Aspects of statistical regression in sensometrics

Abstract A discussion is presented of roles of regression analysis in sensometric studies, distinguishing description, interpretation and prediction purposes. A brief review is given of linear regression methods for prediction in situations with near-collinear explanatory variables, including for example ridge regression and partial least squares, and latent variable models are discussed. Finally problems with statistical and causal inference from regression on covariates in designed experiments are discussed. Illustrations in the paper are based on a sensometric study of apple flavour under varied storage conditions [ Brockhoff, P., Skovgaard, I., Poll, L., & Hansen, K. (1993). A comparison of methods for linear prediction of apple flavor from gas chromatographic measurements. Food Quality and Preference , 4 , 215–222 ], with sensory response data and with gas chromatography measurements as covariates.