interstices

At a post-election panel discussion sponsored by The Blue Review at the downtown Boise State University Center on Main, Idaho Statesman political reporter Dan Popkey stated his theory why the No votes prevailed striking down the three education reform measures:

“I think the reason the (hypothetical) Sugar City mom voted against these measures was because she knew her kids’ teachers and she did not think that they were thugs and she thought that they were really doing the best they could with resources that probably aren’t adequate.”

Popkey’s theory of the race is interesting because it should be possible to test it by comparing the election results with some broad measures of the presence of public education in Idaho: teachers and students.  Far too often it is difficult to take a voting decision or motivation from the quantum realm of the individual voter and apply it to a more general relationship at a larger scale.  In this case to the state and county election results.

Available Census data has some surrogate measures for school teachers so that election results could be compared at a county level.   Perhaps counties where teachers make up a larger part of the community had a higher no vote?

An indirect measure of teacher presence at the county level is Census Bureau data on population by age, and perhaps median age.  But when compared to the election percentages the results were not satisfying.  The r value statistics were low if even barely statistically significant:

• Median age in county and No on Prop. 1:  r= -0.079544101
• Percent of county population < 18 years of age and No on Prop. 1:  r= -0.219040258

Similar results were found for the No vote on Prop. 3, and all were negative values, meaning the relationship was backwards.  Higher median age in a county was unexpectedly more likely to vote No, and counties with a lower percentage of people 18 years old and less were more likely to vote No.  It’s supposed to be the other way around.

However, a refined analysis does show there is something to this theory.

Analysis shows the Obama vote was baked into the cake of the No on Proposition 1 vote.  Proposition 1 had a 57 percent No vote.  The Obama vote of 32.6% leaves about 25 percent of voter turnout came from voters who did not vote for Obama but voted No on Prop. 1.  There appears little to no defection of Obama voters from the No vote on Prop. 1.  So the next question is to look at the crossover vote.  Was the Sugar City Mom Theory in play with this 25 percent?

More robust information on public school teachers and students is available from the Idaho Department of Education website.  Teacher data is by school district but is dated to the 2008-2009 school year, which was before the major budget cuts and teacher layoffs.

A more recent source of data is public school student enrollment for fall 2012.  It was also available by county, thus eliminating the need to add district data and parse out joint districts that serve more than one county.  Of the nearly 282,000 enrolled public school students this fall there are 12,300 state-charter school students that were not allocated by county, but more than 269,000 students are allocated to their public school by county.

A comparison of enrollment numbers and No on Prop. 1 votes (minus the Obama vote) results in most counties close to a regression line:

Again, the high r-square is heavily leveraged by the outlier data points of the larger population counties like Ada and Canyon.  So a look at the data on a percent of vote basis is needed.

Public school student enrollment was divided by total population in a county to arrive at a percentage of county made up of public school students.  These percentages were compared to the percent No vote remaining in each county after subtracting the vote for Obama and the undervote.  Here is the chart:

An outlier data point at the low end of the scatterplot is Blaine County where the No vote on Prop. 1 was only six percent ahead of the vote for Obama.  This is because Obama’s best county in Idaho was won with nearly 59 percent of the vote.  It had by far the lowest Romney Crossover Voter Probability score, half that of the next lowest (Valley County).  And with 59% already voting for Obama, a level that exceeds the statewide 57% No on Prop. 1 vote, there was little upside potential.  Eliminating Blaine County provides some improvement in the statistical relationship:

So approximately one-third of 25 percent of the non-Obama voters who voted No on Prop. 1 can be explained by an association with public school enrollment as a percent of county population.  The Sugar City Mom Theory appears validated for at least that portion of the vote.

About that hypothetical mother in Sugar City with kids in school who voted No because she knows the teachers?  The red dot on the scatterplot is for Madison County where Sugar City is located, and Madison County is where Obama got only 5.8% of the vote and then another 42 percent was added to the No vote on Prop. 1 for a 48.5% No vote on Prop. 1 from that county.  And Madison County ended up well above the regression line indicating the relationship with public school enrollment was an important factor.

