My colleagues and I are currently conducting a survey with Likert-scale items and a portion of the survey allows the respondents to add answers which were not included in the prior items (e.g. What other strategies do you suggest? What other challenges did you encounter?).
Would it be preferable to present the data from the open-ended questions through a Word Cloud or would presenting the summary of the data thematically in the table with the responses be better? (Other suggestions would also be greatly appreciated.)
Word clouds are a visually appealing but rather sloppy way to present quantitative data from qualitative responses, i.e., how often something was mentioned. If you did not code what was mentioned, but instead just looked at the frequency with words were mentioned, this would probably not be acceptable to either a quantitatively or qualitatively oriented reviewer.
So, my opinion is that it might be a good way to start constructing a codebook, but it is too light weight for serious analysis.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size