Lazy Voters, Incompetence, or Election Fraud in Waukesha County?

A  historic week in Wisconsin has left Democrats asking if the county clerk of Waukesha has committed fraud by artificially raising the vote totals for Republican Incumbent Supreme Court Justice David Prosser two days after having been delivered a narrow defeat of about 200 votes.

Conversely, the Republicans should be asking how stupid can this woman be to forget to report all of the votes for the City of Brookefield?  Or should they be worried that members of their party are stealing an election and further enraging an already awoken liberal front?

Either way a friend of mine put it less eloquently, but no less accurately after a few beers “Bitch needs to go to jail, or bitch needs to be fired.”

But what really happened?  Is it possible that Waukesha County vote totals were accurately reported before the “fishy” adjustment?

What does the data tell us?

Based on Nate Silver’s interesting blog in the New York Times, the data points to incompetence and not fraud.  He comes to this conclusion after applying statistical wizardry and regression models using a bunch of information about potential motivating factors and calculating a 95% confidence interval of how many voters should have turned out in Waukesha.  I’m a believer in statistics and can’t find any flaws in his approach, but the math and conceptual tools may be beyond comprehension of many readers.

Using his data and adding some additional population statistics I’ve asked the question: “Is there any relationship between the change in voter turnout (calculated as a percentage of registered voters) and the voter results of the election?”

For example, the November election was a big win for the Republican party in Wisconsin, will they come out and support their Republican candidates as consistently as they did last time?

Take a look, click on the image to open up a larger version of the scatterplot:

The scatter plot shows Voter Support for David Prosser ( percentage of total votes cast in that county) on the x-axis vs the change in voter turnout between the April 2011 Election and the November 2010 election.  The data here is the complete original reporting before Kathy Nickolaus came forward with her “omission.”

Voter turnout is calculated as a percentage of total votes divided by the total number of registered voters in that county.  The y-axis shows the voter turnout for  the April 2011 election minus the turnout for the November 2010 Election.  Data points higher on the y-axis show a more consistent voter turnout from November 2010 to April 2011.  Conversely, data points lower on the y-axis show that there was less consistent voter turnout at the polls.

The size of each data point indicates the size of the population in each county data point.  The color of the data point similar to the x-axis indicates the percentage of voters supporting incumbent Supreme Court Justice David Prosser.  A bright red data point (farther right on the x-axis) indicates strong support for Prosser.  A bright blue point (farther left on the x-axis) indicates strong opposition to David Prosser.

Additionally, I’ve included a “smushed” scatter plot on the left which shows how all of the data points are separated by the difference in voter turnout.  This shows a fairly consistent spectrum of blue point to red points as you move from the top of the voter turnout difference to the bottom of the scale.

What’s going on here?

What I think is most interesting about this plot is the strong correlation of voter turnout difference by support for the Republican candidate.  Imagine a line being fit through that data, drawn more heavily to the larger data points and we have a line starting at the top left and working its way to the bottom right.  It’s fairly reasonable to say that people supporting the republican candidate were not as motivated to head to the polls again as folks that opposed the Republican candidate.

But what about Waukesha county, the large red point at the bottom right of the plot?  It looks kind of lonely down there.  Could Waukesha county voters have been so lethargic on April 5th so as to have by far the biggest difference in voter turnouts by about 5 points?  Is there something else happening?

By contrast, the bright blue points in the top left are a fairly good distance away from all of the other data point populating the center of the plot?  Conspiracy theorist conservatives will likely point at this and claim voter fraud and demand voter id laws and the elimination of same day voter registration laws.  They are on the way, trust me.

The “Correction”?

So what happens when we apply the “correction” or “fraud” depending on which side of the wedge you’re occupying?

With the “adjustment” Waukesha moves significantly up the screen (more consistent turnout, but still at the bottom of the bunch), keeping about the same ratio of Prosser supporters.  Also notable, it moves much closer to its nearest neighbors, both in the sense of data proximity (voting habits) and geographical proximity.

