How do I analyse survey results properly?
Check who answered before comparing any groups. Most survey findings that fall apart do so because the people who abandoned differ systematically from those who finished, and dropout usually correlates with the thing being measured. Then recode reverse-worded items before averaging anything, and report segments below about thirty responses as counts rather than percentages.
Key numbers
- 2 bases
- people who started and people who completed the survey
- <30
- responses in a segment: report counts before percentages
- 1 row
- per respondent before analysing select-all questions
Key takeaways
- Compare starters with finishers before treating the completed responses as representative.
- Reverse-code negative scale items, then check whether each item still behaves like the scale.
- Read free-text themes against the scores and disclose small cells, partial responses, and exclusions.
Section 01
Completion bias beats sample size
A survey with two thousand responses and a forty per cent completion rate among unhappy respondents is less trustworthy than one with four hundred responses and even completion. Size does not fix a systematic absence.
The check takes minutes: compare who finished against who started, split by whatever your headline comparison is about. If dropout correlates with the answer, the headline is measuring dropout.
The decision about partial responses then changes the result, so it belongs in the write-up rather than in a setting nobody looked at. Discarding them is the default in most survey tools and it silently removes the people most likely to be dissatisfied.
In one worked example the tenure gap the survey was run to test disappeared entirely once partial responses were included.
Section 02
Scales need recoding, and then checking
Negatively worded items have to be reversed before anything is averaged, or they cancel out the items they were meant to corroborate. That much is standard.
The step people skip is checking afterwards. Respondents frequently answer a reverse-worded item as though it were positive, and an item that still correlates oddly with the rest of the scale after recoding is not measuring what its neighbours measure. Including it drags the score toward noise.
Section 03
Select-all questions are not one variable
A select-all question is stored as one binary column per option. Treated as separate questions, the percentages sum to something meaningless and every option looks like a minority view.
The base has to be respondents rather than responses, and it should be stated, because forty per cent of respondents and forty per cent of selections are different claims that get reported identically.
Section 04
Free text is evidence, not decoration
Written answers are usually summarised separately from the scores, which wastes both. A theme appearing in forty comments means something different depending on whether those forty people rated you two or five.
Grouping the text by theme and showing the mean score per theme is the most useful thing available, and it takes very little work once the responses are coded.
Section 05
Do you need SPSS to analyse survey results?
Not for counts, completion bias, cross-tabs, scale recoding or descriptive comparisons. A spreadsheet is enough when the question is what this response file says and the groups are large enough to report plainly.
Use SPSS, R or another statistical package when you need weighted survey designs, repeated measures, regression, significance tests, confidence intervals or a reproducible model with diagnostics. The dividing line is inferential statistics, not the number of respondents.
See it on a real project
The apparent 0.42-point gap fell to 0.07 once partial responses were included.
Keep reading
Evidence
Sources
- American Association for Public Opinion Research (2023). Standard Definitions: Final Dispositions of Case Codes and Outcome Rates for Surveys AAPOR, 10th edition.
- Data Analysis App (2026). Survey completion rate: definition and calculation.
- Data Analysis App (2026). When to use an average and when to use a median.