English
previous
Methodology | Page 76
next

Methodology

The Committee of 100 enlisted the services of Harris Interactive to conduct the English-language survey in the United States and Horizon Research Consultancy Group to conduct the Chinese language survey in China. The 2012 survey covered 75 questions, 49 of which were asked in both the US and China, 19 in the United States only, and seven in China only. The 2007 survey had 65 questions, about 80 percent of which were repeated in the 2012 survey and several with slight modification in wording.

  Sample SizeMargin of Error (%)Field Work Dates
General Public1,000± 3.01/19/2012–1/31/2012
Opinion Leaders244 ± 6.212/14/2011–1/31/2012
Business Leaders120 ± 8.912/14/2011–1/31/2012
Policy Makers36 ± 16.2 12/14/2011–1/31/2012

United States

The English-language survey was conducted in the United States between December 14, 2011 and January 31, 2012. A total of 1,400 interviews were completed, including 1,000 among the general adult population and 400 across various business elites, opinion leaders, think tank executives, and Congressional staffers. The survey was conducted by telephone and administered using a computer-assisted telephone interviewing (CATI) system. The sample composition is as follows:

  • General Public: The survey was administered by telephone between January 4 and January 19, 2012 among a national representative sample consisting of 1,000 US respondents, 18 years of age and over. The methodology uses both landline and cell phone sample. Data were collected among each respondent to understand whether they had a landline only, cell phone only, or both, and to what extent they used each type of line to ensure each of these groups was appropriately represented in the data. A Random-Digit-Dialing (landline) sample fails to represent those who only have cell phones and tends to under-represent those who have landlines but use cell phones for the majority of their calls. The cell phone sample frame, in addition to capturing the cell phone only group, allows adequate representation of “cell phone mostly” individuals as well. According to the most recent data from the National Health Interview Survey conducted by the Center of Disease Control and Prevention, approximately 30% of US adults have a cell phone only, and another 20% of the population uses a cell phone for the majority of their calls. The final sample consisted of 500 who only use landlines, 246 who only use cell phones and 254 who also use either. Results were weighted for age, gender, race, education, household size, region, income, and telephone status (landline and cell phone) where necessary to align them with their actual proportions in the population. The margin of error is plus or minus 3.0 percentage points at 95 percent level of confidence.
Elites
  • Business Leaders: The survey was administered between December 14, 2011 and January 31, 2012 with a sample of 120 business elites nationwide. The sample includes 100 executives and senior level decision makers in corporations ranging in size from 500–5000+ employees and 20 respondents randomly selected from trade associations and groups, lobbying firms, chambers of commerce and economic development groups. The margin of error is plus or minus 8.9 percentage points and the response rate is 10%.
  • Opinion Leaders: The survey was administered between December 14, 2011 and January 31, 2012 with a sample of 244 opinion elites nationwide. The sample includes influential leaders from media, academia, non-governmental organizations, interest groups, foundations and the public sector. Within each subgroup, respondents were randomly selected from purchased sample lists and were screened for their level of familiarity on current issues facing China. The margin of error is plus or minus 6.2 percentage points and the response rate is 12%.
  • Policy Makers: The survey was administered between December 14, 2011 and January 31, 2012 with a sample of 36 respondents comprised of 18 Congressional staffers and 18 think tank respondents. Due to office policy and the Congressional calendar, the Congressional Staffer segment experienced lower than projected response rates, making it difficult to achieve the desired number of respondents. Think Tanks were included as part of the “policy community” to supplement the Congressional sample. Congressional staffers are comprised of randomly selected staff including chiefs of staff, senior advisors, legislative counsels, directors and assistants from purchased Congressional staff lists. Think Tanks are comprised of C-suite executives, board members, chairmen, executive directors, presidents, vice presidents, controllers, assistant controllers, chiefs of staff , directors and assistant directors purchased from sample lists. The margin of error is plus or minus 16.2 percentage points and the response rate is 7%.