Could your research recommendations benefit from some amped up data analysis options?
If you said “yes,” then there a handful of techniques you need to be proficient with—or at the very least, have a strong appreciation for. And if you are already in the market research profession, you have likely heard about factor analysis, cluster analysis, conjoint, discrete choice and MaxDiff. These techniques are the bedrock of several of your favorite studies including segmentation, product development, pricing and message testing. But don’t be intimidated! Whether you do your own data analysis or work with a data analysis team, you can learn about what these techniques do, and when to recommend using them—or not!
Factor Analysis & Cluster Analysis: Market Segmentation & More
Factor and cluster analysis are data reduction techniques. Factor analysis is used for reducing question sets into a smaller number of “constructs” or “dimensions.” Market researchers use this procedure for creating attitudinal scales and for developing a smaller set of uncorrelated factors. The latter is useful in regression analysis and segmentation research. Cluster analysis is a classification technique used to reduce the number of survey respondents into smaller groups (aka clusters or segments) based on their responses to other questions or factor scores (a byproduct of factor analysis). In combination, these techniques are often used in market segmentation studies, among others.
Real-life Trade-offs: Discrete Choice & MaxDiff
Life is a trade-off exercise. As consumers, we are making complex choices all the time. You’re at the grocery store, staring at the soup shelf: do you pick the organic soup? The low-cost one? The one with more protein? Or the one from a brand you trust? Even if you make a fast choice, at some level you made a trade-off decision, and in that decision, some criteria mattered more than others.
Trade-off techniques including conjoint, discrete choice and MaxDiff allow us to develop an understanding of what the consumer values when making product choices. In new product research, these techniques are superior to scale measures, such as Likert Scaling, because they minimize several forms of bias. If you use these techniques, you will likely be working with a data analyst or stats pro who knows how to conduct them using tools such as SPSS, Sawtooth or others.
If You Want to Amp Up Your Quant Skills, Have a Great Coach
Want to learn more? Research Rockstar has two upcoming courses which will provide you with a deeper understanding of when and how to use these techniques: Introduction to Factor and Cluster Analysis (starts April 9th for four, 90-minute sessions) and Conjoint, Discrete Choice and Max-Diff: An Introduction (meets April 3rd for one two-hour session). These courses are taught by two highly experienced, hands-on market researchers, who will guide you and answer your questions.
Leading you through the data reduction techniques of factor and cluster analysis will be instructor Julie Worwa. Julie’s diverse experience includes research for healthcare, insurance, finance, IT, manufacturing, and travel and leisure, amongst others. Julie is grounded in a variety of advanced methods used for measuring satisfaction, loyalty, and purchase behavior.
Instructor Jeff McKenna will be taking you deeper into the world of trade-off analysis. For many market researchers, this means conjoint, discrete choice and MaxDiff studies. Jeff brings substantial experience in advanced methodologies for new product development, market segmentation, brand development and customer satisfaction measurement.
Last Call! If you want to amp up your quant knowledge, there is still time to add one or both of these courses to your schedule. Introduction to Factor and Cluster Analysis (starts April 9th) and Conjoint, Discrete Choice and Max-Diff: An Introduction (meets April 3rd).
