When designing survey questions, keep in mind how you will use the survey results
When developing survey questions, keep in mind what methods of data analysis will be used and how you are going to use the survey results for analysis and reporting. Even though you may think you have written good quality survey questions, you may be surprised to discover that there may be a more strategic ways of wording your questions to get the results you really need.
For example. let’s say your company wants to find out the percentage of current software users who are receiving software updates on time. You develop a rating scale question of 1 to 10 where 1 is “very poor” and 10 is “excellent.”
Please rate the timeliness of receiving your software updates
1 2 3 4 5 6 7 8 9 10
Let’s say the analysis to this question reveals a mean score of 6.4. 35% of respondents indicated good timeliness of receiving software updates (answered 8, 9, or 10) and 23% of respondents indicated poor timeliness of receiving software updates (answered less than a rating of 5). How do you interpret the results of this survey data?
Instead, what if the survey question was worded like this:
Were you satisfied with the timeliness of receiving your software update? YES NO
The data from a ‘YES’ or ‘NO’ single response question will provide simpler data to analyze and a clearer idea of software update timeliness.
Another thing to consider is questions that may be structured for multivariate analysis. If a concept is to be a part of a regression analysis, separate response options by the same number interval, such as in a rating scale question. This will maximize the correlations between items.
Learn more about how to design surveys and analyze results using Snap Survey Software.