Mediation in log-linear model

Gloria Gheno ()
Additional contact information
Gloria Gheno: Ca'Foscari University of Venice and ECLT

No 3606137, Proceedings of International Academic Conferences from International Institute of Social and Economic Sciences

Abstract: The analysis of the causality is important in many fields of research. I propose a causal theory to obtain the causal effects in a causal loglinear model. It calculates them using the odds ratio and the concepts proposed by Pearl's causal theory where it is possible. My analysis can be divided into 2 parts. In the first part the effects are calculated distinguishing between a simple mediation model with 1 mediator (model without the multiplicative interaction effect between exogenous variable and mediator) and a mediation model with 1 mediator and the multiplicative interaction effect between exogenous variable and mediator. In both models it is possible also to analyze the cell effect, which is a new interaction effect. Then in a causal loglinear model there are three interaction effects: multiplicative interaction effect, additive interaction effect and cell effect. In the second part the effects are calculated distinguishing between a mediation model with 2 parallel uncorrelated mediators and a mediation model with 2 parallel correlated mediators. In parallel mediation model with correlated influencing variables, Pearl?s theory cannot be used (Pearl, 2014) and and no alternative theory has been proposed. For this reason I propose a new causal concept with relatively formulas to calculate the causal effects in a mediation model with 2 parallel correlated mediators. These types of models are many important in marketing field: for example in customer satisfaction it is important to analyze a model where quality influences the positive and negative emotions and these 3 variables influence the future behavior. Then I show some applications of my causal theory to understand marketing problems.

Keywords: customer satisfaction; direct effect; indirect effect; interaction; loglinear model; mediation model; parallel mediators; total effect (search for similar items in EconPapers)
JEL-codes: C30 C39 M31 (search for similar items in EconPapers)
Pages: 12 pages
Date: 2016-05
References: Add references at CitEc
Citations:

Published in Proceedings of the Proceedings of the 23rd International Academic Conference, Venice, May 2016, pages 543-554

Downloads: (external link)
https://iises.net/proceedings/23rd-international-a ... =36&iid=103&rid=6137 First version, 2016

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sek:iacpro:3606137

Access Statistics for this paper

More papers in Proceedings of International Academic Conferences from International Institute of Social and Economic Sciences
Bibliographic data for series maintained by Klara Cermakova ().