Chapter 7
Module 2Manipulating data
7.2Compute
Femke and Jery were interested in the overall opinion of the entire sample about art house movies. They had, however, four different questions pertaining to participants’ views (the new variable arthous1 – see the previous Assignment 7.1 – and filmint, visit, curious). As you know, one of the aims of statistics is to reduce the information in your data set, or in other words, summarize the results for you and your audience. Here, instead of reporting four different average scores on the four individual items, you may consider reporting just one overall average score. Use the COMPUTE function of the menu TRANSFORM (read about it in Section 7.5.1 of Chapter 7 in case you do not remember what this procedure is for and how it works).
Answer the following question: You might want scores to be on the same scale (1 to 5) to facilitate interpretation of results. How would you do this? Make sure that you include this in your compute procedure. Name the new variable “total” and save the file again.
Say we are interested in the overall opinion of the entire sample about art house movies. We have, however, four different questions pertaining to participants’ views (the new variable arthous1, and filmint, visit, curious). As you know, one of the aims of statistics is to reduce the information in your data set, or in other words, to summarize it for your audience. Here, instead of reporting four different average scores on the four individual items, you may consider reporting just one overall average score. Use the COMPUTE function of the menu TRANSFORM (read about it in Chapter 7 in case you don’t remember what this is for and how it works).
Answer the following question: You might want scores to be on the same scale (1 to 5) to facilitate interpretation of results. How would you do this? Make sure that you include this in your compute procedure. Name the new variable “total”.
Solution
If all went well, this is what you have done:
Note: you divide by four to maintain the scale from 1 to 5. Otherwise the scale would have been a little harder to interpret, because it would run from 1 to 20, while all the other scales in this study run from 1 to 5.