For those of you interested in the Q2 numbers, I have some bad news. Evidently, campaign committees are only required to file FEC reports of donors who give more than $200. True, this is good news for those of us who don't relish the idea of the federal government having lists of where we live, what we do for a living, and which political parties/candidates we support. This is bad news, however, for my work on statistical analysis.
This means that the number of individual donors indicated on the FEC's website only reflects the number of over-$200 donors. Individuals who donated $200 or less are not listed here at all, meaning that a very large segment of the donating public would not be accounted for in my percentages.
For example, if we assume that all $1,423,883 of $200-or-less donations to Ron Paul came in exactly $200 increments, (Admittedly, this is mathematically impossible, and a horrible assumption in any case. Bear with me.) then we see that at least 7,120 donors would not be counted. This is more than thrice the 2,236 reported by the FEC, and could seriously affect percentages for candidates with low numbers. Obama's numbers are even more staggering; with $16,554,783 reported in $200-or-less, at least 82,774 of his supporters would not be counted.
I am left with no real option here. If I really want accurate data, I'm going to have to contact each individual campaign committee treasurer and get either the total number of individuals, or the average donation size for the $200-or-less category. I've already started this, but haven't heard back from anyone yet, which could mean they don't have the data, or it could mean they aren't going to waste their time giving it to someone who doesn't work for Fox, CNN, etc.
At best, the numbers I can come up with now are sketchy. I can either run with the numbers listed on the FEC's website, or I can make the obviously false, very poor assumption of exact $200 donations for the entire category, which will give me the minimum number of uncounted supporters. Neither of these options is particularly attractive to me, and I would consider either of them to be "For Amusement Purposes Only". Incidentally, my Q1 numbers were ran with the FEC-listed numbers, meaning they have the same problem. I will update that post accordingly.
So, what do you guys think? Interested in seeing numbers that you absolutely know are wildly inaccurate? If so, let me know and I'll do the best I can with the numbers I have. If/when I get numbers from the campaign committees, (and believe me, I'll keep trying) I will do this the right way, and we can get a really clear picture of who has the most support in the Primaries. Let's just hope I get the numbers sometime before Super Tuesday, or this whole thing will be pointless.