The Emergence and Efficacy of Longitudinal Data: Brand Asset Valuator a Pioneer in Longitudinal Data
Understanding developmental growth is crucial for explaining how brands grow and decay, and research on brand equity is incomplete without the use of longitudinal methods. Moreover, longitudinal data is the only methodological design that accounts for variation in respondents over time. In an area such as brand equity, where vast variability across respondents is likely, these changes in respondent level data are essential to the comprehension of how brands ebb and flow.
Although much progress has been made in the research designs and methodologies to understand, explain, and predict consumer behaviors, many brand strategy consulting firms, and even academics, still utilize cross-sectional (correlational) data. Cross-sectional data studies have been conducted and will continue to be conducted because they efficiently answer many types of research questions unrelated to longitudinal or causal inferences.
Aside from their utility to answer various types of research questions, they cannot address temporal precedence. Temporal precedence is the concept that a cause must precede an effect in time, so that the effect one variable has on another variable requires some amount of time to elapse, even if the amount of time required to pass is so minimal that the effect appears instantaneous. Thus, cross-sectional data are vulnerable to the lack of clarity about which variable transpired first and may yield uncertainty about which variable is the cause and which is the effect. In other words, cross-sectional data only provides a snapshot of a single moment in time and does not consider what happens before or after the snapshot is taken. Therefore causal factors or mechanisms cannot be established. In brand-consumer research, not being able to establish temporal precedence has huge ramifications for being able to correctly identify drivers and predictors that change consumer perceptions of brands over time.
In sum, the research and methodological design of a brand consumer study is more important than the statistical analysis. A poorly designed brand consumer study cannot be reversed or modified, whereas a flawed or unfit statistical analysis can be reanalyzed. Also, it is important to note that advanced analytics (e.g., Structural Equation Modeling) will not correct for issues with temporal precedence. Thus, whether it be designing a quantitative survey or brainstorming how to answer changes in consumer perceptions of brands over time, always weigh the pros and cons of a cross-sectional versus longitudinal design to achieve the most accurate and evocative results possible.