In my work life, I follow a lot of blogs about social media; one of them is ‘The Network Thinker‘ (in my Bloglines feeds to the right over in the blogroll).
There was a post there today by blog author Valdis Krebs on ‘bundlers’.
I downloaded data of the top bundlers of donations for the 2000 and 2004 Bush campaigns and the 2008 McCain campaign. What’s the overlap of donors between the Bush and McCain campaigns? Will the same people influence both campaigns/administrations? Or will it be starkly different groups? Or something in between?
Below is a map of those who donated to BOTH Bush and McCain. The campaigns are shown as the two red nodes on the left of the map. The green links show donations coming into the McCain 2008 campaign. The blue lines show donations coming into the Bush campaigns of 2000 and 2004. The 128 bundlers, who have contributed to both McCain and Bush, are shown in the arc on the right.
A nice graphic follows, and then the conclusion:
Most of McCain’s 534 large bundled donations [76%] came from donors who did not donate to either of the Bush campaigns. Yet, this kernel of 128 bundlers keeps consistency across all three Republican campaigns in the 21st century.
The Gang of 128 may not allow McCain to wander too far from the current philosophy and approach. If elected, McCain may be different than Bush, but he might not be that different.
Even smart people can be stupid sometimes, and even people who do social network data representation and analysis for a living can be misleading.
So, instead of showing the complete network of bundlers, and highlighting the overlap, Valdis shows only the overlaps – strengthening the conclusion that McCain’s campaign is ‘more of the same’.
Instead of looking at the amounts, and giving some idea of how much money the overlap represents – we get nothing.
And, finally, it would be useful to see how many of the bundlers were also bundling for the other side – as a not-insignificant number of them do.
I love this kind of data analysis, and get pissed off when it’s been done badly. As in this case.
If I get some time this week, I’ll play with this – in fact, let’s make it a group project. Can some of you help out by downloading the bundler database from, say Public Citizen into a csv table and sending it my way?
I’d love to get data from this cycle and ’04, for McCain, Bush, Obama, and Kerry…we can look at the overlaps and relative amounts. Any other analysis ideas?
I think “bad analysis” vastly understates it. That was just a silly graph. I went to his web site and recommended Tufte’s “Visual Display of Quantitative Information.” I tried to be polite about it, but I probably let through a bit too much snark/disdain. Sigh.
I agree if done right–done rigorously, and for both parties–that kind of analysis would be very interesting.
Of course, the analysis of first-level information as is in the campaign finance system doesn’t control for any kind of manipulation of the system (i.e. you could still game the system by changing your “bundlers” and moving money around the system at a level underneath that which needs to be reported.) But it still would be an interesting exercise.
“Follow the money.”
Look at the title of the post — WEEKEND Data Mining. A quick analysis of who overlaps between Bush and McCain. The more complete analysis you describe above is doable, but not quickly, and would cost $$$.
The text of the post describes the results — 24% overlap — and the map shows just those who overlapped and is small enough to show the names[node labels]. The map with the overlaps and non-overlaps was large and names were not readable at the zoom level to display the whole thing, so I chose not to display it. Anyone that READS the text of my post understands 76% to 24%. And anyone that understands sociology knows that a small group may have dominance way beyond their numbers/counts. So it is not simple math that drives behavior.
As for Tufte, he never did social network analysis. If he did, I would like to see his work.
One of the reasons for this post is to excite others — get them gathering and visualizing public data. Go for it! Easy to throw stones and call people names… let’s see your work.
As for your suggestions for further analysis… your focus on the amounts may not work — he who donates $1 more does not necessarily get a $1 more influence. It *would* be interesting to look at overlaps between Repub and Dem bundlers. Go get your data!
Valdis, being purely defensive isn’t an answer to legitimate criticism, and in my professional world tends to result in one’s leaving the room holding one’s head (ask me how I know…).
Your post was misleading for the reasons I set out; it would have been trivial to have included the larger graphic of the complete analysis and shown the detail of those who overlap.
If you believe that amounts donated don’t matter in American politics, I may have an alien spaceship in Nevada that I can show you…
And I can’t imagine that it’s more than a trivial amount of additional work to have downloaded and analyzed the complete dataset.
I’ll be happy to send you a link when I’ve finished looking at the overall bundler spectrum. Perhaps you’ll run it through your software and we can share some graphic representations.
A.L.
You are good at namecalling… first stupid, now defensive. The more you namecall the more it tells me that is all you got.
Like I said before please read the post carefully. I did download the whole set of Republican bundlers for 2000, 2004, and 2008. Otherwise how would I get the percentages for overlap???
I understand that differing donations may matter — someone who donates $100K gets more attention than someone who donates $5. But these bundlers do not fall into such wide categories they have all exceeded high goals. If you think it is important to show the difference between $100,000 and $100,500 go ahead.
So, let’s see your work!
Credibility > sinking…
A.L.
Valdis, a piece of advice. When you’re defending yourself on a public forum, it helps if you take a moment to check your grammar. When you type a response with a lot of missing commas and repeated punctuation, it gives the impression that you’re too busy frothing at the mouth to proofread your text, and that just makes you look more angry and (sorry) defensive.
Specifically, though, the problem with your data analysis is that you forgot to do any. You took a list of 500+ names and found 120+ of them from another list. From there, you… jump straight to concluding that the McCain campaign is in the thrall of a cabal of unrepentant Bush bankrollers. How, then, did you come to this conclusion? Is there a reason to believe that this number is unusually high when compared to “normal” donation levels? (This would seem to be counterintuitive – if anything, you’d expect people who are involved with their respective parties to this extent to be more likely to repeat donations, not less.) Perhaps an analysis of the percentage of Dole donors to Bush donors, or between Bush’s two terms, would put it into perspective. As other commenters have noted, it wouldn’t hurt to look at the Democratic sources of funding in the same light.
Additionally, while you’re correct in pointing out that it’s possible for a group of relatively small percentage to have a disproportionate effect on political actions, why do you believe this to be what is happening in the instant case? The obvious conclusion would be “because the donors I’m talking about are donating a lot more money”; if the Bush-holdover donors were each donating a lot more than the ones who aren’t, then sure, I’d expect them to have proportionately greater influence in the McCain campaign. That’s why people are poking you for not having provided the dollar amounts. In fact, I’m left to conclude that the only reason that you wouldn’t have included that information is that, well, it doesn’t show what you want it to – i.e. I seriously doubt that the repeat donors are donating more money, which inconveniently shoots down your hypothesis.
However, for the sake of comity, let’s assume that you were merely uninterested in the dollar amounts and didn’t see how that affected the issue. Why, then, would you conclude that the Bush-donors were more influential than the larger numbers of non-Bush donors to the McCain campaign? I mean, if you’re saying things like “Gang of 128”, you have at least some kind of basis, right? One you’d like to share with everybody?
In advance, there’s little reason to tell me to crack open my textbooks. I’ve got that degree, thank yez all the same.