ResPro: Exploring the impact of peer networks: a new lever for adopting healthy and pro-environmental behaviours

This study is based on a randomized controlled trial conducted in 10 private and public firms located in the Grenoble metropolitan area. It aims at investigating the direct and peer effects (through co-workers) in the short, medium and long terms of an intervention which mobilizes psychological and behavioral economics techniques to reduce solo car use and increase physical activity during commuting trips.

Employees will be asked to complete several online questionnaires, including two questionnaires before the intervention and three questionnaires after the intervention. Information on individual sociodemographic characteristics, workload and teleworking practices, social interactions within the company, well-being at work, daily life behaviors (such as diet, physical activity, mobility, pro-environmental habits and leisure activities), as well as attitudes and perceptions related to health and the environment will be collected. Information on workplace interactions based on a predefined list including only colleagues who have agreed to participate in the study will also be gathered as part of the online questionnaires, along with additional information to identify shared characteristics among colleagues not present in the list, and attitudes toward different modes of transport. In addition, over a period of 14 weeks, participants will be asked to report their weekly home-to-work mobility using a web-based application.

External Link(s)

Registration Citation

Citation

Chalabaev, Aïna et al. 2026. "ResPro: Exploring the impact of peer networks: a new lever for adopting healthy and pro-environmental behaviours." AEA RCT Registry. March 10. https://doi.org/10.1257/rct.16888-1.0

Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Experimental Details

Interventions

Intervention(s)

This intervention is designed to support employees in adopting healthier and more sustainable commuting habits. It combines personalized guidance, collective motivation, and progressive support over time.

Participants will be invited to reflect on their daily mobility and receive simple feedback to help them better understand the health, environmental and financial impacts of their travel choices. They will also have access to simulation and planification tools. The programme also includes a collective and playful component that encourages physical activity and mutual motivation among colleagues. At some point of the intervention, participants will receive regular feedback and encouragement to help them gradually integrate new habits into their daily routines.

The overall approach is designed to be supportive, engaging, and realistic, with the aim of fostering lasting changes in commuting behaviors that benefit both individual well-being and the environment.

Intervention Start Date

2026-04-03

Intervention End Date

2026-06-26

Primary Outcomes

Primary Outcomes (end points)

The distance traveled by car in kilometers and the minutes of moderate-to-vigorous physical activity will be tracked for either the most recent commuting trip (based on questionnaire responses) or over the course of the week (based on data from the web-based application).

Primary Outcomes (explanation)

Minutes of moderate-to-vigorous physical activity will be calculated by summing the time spent in all active modes of transportation (walking, biking) during commuting trips.

Secondary Outcomes

Secondary Outcomes (end points)

Well-being at work, mode of transport adopted for other trips' motives, attitudes regarding health, environment and transport modes.

Secondary Outcomes (explanation)

Three dimensions of well-being at work will be measured: i) Burn-out will be measured using the "Burnout Assessment Tool", Schaufeli et al. (2019), ii) Professional vigor will be measured using a sub-selection of items from the Professional Vigor Scale, Isoard-Gautheur et al. (2019), iii) Job satisfaction will be measured using the Professional Life Satisfaction Scale (Fouquereau et al., 2002).

The frequency with which different modes of transportation were used during the last 7 days for grocery shopping, other errands, leisure activities, or accompanying/dropping off people will be collected.

Attitudes regarding health, environment or transport modes will be constructed based on 4 to 5 items.

Self-efficacy with regard to adopting alternatives to the car will be measured using three different items, and the motivational stage for changing modes of transportation (alternatives to cars) will also be assessed using three questions.

Experimental Design

Out of the 10 participating sites—representing slightly fewer companies, as some have multiple locations—half will be randomly assigned to receive the intervention. Within these selected "treated" sites, the intervention tool will then be offered to half of the employees who volunteer to participate.

Experimental Design Details

Not available

Randomization Method

Randomization will be conducted using a computer-generated randomization list (e.g., via Stata software) for both site and employee levels.

Randomization Unit

A two-step randomization will be implemented.
First, sites will be allocated to treated or control groups.
Second, participants belonging to treated sites will be randomly selected to benefit from the intervention.

This double randomization allows to identify both direct and peer effects of the intervention.

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

10 sites

Sample size: planned number of observations

600 employees

Sample size (or number of clusters) by treatment arms

Out of the 10 sites, 5 sites will be allocated to the control group and 5 others to the treated group. Among employees of the treated firms, half of them will be allocated to the control group and half to the treated group.

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

We conduct an ex ante power calculation to determine the Minimum Detectable Effect (MDE) for the direct impact of our intervention. We compute the MDE assuming a sample of 150 treated participants and 300 non-treated participants from control firms (total N = 450). The total sample of participants is 600 and corresponds to the addition of 150 non-treated participants from the treated firms. Because we specify two primary outcomes—(i) the reduction in solo car use for commuting trips and (ii) the increase in physical activity undertaken during commuting—we adopt a two-sided significance level of α = 0.025 to account for multiple hypothesis testing, while maintaining 80% statistical power. Under these assumptions, power calculations indicate that a total sample size of 465 individuals (310 control and 155 treated; allocation ratio 0.5) would allow detection of a minimum effect of 15 percentage points (δ = −0.15), corresponding, for example, to a reduction in solo car use from 50% to 35%.

Supporting Documents and Materials

IRB

Institutional Review Boards (IRBs)

IRB Name

INRAE-CIRAD-IFREMER

IRB Approval Date

2025-07-09

IRB Approval Number

IRB00013805-CEPR40

Analysis Plan