Data Donation for Climate Action: Linking Psychological Perceptions to Behaviour via Digital Footprints

Mitigating climate change calls for major changes to the way we live. Many UK residents express concern about climate change, but few studies are able to examine whether such attitudes translate to actual behaviour. This is largely because measuring behaviour is challenging.

Research in many behavioural science fields is often limited to self-reported behaviour measures, which are easily undermined by people's inclination to present a socially desirable image of themselves and other biases (e.g., in remembering behaviour). Digital footprint (DF) data, for example store card logs, could potentially help to address this problem.

There are problems in that the data in digital data sources are hard to translate into standard psychological measures, e.g., how does knowing the quantity of different food products someone buys translate into an environmental behaviour measure? Consequently DF data has been underutilised in developing behavioural insights. This project proposes to build on the latest thinking in Psychology and Computer science, to develop new methods, tools, and infrastructure that harness personal DF data donation to test, and enhance, theories of environmental behaviour.

Specifically, we will develop an Environmental Food Purchasing Index (EFPI) by creating an environmental basket of goods (i.e., a collection of food product categories people often buy) based on assessments of the environmental impact of frequently purchased food products. We will provide these to other researchers to support further research in this area.

In addition, we will make this available for public use in a web portal where people can upload their own purchase card data and use the environmental food purchasing index to examine the environmental impact of their own shopping data.

We will also examine people's understanding of the environmental impact of food purchases. Tasks will ask people to rank the environmental impact of different products, and these perceptions will be compared to objective data to find where people are accurate or misperceive impacts. Data gained will allow us to build an understanding of people's mental models of the environmental impact of food, helping us to identify where communications and labelling might improve understanding if needed.

People who use either a Tesco or Co-op store card regularly, and who are willing to donate their purchase card data to our research, will be recruited to complete a survey examining sociodemographics, environmental perceptions, self-report food purchasing, and also perceptions of data donation. We will match up donated store card data, analysed through our previously developed EFPI, with our survey data.

This will help us validate the index and examine what other constructs relate to real life purchase behaviour. The extent of the relationship (or disconnect) between self-report food purchasing and objective store card data will be of particular interest: in addition to helping us to understand the utility of our measures, exploring relationships with demographics will help us understand how relationships between self report and objective data may differ across different people.

We will also consider how representative our datasets are by comparing our sample of people prepared to donate their personal data to wider consumer and UK demographics.

Finally, a sub-sample of participants in the survey will be invited to take part in interviews where they reflect on their personal data, the validity of the environmental purchasing analysis and conclusions, as well as barriers and ethics of data donation. We will have a methodological focus in our research and pass on our newly developed procedures, algorithms, and understanding of participant needs to other researchers, supporting further research in this area.