Best-Worst Scaling of Food Values
Also known as: Food Values Best-Worst Scaling, MaxDiff Scaling of Food Values, Lusk-Briggeman Food Values, Best-Worst Food Preference Elicitation
Best-worst scaling of food values measures how much consumers care about a fixed set of food attributes — safety, price, taste, nutrition, naturalness, origin, environmental impact, fairness, and so on — by repeatedly asking them to pick the most and least important value from small subsets. Jayson Lusk and Brian Briggeman's 2009 article 'Food Values' introduced this specific application, adapting the best-worst (maximum-difference) scaling method that Finn and Louviere pioneered for food-safety research. Rather than rating each value on a 1-to-5 scale, where everything tends to look important, respondents are forced to trade values off against one another, yielding a discriminating, interval-scaled ranking of what truly drives their food choices and avoiding the scale-use biases that plague conventional importance ratings.
Key highlights
- Forces trade-offs, producing discriminating, interval-scaled importances instead of flat, ceiling-bound ratings.
- Largely free of the scale-use and response-style biases that distort cross-respondent and cross-cultural rating comparisons.
- Cognitively easy for respondents — picking the best and worst of a few items is fast and intuitive.
- Supports both simple count-based scores and full random-utility modelling, with natural segmentation of consumers.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use best-worst scaling of food values when you need a discriminating, comparable measure of how consumers prioritise multiple food attributes and you want to avoid the ceiling effects and scale-use bias of importance ratings. It is well suited to comparing priorities across countries or cultures (where rating styles differ), segmenting consumers by what they value, tracking how priorities shift over time or after an event, and informing product, labelling, and policy decisions. It is less appropriate when you need willingness-to-pay for specific product configurations or the values of attribute levels (a choice experiment or conjoint study is better), when the list of relevant values is unknown or unstable, or when respondents cannot reliably complete repeated forced-choice tasks.
Strengths & limitations
- Forces trade-offs, producing discriminating, interval-scaled importances instead of flat, ceiling-bound ratings.
- Largely free of the scale-use and response-style biases that distort cross-respondent and cross-cultural rating comparisons.
- Cognitively easy for respondents — picking the best and worst of a few items is fast and intuitive.
- Supports both simple count-based scores and full random-utility modelling, with natural segmentation of consumers.
- Yields only relative importance among the items presented, not absolute importance or willingness to pay.
- Results depend on the completeness and framing of the fixed value list; omitted values cannot be scaled.
- Does not directly value attribute levels or product configurations the way a choice experiment does.
- Requires a balanced design and enough tasks per respondent, lengthening surveys relative to simple ratings.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
How is best-worst scaling different from rating each food value on a scale?
Importance rating scales let respondents call everything important, producing ceiling effects and little discrimination, and they are contaminated by individual and cultural differences in how people use scales. Best-worst scaling instead forces respondents to choose the most and least important value in each subset, so they must trade values off against one another. This yields a discriminating, interval-scaled ranking on a common ruler and largely removes scale-use bias, which makes it especially valuable for comparing priorities across people and cultures. The price is that you learn relative, not absolute, importance.
Is best-worst scaling the same as a choice experiment or conjoint analysis?
They share a random-utility foundation but answer different questions. A choice experiment or conjoint study presents whole product profiles defined by attribute levels and estimates the value of those levels, often in willingness-to-pay terms. Best-worst scaling of food values instead scales the relative importance of a list of abstract values (safety, taste, fairness, and so on) by having respondents pick the best and worst items in subsets. Use best-worst scaling to rank what consumers care about; use a choice experiment to value specific product configurations and estimate willingness to pay.
What is the simplest way to score best-worst data?
The most transparent measure is the best-minus-worst score: for each value, count how many times it was chosen as most important and subtract how many times it was chosen as least important, summed across respondents or computed per person. Positive scores indicate generally important values, negative scores unimportant ones, and the ordering closely tracks the more sophisticated logit-based estimates. Analysts often rescale these counts (for instance by a square-root or share-of-preference transformation) for easier interpretation, and reserve full conditional-logit or hierarchical-Bayes modelling for individual-level estimates and segmentation.
Sources
- 1.Lusk, J. L., & Briggeman, B. C. (2009). Food Values. American Journal of Agricultural Economics, 91(1), 184-196.
- 2.Finn, A., & Louviere, J. J. (1992). Determining the Appropriate Response to Evidence of Public Concern: The Case of Food Safety. Journal of Public Policy & Marketing, 11(2), 12-25.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 23). Best-Worst Scaling of Food Values. ScholarGate. https://scholargate.app/food-agriculture-studies/best-worst-scaling-food-values