In fact, there exists a constellation of data points well above the regression line, five counties where the No on Prop. 1 vote was 40% higher than the vote for Obama.  These counties were all very high support areas for Romney and unsurprisingly high percentage LDS population.  This relationship was previously explored and demonstrated.  It does beg the question of the LDS population relating to the No on Prop. 1 vote.  One can infer the Sugar City mom, like 90% of her neighbors, is likely an LDS church member.  Here’s how LDS population relates to the No on Prop. 1 vote:

The relationship of LDS population in a county is stronger with the No vote than the public school enrollment numbers.  The highest Romney-supporting (and LDS) counties show significant crossover to vote No on Prop. 1.  But at the low end of the scale the relationship is less clear.  Low LDS population counties show significant differences in the No vote.

The LDS data elongates the scatter of the data points beyond that for the public school enrollment data.  Percent of county population estimated as LDS members ranges from four percent to more than 90 percent.  Public school enrollment as a percent of population ranges less, only 10 percent to 25 percent.

Taken together, a Sugar City Mom Index (SCMI) can represent both elements of LDS population and public school enrollment.  The two percentages are summed and converted to a ratio (150% = 1.0) to create the SCMI for each county.  The distribution of counties along the range appears more uniform than either single variable.

The SCMI correlates to the No on Prop. 1 vote (above the Obama vote) with an r-square value of 0.7995.  In effect, the SCMI explains 80 percent of the variance in the percent No vote on Prop. 1 for those voters who did not vote for Obama.  In fact, the Blaine County data point anchors the data plot at the bottom left, showing the lowest SCMI score as well as the lowest percent increase in percentage vote for No on Prop. 1.

The Sugar City mom should not be taken as a literal example of female and/or LDS voters, but rather a symbol through which the SCMI illustrates voters who have some connection with public schools as parents, grandparents or having goodwill for local teachers or the school system.  For counties more than 50 percent LDS the SCMI represents those with the largest Romney voter pool, and thus highest percentage of cross over voting.  The counties clustered at the low end of LDS membership showed a percent increase in the No vote on Prop. 1 ranging widely, from six percent to more than 30 percent.  But the role of public school enrollment scattered the vertical alignment of these data points more in a diagonal direction, and thus a stronger relationship for the SCMI and No vote on Prop. 1.  This shows the variables making up the SCMI do measure different elements in a county’s population and social structure.  The combined effect arrays the counties closer to a more predictable regression line.

Keep in mind these counties are of different population and voting numbers and the strong relationship does not necessarily mean 80 percent of the cross over voters fit the SCMI - No vote on Prop. 1 profile.  The twelve counties that are 50 percent or more LDS population represent 23 percent of the voter turnout (and 26 percent of election day registration), so most voters are resident of the 32 counties with less than 40 percent LDS population.  The relationship does pick up an important dynamic that played out in the election at both ends of the SCMI, though it appears school enrollment played a more important role in the lower percentage LDS counties and Romney cross over voters in the high percentage LDS counties.

So back to the original theory.  When controlling for the Obama voters, those for whom there is a high confidence they all pretty much voted No on Prop. 1, the lion’s share of the remaining voters appear to fit the profile of the Sugar City mom, or her spouse, family and friends.  Veteran political reporter Dan Popkey was on the money.

An earlier post identified the ticket splitting of Idaho voters, those who voted for Mitt Romney and then voted No on Propositions 1, 2 & 3 in the face of Republican establishment support for the ballot measures.  A Romney Crossover Voter Probability for each county can be calculated as a function of the two variables below, and shows significant differences across the state as illustrated on this chart:

Before trying to understand the subspecies labeled the Romney Crossover Voter, let us first account for the Obama voters who voted No on Prop. 1.  

