Methodology and How to Interpret Survey Data
The information collected in this report was obtained from in-home personal interviews conducted by The Gallup Organization in the months of May, June and July 1999 with a representative national sample of 2,553 adult Americans 18 years of age or older. This survey included oversampling of African-Americans, Hispanics and affluent Americans with household incomes over $70,000 in order to have enough respondents to carry out statistically reliable analyses of these groups. Interviews were conducted in English, and, when necessary in Spanish. The sampling procedure did not target the very wealthy (those with incomes above $200,000) because they constitute such a small percentage of the population. Weighting procedures were used to ensure that the final sample was representative of the adult population in the United States in terms of age, race/ethnicity, education, marital status, size of household, region of the country and household income. The error rate for the total sample is +/- 3%.

In this report, findings relating to contributions are for households, and findings relating to volunteer time are for individuals from May 1998 to May 1999. Several persons may work or live in a household, and since they usually think of income in relation to their household, respondents were asked to report on aggregate household contributions and income. In contrast,volunteer time was collected for individuals because research has shown that individuals know how much time they volunteer, but do not respond very accurately about the volunteer time of other members of their households. In cases where an individual is asked for his or her opinions, those opinions were reported for individuals and not for households or families. All other questions relating to behaviors, attitudes, or other opinions were for individuals.

The INDEPENDENT SECTOR Giving and Volunteering National Survey has been conducted biennially since 1988. Please note that the most recent survey broke with that tradition. By changing the survey year to 1999, we hoped to move the survey away from traditional election years, when people are inundated with surveys. The survey series will now resume the biennial schedule, with the next scheduled for 2001. The 1999 survey was conducted in May, June and July, at a time when people were most aware of their giving during the past calendar year due to the April 15th deadline for federal income tax returns. Therefore, all giving figures represent calendar year 1998. However, because of issues of recall, respondents were to report on their volunteering in the past 12 months (May 1998 - May 1999). For purposes of this report this time frame will be referred to as 1998. All questions reflecting the respondent’s personal attitudes are those held by the interviewee at the time of the survey.