Evaluating the Relative Effectiveness of Two Front-of-pack Nutrition Labels

Research Aims & hypotheses: To use a three arm randomized controlled trial (RCT), and an experimental online grocery store developed by the team, to test two competing approaches of front‐of‐pack (FOP) nutrition labelling on measures of diet quality: 1) Multiple Traffic Lights (MTL) and 2) Nutri‐Score (NS). All participants will experience three different shopping conditions: 1. Arm 1: A control arm that mirrors a traditional web‐grocery store with no HCS or other FOP labels. 2. Arm 2: Similar to Arm 1, with additional MTL labels displayed on all products FOP. A 60‐second introductory video briefly explaining the MTL scheme will be shown before participants shop as they are assumed not to have pre‐existing knowledge of the label. 3. Arm 3: Similar to Arm 2, with Nutri‐Score labels instead of MTL labels displayed on all products FOP. As with the MTL, a 60‐second introductory video briefly explaining the NS scheme will be shown before participants shop. The investigators hypothesize the following: Primary Aim: Diet quality as measured by Alternative Healthy Eating Index Scores (AHEI‐2010) will be highest (best) in Arm 3, followed by Arms 2 and 1 (worst). The investigators further hypothesize a similar ordering (Arm 3 best and Arm 1 worst) for the following secondary outcomes: Total and per serving values of the following: 1. Energy (kCal) 2. Sugar (g) 3. Total fat (g) 4. Saturated fat (g) 5. Sodium (g) 6. Fiber (g) 7. Protein (g) And 8. Average Nutri‐Score across all products purchased where A=5 and E=1, and 9. Cost per calorie and per shopping trip. The investigators will test these hypotheses in total, and separately for foods and beverages given Singapore's focus on reducing intake of sugar‐sweetened beverages. The investigators will also test whether the impact of FOP labelling is moderated by mood and level of hunger at the time of purchase, and by education or income level.