The No vote on Proposition 1 exceeded the vote for Obama in all counties by a wide margin.  Statewide Obama recevied 32.6% and No on Prop. 1 got 57 %.  The two votes do relate to one another:  as the percent vote for Obama increased, so did the No vote on Prop. 1:

An additional 14,000 voters skipped the presidential race, and when the undervote is added to the Obama vote the correlation improves from an r-square of 0.6286 to the 0.6414 shown above.  This exercise was part of a sensitivity analysis looking at the votes for the minor party candidates.  By the way, votes for the Green Party and leftist candidates or the Libertarian Party and Constitutional Party Presidential candidates did not improve the correlation with either the No vote when added to Obama or the Yes vote when added to Romney respectively.  Only the undervote was a factor.

The county level data shown above is validated by comparing Obama votes with the No on 1 votes using recently released precinct-level data.

First, looking at the raw votes there are 967 precincts across the state, and in only seven did Obama get more votes than the No vote on Prop. 1.  They were:

• North Blaine County - Obama 624 or 63%; No 575 or 60%
• North Ketchum - Obama 546 or 65%; No 538 or 67%
• Fort Hall - Obama 565 or 89%; No 518 or 83%
• Lapwai - Obama 501 or 76%; No 496 or 74%
• Dietrich - Obama 82 or 25%; No 79 or 33%
• Slickpoo - Obama 3 or 33%; No 2 or 25%
• Riddle - Obama 46 or 69 %; No 45 or 69%.

Five are precincts where Obama polled from 60 to 89 percent.  Another precinct Obama received 3 votes and No on 1 got 2 votes.  Seven precincts out of 967 means in only 0.72% of cases did Obama outperforme the No on Prop. 1 vote.  This represents a very low defection rate.

A plot of the 967 precincts and raw vote comparisons looks like this:

This results in a huge r-square value, heavily leveraged by the hand full of precincts where a thousand or more votes were tabulated.

When the precinct data for Obama vote and No on Prop. 1 vote are converted to percent of vote for each precinct a slightly different picture emerges:

The r-square value of 0.58 is closer to the county-level comparison above, so the precinct data validates the relationship established with the county data.  And because only seven precincts had higher raw votes for Obama than for No on Prop. 1 it indicates there was little defection among the ranks of Obama voters from the No on Propositions 1, 2 & 3 campaign.

It appears valid to conclude the Obama vote was baked into the cake of the No on Proposition 1 vote.  Proposition 1 had a 57 percent No vote.  If you subtract the roughly one-third of Idaho voters who supported Obama (32.6%) to the No on Prop. 1 vote, it means an additional 25 percent came from voters who did not vote for Obama.  The next question is to look at the crossover vote.  And how was Dan Popkey’s ”Sugar City Mom Theory” in play with fully one quarter of the voters?

Despite public and media attention to the school-related ballot initiatives and Mitt Romney as the first LDS member to lead a major party ticket, overall voter turnout on 2012 was down 1,216 votes in Idaho when compared with 2008.

For fifty years every presidential election year the numbers of voting age population has increased, as has the voting eligible population since they starting counting in 1980. Only in 1988 was the number of registered voters less than the previous Presidential election.  That exception aside the numbers climb every four years.

But voter turnout shows a different pattern where a high turnout election (1960, 1980, 1992, 2008) is followed by years where turnout does not grow.  Lack of population growth in the 1980s probably explains the 1984 and 1988 turnout being no larger than 1980, plus the latter had a large turnout driven by the US Senate contest between Frank Church and Steve Symms.  So too the 1996 election showed only slight growth over the big turnout in 1992 when Ross Perot nearly bested Bill Clinton for second place in Idaho.  The Kennedy-Nixon contest of 1960 was the year Idaho led the nation in voter turnout.  The next two elections of 1964 and 1968 voter turnout in Idaho did not increase.

A September forecast had projected an 80 percent turnout among registered voters, and in October the thinking was turnout would hit 78 percent.  Either prediction was a high bar: the actual result was shy of 75 percent.

There was no US Senate election in Idaho in 2012, a race that in past has coincided with larger turnout (when there is a competitive race).  Thus the data points above show the years where there was no US Senate race have been among the lower for voter turnout. Since the voting age was lowered to 18 year olds voter turnout in Presidential election years without a US Senate race never exceeds 75 percent.

Results did not match up well with the early hype about fellow church goers of Mitt Romney turning out in large abundance that would move the dial at the statewide level.  Some small population LDS-dominated counties did show a bump in voter turnout, but their numbers were too small to make much of a difference.

This year’s election is the second worst turnout among voting age population since they started keeping track in 1960 of total votes cast.  Only the 2000 election was worse.  Same goes for voting eligible population (as compiled by George Mason University) where data goes back to 1980.  Again, the 2000 election rates at the bottom on that turnout measure.

When measured against the registered voter data the 2012 turnout of 74.38% looks relatively better.  There have been 14 Presidential elections over the past 52 years and 2012 ranks 9th in percent turnout of registered voters, so it finishes ahead of five other elections:  1976, 1984, 1988, 1996 and 2000.  One factor that may be at work here is following years of large voter interest and turnout the voter registration rolls get bloated with additional names of one-time voters.  This deadwood is eventually thinned from the ranks if a person does not vote over a period of four years.  Voter registration appears to have increased faster in 2004 and 2008 and 2012 may be a reversion to the mean.

There was no appreciable drop off of voters in the Presidential election (bottom two lines on graph above).  Some 98.2% of those who turned out voted in that election.  The 1.8% would include both undervotes and over votes, the latter where a person mistakenly votes for two candidates for President.

It still holds true that a good way to project voter turnout is based on the increase in registered voters from the primary election to the general election.  With data going back to the late 1960s there is a data set of some two dozen election years from which the above scatterplot was developed.  The 2012 data point falls nearly on the regression line.

Sadly, it is now virtually impossible to use the voter registration number as a predictor because beginning in 1994 Idaho shifted to election day registration and over time fewer people bother to register prior to the election and they show up at the poll on election day.  This pattern was repeated in 2012 as shown below:

Just using voter registration between the primary election and the cutoff date prior to the general election is a poor predictor of voter turnout.  And over time it appears campaign strategy and government information has stressed the ability to register on election day thus changing voter behavior to kill two birds with one stone.  So what may be a loss for data analysis is a gain for good public policy that encourages people be enfranchised to participate in elections. 

Finally, the chart below shows the ballot measures Propositions 1, 2 & 3 all received high voter attention and little “fatigue” or drop-off among voters.  More than 97 percent of voters participated in voting yes or no on each of the ballot measures, and compared to the total vote in the Presidential election the drop off was less than one percent.  Historically there used to be more ballot fatigue, something more common on less well known ballot measures or on topics that may seem complicated or not that all compelling.  

A so-called great debate over Bronco Stadium breaks out as the seating capacity was increased this year to 37,000.  This was followed by some of the highest-record attendance in stadium history.  So should additional seating be planned for the facility?

Over the long term as stadium seating has increased, so has attendance.  Even with this year’s expansion the attendance figures show the seating fairly efficiently utilized - 95 percent or so.  Here is a plot of capacity with information on game attendance, at least what’s readily available out there on Wikipedia.  Which means we have all game attendance since 1999, plus 1980-81 and 1994.  In addition there are attendance records for the Boise State v. University of Idaho games from 1971.

When the second deck was added on the east side and 20,000 seats were made available more fans came to the game.  Average attendance in 1980 and 1981 show the peak game, usually the game against University of Idaho, exceeded capacity.  Average season attendance was also close to capacity.  End zone bleachers were brought in to increase seating to 22,500 through the 1980s.

In the late 1990s seating increased to 30,000 from 22,500 and it took a few years before average game attendance came close to filling capacity.  Of course there is always a peak game every year, and from data available Bronco Stadium would be filled at least once a season.

To get a closer look at utilization of seating capacity the raw data in the chart above is adjusted below into percentages:

Nearly every year the Idaho Vandals played in Bronco Stadium, starting in 1971, the place was full, maybe above official capacity.  The exceptions were 1982, 1986.  Both were closely contested games, won by Idaho.  The 1992 and 1996 games were just under the 100 percent line, both games were big wins by the Vandals.  After Boise State began its long domination in 1999 the games still attracted a sell-out crowd, thus the black and yellow triangles are above the 100 percent line.

