Evaluator: Kearney, Matthew D.; Maheu, Arlene R.; Booth, Madalyn; Newberg, Andrew B.; Cronholm, Peter F.; & Ayubcha, Soussan. University of Pennsylvania; Thomas Jefferson University.

Evaluation Methods:

In this 2024 longitudinal observational study, researchers analyzed matched baseline and post-intervention surveys from 171 of 904 participating students enrolled in Vetri Cooking Lab during the 2018-2019 academic year. A 46-item survey was administered at the start and end of the program, measuring culinary confidence, culinary knowledge, dietary consumption behaviors, and preferences for 20 vegetables.  For the survey analysis, researchers excluded students skipping more than one-quarter of either survey and used median imputation to account for any missing responses. Researchers used paired t-tests to assess pre-post changes, while employing Pearson correlations and Cohen’s d to quantify associations and effect sizes. Using multivariable logistic regression, researchers examined whether changes in knowledge and confidence predicted positive consumption changes after adjustment for age, race, ethnicity, gender, language, and program site. The study acknowledged limitations such as lacking a comparison group, a low completed survey response rate, and relying on self-reported outcomes; thus, researchers established findings as pre-post associations rather than causal effects, noting that they may not be generalizable beyond similar populations.

Evaluation Type: Non-experimental

Summary of Outcomes:

A 2024 study evaluating the impact of a nutrition education program on young people’s eating habits found statistically significant pre-post gains in cooking confidence and nutrition knowledge. Mean confidence among Vetri Cooking Lab participants increased from 14 to 15.52 on a 0-20 scale, and mean knowledge rose from 11.65 to 13.29. Knowledge and confidence were positively correlated, meaning that as nutrition knowledge increased, so did students’ cooking confidence. Although the overall consumption behavior score did not significantly change, nearly half of students reported at least some change in consumption behaviors (49.7%). Additionally, increased confidence was positively associated with consumption change. Students who became more confident about adding a vegetable to each meal or understanding the benefits of fruits and vegetables had nearly three times higher odds of reporting consumption changes. Knowledge change predicted positive consumption change only before adjustment.

Changes in vegetable familiarity and preferences were also statistically significant. The mean percentage of vegetables liked increased significantly, from 36.1% to 43%, while the mean percentage of students who had not tried them declined from 33.5% to 24.7%. The percentage who disliked increased modestly from 30.4% to 32.3%. Vegetables presented during VCL showed significant increases in liking and significant declines in “not tried” responses than vegetables not presented during programming. Among vegetables presented in the program, chickpeas, bell peppers, and mushrooms had the largest relative gains in liking.

Age and gender were not associated with consumption changes. White students had higher adjusted odds of reporting positive consumption changes than Black students. And in unadjusted models, students identifying as Hispanic/Latino and speaking Spanish had three times the likelihood of consumption behavior changes.