This is consistent with analysis looking at support for Scott Walker in the 2010 election vs the 2008 presidential election.  The link opens up a Tableau workbook with the Walker scatterplot and the Prosser Scatterplots used in this post.

Bottom line, this data doesn’t prove that the voter turnout in Waukesha is abnormally low.  Its also shows that the adjusted point is not unreasonable.  I’ll leave it to the Government Accountability Board and the US Attorney General to prove that Waukesha’s reporting is on the up and up.

Update: Sep 28, 2011

The GAB has concluded that Kathy Nickolaus did not commit a crime:

“her conduct does not appear to rise to the level of conduct that can be described as willful neglect or a refusal to comply with the law.”

She will however have to start playing by the rules.

“Your actions following the April 5, 2011 Spring election did not conform to the legal requirements imposed on county clerks,” G.A.B. Chairperson Thomas H. Barland said in a letter to Clerk Nickolaus.  “When one election official fails to act consistent with those responsibilities, steps must be taken to correct the failure in order to prevent it from recurring, and to restore public confidence and trust in the administration of elections.”

WI 2011 Supreme Court Election Results – Pre Fishy Correction

I’m putting together an analysis of the recent Wisconsin State Supreme Court Election results and I would like to compare it against voter turnout from the 2010 Gubernatorial election.

Can someone please point me in the direction of county by county voter results for the 2010 Gubernatorial election? Preferably in text, html table or spreadsheet form. Found it:

http://gab.wi.gov/elections-voting/results/2010/fall-general/county

In the mean time: here are the data sources that I am using for the current analysis:

AP Election Results

Wisconsin Census Estimates by County

Wisconsin State Demographics – Department of Health Services

Here are the Supreme Court Election Results by County prior to the announcement from Waukesha County earlier today combined with data sources above:

Google Doc Spreadsheet with Election/Census Results

2011 Wisconsin Supreme Court Election Results by County

At this point I’m very suspicious of the announcement made today in Waukesha County which adds 10,859 to the Plosser total and 3,546 votes to the Kloppenburg total.

I’d like to do a full comparison of the difference in voter turnout between this election and the recent gubernatorial election in 2010.  Conservative friends of mine are simply saying “look, conservatives were passionate about this race too.”  The enormity of this error and the fact that it happened in a county which happens to be the third largest by population and second largest by Prosser vote differential sounds like the perfect data point to turn the election in the Incumbent Republican’s favor.

Take a look at the data.  Make your own decisions.  What do you see?

Who Is Buying Government Influence: WI Query Tool

Utilizing the Wisconsin Democracy Campaign data I have created a searchable dashboard that shows Employer’s and PAC’s contributions to all Wisconsin State races reported to wisdc.org.

Click here or on the graphic to open the dashboard:

Use the Employer search box to find the company you want to investigate.  Data is aggregated by major race category and details are shown below for specific races including whether the giving came from individual contributions or Political Action Committees.

To make the query tool more efficient, only employer/PACs with contributions over $10k were included in this dashboard.

Find something interesting?  Like to see something added?  Please drop a note in the comments.

More Campaign Contribution Data: Two Sides to the Data

Earlier posts in this series examined campaign contributions particularly looking at contributions to the Fitzgerald brothers and comparisons of giving to the Scott Walker and Tom Barret campaigns respectively.

Before this analysis proceeds any further, I want to clear up a quick misconception of this data.  Some people on the “Boycott Scott Walker Contributors” group on facebook have taken a one-sided look at this data and said something like “Look, people employed by Kwik Trip contributed over $10k dollars to Scott Walker’s campaign.  I’m never shopping there again.”  However with a closer look, data can be found showing that these same employees donated just about the same amount to the Tom Barrett campaign.  So should you boycott Kwik Trip?

Personally, I’ll still shop at Kwik Trip.

Will I buy Johnsonville Brats again or attend Bratfest this year? They gave about $37k to Walker in 9 contributions.

No, its more likely that I will buy Klements and donate money to one of the charities that benefitted from Bratfest last year (Family Support and Resource Center in Dane County is a good one.)

In this post I am attempting to provide a comprehensive look at Political Action Committee (PAC) and Individual Donations for every Wisconsin race from the 2010 elections with data reported to the Wisconsin Democracy Campaign (wisdc.org).  Another change in this post is that I have manually edited employer names so that Wal-mart and Walmart show up under the same employer label.

First off, the Vector Table.  This shows giving to both parties and is sorted by the difference.  The important piece is the difference or the vector.  A positive number indicates that the total amount of donations favors Republican candidates.  A negative number indicates that more money was given to Democrats.

Click here or on the graphic to open the Vector table:

You will notice lots of checkboxes that allow the user to choose which data to include in the display allowing one to view data by a specific race or government branch (Senate, Assembly, etc.)  Additionally there is a checkbox that allows users to control whether they look at Individual contributions, PAC contributions or both.

Click on the image to pop up the new data:

Interactive Scatterplot:

Click here or on the graphic to take a closer look at all of the contributions in one place, on one page.

Lastly a table is provided showing Employer / PAC giving across all of the different government branches and the 2010 races. Open the workbook and filter by companies, candidates and races.

As always, explore the data and make your own decisions. If anyone finds anything particularly interesting or would like to see something specific, please leave a note in the comments.

Dastardly Duo?

In all of this hoopla concerning the governor of Wisconsin and his nearly fascist regime thus far, two key players being left out of Walker’s direct spotlight are Assembly Speaker Jeff Fitzgerald and Senate Majority Leader Scott Fitzgerald.  These guys are playing dirty politics with the very worst of them as exemplified by the illegal special session vote witnessed in this video of March 10, 2011.

Being an outraged citizen I thought it appropriate to expose the financial contributors to both Jeff and Scott Fitzgerald’s campaign.  Again, this is public data pulled from wisdc.org database based on personal contributions and not PAC spending.

Scatterplot of Fitzgerald Campaign Contributors: Click on the link or the image to open up an interactive data viz hosted by Tableau Public.

Who is funding dirty politics?

The plot shows number of contributions (x-axis) by contribution amount (y-axis).  The data is aggregated at an employer level.

The x-axis is reversed and placed on logarithmic scale (spaces out the data) so that it reads left to right as <more contributions> to <less contributions>.  The marker size indicates the size of the average contribution.  A larger marker indicates more money from less people.  The two types of points that interest me are the points high up on the y-axis and points with a large marker sitting above the rest with same relative number of contributions.

The chart is fully interactive and you can control what recipients, donors, employers and years the contributions were made in are displayed.

It really shouldn’t come as much of a surprise to folks who have been tracking Governor Walker’s contributions, but M&I bank as well as Northwestern Mutual Life factor in as major donors in these campaigns, again.  A couple of payday loan companies also contribute with high averages but I suppose that shouldn’t surprise many folks that these type of companies would back Republicans.

What is surprising to me is that by far the largest donors found here are several employees of Sentry Insurance contributing nearly $7,000 at $100 a pop per employee.  This is relatively small average but what would move so many people to contribute to two political campaigns in so many numbers.  Also interesting to me is that many of these donors are from far outside either of the Fitzgeralds’ districts.

I’m not a necessarily a fan of boycotting companies’ products and services because of the incidents that have occurred in Wisconsin early in this year.  We’re talking about real people with real jobs whose livelihood could be at stake if a significant impact is made to these companies.  Some of these people have no bearing on the donations being made and labeled as employer interests.

On the other hand the legislators at the state are not listening to the voices of the people yelling out.

Do with this what you will.  Make your own decisions.  Please comment if you find anything interesting you would like to share.

Full Disclosure:

I was once an employee of a company that was purchased by Sentry Insurance after I had left.  I am still in touch with many of my ex-coworkers and have no ill will for them or the company other than my opposition to the candidates that they supported.

Walker/Barrett Contributors: Playing Both $ides.

