Hi All
I mesured two constructs, intention to book a room (4 items) and intention to book from a web site (3 items), pearson correlation = 0.344**) one of the reviewers recommended me to merge the two scales. Please tell me what's the right way to do that! (both come from the same questionnaire) Factorial analysis of the 7 items at time showed One factor with 60.6 cumulative var
thanks
If they are part of the same factor then you can treat this as a latent variable. If this is the case then you should be able to just calculate the mean score for the factor (which combines these two).
Your factor analysis results are great. As reviewer suggested these two constructs are basically one and only one construct, as the factor analysis yields one factor with 60 Percent variance explained by them, you need to computed / impute only one variable that would seem like intention to book a room.
Do whatever you did to create the 3 and 4 item scales, and apply that to the 7 items. The simplest approach is add all the 7 items together -- which will work well if all 7 items had similar loadings on the single factor you found.
Like David said, since all the things are there (correlation between the items, 60.6 CVE and factor analysis shows loading of 7 items on one factor), you can simple treat 7 items as one variable and do the rest of analysis (whatever you are doing).
There could be an issue if there were no correlation among the items and factor analysis was showing a different result.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS