2 The findings come from surveys conducted in 2018 on the IpsosGlobal Advisor platform using the Ipsos Online Panel system.The survey comprising questions A1-A5 was conducted betweenApril 20 and May 4, 2018 with 20,767 adults across 27 countries:Argentina, Australia, Belgium, Brazil, Canada, Chile, China, France,Great Britain, Germany, Hungary, India, Italy, Japan, Malaysia,Mexico, Peru, Poland, Russia, Saudi Arabia, Serbia, South Africa,South Korea, Spain, Sweden, Turkey and the United States. The survey comprising questions B1-B13 was conducted betweenMay 25 and June 8, 2018 with 23,249 adults across 28 countries(the same as above plus Colombia). All survey respondents are aged 18-64 in Canada and the U.S. and16-64 in all other countries . The sample size per country in each survey is approximatelyN=1,000 for Australia, Brazil, Canada, China, France, Italy, Japan,Malaysia, Spain, Germany, Great Britain, and the U.S., andapproximately N=500 for Argentina, Belgium, Colombia, Chile,Hungary, India, Mexico, Peru, Poland, Russia, Saudi Arabia,Serbia, South Africa, South Korea, Sweden, and Turkey. Weighting has been employed to balance demographics andensure that the sample’s composition reflects that of the adultpopulation according to the most recent country census data. A survey with an unweighted probability sample of this size wouldhave an estimated margin of error of +/-3.1 percentage points fora sample of 1,000 and an estimated margin of error of +/-4.5percentage points for a 500 sample 19 times out of 20. In 17 of the countries surveyed ,internet penetration is sufficientlyhigh to think of the samples as representative of the national population within the age ranges covered: Argentina, Australia,Belgium, Canada, France, Germany, Hungary, Italy, Japan, Poland,Serbia, South Korea, Spain, Sweden, Great Britain, and U.S. Brazil, Chile, China, Colombia, India, Malaysia, Mexico, Russia,Peru, Saudi Arabia, South Africa, and Turkey have lower levels ofinternet penetration. Samples from those countries should not beconsidered to be fully nationally representative, but instead torepresent a more affluent, connected population, representing animportant and emerging middle class. Results may not always sum to 100% or may be 1 pointhigher/lower than the actuals due to rounding, multiple responsesor the exclusion of don't knows or not stated responses. 2016 Ipsos.All rights reserved. Contains Ipsos' Confidential and Proprietary information and maynot be disclosed or reproduced without the prior written consent of Ipsos. 。。。。。。