Understanding the Impact of Online Food Advertisements and Emotions on Adolescents’ Food Choices

How to cite

Vecchi, Martina; Fan, Linlin; Myruski, Sarah; Keller, Kathleen Loralee; Nayga, Rodolfo; Yang, Wei. 2022. "Understanding the Impact of Online Food Advertisements and Emotions on Adolescents’ Food Choices." AEA RCT Registry. AEARCTR-0009134. https://doi.org/10.1257/rct.9134-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Martina Vecchi* , Linlin Fan , Sarah Myruski , Kathleen Loralee Keller , Rodolfo Nayga , Wei Yang

* Primary investigator

Registry ID
AEARCTR-0009134
DOI
https://doi.org/10.1257/rct.9134-1.0
First registered
Last updated
Status
In development
Trial dates
2022-05-06 – 2022-08-01
Countries
Private
JEL codes
C99, I12, M37, Q13
Registry record
http://www.socialscienceregistry.org/trials/9134

Abstract

In this experiment, we conduct an online survey with adolescents between 13-17 years old to study the impact of online food advertising on food choices, and the interaction with positive and negative emotions induced through film clips identified through a prior study conducted by the researchers.

Intervention

We will employ a 3X2 between-subjects design where we randomly assign participants to three alternative emotion conditions (negative, neutral, and positive) and two alternative online advertisements (unhealthy food advertisement and non-food online advertisement). Participants are asked to carefully watch two two-minute film-clip validated to elicit the targeted emotion (positive, neutral, or negative). Before and after the two film clips, participants are randomly assigned to one of the two advertising conditions: food advertisements and non-food advertisements. Participants watch three 30-second advertisements either on unhealthy food items (Hershey Kisses, Lay’s potato chips, and Oreos) or on a non-food item (e.g., Nintendo Switch, shoes, Spotify).

Participants then begin a food decision phase. Twenty food items of similar prices are displayed on the screen in random order, to avoid order effects. We present participants with both healthy and unhealthy options: among the twenty food items, ten will be healthy (five sweet and five salty) and ten will be unhealthy (five sweet and five salty). Participants are asked to select the five items they would like to eat. They are also informed that one of every seven respondents will be randomly drawn to receive their selected food items via mail. The food choice is incentivized to motivate participants to make choices representative of their actual preferences, allowing us to identify the impact of advertising on non-hypothetical food choices.

Experimental design

This online experiment will take place in the Spring of 2022. Each participant will be asked to complete an around 30 minutes online survey. We aim to recruit approximately 750 respondents with the help of Qualtrics.
Before initiating the survey, the legal parent or guardian will provide consent for their child to participate in the study, if they choose to proceed with the survey, implied consent is given. Then, parents or legal guardians will be asked screening questions about their child (age and race), their home address, and their email address. Their home address would be used only to deliver the food selected by their child in the survey if randomly selected to receive the food delivery.
Following parent consent, adolescents will give their assent to participate and indicate their race and age before proceeding to make sure they meet our participants’ criteria. Again, if they choose to proceed with the survey, implied consent is given. Firstly, adolescents will be asked about their sociodemographic characteristics (age, gender, school grade, race, state).
At the beginning of the survey, adolescents will report their emotions using the PANAS and their hunger level. We will randomly assign adolescents to one of the three emotion conditions: positive, neutral, and negative. Participants will be asked to carefully watch two 2-minute film clips validated to elicit the targeted emotion (positive, neutral, or negative). At the beginning and at the of the film clips, and between the two emotion-eliciting film clips, participants will be randomly assigned to one of the two advertising conditions: food advertisement and non-food advertisement. Participants in the food advertisement conditions will watch three 30-seconds advertisements about food products. Participants in the non-food advertisement conditions will watch three 30-seconds advertisements about non-food products.
We will then collect again participants’ emotions using the PANAS and hunger level. Participants will then begin a food decision phase. Twenty food items of similar prices will be displayed on the screen in random order, to avoid order effects. We will present participants with healthy and unhealthy options: among the 20 food items, 10 will be healthy (5 sweet and 5 salty) and 10 unhealthy (5 sweet and 5 salty). Participants will be asked to select the 5 items they would like to eat. They will also be informed that 15% of the respondents will be randomly drawn to receive their selected food items via mail.
Before leaving the study, we also collect the perceived healthfulness and taste of the foods offered, the presence of restrained eating, food eaten and drank during the last seven days, emotion regulation, online time use, and self-reported height and weight.

Design details

Randomization method
Randomization was done by Qualtrics
Randomization unit
Individual
Clusters
750 respondents
Observations
125*6=750 respondents
Arms
125 respondents per treatment
Primary outcomes
1. Number of healthy choices 2. Calorie, saturated fat, and sugar content of the chosen foods

Cited by

No citation data available for this registration yet.

Keywords