International Public Opinion on Climate Change: Household Climate Actions – Adoption and Barriers, 2023

Appendix I: Survey Method

The data in this report are based on a survey of 139,136 Facebook monthly active users, aged 18 and older. Responses were collected from 187 countries and territories worldwide, including 107 individual countries and territories and 3 geographic groups comprising 80 additional countries and territories.

This report includes a data sub-sample of 37 high-emission countries and territories with 39,683 observations. The countries in this list are both: (1) above the global average in emissions (4.4 tons CO2e) and (2) have a GNI per capita that is higher than 12,535 $US and therefore are classified as high-income by the World Bank. The survey was conducted August 3 – September 3, 2023. Sampled Facebook users received an invitation to answer a short survey at the top of their Facebook News Feed and had the option to click the invitation to complete the survey on the Facebook platform.

The sample was drawn from the population of Facebook monthly active users, defined as registered and logged-in Facebook users who had visited Facebook through the website or a mobile device in the last 30 days.The Facebook monthly active users figure is reported in the company’s quarterly earnings report found on As of June 30, 2023, there were 3.03 billion monthly active users globally. Within each country or territory surveyed, we drew a sample in proportion to publicly available age and gender benchmarks.

All countries and territories were sampled in proportion to data from the United Nations Population Division 2019 World Population Projections. The sample population in the United States was drawn in proportion to the U.S. Census Bureau Current Population Survey 2018 March Supplement.

Data were then weighed separately for each country, territory, and group using a multi-stage, pre- and post-survey weighting process based on census and nationally representative survey benchmarks, Facebook demographics, and Facebook engagement metrics, balanced to the total number of survey completions.Weights were generated in three stages. First, we adjusted for sampling in proportion to age and gender census benchmarks. Second, we adjusted for non-response error using Inverse Propensity Score Weighting (IPSW) on Facebook demographics and Facebook engagement metrics to make the sample more representative of Facebook monthly active users. The Facebook monthly active user sampling frame constitutes a sizable proportion of the global population, but the proportion of adults who use Facebook differs from country to country. To more closely reflect the entire adult population in a country or territory, we used poststratification based on publicly available nationally representative survey benchmarks, using weights from the first stages as inputs. The final weights were balanced to the total number of survey completions.

This project is a partnership between the Yale Program on Climate Change Communication, Data for Good at Meta, and Rare’s Center for Behavior & the Environment. The survey instrument was designed by Anthony Leiserowitz, Seth Rosenthal, Marija Verner, and Jennifer Carman of Yale University, with support by Emily Wood, Natalia Ordaz Reynoso, and Nicole Buttermore at Meta; and Erik Thulin at Rare. The survey was fielded by Meta. Survey weights were generated by Nicole Buttermore and Natalia Ordaz Reynoso. The charts and maps were designed by Emily Goddard and Jennifer Marlon of Yale University.

Rounding error and tabulation

Percentages are weighted based on general population parameters for each surveyed country and territory.

For tabulation purposes, percentage points are rounded to the nearest whole number. As a result, percentages in a given chart may total slightly higher or lower than 100%. Summed response categories (e.g., “a lot” + “a moderate amount”) are rounded after sums are calculated (e.g., 25.3% + 25.3% = 50.6%, which, after rounding, would be reported as 25% + 25% = 51%).

Sample size for surveyed countries, territories, and groups

Our study collected data from a total of 187 countries and territories,Samples were drawn from two additional areas (Eritrea and Tuvalu), but no respondents participated in the surveys in these areas, so they are excluded from the final list. including 80 countries and territories that were collapsed into three geographic groups due to sample size (i.e., Caribbean, Asian & Pacific Islands, and Sub-Saharan Africa). In this particular report, we use a data sub-sample of 37 high-emission countries and territories with 39,683 observations. The unweighted sample sizes for each country and territory in this data subset are as follows:

  • Australia (n = 936)
  • Austria (n = 1,157)
  • Belgium (n = 1,027)
  • Canada (n = 1,010)
  • Chile (n = 1,192)
  • Croatia (n = 669)
  • Cyprus (n = 571)
  • Czechia (n = 1,104)
  • Denmark (n = 1,020)
  • Finland (n = 631)
  • France (includes Saint Pierre and Miquelon) (n = 1,497)
  • Germany (n = 2,023)
  • Greece (n = 1,024)
  • Hong Kong (n = 467)
  • Hungary (n = 620)
  • Ireland (n = 1,080)
  • Israel (n = 988)
  • Italy (n = 1,093)
  • Japan (n = 1,116)
  • Lithuania (n = 1,301)
  • Netherlands, The (n = 927)
  • New Zealand (n = 1,029)
  • Norway (n = 1,134)
  • Oman (n = 483)
  • Poland (n = 897)
  • Portugal (n = 948)
  • Saudi Arabia (n = 1,440)
  • Singapore (n = 378​​)
  • Slovakia (n = 1,050)
  • South Korea (n = 1,038)
  • Spain (n = 912)
  • Sweden (n = 1,299)
  • Switzerland (n = 1,007)
  • Trinidad and Tobago (n = 539)
  • United Arab Emirates (n = 1,271)
  • United Kingdom (n = 1,969)
  • United States of America (n = 2,836)