IBM SPSS Statistics is an integrated family of products that addresses the entire analytical process, from planning to data collection to analysis, reporting and deployment. With more than a dozen fully integrated modules to choose from, you can find the specialized capabilities you need to increase revenue, outperform competitors, conduct research and make better decisions.
The following passage explains the concept of Checking for a Response Set. If you would like to know more about training courses, visit the IBM Training Course Catalog.
When a questionnaire is lengthy, or when a respondent is pressed for time or not that motivated, he or she may show evidence of a response set. A response set is when the same response is given to a set of items, for example always responding “strongly agree”. When you find a response set, you may decide to discard the responses, since they would not be valid. One way to identify response sets is to use the Compute Variable procedure in IBM SPSS Statistics.
In this example, the first set of 15 questions on an SPSS customer survey lists various SPSS products and services and asks respondents to rate them on a scale of 1 to 5. In the figure below, we are using the Compute Variable procedure to create a new variable called “rset” and this variable will be created by calculating the standard deviation of the first set of 15 questions.
We are interested in anyone with a standard deviation of zero because that means that their responses did not vary, indicating that the same response was given for all 15 questions. There are 10 customers who have a value of zero on the variable rset. We can see that these 10 customers gave the same response on all 15 questions. By definition, these people have a response set on this series of items, and they may not have bothered to carefully answer the questions.
At this point, we as the researchers need to determine if we are going to keep the data from these 10 people (if we believe their responses are valid) or we might remove these 10 people from the dataset if we believe that these respondents were not taking our survey seriously.
Follow @basupp on Twitter
for updates, news, articles and videos related to Business Analytics Support.