Level: Advanced
The main purpose of selective two-sided causal analysis in CS (customer satisfaction) surveys is to "objectively clarify which targets should be addressed." For this reason, we focus on cause-related items (services, product characteristics) that have a particularly strong impact on result-related items (satisfaction evaluation), and actively exclude other items during the analysis process, taking a priority-oriented approach (the vital few and the trivial many).
However, it is valuable to use all the data collected in the CS survey after completing the selective two-sided causal analysis to understand the overall causal structure of CS, as this will enable advanced and useful knowledge to be gained. As a useful method for this purpose, we propose a method based on SEM (structural equation modeling) that restores and utilizes the data excluded in the selective two-sided causal analysis process.
Even if they do not have a strong direct effect on the outcome system, those that influence factors in the causal system (causal items and factors) can provide valuable information in understanding the overall causal structure. The creation of a path diagram for SEM after a selective two-sided causal analysis begins with processing the structural model diagram that is the final result of the selective two-sided causal analysis. Items that were excluded during the analysis process are then gradually added back to this to gradually build an overall causal structure. As is clear from the above, the essence of this method is reverse causal analysis, which uses the causal structure obtained from the selective two-sided causal analysis as its core and reverses the diagram to focus on the factors behind it. Although it is not SEM itself, it is essentially an approach that conforms to SEM and can therefore be positioned as quasi-SEM.
*No video will be released for this announcement.
