The purpose of this technical note is to document data sources and survey instruments, address issues of statistical reliability, and define terms used in the report. Statistics and data reported here come from the Volunteer and Civic Engagement supplements, which have been conducted by the U.S.. Bureau of the Census each fall since 2002 and 2008 respectively as part of the annual Current Population Survey (CPS).
Bureau of Labor Statistics Technical Notes
The CPS is a monthly survey of about 60,000 households (approximately 100,000 adults), conducted by the U.S. Census Bureau for the Bureau of Labor Statistics. The CPS focuses on obtaining information on employment and unemployment for the nation's civilian non-institutionalized population ages 16 and older. The purpose of the September and November supplements is to obtain information on the incidence of volunteering, the characteristics of volunteers, and civic life indicators in the United States.
Since 2002, the Census Bureau has introduced adjustments to the population controls for the CPS as part of its annual update of population estimates. In addition to the population control adjustments, new questions on race and ethnicity were introduced into the CPS in January 2003. For a discussion of the changes introduced into the CPS in January 2003, see "Revisions to the Current Population Survey Effective in January 2003," in the February 2003 issue of Employment and Earnings, available on the BLS Web site at http://www.bls.gov/cps/rvcps03.pdf (PDF 73 KB). For a discussion of the changes introduced in January 2004, see "Adjustments to Household Survey Population Estimates in January 2004," in the February 2004 issue of Employment and Earnings, available on the BLS Web site at http://www.bls.gov/cps/cps04adj.pdf (PDF 12 KB). For a discussion of the changes introduced in January 2005, see "Adjustments to Household Survey Population Estimates in January 2005," in the February 2005 issue of Employment and Earnings, available on the BLS Web site at http://www.bls.gov/cps/cps05adj.pdf (PDF 32 KB). For a discussion of the changes introduced in January 2006, see "Adjustments to Household Survey Population Estimates in January 2006," in the February 2006 issue of Employment and Earnings, available on the BLS Web site at http://www.bls.gov/cps/cps06adj.pdf (PDF 239 KB). For a discussion of the changes introduced in January 2007, see "Adjustments to Household Survey Population Estimates in January 2007," in the February 2007 issue of Employment and Earnings, available on the BLS Web site at http://www.bls.gov/cps/cps07adj.pdf (PDF 53 KB). For a discussion of the changes introduced in January 2008, see "Adjustments to Household Survey Population Estimates in January 2008," in the February 2008 issue of Employment and Earnings Online, available on the BLS Web site at http://www.bls.gov/cps/cps08adj.pdf. For a discussion of the introduction of the revised population controls and the impact of their introduction on CPS data, see "Adjustments to Household Survey Population Estimates in January 2009," available at http://www.bls.gov/cps/cps09adj.pdf. For a discussion of the introduction of the revised population controls and the impact of their introduction on CPS data, see "Adjustments to Household Survey Population Estimates in January 2010," available at http://www.bls.gov/cps/cps10adj.pdf. For a discussion of the introduction of the revised population controls and the impact of their introduction on CPS data, see "Adjustments to Household Survey Population Estimates in January 2010," available at http://www.bls.gov/cps/cps11adj.pdf. For a discussion of the introduction of the revised population controls and the impact of their introduction on CPS data, see "Adjustments to Household Survey Population Estimates in January 2011," available at http://www.bls.gov/cps/cps12adj.pdf.
The CPS response rate at the household level varied between 92 percent and 94 percent between 2003 and 2005. For individuals in households that completed the basic CPS, the person-level response rate for the volunteer supplement varied between 86 percent and 88 percent over this same time period. As of 2006, the Census Bureau no longer calculates the overall response rate for the CPS volunteer supplements. According to the documentation for the September 2006 CPS: "Since the basic CPS nonresponse rate is a household-level rate and the Volunteer Service supplement nonresponse rate is a person-level rate, we cannot combine these rates to derive an overall nonresponse rate. Since it is unlikely the nonresponding households to the basic CPS have the same number of persons as the households successfully interviewed, combining these rates would result in an overestimate of the 'true' person-level overall nonresponse rate for the Volunteer Service supplement" (pages 16-3 and 16-4).
Volunteer Questions and Concepts
Volunteers are defined as persons who performed unpaid volunteer activities at any point during the 12-month period, from September 1 of the prior year through the survey week in September of the survey year. The count of volunteers includes only persons who volunteered through or for an organization - the figures do not include persons who volunteered in a more informal manner.
