The data data was collected using questionnaire asking professional opinion about importance of variables , then EFA (using SPSS) was conducted to to explore the latent variables
The argument raised was :
The data from the questionnaire do not measure the variables themselves. They just measure people views on whether the variables are important
THE METHODOLOGY IF IT WAS BASED ON ACTUAL MEASUREMENTS OF THE VARIABLES RATHER THAN PEOPLES VIEWS ON WHETHER THEY WERE IMPORTANT.
Is this argument true ? or it is ok to use the questionnaire ?
I would appreciate if the answers have a supporting theoretical background with citation ,
The argument is correct because we know that the best possible explanatory model is not necessarily the best possible predictive model. See the proof in the appendix of the attached paper. Thus there is often no single best model and hence no one can predict what the best variables are. One can only measure and compute based on the purpose of the model. This assumes that I understand your question, the response of the reviewer and the attached paper. Best, David Booth
Opinion-type data on a Likert scale are routinely subjected to EFA. I suspect that EFA was originally invented for opinion-type data. In Engineering and the exact sciences the scientists make precise measurements. They will not do factor analysis with rotation etc. but they use Principal Components Analysis to reduce the number of variables with a view to graphical display.
Dear Francois ,
Can we say that in Engineering the variables should be measured on actual basis, where in social science/Business mgt can be based on professionals and experts opinions?
Do you have a supporting references ? It will be highly appreciated as I need it to support my research ...I conducted the EFA based on experts opinion , I asked them to rate from 1 to 10 the importance of the measures , then the Factor analysis was used ...
Dear David ,
Thanks for your feedback
I am doing it for Business management not engineering and particularly to develop performance measurement model
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS