RK ARCHIVE - The Petition by Senate District

The Petition by Senate District

By: Eric
Published On: 7/31/2007 3:21:08 PM

District (Incumbent)	Signatures	Estimated Voter Turnout	Percent of Angry Voters
011 (Martin, Stephen H.)	5649	25100	22.5%
008 (Stolle, Kenneth W.)	3503	17591	19.9%
012 (Stosch, Walter A.)	5607	28236	19.9%
014 (Blevins, Harry B.)	3293	18094	18.2%
033 (Herring, Mark R.)	7793	42932	18.2%
001 (Williams, Martin E.)	2322	13768	16.9%
029 (Colgan, Charles J.)	5701	35441	16.1%
030 (Ticer, Patricia S.)	4076	25967	15.7%
021 (Edwards, John S.)	5115	33377	15.3%
009 (Lambert, Benjamin J., III)	4036	27031	14.9%

After a bit of number crunching and a large amount of data gathering and sorting, I've finally got a mapping between the zip code based data from the Civil Remedial Fee Petition (aka the abusive fee bill) and the Senate/House districts. While this is by no means a perfect, I do believe this data will shed light onto which races the CRF's will play a more (or less) significant role. I've included a detailed explanation of the assumptions and calculations at the end for those who care about such things.

This table shows the number of petition signatures per Senate district and as a percentage of an estimated voter turnout for the district, sorted inversely by percentage. The percentage (i.e. points) should not be taken literally because of a number of unknown factors such as the real turnout (it could be different than recent off-off year elections), the number of people who signed the petition who actually vote, and who those people blame for the CRFs. In reading the comments I found that blame, while more toward the Republicans, was shared with Democrats or incumbents in general.

Caveats aside, I think this information should cause concern among a number of candidates. Many of these percentages are in the double digits or upper single digits - which is a significant percentage when dealing with elections. While a 20+ point swing is extremely unlikely, based on these numbers there could realistically be a 5 point swing in some races - putting incumbents who are in tight races in some trouble.

The full Senate table (sorted by district) is below the fold. The Delegates list is on it's way.
The full Senate list:

District (Incumbent)	Signatures	Estimated Voter Turnout	Percent of Angry Voters
001 (Williams, Martin E.)	2322	13768	16.9%
002 (Locke, Mamie E.)	2128	21447	9.9%
003 (Norment, Thomas K., Jr.)	3315	38221	8.7%
004 (McDougle, Ryan T.)	4670	42571	11%
005 (Miller, Yvonne B.)	2263	16699	13.6%
006 (Rerras, Nick)	1761	29304	6%
007 (Wagner, Frank W.)	3440	23342	14.7%
008 (Stolle, Kenneth W.)	3503	17591	19.9%
009 (Lambert, Benjamin J., III)	4036	27031	14.9%
010 (Watkins, John)	5408	40482	13.4%
011 (Martin, Stephen H.)	5649	25100	22.5%
012 (Stosch, Walter A.)	5607	28236	19.9%
013 (Quayle, Frederick M.)	2703	35895	7.5%
014 (Blevins, Harry B.)	3293	18094	18.2%
015 (Ruff, Frank M., Jr.)	2275	52066	4.4%
016 (Marsh, Henry L., III)	3155	21302	14.8%
017 (Houck, R. Edward)	4011	42641	9.4%
018 (Lucas, L. Louise)	1898	32038	5.9%
019 (Hawkins, Charles R.)	3089	37088	8.3%
020 (Reynolds, Wm. Roscoe)	2941	48873	6%
021 (Edwards, John S.)	5115	33377	15.3%
022 (Bell, J. Brandon, II)	6852	46653	14.7%
023 (Newman, Stephen D.)	4068	46235	8.8%
024 (Hanger, Emmett W., Jr.)	2602	43257	6%
025 (Deeds, R. Creigh)	2868	36976	7.8%
026 (Obenshain, Mark D.)	2634	47162	5.6%
027 (Potts, H. Russell, Jr.)	4959	52510	9.4%
028 (Chichester, John H.)	4191	32285	13%
029 (Colgan, Charles J.)	5701	35441	16.1%
030 (Ticer, Patricia S.)	4076	25967	15.7%
031 (Whipple, Mary Margaret)	4384	37430	11.7%
032 (Howell, Janet D.)	4787	39837	12%
033 (Herring, Mark R.)	7793	42932	18.2%
034 (Devolites Davis, Jeannemarie)	5613	46161	12.2%
035 (Saslaw, Richard L.)	3638	25969	14%
036 (Puller, Toddy)	4695	33072	14.2%
037 (Cuccinelli, Ken, II)	4766	34522	13.8%
038 (Puckett, Phillip P.)	1922	50564	3.8%
039 (O'Brien, Jay)	4950	36264	13.6%
040 (Wampler, William C., Jr.)	1308	44588	2.9%

Calculations and Assumptions
Comments, questions, suggestions, etc are welcome.
1. This data represents signatures through approximately the 159,000 mark.
2. Data was first aggregated to county/city level by mapping zip codes to county/city and adding in signatures which provided county/city name instead of zip code. Some data (approx 4-5 thousand signatures) was lost due to location information that didn't map to a Virginia county/city.
3. Senate, House, and Congressional voter registration counts (as of Jan 2007) by county/city was pulled from the SBE website. Based on voter registration count, a percentage allocation per district within each county/city was determined. For example, suppose a county covered two senate districts with the first senate district having 3000 voters and the second having 6000. Dividing each by the total yields 33% allocation for the first and 66% allocation for the second.
4. The percentage allocations were then applied to the signature total per county/city, resulting in the number of signatures per district within each county/city.
5. For districts that cross county/city borders, the signature counts for each of these districts were added together, resulting in a total number of signatures per district.
6. The voter turnout percentages from the last off-off year election (2003 - data also from SBE) per county were applied to the Jan 2007 voter registration data to determine the estimated voter turnout.