The survey was introduced as follows: "This month, we are interested in volunteer activities, that is, activities for which people are not paid, except perhaps expenses. We only want you to include volunteer activities that you did through or for an organization, even if you only did them once in a while."
Following this introduction, respondents were asked the first supplement question: "Since September 1 of last year, have you done any volunteer activities through or for an organization?"
If respondents did not answer "yes" to the first question, they were asked the following question: "Sometimes people don't think of activities they do infrequently or activities they do for children's schools or youth organizations as volunteer activities. Since September 1 of last year, have you done any of these types of volunteer activities?"
Respondents were considered volunteers if they answered "yes" to either of these questions. This is the same method of identifying volunteers that was used in each of the volunteer supplements since 2002. CNCS defines regular volunteers as individuals who served 100 hours or more at one or more organizations during the previous year surveyed.
Respondents classified as volunteers were asked further questions about the number and type of organizations for which they volunteered, total hours spent volunteering, how they became involved with the main organization for which they volunteered, the type of activities they performed for the main organization, and what their main activity was.
Organizations are associations, societies, or groups of people who share a common interest. Examples include youth groups, civic organizations, churches, synagogues, and other religious institutions. For the purpose of this CPS supplement, organizations are grouped into major categories, including religious, youth, and social or community service organizations. The main organization is the organization for which the volunteer worked the most hours during the year. If a respondent volunteered for only one organization, it was considered the main organization.
Activities are the specific tasks the volunteer did for an organization. Examples include tutoring, fundraising, and serving food. For the purposes of providing more detailed information, the category "Any other type of activity/specify," was not included in the list of the main volunteer activities.
Starting in 2006, individuals were asked, "In the last year, have you attended any public meetings in which there was discussion of community affairs?" and, "In the last year, have you worked with other people in your neighborhood to fix or improve something?" Starting in 2008, individuals were also asked, "During the past 12 months, did you donate money, assets, or property with a combined value of more than $25 to charitable or religious organizations?"
Civic Engagement Indicators
The November Civic Engagement Supplement is designed to measure participation in organized groups, nonelectoral political activities, social connections with family and community members, and trust and confidence in people and institutions.
In November 2009, the Civic Supplement was administered to a cut-down sample of about 14,000 outgoing CPS households (that were either finishing their obligation altogether or beginning their scheduled six-month break). Several questions were also eliminated from the November 2008 CPS survey instrument, including three questions about nonelectoral political participation, and the questions about sources for news and current events. This shortened version of the survey was also administered in November 2010, along with the CPS Voting Supplement.
Changes to the 2011 Civic Supplement:
In 2011, BLS did not administer the voting supplement however it did include seven new questions. One item pertains to voting in local elections, and an additional question asked about using the internet to express opinions about political or community issues. A question on the 2010 supplement that asked about "contact with family and friends via the internet" was modified on the 2011 supplement to "see or hear from family and friends." BLS also added four additional questions to the 2011 supplement pertaining to trust in people and confidence in institutions (See the Glossary for item phrasings). A final modification involves how BLS coded frequency on several items. On the 2011 supplement BLS changed how frequency was coded to include an additional category, "Less than once per month."
Reliability of the Estimates
Statistics based on the CPS are subject to both sampling and nonsampling error. When a sample, rather than the entire population, is surveyed, there is a chance that the sample estimates may differ from the "true" population values they represent. The exact difference, or sampling error, varies depending on the particular sample selected, and this variability is measured by the standard error of the estimate. There is about a 90% chance, or level of confidence, that an estimate based on a sample will differ by no more than 1.6 standard errors from the "true" population value because of sampling error. BLS analyses are generally conducted at the 90 percent level of confidence.
The CPS data also are affected by nonsampling error. Nonsampling error can occur for many reasons, including the failure to sample a segment of the population, inability to obtain information for all respondents in the sample, inability or unwillingness of respondents to provide correct information, and errors made in the collection or processing of the data. For a full discussion of the reliability of data from the CPS and for information on estimating standard errors, see the CPS "Explanatory Notes and Estimates of Error."
