Description
Forrester conducted an online survey fielded in November and December 2009 of 4,000 individuals ages 16 to 75 in top urban cities/states of Mexico including México City (Distrito Federal), Guadalajara (Jalisco), Monterrey (Nuevo Leó n), Puebla City (Puebla), Toluca (Mexico State), Tijuana (Baja California), Leó n (Guanajuato), Juárez (Chihuahua), Veracruz City (Veracruz), and Mérida (Yucatán) and Brazil including São Paulo city (São Paulo), Rio de Janeiro City (Rio de Janeiro), Recife (Pernanbuco), Porto Alegre (Rio Grande do Sul), Belo Horizonte (Minas Gerais), Salvador (Bahia), Curitiba (Paraná), Brasilia (Distrito Federal), Ribeirão Preto (São Paulo inland), Presidente Prudente (São Paulo inland), São José dos Campos (São Paulo inland), and São José do Rio Preto (São Paulo inland). For results based on a randomly chosen sample of this size (N = 2,000 for Mexico; N = 2,000 for Brazil), there is 95% confidence that the results have a statistical precision of plus or minus 2.2% of what they would be if the entire urban population of individuals ages 16 and older had been surveyed. Forrester weighted the data by age, gender, and socioeconomic level (in Mexico, ABC+, C, and D+ levels are represented; in Brazil, AB1, B2, and C1C2 levels are represented). The survey sample size, when weighted, was 4,002. (Note: Weighted sample sizes can be different from the actual number of respondents to account for individuals generally underrepresented in survey data.) Please note that this was an online survey. Respondents who participate in online surveys have in general more experience with the Internet and feel more comfortable transacting online. The data is weighted to be representative for the total online population of Mexico or Brazil on the weighting targets mentioned, but this sample bias may produce results different from Forrester's Latin American benchmark survey. The sample was drawn from members of Livra's online panel, and respondents were motivated by a prize drawing incentive to participate. The sample provided by Livra is not a random sample. While individuals have been randomly sampled from Livra's panel for this particular survey, they have previously chosen to take part in the Livra online panel.