I was discussing the matter of campaign finance with a friend tonight and he admitted that a family member of his supported both candidates in last year’s 2010 Wisconsin gubernatorial election.

What?

The  way the data is presented in dimensional / categorical form sometimes obliterates inherent links and relationships found in the data.

Similar to the divisive demarcation of our two party political system, we sometimes forget everything that we have in common.  Its just red data or blue data and nothing in between.

Take a look at the data by clicking on this link to pop up the Tableau workbook that I have shared.

Both Sides?

Admittedly, the data comes from private donors who are required to disclose their employers.  Sometimes you just have to see the data.

Walker’s Wiscon$in Campaign Contributions: What’s In There?

So I pulled some data from wisdc.org detailing private donations to both the Scott Walker and Tom Barrett camps in the Wisconsin Gubernatorial election of 2010. Given the recent brew-ha-ha in Wisconsin (read absolute bullshit), I thought it might be worth revisiting with new tools to see if there is anything really interesting tucked away in this public data.

The most interesting insight to be had so far is the specific employers of contributors popping up to the top of the heap.  The first look aggregates the data at a listed employer grain and plots the total number of contributions(x-axis) by contribution amount (y-axis).  By reversing the x-axis and plotting on a logarithmic scale, the data spreads out and reads from left to right as <more contributions> to <less contributions>.  Note that the blue points represent contributions to Democratic candidate Tom Barrett, and the orange points represent contributions to our current governor, Scott Walker.

With all of the plots presented here, click on the graphics to open an interactive data viz.

The points that jump out at me are the larger contributions (higher on the y-axis) and especially the higher contributions that come from a smaller number of contributions (farther to the right – reverse axis).

Click on the image to bring up an interactive data viz powered by Tableau.  Hovering over a data point will provide more details including the specific numbers and the data label.  Additionally you can control the data selected by interacting with the filter check boxes on the right.

My very first post highlighted significant contributions from Miller, Johnsonville, and Sargento (Beer, Brats and Cheese).

All kidding aside, the Johnsonville point ( ~$37k from 8 contributions) sticks out as big dollars from just a few donors, meaning significant corporate interest.  I’m not suggesting that anything is amiss, but holding my personal objections to what’s happened around Wisconsin in 2011 has me thinking that maybe I should start buying Klement’s instead.

Other notable points for Walker un-enthusiasts:

M&I Bank, Aurora Health Care Systems, Hy Cite Corp, Wausau Homes, Miller Brewing, Sargento Inc, Northwestern Mutual Life.

Another interesting plot looks at the same data, grouped at a categorical interest instead of the donor’s employers.  This type of categorical interest has been covered in several media outlets, however I like the scatter plot approach to understand the relationship between the contributions and how many people are making these contributions.  Again a logarithmic reversed x-axis was used to display the number of contributions.

By far  the most noticeable point here is $1.4M contributed to Barrett by interest groups identified as “Lawyers/Law Firms/Lobbyists”.  After that its all Walker with heavy interests from Manufacturers, Construction, and Banking.

Perhaps you would like to know more information about how each candidate faired by interest category.  The following chart will allow you to compare both candidates total contribution amount, total # of contributions, and average contribution amount by categorical interest.

Here you can see how different each candidates contributions are in each category, sorted by total contributions.  The thickness of the bars indicate how large the average contribution in each category is. That information is also displayed on the far right hand side group of bars.  Again there’s nothing really new here, but it is a cute way to compare the differences.

Lastly geting to play with these new tools in Tableau, I created a map geographically showing where the money is coming from.  The counties are filled by Median Household Income and the size of the data points indicates the total amount of contributions by each candidate.

There isn’t anything really surprising here (at least not that I found).   Donor zip code was used to geocode the data.  However, I wish there was a way to wiggle the coordinates so that data points in the same zip wouldn’t always lay over the top the of a different data point (ie, Milwaukee and Madison).

"Where's the money?"