Many of the statistics we report are calculated using multiple years of data to increase the reliability of the estimates. As single-year estimates for small segments of the population may be unreliable, we most often use a three-year average that combines responses from the 2009, 20010, and 2011 volunteer supplements to increase the reliability of the estimates. For medium-sized metropolitan areas with samples large enough to meet our data reliability standards, we report four-year averages using data from the 2008, 2009, 20010 and 2011 CPS supplements; for the "additional" metropolitan areas, which tend to have smaller annual samples, we use five years of data (2006-2010) to calculate volunteer statistics. For statistics where the sample size is too small to yield reliable estimates, the estimated values are suppressed and replaced by an asterisk ("*"). Additionally, some values for median hours volunteered by subgroups of the volunteer population have been topcoded. While using multiple years of data increases the reliability of the estimates, caution is advised when interpreting rankings, as differences between some areas may not be statistically significant.
In 2011 the Census Bureau improved the age masking technique applied to public use data files to protect the confidentiality of its respondents. Prior to 2011, BLS tabulations and estimates derived from the public use data file always agreed. Prior masking technique allowed 16+ population estimates to be equal to those generated by the public use file. With the new age masking technique, this is no longer the case. Beginning in 2011, estimates using the new masking technique will differ from BLS population estimates.
For some single-year statistics, readers may find small differences between this report and the BLS Volunteering in the United States report. These differences are attributable to restrictions placed on the public use data files.
Historical Volunteer Data from 1974 and 1989
Earlier versions of the CPS Volunteer Supplement were conducted in May 1989 and April 1974. For all three periods, we define an adult volunteer as someone age 16 and older who did work through an organization in the previous 12 months for which they were not paid. This adult definition excludes respondents age 14 and 15 from the analysis, consistent with the definition used by the Bureau of Labor Statistics (BLS) since 1989. The definition used in 1974 was based on all respondents aged 14 and older.
For determining organization information, the organizational type categories have remained relatively consistent between 1989 and the present. The 1974 survey did not ask respondents which organization(s) they volunteered with. Therefore, we are not able to include the organization type from 1974. When necessary, the labels of the 1989 categories have been revised to match the present-day category titles. The present-day BLS coding of organizational types includes two categories that were not available to respondents to the 1989 survey: "Environmental and animal care" and "Public safety." These categories have been recoded as "Other" for comparability with 1989.
During the 1974, 1989 and current volunteer surveys, the survey instrument used a different number of prompt questions to determine the volunteer rate. The 2002, 2003, 2004, 2005, 2006, and 2007 volunteer rates are calculated based on responses to two volunteer prompts in the volunteer survey instrument ("Since September 1st of last year, [have you] done any volunteer activities through or for an organization?" and "Sometimes people don't think of activities they do infrequently or activities they do for [ a ] children's schools or youth organizations as volunteer activities. Since September last year, [have you] done any of these types of volunteer activities?") The 1974 volunteer rate was based on responses to a question similar to the first volunteer prompt on today's September supplement. The 1989 volunteer rate was calculated based on responses to two volunteer questions: "Last week, [have you done] any unpaid volunteer work?" and "Even though [you] did not do any unpaid volunteer work last week, did you do any unpaid volunteer work over the last twelve months, that is since May 1, 1988?"
In 1974, some states did not have a sufficient sample size to determine the volunteer rate. In the state profiles, where there was insufficient data to generate a volunteer rate for 1974, there is an N/A or Not Applicable in the column for 1974.
To measure volunteer retention, we analyzed panel data from the CPS. Each September, 50 percent of all households that participate in the CPS are also selected to participate the following September. Among members of these households, almost 90 percent of respondents who answer the questions about volunteering in the first year also answer the volunteering questions in the second year. We analyzed the changes in the responses to the volunteering questions from respondents in the 2009-2010 and 2010-2011 volunteer supplements to calculate the national volunteer retention rate, as well as retention rates for regions and states. For large metropolitan areas, we report retention rates calculated from the 2007-2008, 2009-2009, 2009-2010 and 2010-2011 CPS panels.
