TL;DR
This paper introduces a massive, cross-cultural dataset of over 59,000 participants from 63 countries, collected through an online survey designed to understand the psychology of climate change. The study tested 11 different interventions aimed at promoting climate action, providing a rich resource for researchers to explore the complex interplay of psychological, demographic, and national factors influencing climate-related beliefs and behaviors globally.
What Problem Does This Paper Address?
Climate change is a pressing global issue, yet understanding the psychological drivers of individual and collective action to mitigate it remains a significant challenge. Much of the existing research is limited to Western, educated, industrialized, rich, and democratic (WEIRD) populations, hindering the generalizability of findings. Furthermore, correlational studies often struggle to establish causal relationships between psychological factors and climate action. This paper addresses these limitations by providing a large-scale, internationally diverse dataset that tests the efficacy of various interventions designed to promote climate change mitigation across a broad spectrum of outcomes and demographics.
Methods
The study employed a quasi-experimental, between-subjects design. Participants were randomly assigned to one of 11 different interventions or a control condition. These interventions were developed through a collaborative process and tested for their ability to influence four key outcomes: climate change belief, support for climate policies, willingness to share climate information on social media, and performance on a pro-environmental behavioral task (the Work for Environmental Protection Task - WEPT). In addition to the intervention effects, participants provided demographic information and reported on other relevant variables like political orientation, environmental identity, and trust in climate science. The data collection was conducted online via Qualtrics, with rigorous attention checks and data cleaning procedures implemented to ensure data quality.
Datasets
The primary dataset introduced in this paper consists of data from 59,508 eligible participants collected between July 2022 and July 2023 across 63 countries. This dataset includes:
- Raw and cleaned survey data: Containing responses to interventions, outcome variables, demographic information, and additional independent variables.
- All survey materials: Including consent forms, intervention descriptions, and questionnaires.
- Relevant analysis scripts: To facilitate replication and further analysis.
- Data visualizations: To aid in understanding the data distributions and findings.
Additionally, the paper describes data from an "intervention tournament" where collaborators submitted and ranked potential interventions, providing insights into the perceived practical and theoretical support for different approaches.
Key Insights and Ideas
- Global Reach of Climate Psychology: The study highlights the feasibility and importance of conducting large-scale, international research on climate psychology, moving beyond WEIRD samples to capture a more nuanced global perspective.
- Intervention Efficacy Across Cultures: By testing 11 distinct interventions across 63 countries, the dataset allows for the examination of how different psychological framing and nudges impact climate beliefs and behaviors in diverse cultural contexts.
- Multi-faceted Measurement of Climate Action: The inclusion of climate beliefs, policy support, social media sharing, and a behavioral task (WEPT) provides a comprehensive view of climate action, acknowledging that it is not a monolithic construct.
- The Power of Collaboration: The paper showcases the success of a large, international collaboration in generating a substantial and valuable dataset, demonstrating a model for future large-scale social science research.
Key Results
While the paper primarily focuses on describing the dataset and its collection, it does present some preliminary findings:
- A total of 83,927 individuals participated, with 59,508 meeting the inclusion criteria (passing attention checks and the WEPT demonstration).
- The dataset includes a wide range of demographic variables, with varying degrees of matching to national population statistics across countries (detailed in Table 2).
- The WEPT task resulted in the planting of 333,333 trees through a collaboration with The Eden Reforestation Project, estimating a significant sequestration of carbon dioxide.
- Reliability measures (Cronbach's Alpha, McDonald's Omega) for the key outcome variables (climate beliefs, policy support, political orientation) indicate good to excellent internal consistency across countries.
Limitations and Future Work
The authors acknowledge several limitations:
- Sampling Heterogeneity: While efforts were made to match samples to national demographics, sampling procedures varied across countries, leading to heterogeneity. However, they also cite research suggesting convenience samples can be adequate for estimating treatment effects.
- Translation and Adaptation Challenges: Despite rigorous translation and adaptation protocols, some interventions required country-specific adaptations, and the paper notes that some teams had issues with full translation and adaptation, with participants from those teams being removed.
- Attrition Rates: While the cleaned dataset aims to maximize utility, there is a small portion of participants who did not finish the entire survey or respond to all items.
Future work can leverage this rich dataset for a multitude of secondary analyses, exploring the effectiveness of interventions across socio-political variables, cultural dimensions (like individualism-collectivism), and other demographic factors. The authors also provide extensive resources, including code and data, to facilitate such future research.
Summary updated