Researchers Spar on Question of How Much Americans Give to Charity

The elusive answer to the question of how much money Americans give to charity is at the core of a flap involving teams of researchers who study that problem.

Two scholars at Boston College have concluded that the average U.S. household gave more than $1,000 to charity in 1995 — the latest year for which data are available. By contrast, Independent Sector, in its widely circulated biennial survey on giving and volunteering, estimated household giving that year at $696.

The main reason for the gap, the Boston scholars say, is that the Independent Sector figure, based on polling by the Gallup Organization, does not fully reflect giving by the nation’s wealthiest citizens and those with the highest incomes. Better data-collection procedures, they suggest, would yield improved results.

“Our findings indicate that a substantial upward revision in the amount of annual charitable giving is in order,” write Paul G. Schervish, a sociology professor who directs the Social Welfare Research Institute at Boston College, and John J. Havens, a senior research associate there, in the June issue of the journal Nonprofit and Voluntary Sector Quarterly.

They base their conclusions in part on a year-long study of 44 Boston-area households. By carefully documenting charitable contributions on a weekly basis, they concluded that those households reported at least 30 per cent more in total donations than did their demographic counterparts in previous Independent Sector surveys.

Independent Sector, for its part, does not claim that the aggregate figures in its Giving and Volunteering Survey are on the money.

“You don’t have to be an expert to realize that the estimate we put out is low,” says Murray S. Weitzman, a senior consultant who has worked on the survey since 1963. “The survey’s primary purpose is to show behavioral patterns of the American population,” such as the fact that bigger donors are likely to be found among regular churchgoers and volunteers. It makes clear that the country’s wealthiest households were excluded.

“An accurate analysis of giving by the wealthy using an appropriately drawn sample would be very expensive and would need the approval and oversight of the U.S. Department of the Treasury,” write Mr. Weitzman and Virginia A. Hodgkinson, Independent Sector’s Interim Vice-President for Research, in a draft response submitted for publication in the Nonprofit and Voluntary Sector Quarterly.

They write that Independent Sector, a national coalition of non-profit organizations, lacks the estimated $3-million to $5-million it would cost to make use of trained interviewers, have survey respondents keep detailed diaries, and collect information about the richest Americans.

Independent Sector will not conduct a survey this year. Not only is it rethinking its survey methodology, but it also plans to switch its data collection to odd years, since polling organizations are preoccupied with national electoral issues during even years.

The Boston College scholars say that three other national surveys conducted for 1995 tend to confirm their conclusions. The Harvard Survey of Health and Life Quality found an average household contribution of $969 for that year. The Survey of Consumer Finances, which is conducted for the Federal Reserve and which deliberately oversamples wealthy households, listed the figure as $972. And the General Social Survey conducted by the National Opinion Research Corporation estimated household giving at $1,099.

Figures published by the American Association of Fund Raising Counsel in its annual yearbook of philanthropy also support the higher figure, they say. Its original estimate for household giving in 1995 was $1,167 — although it has since been adjusted down to $961 to reflect a new methodology. The figure is based not on survey data but on charitable deductions itemized on tax returns, with a value imputed for non-itemizers.

“The A.A.F.R.C. numbers have been higher by 20 to 30 per cent going back 5 or 10 years,” notes Mr. Havens, who says the discrepancy has long puzzled researchers. Improving the field procedures of the Gallup poll may help close that gap, he suggests.

Adds Mr. Schervish: “We believe that the more carefully the survey is done, the more repeatedly people are asked, and the greater the ability to cover all members of a household, the higher the giving figure will be.”

The Boston survey supports the premise that people tend not to remember — and therefore to understate — some of their charitable donations when they are asked about a year’s worth of giving at a time, the researchers say. “If you include lots of informal giving that people seem to forget — the $5 here, the $10 there — there is more charitable giving,” Mr. Schervish says.

The implication for charities, he adds, is that “people really are giving, especially at the upper end” of the income spectrum. “If this is the case, to inspire people to engage more in what they’re already doing is a different kind of social-policy problem” than motivating people who do not already engage in charitable giving.