Sunday, February 7, 2016

Probing the bellwether status of Vigo County voters

Previously published in the Terre Haute Tribune Star, 7 February 2016


Two weeks ago I posted a link to an Internet poll for Vigo County voters’ presidential preferences. As promised, here are the results. Thank you to those of you who participated.


This poll cannot be used to make any claims about Vigo County because the sample size is too small and it’s not representative of Vigo voters. Below I discuss the small sample, who responded, and speculate on results.


First, only 42 registered Vigo voters took the survey. This leads to the only significant finding: few people read this column.


After 24 hours I began trying to get the link out to more venues but that didn’t increase the sample size much. There are 19 zip codes in Vigo County, but poll responders only came from nine. The zip codes responders came from were from Terre Haute and its metro area and from West Terre Haute.


The results also suggest that significantly more Democrat voters responded than Republican voters. Once I saw this trend, I tried to get the poll in front of more likely Republican voters, without much success.


Thirty-three of the 42 responders indicated they were likely or very likely primary voters. Among them, when asked who they would vote for if the election were today, I gave a list of almost all Republican and all Democrat candidates. Those 33 voted this way: Sanders 14 (42 percent); Clinton 7 (21 percent; Trump 4 (12 percent); Rubio 3 (9 percent); Paul 2 (6 percent); Cruz 2 (6 percent); and Carson 1 (3 percent). The lineup doesn’t change when we look at those 36 voters who indicated they were likely or very likely to vote in November. The rank order stays the same.


Other (better) polls tell us that Sanders has the youth vote. I asked Vigo voters if they had voted in the last four elections in Vigo. Thirty-four said yes, so that means these are not new voters and would at least be 34 years of age. The lineup is the same, Sanders, Clinton, Trump, Rubio, Paul, Cruz, Carson. Among the six voters who did not vote in each of the last four elections, the lineup is Sanders (4), Clinton (2). If the other polls are accurate, this means there are a lot of “white progressives” responding to my poll. There are many “white progressives” among the professoriate.


I asked who Vigo voters voted for in the last four elections. Among the nine consistent Republican voters (meaning they voted Bush, Bush, McCain, Romney), if the election were held today they voted Trump 4, Rubio 3, Cruz 1, Carson 1. Among the 15 consistent Democrat voters, it was Sanders 8, Clinton 7.


I hoped to identify “bellwether” voters which is why I asked about who Vigo voters voted for in the last four elections. Among the 42 eligible possible bellwether voters, there was only one. And the single bellwether Vigo voter indicated if the election were held today, s/he would vote for Sanders.


Can we learn anything from this too small, unrepresentative poll? No. We can still speculate, however. Given that the Iowa caucuses were last Monday, this small sample of consistent Republican Vigo voters do not like Ted Cruz as much as Iowa Republicans nor as much as the polls suggest New Hampshiricans do. This sample likes Trump and Rubio. Iowa Democrats were split between Clinton and Sanders and so is the small sample of Vigo Democrat voters only they slightly like Sanders more than Clinton. The polls suggest New Hampshirecrats like Sanders far more than they do Clinton. So, the bellwether is pointing in the general direction on the Democrat side, and is pointing more toward Trump and Rubio (who finished in a tight 2, 3 finish, just as they are a tight 1, 2 finish among the small Vigo sample). New Hampshire votes Tuesday. Polls suggest that Trump will win big, followed by Cruz and Rubio and that Sanders trounces Clinton.


The populist candidates, Trump and Sanders, are the most popular among the respective parties in this sample.


This poll uncovered one bellwether voter. That is 2.3 percent of the valid sample. At least this is an estimate of how many bellwether voters there might be, 2.3 percent. There are 70,000 registered Vigo voters, so that would be 1,610 bellwether voters. If half vote but all bellwethers do, that doubles their proportion. Is 4.6 percent of presidential Vigo voters who are bellwether voters enough to decide the election? Seems too small to me. What do you think?
Blog Directory - Blogged The Steiger Counter at Blogged