Compare methods

	Best-Worst Scaling of Food Values	Dietary Pattern Analysis
Field	Food Agriculture Studies	Food Agriculture Studies
Family	Process / pipeline	Process / pipeline
Year of origin	2009	2002
Originator	Jayson L. Lusk & Brian C. Briggeman (food values application); Adam Finn & Jordan Louviere (BWS method)	Frank B. Hu; P. K. Newby & Katherine L. Tucker
Type	Maximum-difference choice-based scaling pipeline for food values	Multivariate pipeline for deriving empirical dietary patterns from food intake
Seminal source	Lusk, J. L., & Briggeman, B. C. (2009). Food Values. American Journal of Agricultural Economics, 91(1), 184-196.	Hu, F. B. (2002). Dietary pattern analysis: a new direction in nutritional epidemiology. Current Opinion in Lipidology, 13(1), 3-9.
Aliases	Food Values Best-Worst Scaling, MaxDiff Scaling of Food Values, Lusk-Briggeman Food Values, Best-Worst Food Preference Elicitation	Empirical Dietary Patterns, A Posteriori Dietary Patterns, Data-Driven Dietary Patterns, Eating Pattern Analysis
Related	3	4
Summary	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.	Dietary pattern analysis is the nutritional-epidemiology application of multivariate statistics that identifies how foods are actually eaten together, summarizing the whole diet into a few empirical patterns rather than studying single nutrients in isolation. Introduced as a research direction by Frank Hu in his 2002 Current Opinion in Lipidology review and surveyed methodologically by Newby and Tucker in 2004, the approach takes a matrix of food-group intakes and applies factor (principal component) analysis, cluster analysis, or reduced-rank regression to extract a posteriori patterns such as a 'prudent' pattern rich in fruits, vegetables, and whole grains and a 'Western' pattern high in red meat and refined foods. While the underlying algebra is generic principal component or cluster analysis, what makes this a distinct method is its substantive construction: the input is the food-group intake matrix of the whole diet, and the output is interpretable eating patterns linked to disease.