THE INFLUENCE OF TESTING CONTEXT ON TEA BAG PRODUCT ACCEPTANCE IN CENTRAL LOCATION TESTS

Abstract

ABSTRACT We investigate the reliability and predictive power (correspondence with home‐use test [HUT] outcomes) of central location tests (CLTs) using a tea bag product. The setting involves three different CLT testing contexts reflecting “natural versus controlled” approaches: (1) traditional CLT, using controlled testing with ad libitum consumption (n = 47); (2) dosing CLT (DCLT), offering dose choices (sugar and milk combinations) that are different across consumers but similar across samples for individuals (n = 47); and (3) free choice CLT (FCLT), simulating natural tea‐making circumstances and individual brewing preferences (n = 44). CLT results are compared with outcomes from HUTs, employing Pearson correlation coefficients on CLT‐HUT scores as indicators of the ability of CLT outcomes to predict HUT outcomes. Analysis of variance and internal preference mapping methods are also applied. FCLT is found to better discriminate between samples than the other CLTs. However, only the DCLT group revealed a statistically significant (though weak in magnitude) relationship (r = 0.234) between CLT and HUT.PRACTICAL APPLICATIONSCentral location tests (CLTs) are used extensively in sensory research because of their cost‐effectiveness and ease of implementation. However, questions remain about whether CLT responses are consistent with responses from more natural, longer term testing environments. This research studies consumer liking in CLT and compares it with choices of tea made and consumed at home in home‐use tests (HUT). It aims to address two major outcomes from three different CLT testing contexts: (1) the degree of relationship to liking scores derived from long‐term testing in HUT; and (2) ability to distinguish between samples. The testing product chosen here involves complexity in preparation and significant cultural contextual consumption. Thus, the results are expected to contribute to improving design of CLTs to predict long‐term acceptance and choices in food research.