I am working on a career interest test which is based on paired comparison data. There are 12 dimensions in 6 groups. Each dimension from each group is compared with all dimensions of all other groups one by one (but not compared with the other dimension within its own group).
I am looking for a R package that can help me with:
1. Examining if the groups and dimensions within the groups allocation is a suitable model (confirmatory factor analysis on hierchical factor modeling)
2. The 12 dimensions are supposed to occupy different axis in a sphere. I am looking for a pack that can help me place the dimensions in a circle and show that to me visually along with fit statistics.
Any suggestion will be highly appreciated.
You have a large choice, I would start with this Link:
sem packages in r - Bing
Best wishes, David Booth
Hello Vijai,
For the sem, have a look at lavaan (https://lavaan.ugent.be/)
For visual display, it wasn't clear what you're asking for. There are a number of libraries and functions that might apply.
Good luck with your work.
If I understand you correctly, you have twelve dimensions that define a hyper-sphere (of 12d). How many variables do you have measurements of to evaluate if this 12-d model is appropriate? Do you hypothesize these are orthogonal dimensions??? To suggest an appropriate package it is worth knowing more about your model and your data.
Daniel Wright thanks for your answer. I think I was not clear enough in my questions. So, here, I am trying to rephrase it. I have 12 observed variables (I called them dimensions) and they are supposed to be in 6 groups (2 in each). These 6 groups of supposed be in a 2 dimensional space (both dimensions are supposed to be orthogonal). Now, I have data collected from ~2000 individuals on this questionnaire. I want to see if the hypothetical structure (12 variables, 6 groups in a 2 dimensional space exists in the data). In Mplus, I can do an exploratory factor analysis or confirmatory factor analysis by creating the diagram and get the numbers to see if the model fits. How can I do that in R?
The data is little complicated in the structure. There are 12 items measuring each of the 12 variables (144 items total), these 144 items are in 72 pairs of items in such a way that items from one group (2 variables in that group) and put with the items form opposite group (G1 is Juxtaposed to G3, G2 to G4 and G3 to G6). It is Ipsative comparison forcing the choice between two opposites.
Vijai Pandey, you may find the xxM package in R relevant and useful. Here is the link to this package: https://xxm.times.uh.edu/
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling