Structural Equation Modeling to examine the role of emotions in advertising and consumer choice: a case history of partnership activity
Ilaria Parisi | Università degli Studi di Milano-Bicocca (16/11/2011)
Corso di Laurea Magistrale SSE
Relatore: prof. Paolo Mariani
Correlatore: dott. Gianluca D'Innocenzo
Contest and Objective
For a communication agency is still more difficult to explain and capture the effect of below the line media such as events or sponsorship activities through the use of traditional communication models, as the consumer's choices often appear irrational and based on emotional responses. Thus, there is a need for models about event and sponsorship effectiveness. The objective of this research is to design a standardized model to examine how the latent variables, such as involvement, emotions and event attitude, can influence the brand attitude and buying intention and to measure these latent variables by a standard questionnaire, allowing to be used across brands, companies and different types of event or sponsorship.
The conceptual model for the effectiveness of an event has been developed with inspiration from the neuropsychological theory. It has been used a Structural Equation Modeling approach that combines Factor Analysis and Path Analysis into one; the ML method is used for estimating the model and Cronbachs Alpha test is used to validate results. The model delivers three key outputs: 1- KPIs: the overall performance of the event/partnership and impacts on brand health. 2- Impact: the level of influence the event/partnership has on partner brand buying intention. 3- Optimiser: a road map to improve the event/partnership
On the basis of this model, MEC, an international media agency, has created a specific consumer-base tool: ROSE (Return On Sponsorship & Event). This tool offers a synthetic indicator on the marketing manager's desk, allowing him/her to optimize investments and decide if and how to re-invest on an event or a sponsorship. The unique metrics provided together with the possibility to repeat the analysis across time allows the Client to leverage on its own sponsorship results as means of dealing.
SAS Product/Software Used
The software SAS has strongly been used for the whole research project: to import data, to check and clean data, to recode and to standardize latent variables as indicators initially; therefore for a data analysis step using the PROC CALIS designed for Factor and Path Analysis models and finally to manage results by creating, merging and exporting outputs