For years with annual average attendance data available capacity was utilized at 95 percent or better except for 1999 - 2003 when the stadium had recently expanded by 33 percent, from 22,500 to 30,000.  It took some time to grow into the capacity.  Thus when we have the expansion to 37,000 seats in 2012 there is a dip in annual average attendance as a percent of capacity, but only to 95 percent.

One argument is that because not all the games were total sell outs there is some question whether additional seating is a good investment.  Reader comments on the Idaho Statesman story linked above are amusing (though there are some some good ideas), as everyone has an opinion what would lead to better attendance:  

• a more exciting offense, 
• better opponents (not understanding Colorado State is in the conference), 
• more parking nearby, 
• a larger jumbotron, 
• easier egress, 
• access to alcoholic beverages, and yes 
• cheaper ticket prices.

Those who called for more, easy parking were rebutted by those who said many people going to the games are overweight need to go for a walk now and then.

For a smarter discussion there is the Blue Turf Board (registration required) where for example there is the observation about the need to “defrag” season ticket seating to free up isolated single seats that are going unsold.

Aside from the focus on the individual fan and what might make for better attendance there is the need to look at some broader fundamentals, and I see three:

• stadium capacity given potential market
• ticket price increases related to household income
• high costs of additional capacity

Principles of efficiency dictate that capacity not be overbuilt.  But do you plan capacity for peak use, like an electric utility, or more towards average utilization rates?  The chart below shows that over the long term the capacity of Bronco Stadium appears well in line with local population growth:

If anything using Ada County population is conservative because the Broncos draw fans across southern Idaho and eastern Oregon (best anecdotal evidence can be gathered at the Gear Jammer at exit 95 at Mountain Home about 1 to 2 hours after a ball game.  There are lots of people returning home to south central and eastern Idaho).  In general the seating capacity growth trend is slightly less than the county population trend.

Finding that stadium capacity is in line with area population does not in itself justify stadium expansion, but it is a factor to keep in mind.

Little solid data is readily available on ticket prices.  There is much comment about it, and the annual price increases could be a factor that at some point could affect ticket sales.  So far however, for every internet comment posted from a former season ticket holder who walked away due to high ticket prices and affordability there has been someone to replace him.  But there is little debate that ticket sales income for Boise State football games has grown well beyond that of income at the household or local economic level.  In fact, it could be argued that people are allocating their discretionary spending towards Boise State football and away from other activities (e.g., hunting and fishing license sales are reported to have remained relative flat or declined).

The final factor to wrestle with is any significant expansion will come in large increments in seating and such a structure will have huge up front costs.  Ticket prices on new seats for six home games per year have a limited ability to pay back the construction costs, so the costs will be spread to the existing 37,000 seats.  Yet ticket prices are increasing every year to feed the ongoing operations of the football program.  This appears to be the limiting factor to expansion.

Idaho voters gave 73.3 percent approval to a constitutional amendment to protect the right to hunt, fish and trap.  Of the 1.15 million Idahoans 18 years of age or older about 360,000 purchase a hunting or fishing license (or both).

Whether it’s friends, relatives or business associates of people who hunt or fish, or people who used to hunt and fish but no longer do, yet they support the activity, the election returns make it clear that more than 456,000 voters colored in the “yes” oval on their ballot.  This shows unmistakeable support well beyond just those sportsmen who participate in hunting and fishing activities.

Looking at county level data on sportsmen license data and comparing it to the voting results on the constitutional amendment it’s pretty clear that one factor did not really relate to the other in terms of counties that have a higher percentage of hunting and fishing license holders gave a higher percentage of vote in favor of the constitutional amendment.

The data points (shown in hunter orange) are all over the place, including one where less than half the voters supported the constitutional amendment.  This was Blaine County.

Under a hypothesis that perhaps fishing license holders, who are more numerous, may swamp the hunting data, information on those who purchase just a hunting license were compared to the election results as seen below.

The line is somewhat steeper but the data points are well scattered.  Having information about one - licenses - does not really help predict the other: votes.