Definition of Cities and Metropolitan Statistical Areas
All volunteer data for cities are collected from the metropolitan area in which the city is located, including metro area residents outside the city limits. City volunteer statistics are divided into three groups: large, mid-size, and additional. The groups are defined using two criteria: Census population estimates for metropolitan areas, and number of metropolitan-area residents who responded to the Current Population Survey (CPS) Volunteer Supplement. Large cities represent the 51 largest metropolitan areas, according to 2009 Census population estimates. Mid-size cities represent metropolitan areas with populations generally greater than 100,000 and less than 1 million people and a sufficiently large number of CPS respondents. Additional cities also represent metropolitan areas with populations of less than 1 million people but greater than 100,000, and have CPS samples large enough to provide reliable estimates of a limited number of volunteer statistics. Cities of any population size that did not have the requisite CPS sample size were not included on this Web site.
Details about the Metropolitan Statistical Areas (MSAs) from which city data are collected can be found on the city profile pages. MSAs carry the name of one or more principal cities, the most heavily urbanized cities in the area; the names of the principal cities are used as designations for the city data published here. The federal Office of Management and Budget (OMB) is responsible for determining and publishing the boundaries of MSAs. In order for an area to be designated an MSAs, the area has to have at least one urbanized area of 50,000 or more in population, plus adjacent territory that has a high degree of social and economic integration with the urban core as measured by commuting ties. The MSAs definitions used in the CPS Volunteering Supplements can be found in the Appendix to OMB Bulletin #03-04, issued June 30, 2003. Since then, OMB's periodic updates have changed the names and principal cities for some MSAs, but the same county-based geographic definitions described in OMB Bulletin #03-04 remain in place today.
The Current Population Survey changes its MSA definitions once every 10 years, to reflect population changes documented by the decennial Census. For the New England states (Connecticut, Rhode Island, Massachusetts, Vermont, New Hampshire, and Maine), the CPS uses NECTAs (New England City and Town Areas), which are composed of towns and cities, to describe metropolitan areas, rather than MSAs, which are composed of counties. Since the available CPS data does not allow respondents to be identified by county, all New England volunteer statistics are reported for NECTAs. Boundary definitions for NECTAs can be found in the Appendix to OMB Bulletin #03-04.
Data Sources for Other Indicators
For most of the indicators on the "Factors that influence the volunteer rate" page, the original source was the 2006 State and Metro Area Data Book. The data book, a Census Bureau publication, has data on population density (Table B-1), educational attainment and poverty rate (Table B-4), and homeownership rates and housing units located in multi-unit structures (Table B-7), as well as details about the original data sources, which, in many cases, was Census 2000.
The American Community Survey (ACS) provides updated data at all geographical levels for several indicators. American FactFinder, accessed July 2012, served as the source for single-year (2010) and pooled three year (2008-2010) ACS data for several indicators. Additionally, home ownership rate, units in multi-unit structures, average travel time to work, high school grad rates, poverty rates, and bachelor’s degree rates are from ACS 2010 3-year estimates. Population estimates for 2010 and 2011, land area (2010) and population density (2010) are derived from the 2010 Census. Unemployment rates are from the Bureau of Labor Statistics.
Data on nonprofits per 1000 residents were originally taken from the Internal Revenue Service's Exempt Organizations Business Master File, August 2010. Churches and congregations are excluded from the totals. State and county data from this master file were downloaded from the Urban Institute's National Center for Charitable Statistics in July 2011, and aggregated to form organizations per 1000 residents for metropolitan areas, regions, and the USA.
Data on unemployment rates were taken from the Bureau of Labor Statistics' Local Area Unemployment Statistics (LAUS) program (http://www.bls.gov/lau/). The LAUS program produces monthly and annual employment, unemployment, and labor force data for Census regions and divisions, states, counties, metropolitan areas, and many cities, by place of residence. Monthly data from October 2010 through September 2011 were used to compute annualized unemployment rates that match the reference period for the CPS Volunteer Supplement.
Due to cost restrictions, data on foreclosure rates were omitted from this year's report. Prior to 2011 data were compiled by RealtyTrac ( http://www.realtytrac.com/trendcenter/ ). RealtyTrac's county-level "inventory" datasets show the number of housing units in some status of foreclosure as of a given date. Inventory data from September 30, 2009 and September 30, 2010 were used for Volunteering in America 2011. Under the licensing arrangements, we publish the difference in foreclosure rates between 2009 and 2010, but not the rates themselves for either year.
For metropolitan areas in New England, volunteer rates and volunteer retention rates come from the CPS and thus are reported at the NECTAs level. Data for all other indicators are reported for the geographical area that corresponds most closely to the NECTAs's boundaries.