December 2009
M T W T F S S
« Nov   Jan »
 123456
78910111213
14151617181920
21222324252627
28293031  

Announcing The Project

For the past month and a half or so, I’ve been mentioning this thing I called “The Project” on twitter and Facebook occassionally. I’ve told a few people what it is already, but I am ready now to officially unveil it.

The Project took quite a bit of work. First, I had to set everything up in a table which would eventually have nearly 121,000 individual data points. (120,944 to be exact). I manually entered the data for the first 26,000 data points. I found a way to automate the next 95,000, but even this was a time consuming process – just not anywhere near as time consuming as the first 26,000.

But why would I spend so much time entering data – on top of everything else I already do? Why did I deem it so important? What is The Project?

Quite simply, The Project is quite possibly the single most powerful tool I’ve yet deployed other than this site in general.

At its heart, as the ‘table’ reference above suggests, it is an Excel file. Nothing too fancy there, and really there isn’t anything overly fancy within the file. It is the data within the file that makes this tool so powerful.

What is the data?

Every single vote cast in the 2009 legislative session from every single member of the General Assembly, directly from LEGIS.

The 26K votes I manually entered were the Senate, 56 members x 469 votes in the 40 days of the 2009 Legislative term. The 95K I managed to automate to a certain degree were the votes from the House, 180 members x 526 votes in the 40 day Legislative term.

Now, I have the ability to do any number of analyses. For those members of the General Assembly who are currently running for statewide office in 2010, I can now examine their 2009 voting records in detail very easily – and I plan to. I can look for patterns of who bucked their Party and proved a ‘maverick’ streak. (Two notables from the Senate who regularly did this were Jeff Chapman and Preston Smith.) I can look for patterns of attendance and who appeared to be too busy doing other things to worry about actually casting votes in the Session – and there are a couple of people in both the House and the Senate who fit that mold.

With future planned additions to how I encode names, I will be able to easily track various groups of people, such as the former 216 Group or Committee Chairmen or Atlanta vs Rural or North Ga vs South Ga or any number of other things.

One caveat to the data I have is that the Senate recorded roll call votes in LEGIS, while the House did not.

For the moment, I’ll leave you with some basic stats as far as general numbers:

In the Senate, 56 Senators voted on 469 individual items, for a total of 26,264 data points. Subtracting out the roll call data points yields 24,136 actual votes – there was no roll call vote on the last two days of the term. Of these, 1,323 were members not voting on the issue at hand, 1,268 were members being excused from voting on the issue at hand, and 3,784 were members voting against the issue at hand. As percentages of actual votes, this means that members did not vote at all 5.5% of the time, were excused 5.3% of the time, and voted against the measure at hand 15.7% of the time. As averages per member, this represents the average member not voting 23.6 times in 40 days, being excused an additional 22.6 times in 40 days, and voting against the measure at hand 67.7 times in 40 days.

In the House, 180 Representatives voted on 526 individual items, for a total of 94,680 data points. Subtracting out the roll call data points, there are 486 actual votes, indicating 87,480 votes other than roll call. Of these, 5,724 were members not voting on the issue at hand, 4,102 were members being excused from voting on the issue at hand, and 16,188 were members voting against the issue at hand. As percentages of actual votes, this means that members did not vote at all 6.5% of the time, were excused 4.6% of the time, and voted against the measure at hand 18.5% of the time. As averages per member, this represents the average member not voting 31.8 times in 40 days, being excused an additional 22.8 times in 40 days, and voting against the measure at hand 89.9 times in 40 days.

Note that there are a variety of perfectly legitimate reasons to not vote or be excused, including conflicts of interest among others.

Leave a Reply

 

 

 

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>