Findings presented above are based on CIRCLE'S analysis of the Census Current Population Survey (CPS) data. Any and all errors are our own. Volunteering estimates are from CPS September Volunteering Supplement, 2002–2010, and the Volunteering in America website at www.volunteeringinamerica.gov. Voting and registration data come from the CPS November Voting/Registration Supplement, 2004–2010, and all other civic engagement indicators, such as access to information and connection to others, come from the CPS 2010 November Civic Engagement Supplement. For some indicators, the 2008 and 2009 data were combined whenever possible, to achieve the largest possible sample size and to minimize error.

CIRCLE uses CPS data because it is rigorous, has a large sample, and is conducted within two weeks after each election, when people are still likely to remember whether or not they voted. The CPS supplements have a large sample of over 100,000 and can therefore be used to estimate voting trends among various groups. For example, the CPS can be used to estimate changes in voting patterns for young women and men, for racial and ethnic groups, and for young people of different education levels. Finally, the CPS is a good source for tracking voting trends over time because the CPS has used a consistent methodology throughout the years so trend lines can be created for voters back to 1972. Most states do not collect demographic information about their voters.

Estimates for the volunteering indicators (e.g., volunteering, working with neighbors, making donations) are based on U.S. residents ages 16 and older. Estimates for civic engagement and social connection indicators (e.g., exchanging favors with neighbors) are based on U.S. residents ages 18 and older. Voting and registration statistics are based on U.S. citizens who are 18 and older (eligible voters). Any time we examined the relationship between educational attainment and engagement, estimates are only based on adults ages 25 and older, based on the assumption that younger people may still be completing their education. The sample size for the 2010 September Volunteering Supplement for Arizona is 1224. The sample size for Arizona for the 2010 November Civic Engagement Supplement is 1261.

Because we draw from multiple sources of data with varying sample sizes, we are not able to compute one margin of error for the state across all indicators. In Arizona, the margins of error for major indicators varied from ± 1.5% to 3.2%, depending on the sample size and other parameters associated with a specific indicator. Any analysis that breaks down the sample into smaller groups (e.g., gender, education) will have smaller samples and therefore the margin of error will increase. It is also important to emphasize that our margin of error estimates are approximate, as CPS sampling is highly complex and accurate estimation of error rates involves many parameters that are not publicly available.

www.ArizonaFuture.org
www.TheArizonaWeWant.org