If you are assessing the satisfaction of a group of people (customer satisfaction, employee satisfaction, etc.), you are probably interested in knowing how important the different elements are to the group that is being polled. This information is critical to understanding exactly what you need to fix. For example, if your customers are particularly dissatisfied with some aspect of your service, but you discover that the thing they are unhappy with is not important to them, then you know that you do not need to worry about it as much. Similarly, if you find that your customers are moderately satisfied with some aspect of your service, but you also know that this is the most important thing to them, you will know that this is an area you need to focus on improving.
It is tempting to include a set of items that asks people to rate the "importance" of the things they are being asked, but the problem with doing this is that it can double the length of your survey and thereby reduce the response rate.
A better solution is to conduct a simple statistical analysis of the results in which you correlate responses to each "satisfaction" item with the overall level of satisfaction. This would work as follows...
Suppose you have a survey that measures 15 aspects of customer satisfaction and you want to know which are the most important to your customers. As the final question in the survey, ask a general question about overall level of satisfaction.
Then, once you have gathered all the data, use our Correlation Engine to look at the correlations between each item in the survey and the overall level of satisfaction. Items with the highest correlations are the ones that are most important to people and items with a low correlation are less important. With this information, you will be able to target your improvements in the areas where they will be the most effective.
One final note on correlating overall satisfaction with the items in your satisfaction survey: If your respondents come from distinct demographic groups and there are enough respondents in these groups, calculate the correlations for the key subgroups of respondents. You might find that these different respondent groups consider different things more important. (You must define these groups as demographic questions when you create your survey.)
Sample output from our Correlation Engine for an employee satisfaction survey.
