Dataset of the Day: Who is more Generous? Republicans or Democrats?
January 7th, 2009by Kevin Burke
About the Author: Kevin Burke is a member of the FortiusOne analysis team. Read more from this authorI came across an article that appeared in the New York Times called “Bleeding Heart Tightwads” by Nicholas D. Kristof. Kristof, a liberal, goes on to say that liberals tend to be stingy when it comes to giving to charity. I found it to be very interesting and also contradictory to what I originally believed. I had always in the past believed liberals to be the group that is extremely generous and more prone to give to charity than Republicans. This is because liberals tend to push for policies that focus on using government spending to increase opportunities for the needy, something that is consistent of a charitable nature. After reading this article and doing some further research I decided to use Finder! and Maker! to display some of my findings on the subject.
In Kristof’s article he says he used the source, The Catalogue of Philanthropy to measure generosity by state. I went to this source and decided to map the inverse of what they call their “Generosity Index Ranking”. The site describes the index as rank of each state’s average adjusted gross income (AAGI) to the rank of each state’s average itemized charitable deductions (AICD). The arithmetical differences between these two rankings are then themselves ranked, resulting in the Generosity Index rank. So to basically sum it up the Generosity Index measures who spends a greater percentage of their income on charitable causes.
I mapped the Catalogue of Philanthropy’s figures for 1996, 2000, and 2004 because these years coincided with Presidential Elections were you would be able to see which states were red (conservative) or blue (liberal). The following combined datasets are mapped below. The orange states represent states that were red (conservative) and the blue states represent states that were blue (liberal). Also, the white circles represent the Inverse Generosity Index Ranking for the state. A larger white circle means a top ranking or that residents of the state contribute a larger percentage of their income to charity. Pay close attention to the average generosity scores at the bottom of each map. (Please click on images for a larger view or go to their Maker! weblink below the picture for an interactive view)
maker link
Average Red State Generosity Index Score = 33.84, Average Blue State Generosity Index Score = 11.89
maker link
Average Red State Generosity Index Score = 32.80, Average Blue State Generosity Index Score = 14.55
maker link
Average Red State Generosity Index Score = 35.37, Average Blue State Generosity Index Score = 19.45
During all three years states that voted republican had higher Inverse Generosity Index Scores as a whole over states that voted democratic according to the average index scores. Just by looking at the maps you can see that the orange colored states (conservative states) have the larger white circles (high rankings) and the blue colored states (liberal states) have the smaller white circles (low rankings). The next map shows states that during all three elections in 96, 00, and 04 voted for the same party versus an average of the state’s Inverse Generosity Index score for those three years. This is to give a good overall average of how very democratic states compare to very republican states when it comes to giving to charity.
maker link
Average Red State Generosity Index Score = 35.35, Average Blue State Generosity Index Score = 12.46
As you can see from the above map the data from past Presidential Elections the truly republican states have higher generosity rankings than the states that have solely voted for democrats. So maybe it is true that republicans are more generous.
Kristof in his article says that liberals often claim these findings are misleading because conservative states have higher religious populations. This causes their charity to go toward building big churches which is not accurate of measuring charity. I decided to take a close look at this accusation.
Below I mapped percentages of state populations that say they practice no religion (Dark Blue = high no religion population, dark orange = low no religion population) from a 2001 study by a group from Graduate Center of the City University of New York. I then compared these figures with the Inverse Generosity Index rankings from 2000. By doing this I figured it would give a look at how states with high religious participation gave to charity and how it compared to how states with low religious populations gave to charity.
Do liberals have a valid accusation? According to this map it appears so. For the most part it seems that dark orange states, the states that have high religious populations, have large black circles which indicate that they have a higher Inverse Generosity Index. Also most of the dark blue states have small black circles. But there are still a few outliers and when comparing correlations between no religion vs. charity to politics vs. charity we see that there is a stronger correlation between politics vs. charity (see below).
Values closer to -1 or a slope that is almost a straight line, show strong correlations. Therefore we see that low activity in church vs. generosity does not show as strong of a correlation as politics vs. charity. So can liberals really say that their lack of giving is because they are not giving to churches as much as conservatives? Are the facts about religious participation strong enough to discredit the facts on political participation?
So are liberals stingier than conservatives?. The data has suggested that they are. One thing that I am unhappy with is that this data is somewhat dated with the most recent year being in 2004 for philanthropy stats from the Catalogue of Philanthropy. I am anxious to see if the trend has continued into the present and am eager to compare 2008 charity figures with red and blue states from the 2008 Presidential Election.
Overall, I like how Kristof does not see the data as a negative, but a way to encourage more of his fellow liberals to contribute more. He states in his article, “Come on liberals, redeem yourselves, and put your wallets where your hearts are.”
Popularity: 28% [?]
