Statistics Blog

Stated vs Derived Importance: What’s the Difference?

When derived importance is reported alongside stated importance, clients often find the results contradictory and confusing. Knowing the difference between the two and understanding the end goals can help marketing researchers decide when to report stated versus derived importance.

What is stated importance?

Stated importance are preferences that are directly articulated by the respondent. Types of survey questions that measure stated importance include:

• Multiple response: All important features are selected from a list.
• Rating scale: Each feature’s importance is rated on a Likert scale.
• MaxDiff: The most important feature and the least important feature are selected from a set.

Stated importance is not limited to just importance measures but also includes satisfaction, impression, likelihood to purchase, likelihood to consider, and likelihood to recommend, among others. Stated importance is also not limited to the rating of features; it also includes ratings of claims, benefits, characteristics, and attributes, among others. Typically, researchers will create a ranking of the features based on the percentage of people who selected either ‘extremely important’ or ‘very important’ (Top Two Box) or just ‘extremely important’ (Top Box). This reveals the most preferred or important features based on what the majority of respondents have directly articulated.

What is derived importance?

Derived importance are hidden drivers of an outcome, which reveal additional information about the unconscious factors that influence behavior. These silent influencers are discovered through analytical techniques such as:

• Correlation.
• Key driver/Regression/Relative importance.
• Choice modeling.

Like stated importance, derived importance is not limited to questions about importance, and ratings are not limited to features.

Derived importance results often do not match the stated importance results. This mismatching is likely due to psychological effects that are not conscious. Often, what people say they want is different from what they show they want; what people say they do is different from what people actually do. In other words, people often say one thing but behave very differently, and this shows up in the differences between stated importance (saying what is wanted) versus derived importance (showing what is wanted).

Which one should you report?

If the end goal is to increase membership or purchase, for example, then using the derived importance results to show what is most impactful to increasing the outcome is likely more helpful than showing stated importance. If the budget doesn’t cover the analytics required to determine derived importance, then stated importance can be good enough to report. Reporting both stated importance and derived importance may be confusing and unhelpful. If both are reported, it may help to explain that the stated importance results are things the respondents know they want, whereas the derived importance results are things the respondents don’t know they need.

Conclusion

Stated importance often does not match derived importance results. For stated importance, respondents directly tell us which attributes they find to be important. With derived importance, we learn which attributes are important to increasing a specific outcome—and they are ones that respondents may not know impact their decision-making.

Navigating stated versus derived importance doesn’t need to be difficult. Try to avoid reporting both to prevent confusion. If both are requested, explaining the nuance between the two may be most helpful.

Audrey Guinn

Statistical Consultant, Advanced Analytics Group

Audrey utilizes her knowledge in both inferential and Bayesian statistics to solve real-world marketing problems. She has experience in research design, statistical methods, data analysis, and reporting. As a Statistical Consultant, she specializes in market segmentation, SEM, MaxDiff, GG, TURF, and Key Driver analysis. Audrey earned a Ph.D. and Master of Science in Experimental Psychology with an emphasis on emotional decision-making from The University of Texas at Arlington.

• 
•