Meaningful Numbers: Consumer Response to Verbal Reaffirmation of Numerical Nutrition Information

Although the types of front-of-package (FOP) nutrition labels that best engage and influence consumers’ product evaluations and decision-making processes are still poorly understood (Newman, Howlett, and Burton 2016), research seems to converge on the idea that some combination of numerical and easy-to-process information is most effective across a variety of situations (Newman et al. 2018; Sanjari, Jahn, and Boztug 2017). The GDA-Traffic Light (GDA-TL) is one label scheme that combines condensed numerical information about key nutrients (e.g., fat, sugar, and salt) with color indicating whether the amount of each key nutrient is below a defined value (FSA 2016). Due to this combination of numerical information and simple color coding, GDA-TLs are often regarded as the superior FOP nutrition labeling scheme (Borgmeier and Westenhoefer 2009; Grunert, Bolton, and Raats 2012; Siegrist, Leins-Hess, and Keller 2015; van Herpen, Hieke, and van Trijp 2014). Despite the promising features of the GDA-TL label, it is not without its shortcomings. For example, this labeling scheme does not assist consumers in distinguishing between multiple products with the same color profile. Color profiles often do not differ within a category, limiting their usefulness in decision making. Moreover, the color cue is ambiguous when it does not directly reaffirm the numerical information. For instance, three products that contain 6g, 21g, and 23g sugar would be color coded as amber, amber, and red, respectively (FSA 2016). As this example shows, the numerical information would lead to a different inference (i.e., product 1 contains substantially less sugar than the other two) than the color profile (i.e., product 3 contains substantially more sugar than the other two). Against this background, it is not surprising that the GDA-TL does not always work better than a purely numerical label such as the GDA (Crosetto, Muller, and Ruffieux 2016; Koenigstorfer, Groeppel-Klein, and Kamm 2014). From a consumer research perspective, this means that some combinations of numerical and easy-to-process information—especially if the cues do not perfectly align—are not always able to keep their promise of assisting consumers with nutrition information processing.