Blogs by Audrey Guinn, Ph.D.


  • 14Aug
    Fancy Statistics do not equal causation

    In research, understanding cause is often the goal. What is causing a product to sell? What is causing a decrease in subscriptions?

    Frequently, though, data has been collected using typical surveying methods that will not render answers about causation no matter which robust and fancy statistics are used. The only way to truly determine cause and effect is to control for all extraneous variables by utilizing an experimental design. That way, all other possible explanations for the outcome have been ruled out.

  • 1Aug
    Optimizing Segmentation

    What is the IAT? The Implicit Association Test (IAT) was created by psychologists in 1998 and is believed to measure implicit associations about topics such as race, sexuality, weight, gender, nationality, age, skin tone, religion, and disability (among others).

    Since marketing researchers have begun to use it in their research, @Audrey has put together this blog that helps marketing researchers draw their own conclusions about whether or not to use the IAT in their research.

  • Strategic Impact

    In market segmentation, the distinctiveness of the segments depends on the types of questions used in the segmentation analysis.

    Typically, market segmentation uses 5 question types in the analysis so that segments differ on many facets (needs, behaviors, psychographics, personality characteristics, and demographics), not just needs. Analyzing the data using a variety of these 5 question types gives a holistic view of the consumer market.


  • Structural Equation Modeling

    Structural Equation Modeling is a flexible multi-use tool in the marketing researcher’s pocket.

    Researchers benefit from using SEM due to its multi-functional capabilities. SEM’s benefits are many, such as managing many independent and dependent variables, examining different types of models, accounting for measurement error, and analyzing all relationships simultaneously.

  • Home Comfort Data

    What is Multicollinearity?

    Multicollinearity (also known as collinearity) occurs when two or more variables are very highly correlated. Singularity, a more serious form of multicollinearity, occurs when two or more variables are redundant, where one variable is a linear combination of the others.


  • 6Jul
    Bridging Model

    Republicans, Democrats, and Independents seem to be drifting further apart and these differences are noticeable not only within political ideology but also within more mundane aspects of life.

    Decision Analyst’s monthly “Consumer Reactions to COVID-19” tracker finds that these divisions exist within beliefs about COVID-19 and the vaccine, feelings surrounding the pandemic, concern about the pandemic, and even comfort levels with gathering in different situations.

  • Marketing Research

    Personal characteristics and situational circumstances are potential explanations for why some people receive the vaccine while others do not.

    Therefore, we wanted to understand differences in ethnicity, age, political affiliation, income, gender1, area lived in, and occupation with regards to vaccination. To examine these potential demographic differences, we analyzed the data from Decision Analyst’s monthly “Consumer Reactions to COVID-19” tracker.

  • Marketing Research

    As of May 20th, 48% of the population has had at least one dose. However, that leaves a little over 50% of the population unvaccinated.

    Using the data collected in Decision Analyst’s monthly “Consumer Reactions to COVID-19” tracker, we did examine what impact, if any, beliefs about COVID-19 and its vaccine have on the decision to get vaccinated.

  • B2B Research

    In an effort to catch survey cheaters, researchers use negatively worded attributes placed in groupings of positively worded attributes.

    This context switching causes respondent confusion, which creates error. It may be time for researchers to relinquish negatively worded attributes. So, how can researchers catch cheaters, speeders, and straight-liners if negatively worded attributes are no longer included in the survey?


  • 10Feb
    3 Avoidable Statistical Mistakes by Audrey Guinn, PH.D.
    Avoiding Type 1 Error

    Marketing research is grounded in the scientific method: answering questions by generating a priori hypotheses, collecting data to test hypotheses, and analyzing data to draw conclusions. Adhering to the rules of the scientific method is important to ensure that results are valid and unbiased.

    Sometimes marketing researchers are tempted to use undesirable methods, like conducting many single significance tests, performing statistical tests without hypotheses, and rerunning statistical tests until desired results are discovered. Unfortunately, engaging in these methods has unintended, detrimental consequences: namely, an increase in Type I Error.


  • Strategy Research

    Suppressors are rarely talked about in the marketing research community. They are viewed as the “red-headed stepchild” of statistics: rejected, neglected, and outcast.

    Suppressors are variables that when added to a regression model, change the original relationship between X (a predictor) and Y (the outcome) by making it stronger, weaker, or no longer significant—or even reversing the direction of the relationship (i.e., changing a positive relationship into a negative one). What can researchers do when encountering problem suppressors?

  • Questionnaire Bias

    Biased survey questions wreak havoc on the reliability and validity of the survey which produces junk data.

    Biased questions increase respondent confusion which then increases error in their responses. This in turn reduces the strength of the relationships between variables. In worse case scenarios, biased questions can return results that may be untrue which favor a specific outcome. So what can we do to avoid bias in surveys?

Contact Decision Analyst

Audrey Guinn, Ph.D. ( is a Statistical Consultant in the Advanced Analytics Group at Decision Analyst. She may be reached at 1-800-262-5974 or 1-817-640-6166.


Library Sections


Blogs By Decision Analyst Researchers


Blogs By Advanced Analytics Team


Blogs By Qualitative Research Team