Monthly Archives: May 2014

2014 Danville Public Issues Survey data file now available

For interested parties, the full dataset from the 2014 Danville Public Issues Survey is now available for download. The data file can be accessed here:

As with other previously-released data files, this is available for examination and analysis by interested parties. Any public documents (blog posts, academic articles, conference papers, newspaper editorials, etc.) produced using this data should cite and reference the poll and its author. Also, the document should include a disclaimer that neither the author nor Centre College are responsible for the interpretations or conclusions of the analysis.

Commission voting and voter issue importance

In a recent letter to the editor, Danville Mayor Bernie Hunstad issued some corrections to a newspaper article that had recently been published about the upcoming Boyle County Republican Judge-Executive primary race.

First, he writes that “our commission has voted together 97-plus percent of the time.” He is generally correct in this assertion, although it’s not quite 97%+. An analysis that I published in 2011 shows that the commission voted unanimously on 80.7% of their votes during the summer of 2010 through summer of 2011. A year later that figure had changed to 82.6% for 2011-2012. The latest analysis published last year shows unanimous voting 87.3% of the time from 2012-2013. Regardless of the exact percentages, Mayor Hunstad is correct in that there is much more agreement than disagreement in the Danville City Commission under his tenure, at least measured in terms of roll call voting by commissioners.

He also writes that “the next board will not likely focus on the number one issue with the voters, which is the creation of better jobs and improvement in our local economy.” This is also correct, as evidenced by our 2012 Boyle County Exit Poll which showed a plurality (30%) of Boyle County voters listing “jobs” as the most important problem facing our local community. 38.3% said the same in the 2011 Exit Poll

Three cheers for epistemic humility

In response to today’s conflicting jobs reports, David Leonhardt explains:

Any one jobs report contains a fair bit of statistical noise, as Neil Irwin and Kevin Quealy have explained. It’s a mistake to pretend otherwise. The best approach is to take all the evidence — both the household and business survey, as well as multiple months of data — and use it all to tell the most sensible story we can, based on the evidence. [emphasis added]

Right now, that story looks something like this: The labor market appears to be gaining strength. But there are enough conflicting signals that we will need more months of data before we can be sure.

I’m a big fan of epistemic humility, especially in today’s world that values certainty and absolutes. One of my guiding philosophical approaches to life is that since all human sources of knowledge are fallible, the closest any of us can get to Ultimate Truth is obtainable only through taking as wide a sample as possible of as many different perspectives as possible on a particular question or topic, then taking a weighted average of them all to try to come to a tentative (although never final) personal conclusion or belief in some truth, whether empirical or otherwise.