Structural Equation Models are confirmatory models that allow complex relationships to be modeled and statistically evaluated. SEMs simultaneously perform confirmatory factor analysis and linear regression analysis.
Hypothetical models are specified based on current category understanding. Survey data are fitted to these models and the fit is evaluated using classical statistical measures. In this way, hypothetical customer behavior dynamics can be confirmed or denied by validating or rejecting a specific hypothetical model structure. The final model structure reflects valid causal relationships between attitudinal factors, product attributes, touchpoints and brand preferences creating genuine insight into key drivers that motivate relevant behaviors and beliefs, such as customer satisfaction or brand sales.
Attitudinal and belief latent factors can be derived from attitudinal statements in the survey instrument. By using multiple attitude statements to derive latent factors, reliability and validity of the attitudinal factors are greatly enhanced. Overall customer satisfaction ratings, past purchase history or brand preference ratings can be used in the SEM as dependent variables. The attitudinal factors, as well as other variables directly taken from the survey or derived from data from the survey, can be regressed against overall customer satisfaction in a series of simultaneous linear regression models.
SEMs have the potential to identify and quantify extremely complex relationships between attitudes, perceptions, beliefs, past experiences and overall satisfaction or sales. Understanding these relationships can create rich interpretations of market dynamics that could have significant impact on business strategies.
MACRO Consulting, Inc. has extensive experience building SEMs for a variety of applications, including market structure models, media mix models, brand tracking studies, customer satisfaction and more.