Hi~
I was trying to run a multidimensional unfolding with the Prefscal function of SPSS, but it wouldn't execute because:
“Invariant part of the data found. Check, depending on conditionality chosen, your data for constant parts. ”
The variables entered were 10 colums of dissmilarity scores. I don't know if I understand what the warning means, but these scores surely are not constant.
Could you give me some advice on that？
I got it! Thanks to Xiaomin!
Sorting case by the key variables, it's easier to identify the region where a bulk of repeated numbers lie.
In my case, this invariant came from single-number ratings of some participants (e.g. choosing 7 for all items of a scale). I centered the ratings around the means, which made all these single-number ratings 0.
Hopefully someone would find this helpful.
The Multidimensional Unfolding procedure a method to find a common quantitative scale that allows you to visually examine the relationships between two sets of objects. Please check the data is rectangular proximity matrices. and consider each separate column object similar way row of a proximity matrix is considered a separate row object and the matrices are stacked because proximities from multiple sources. it should have two variable and dimensions should not exceed number of ojects-1
A
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS