Colleagues -
How do I label items on a sub-scale that were first (1) reversed scored by recoding into a different variable, (2) then were averaged (using the Transform > Compute, etc)?
Do I use the label from the reversed scored items, or the labels from the not-reversed scored items? For example, of my 10 items on the IPIP 50-item 5-point Likert scale, for “Extroversion” 5 items were reversed scored. After I have obtained the MEAN of all of those items, in the Variable view, under “Label”, how do I label that new composite scale, on a scale of 1 – 5? Remember - 5 items were reversed scored. Thanks so much for your help!
Ruth
When I have survey items that should be reverse scored, I recode them appropriately and label them the same variable name but with an R --- e.g. Item10R (versus Item10). I do the recoding BEFORE I run reliabilities, computer statistics, etc. After recoding, I only use the reverse-coded variable thereafter.
Thank you, Diana - that is exactly what I have done.
It is the next step that I am confused about; that is, after you have done what you outlined in your post.
For example, if you have a sub-scale with 10 questions and you sum or average the responses by using the Transform > Compute function, after you click on "OK" to complete the action, you now have to label the summed or averaged items .
When you sum or average sub-scale responses, you will name the computed scores in the "Target" box. Then you will save the new command after you have executed it. It is the next step that is not clear to me: that is, you will have to go to the Label column in the data view and label the newly transformed variable. Which scoring convention do you use: the reverse-scored range, or the normal score range?
You see, in the IPIP scale that I am using, there are 10 items under each personality trait; these are the sub-scales. Within each sub-scale there are reversed scored items. Some items will be scored on a scale of 1 to 5, and others will be 5 -1. Which labeling convention is used after a summing or averaging function has been completed? That is my dilemma :)
Thanks -
Ruth
Colleagues - I have solved my dilemma!
It occurred to me that the reverse scored items take on the values of the other items, so I will simply use the label of the non-reversed items.
After relevant variables have been reverse-coded, then their scale is the same as the other variables in the study. As a generalization, higher numbers typically equate to a more favorable response.
Did you transform the reverse score items so that all the items are going in the same direction before aggregating? For example, if you have 5 items where higher numbers mean more of the trait, and 5 items where higher numbers mean less of the trait, when you average them together what you'll get is mush. So you need to get them all going in the same direction first (doesn't matter which way, you just need to pick one; usually I go higher numbers equal more of the trait) by subtracting out the max from the score (so if you have items that are scored 5-1 and you need them going the other way, Xtransformed = 5-X).
Once you've got everything going in the right direction, just label it something to let you know that those are the ones with the correct transformations done (e.g., in the case of your example, maybe Extroversion_trans).
Hi, Sylvia -
Thank you for your insight :) You are right.
What I ended up doing was to save the original file and use the Transform into Same Variables function. I then reverse-scored the items by entering the Old value as 5 and the new value as 1, etc. Then labeled them from 1 to 5. At first I had transformed the items using the Transform into Different variables function, but that got to be too confusing.
Thank you so much :)
This dissertation is going to be the death of me!
Is it possible to use this method :
Transform>Compute variable
SUM(Q1,Q3,Q4Rev,Q5,Q6Rev)
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS