To empower online grocery shoppers in making nutritionally and environmentally informed decisions, we investigate the eficacy of the Scale-Score, a label combining nutritional and environmental information to highlight a product's benefit to both the consumer's and the planet's health, without obscuring either information. We conducted an experimental study in a mock online grocery environment, and assessed label eficacy. We find that the Scale-Score supports nutritious purchases, yet needs improving regarding sustainability support. Our research shows first insights into design considerations and performance of a combined yet disjoint food label.
Labels support consumers in making nutritious and sustainable decisions by transforming
complex information about food e.g., nutritional values, animal welfare standards, or
environmental aspects, into simple logos or diagrams [
Our research focuses on the design space of labels which comprise of both, health and
environmental information; when it matters most, while online grocery shopping [
The Scale-Score (Figure 1, right) combines Nutri- and Eco-Scores into a single label represented by a classic beam scale. It also shows an overall rating based on the product’s Nutri- and Eco-Scores by computing their mean value; in case of an uneven result, we opted to go in favour of the Nutri-Score, prioritising nutrition over sustainability, in accordance with previously gathered user requirements.
We employed a within-subjects design with a single independent variable, visualisation, and three factors, Scale-Score, Nutri-/Eco-Score, and baseline with no visualisation. Dependent variables (i.e., shopping behaviour) were: 1) average environmental value of chosen products (based on Eco-Score calculations) and 2) average nutrition value of chosen products (based on Nutri-Score calculations). In each condition, participants shopped according to a shopping list with three items (cereal, milk, and peanut butter). Participants were entered into a random draw to win their shopping basket, as an incentive to encourage normal purchase decisions. The study lasted one hour.
We recruited 12 participants (5f, = 39 years, = 22.9 years) through our institution’s forums. Regarding demographics, all participants stated having seen the Nutri-Score prior; the Eco-Score was seen by 16.7 % prior.
We used a one-way ANOVA with post-hoc Tukey-tests via IBM SPSS Statistics (v. 28.0.1.0). For this study an alpha ( ) of 0.05 was used.
There was no statistically significant diference between the conditions on ‘nutrition’ as determined by one-way ANOVA ( (2.35) = 0.8 = 0.458) . However, a Tukey-post-hoc-test revealed that Scale- and Nutri-/Eco-Score means did not difer from each other ( -value = 0.994), but both Nutri-/Eco-Score ( -value = 0.557) and Scale-Score ( -value = 0.496) difered compared to no score. Looking at the descriptive statistics, Scale-Score resulted in lower nutrition values (mean = 2.89) compared to Nutri-/Eco-Score (mean = 3.06) and no score (mean = 4.78).
There also was no statistically significant diference between the conditions on ‘sustainability’ as determined by one-way ANOVA ( (2.35) = 1, 301 = 0.286) . Tukey-post-hoc-test revealed a non-significant diference between Nutri-/Eco-Score ( -value = 0.595) and Scale-Score ( -value = 0.810) compared to no score. Looking at the descriptive statistics, Scale-Score resulted in lowest sustainability values (mean = 53.11), Nutri-/Eco-Score in highest (mean = 59.78) and no score in intermediate (mean = 55.69).
To achieve a global and successful transition to healthy and sustainable diets, systems and tools are needed to support consumers in this. We designed Scale-Score, a label that displays nutritional and environmental information.
We did not find significant diferences in support provision of either visualisation compared to baseline, however there is a trend showing the Nutri- and Eco-Score combination may support consumers in sustainable and healthy decision making. The Scale-Score may support nutritious choices compared to no visualisation, however, worsened environmental impact of the basket compared to baseline. First, this may be due to the make-up of the Scale-Score: nutritional aspect weighted more into the final score. Consequently, a product that is marked with a good Scale-Score rating (e.g. B) may well contain an environmental ‘D’ rating. As a result, the average sustainability score was worse compared to Nutri- and Eco-Score representation, resulting in Scale-Score’s poor environmental performance. Second, participants may have ignored the multi-level information provided, given the small sizes of Nutri- and Eco-Score labels within the Scale-Score, contributing further to the de-valuation of environmental information.
In future, we plan to re-design the label taking advantage of the interaction modalities available in web-based interfaces, where meta and multi-level information can efectively support sustainable and nutritious grocery shopping. At ICT4S, we hope to inspire and engage in conversations on user-centred food label designs embedded in personal informatics.
Longer version of this submission is available on http://oops.uni-oldenburg.de/5512 (German). Source code for the online supermarket is available via GitHub.