Sunday, December 19, 2004

Math help needed

Mark Kelly's article in the Ideas section of Today's Globe questions the Generosity Index as published by the Catalogue for Philanthropy. I don't think he goes far enough to challenge the bad math involved here.

"One flaw in studying itemized charitable deductions: An estimated 70 percent of taxpayers don't itemize, so what most of the population gives to charity is unknown. But the Generosity Index's logic is so flawed that the index is almost meaningless anyway. For example, Massachusetts could rank first both in average income and average charitable deduction with, say, $100,000 in each. But subtracting 1 from 1 leaves you a generosity gap of zero; we would still trail Mississippi, despite giving all our money away"

...

To be fair, one of the first people to admit the shortcomings of the Generosity Index is the creator of the index himself: George McCully, president of the Catalog for Philanthropy. He insists he only wanted to create a tool that drew attention to patterns of charitable donations and, ideally, prodded people to give more.

McCully calls his Generosity Index "crude but telling." He's right about the crude part, although the "telling" remains to be seen. With data so slippery and definitions of "generosity" so elusive, it's hard to say how stingy, cheap, or average Massachusetts truly is.

I wholeheartedly agree about the crude part. I don't believe you can work the numbers like they have to come up with anything more meaningful than a conversation starter. Mark's article spends some time on another approach to this but it seems to be far more analytical and complicated than it needs to be. There should be a simple solution for presenting the numbers to respectfully indicate the reality of charitable giving. I will spend some idle moments on it. Raising the question, then providing time to think about it, is generally a good way to come up with a solution.

What are your thoughts on this? How would you come up with an index of generosity?

No comments:

Post a Comment