To examine the association between food insecurity and emotional eating (EE) in US Latinxs and explore the mediating role of perceived stress.Cross-sectional analysis. Food insecurity was measured with the six-item US Department of Agriculture Household Food Security Scale; EE with the Three-Factor Eating Questionnaire R18-V2; and perceived stress with Cohen's Perceived Stress Scale-10. Covariates included age, sex, education, marital status, household size and country of birth. Mediation was tested using the Baron and Kenny method and the mediated proportion was calculated. Analyses included multivariable linear regression and multinomial logistic regression.A largely Latinx city in Massachusetts, USA. Participants were recruited from a community health centre serving a large portion of this Latinx community.Latinx individuals (n 580), aged 21-84 years.Overall, 34·4 % were food insecure and 33·8 % experienced High EE. Food insecurity was associated (adjusted OR; 95 % CI) with higher odds of High EE (1·96; 1·28, 3·02) but not Low EE (1·27; 0·82, 1·99). Food insecurity was associated (β; 95 % CI) with higher perceived stress (5·69; 4·20, 7·19). Perceived stress was associated (adjusted OR; 95 % CI) with High EE (1·09; 1·06, 1·12) but not Low EE (1·00; CI 0·97, 1·02). When perceived stress was added in the main effects model, food insecurity was no longer associated (OR; 95 % CI) with High EE (1·31; 0·83, 2·07) and explained 69·9 % of the association between food insecurity and High EE.The association between food insecurity and high EE among Latinxs may be largely mediated by perceived stress. Longitudinal studies are needed.