As it stands, the plotting will always stick Barrett on top of Walker, so if Barrett won the area, you can’t see what Walker collected.  However, there are filters at the right that you can use toggle the specific candidate’s data points.

Predictably, the money follows population density and median household income. Walker did better in areas surround Milwaukee.  Barret did much better than Walker in Dane County.

I’m admittedly a novice at Tableau.  If you have any tips or questions I would welcome them both in the comments.

Campaign Contributions by Beer, Brats and Cheese.

An article at Dane101.com caught my attention today:

When the firefighters arrived at the Capitol this morning they started the chant “MOVE YOUR MONEY!” Firefighters Local 311 President Joe Conway told the audience they should move their money out of M&I Bank. The bank was one of the leading contributors to the Walker campaign due to contributions by current and former executives and board members.

So out of curiosity I decided to pursue the campaign contributions listed at Wisconsin Democracy Campaign.  The simple query returned 193 records, 1o records per page of M&I bank employees supporting Scott Walker’s campaigns between 1993 and the most recent data added in October 2010.

Then the light turned on.

Sure we can shine the light on a couple of records and form whatever biases you want from a small sample, or we can look at the whole damn thing and maybe learn something.

“Damn it¡ (<– that’s the sarcasm mark) The query tool on the site only allows me to pull 100 records at a time¡” A quick look at the source (php) and it appears that the query is submitted via the webpage address … okay.

http://www.wisdc.org/index.php?filter=+Search+&from=–&to=–&name=walker&module=wisdc.websiteforms&cmd=searchadvanced&amp;qty=100

Did you notice that “qty=100” bit in there too?

I wonder what happens if we add a couple of zeroes to the end of that?

Surely they would have written in a bit of logic allowing only 10, 25, 50 and 100 records to be pulled at a time?  Or maybe they want us to be able to access as much of this data as we want since the idea is to make campaign contributions transparent?

With a tip of the cap to the folks at wisdc.org I ran the query with a few more zeroes and downloaded to text 20k campaign contribution records (sorted by amount) for Scott Walker and what ended up being all of the campaign contribution records for his opponent in the 2010 gubernatorial race, Tom Barrett.

I dumped the tab delimited text file into Tableau Public (my new best friend, but more on that later), filtered for the last two calendar years, and created a couple of quick scatterplots looking at how many contributions employees of certain companies are making.

Click on the graphic below to open up an interactive data plot and decide for yourself:

Wisconsin Campaign Contributions

 

An interactive data viz should pop up showing a scatter plot with the number of contributions on the x-axis and the total amount of campaign contributions on the y-axis.  Walker and Barret’s respective data is color coded (the red is drying out) and a number of filters are available on the right.  Admittedly, this can be a bit confusing to look at, but a lot of information can be obtained from looking closely.

On the x-axis the number of contributions is plotted in a reverse order and on a logarithmic scale.  This spaces out the data and reads from left to right as <more contributions> to <less contributions>.  The y-axis displays the total contribution amounts by the employer of the donors.

Hover over a marker and you can see the specifics of that data point.

Some interesting observations and companies that won’t be getting any business from me (as I am decidedly anti-walker):

Miller/Coors (39 contributions, $23k) {Beer}

Johnsonville Foods ( 8 contributions, $37k ) {Brats}

Sargento Inc ( 23 contributions, $25k) {Cheese}

Aurora Bay Care Health Systems (43 contributions, $28k) {Healthcare: for the Obesity/Heart Disease}

M&I Bank ( 74 contributions, $55k) {my savings paying for said Healthcare}

Northwester Mutual Life (74 contributions, $27.5k) {collected by my loved ones after i die to support Walker’s recall election}

Okay, so I’m shining a fairly narrow blue light at those results, but I found this “correlation” humorous.

Seriously, take a look at the data and let me know what you see.  It’s kind of like a Rorshach test.

Full disclosure:

The top 9 data points are chopped off of the plot to allow a close up look at the bunched up data close to the base.  The employers listed include 3 law firms (all for Barrett), 2 “no employer listed”, and 2 “retired”.