I am conducting an analysis of customer satisfaction survey data. Many of the variables in the data are measuring the same concept. For example, several of the questions in the customer questionnaire measure the degree to which the service person with whom the customer dealt was competent. In the analysis I want to measure the strength of influence of that concept (competence) on an outcome variable (overall satisfaction). I also want to measure the degree to which each of those observed variables are associated with competence.
My question is this: when I create the node to represent competence, how do I decide how many states it has? Do I simply assume it has the same number of states as the observed variables that comprise it?
The front end.
1 post • Page 1 of 1