lower-use group, choose a relevant practice task and record an owner, review date and observable measure. Repeat the same use question after a comparable opportunity period. A follow-up comparison should distinguish the same participants from a different responding group.
Check a percentage yourself, bring your research task to a walkthrough or create a survey. Do not use this fictional example as evidence of training impact or customer results.