January 7th, 2009 at 11:46 am
“This is because liberals tend to push for policies that focus on using government spending to increase opportunities for the needy, something that is consistent of a charitable nature.”
This statement is flat wrong. Charity is putting your hand in your own pocket to giving to someone in you decide is in need.
Liberals advocate the government putting its hand in the pocket of someone else to give to someone it decides is in need (very often this decision is based not on genuine need but political gain).
January 7th, 2009 at 12:08 pm
this interpretation has been a virtual truism for the last couple of decades (i remember reading about this in the 90s). one interpretation may be that often those on the left of the spectrum also find employment with organizations that do ‘good’, and thus contribute their time and sacrificed pay: two things such studies don’t attempt to quantify.
i don’t have any data to support my theory, but i think it’s one approach to the issue that i haven’t read about yet. in sum, if someone works for a non-profit for 35K, s/he doesn’t have a whole lot to give away monetarily speaking. but the time, thought, and energy spent at work is her/his contribution.
of course, there are more complicating factors as to why one chooses to work for socially progressive organizations–not all of them altruistic.
another spin on these data is that ‘liberals’ want to spend collective money, not their own–they want to socialize charity, whereas ‘conservatives’ want to individualize charity. that, too, sounds like a legitimate interpretation of fundamentally different world-views. ‘liberals’ want us all to chip in (through taxes), and ‘conservatives’ want to pick their own charities (and pay fewer taxes). of course, in practice, republican and democrat politicians both want citizens to pay taxes–it’s just how they spend it that differentiates them sometimes (i say sometimes, b/c both have their own lobbyists to appease, and the lion’s share of money spent is rarely in accordance w/ their average constituent’s wishes).
‘liberals’ seem to grasp that economies of scale are more efficient (larger pot of $ can do more for R&D, schools, healthcare, etc.), whereas ‘conservatives’ are more concerned about individual liberty. it’s a ‘we are in this together, so let’s all pitch in and help’ vs. ‘your liberty ends where mine begins, so i won’t tell you how to spend your money, and you leave my money alone.’ i don’t think one side is more generous than the other: i think their worldviews are distinct and incompatible. the only way to move past the impasse is for one side to convince greater numbers of the cogency of one’s argument (instead of the usual demagoguery that passes for political discourse in the US)
January 7th, 2009 at 2:03 pm
My only comment is…..I think we are mixing apples and oranges here. Not all Republicans or Democrats can be pushed into the conservative vs. liberal camps. There are conservative democrats out there, just as there are liberal republicans.
There should be two comparisons. Republican vs. Democrat and liberal vs. conservative. Not necessarily mixing the two.
Great post.
January 7th, 2009 at 3:54 pm
I enjoyed reading your analysis.
I think that there might be an additional factor. People in blue states might have less disposable income to give to charity because they pay more in taxes. It’s not that they like to pay taxes, but most blue states get less back in federal money than they pay, while most red states receive more than they pay. So, blue states (or donor states) have higher state income taxes to make up the difference, while most red states (or welfare states, if you will) have lower state income taxes.
January 7th, 2009 at 4:48 pm
This analysis suffers from the ecological fallacy. You cannot connect a states voting pattern to state charity donation data and say anything meaningful about liberal or conservative philanthropy. You may be able to say that people in Mississippi are, on average, more generous than people in Wisconsin. But you can’t say why with any certainty.
Consider a scenario in which a state’s minority ‘liberals’ make a majority of the donations and donate heavily. The state would have both a high ‘Generosity Index’ and be red but you would know nothing about liberal vs. conservative giving patterns.
The overall theme may hold true — that conservatives donate more than liberals — but this analysis does nothing to bolster that conjecture. It may be useful as exploratory statistics but it fails as real, rigorous analysis.
January 8th, 2009 at 7:30 pm
yep, ecological fallacy is a big issue with that analysis; theoretically (very unlikely though) it would be possible that no single “conservative” donated anything at all and still get the same outcome, but there is a far bigger problem with that analysis:
this philantrophy organisation builds an ordinal scale of 2 Values and then does builds the arithmetic difference of it to determine the most generous states. This is total bogus. Imagine theoretically the people of the richest state meaning Rank 1 in AAGI donates 100$ of their money their money so they are also Rank 1 in AICD. Now what score do they get? 1-1 = 0
Now take a state which has Rank 40 in AAGI and the people there donates 10% of their income which makes them lets say place 15 in AICD. Now they get a score of 40-15 = 25. This means they would have a much more favourable Rank in the Generiosity Index.
That Index totally favours the poorer states and as most rich states vote democrats the findings are not surprising.
