# Testing the effect of a dynamic descriptive social norm message on meat-free food selection in worksite cafeterias: a randomized controlled trial

## TL;DR

This study tested if social norm messages in cafeterias could increase meat-free meal choices but found no significant effect.

## Contribution

The study evaluates the effectiveness of co-created dynamic social norm messages to promote meat-free food selection in real-world settings.

## Key findings

- Dynamic social norm messages did not increase meat-free meal sales in worksite cafeterias.
- Most customers did not notice or recall the messages, suggesting low intervention salience.
- The intervention had no impact on overall meal sales or dietary behavior change.

## Abstract

Overconsumption of meat is a threat to planetary health. Meat consumption is socially and culturally patterned, and interventions using social norms could be a promising strategy to encourage meat reduction.

We developed and tested the effectiveness of a dynamic descriptive social norm message displayed in worksite cafeterias (N = 25, intervention = 12, control = 13) to increase meat-free meal selection. The message was developed based on existing evidence and in collaboration with the catering company operating the cafeterias. The message communicated a specific change in target behavior, using a relevant and relatable referent group, grounding the desired behavior change in time and place, and included a clear call to action. The social norm messages were displayed in each intervention cafeteria for 8 weeks on free-standing banners, posters, and floor stickers. We compared the change in weekly percentage of meat-free meal sales (measured as number of meals sold) between intervention and control cafeterias through linear mixed-effects models. We conducted fidelity checks in intervention cafeterias and interviewed customers to assess perceptions of the intervention.

There was no evidence that the intervention led to an increase in sales of meat-free meals (− 2.22 percentage point change, 95% CIs [− 7.33, 2.90], p = 0.378). Pre-intervention baseline sales of meat-free meals varied by site, but there was no evidence the intervention was differentially effective for sites with higher vs. lower baselines. There was also no evidence that the intervention changed overall meal sales. The intervention was implemented with high fidelity, though out of 155 customers interviewed, 57% reported that they did not notice the messages, and only 2% correctly recalled the message.

There was no evidence that empirically informed and co-created dynamic descriptive social norm messages increased selection of meat-free meals in worksite cafeterias. This could be due to low salience of the intervention in a busy, fast-paced environment, or the strength of existing eating habits in a workplace cafeteria. The findings suggest that norm messaging interventions, when delivered as an isolated intervention, may not be effective to change a complex and socially grounded dietary behavior such as meat consumption.

OSF Registries, Registered September 23, 2022, https://osf.io/h7zkf

The online version contains supplementary material available at 10.1186/s12916-025-04302-9.