Seriously I haven’t seen such a hilarious statistical prank for a long time. I got suspicious as soon as I couldn’t understand immediately how the Ranking works, first I thought I was just tired and my english skills are too bad, but then I asked myself why they made such a complicated construct instead of just either taking the ratio of donated money / income – either arithmetical middle or maybe even better the median values if that non-aggregated data is actually available.
If I recall correctly most states in the last election voted something like 45-55% for either candidate and that difference would also be too small to really explain big chances very likely
January 8th, 2009 at 7:49 pm
the 100$ in the 2nd paragraph are supposed to be 100%, sorry for that.
P.S: Am I blind or is there no way to edit a comment?
January 8th, 2009 at 8:22 pm
sorry for posting 3 comments in a Row , I intented to go to bed after each single one of them, but it let me no rest.
I quickly calculated the percentage of income spend for charitable deductions for each state for 2000 and 2004 and matched against these 2 specific elections it surprisingly doesn’t change the result too much. People in states the Dems one spend about 6,5% in 2000 and 2004 and Republicans spend 9,5%. That doesn’t how incredibly stupid the GI is and for 1996 and
But for the 1996 and 2008 the results would probably look very different.
How is the AICD measured by the way?
There is a lot more to this topic but its already 2:22 AM here now so I’ll finally go to sleep.
January 9th, 2009 at 12:27 pm
Aurel,
Great Posts. Sorry if they kept you up late last night
The link below might help you with some of your questions. It is technical notes from the Catalogue of Philanthropy on their study. Hope it helps.
http://www.catalogueforphilanthropy.org/natl/generosity_index/2004_us_notes.php
Thanks for all the great comments everyone!
January 9th, 2009 at 6:58 pm
[...] was reading Kevin Burke’s post today on the relationship between political affiliation and charitable giving, and thought it was a [...]
January 22nd, 2009 at 9:25 am
The problem with this is that contributions to churches, which are not really charities are counted here. People who are swindled into paying huge salaries to church leaders are distorting these statistics. Basically this shows that people in red states have less ability to think on their own, and feel obligated to pay people to lots of money to feel good about their own existence.
February 11th, 2009 at 1:55 pm
Hello Kevin,
I’m curious about the correlation graphs and coefficients you show above. Were they generated via GeoCommons, or with an external tool?
Thanks,
Dan
February 12th, 2009 at 10:20 am
Hi Dan,
They were done with GeoCommons, but it is not something we have on the public site. It is easy to come to misleading conclusion with the tool for those without a background in statistics, so still trying to figure how to deal with that. The tool and several related data analysis tools are available for our enterprise customers where it is a more controlled environment.
best,
sean
November 16th, 2009 at 1:42 am
mattmc, you’re a bigot. Next time try a reasoned argument bolstered by facts and maybe you won’t expose your ignorance.
November 20th, 2009 at 9:39 am
What it seems a few people above are not recognizing is that the article states that it is based on a percentage of income, not the actual dollar amount. If I earn $50,000 and give $5,000 to charity that is 10%. If I make $500,000 and give $20,000 that is more money but only 4%. I think most people would agree that the person giving 10% was more generous. As for mattmc; what a load of rubbish. Your personal hate for religion does not make those with faith wrong. With ideas like yours, it is obvious you have no clue who are the main supporters of food pantries are and the groups that provide the bulk of workers at agencies that provide meals to the needy.
November 20th, 2009 at 9:40 am
What it seems a few people above are not recognizing is that the article states that it is based on a percentage of income, not the actual dollar amount. If I earn $50,000 and give $5,000 to charity that is 10%. If I make $500,000 and give $20,000 that is more money but only 4%. I think most people would agree that the person giving 10% was more generous. As for mattmc; what a load of rubbish. Your personal hate for religion does not make those with faith wrong. With ideas like yours, it is obvious you have no clue who are the main supporters of food pantries and the groups that provide the bulk of workers at agencies that provide meals to the needy.
January 27th, 2010 at 8:34 am
[...] Southern Hemisphere having an affair on the tax payers account? "Just wondering" Dataset of the Day: Who is more Generous? Republicans or Democrats? | Off the Map – Official Blog of… Reading this article on charitable giving, it is clear that the right give much more to charity [...]
February 11th, 2010 at 8:41 pm
[...] have always been more generous and helpful to their fellow man than you liberal tightwads. And the data backs this up. It has never been about us not caring about people getting health care. Ever. That’s just you [...]
April 15th, 2010 at 8:37 am
[...] Burke in “Who is More Generous? Republicans or Democrats?” actually gives more historical data to support Kristof’s [...]
July 28th, 2010 at 9:37 am
[...] a liberal, goes on to say that liberals tend to be stingy when it comes to giving to charity. Dataset of the Day: Who is more Generous? Republicans or Democrats? | Off the Map – Official Blog of… Quote: — Although liberal families' incomes average 6 percent higher than those of [...]