In the summer of 2012 Idaho Department of Fish and Game contracted with Responsive Management to conduct a public opinion survey of Idahoans on fish and wildlife issues. One set of questions asked people’s approval or disapproval of hunting, fishing and trapping.  The chart below shows the survey response, and compares it to the election results:

Both hunting and fish are quite popular, with fishing be the more popular of the two, with 97 percent approval.  Trapping has 61 percent approval but about 24 percent disapproval.  The poll results on disapproval of trapping align with the “no” vote on the constitutional amendment.

Here are a series of scatterplots showing LDS membership, vote for Romney, comparisons with vote for McCain in 2008, relationship with voter turnout, Romney vote compared to the “No” vote on Proposition 1 (a measure of crossover voting or ticket splitting), and an indication of probability of crossover voting by Romney voters based on LDS membership by county.

For starters, here is what McCain’s vote in 2008 looked like compared with the percent of county population that is LDS membership, by county:  

As the scattering of data points in the above chart trend upward with percentage of a county population LDS members, the relationship is more pronounced in counties nearing 60 percent or more LDS members.  McCain received 61 percent of the vote statewide in 2008 and in counties with less than 20 percent LDS population there is not much of a relationship.

Romney in 2012 improves on McCain, both in percentage of vote and in the relationship with percentage of county population that is LDS membership.  The r-square value 0.53 means about half of the variation in the percent vote for Romney can be explained by the LDS membership in the county.  As the first major party nominee for President who is LDS, Romney benefitted from fellow church members giving him hefty majorities of the vote.

The chart below looks at the percent of vote in 2008 and 2012 for both candidates and compares the difference in percentage of vote:

This is a simple subtraction of the Romney percent of vote minus McCain percent of vote in 2008.  As LDS membership increases the benefit to the Romney vote is more pronounced.  A difference of eight percent or better is noted for the counties where LDS population is 80 percent or higher.

Clark County was removed as an outlier because the county only has two or three precincts and a hand full of voters did not vote in 2012 compared to 2008 and drastically changed the percent of vote carried by Romney.  President Obama recorded 66 votes in Clark County, two more than his 64 votes in 2008.  It’s the only county in Idaho where he got more votes this year.  

So did Romney being on the ballot drive an increase in voter turnout?  The next chart shows a pretty weak relationship:

Data points are scattered across the chart.  For most counties the total vote for President in 2012 is actually lower than that in 2008.  This is because voter interest in the 2012 election must have waned compared to four years ago.  The decrease in voter turnout for President Obama is an important factor.  Only a couple of counties have a 10 percent increase in turnout, both at the high end of LDS membership as a percent of county population.  But there are also high percentage LDS counties where turnout was down.  Overall it’s a mixed bag, but only four counties show more than a five percent increase in votes in the Presidential election.  That’s not too impressive.  More precision can be brought forward in an analysis once the votes are canvassed and official numbers for total ballots cast are finalized.

What about ticket splitting the vote for Romney and then voting “No” on Propositions 1, 2 & 3?  The next chart shows the Romney percent of vote by county and the No vote on Proposition 1:

The trend line is opposite the other charts, meaning it’s a negative relationship. As Romney’s percent of vote increased the percent of vote for “No” on Proposition 1 decreased.  But the shape of the curve starts to flatten, meaning a higher proportion of voters were crossing over to vote No on Proposition 1.  Overall there is a high r-square value of 0.61 meaning a strong correlation between the two.  As county vote percentages for Romney piled up, generally the percent of vote “No” on Proposition 1 went down.  But there is wide variation and most counties were more than 50 percent on the No vote side of Proposition 1, even in some of the highest Romney supporting counties.

Generally, as the vote for Romney grew more dominant the percentage of voters who would become ticket splitters had to increase as well, otherwise both could not have prevailed.  Here is how a measure of probability of ticket splitting by county compares with the LDS membership as a percent of a county’s population:

Overall, the counties with high LDS membership as a percent of population had a higher probability for ticket splitting, where a given Romney voter would then vote “No” on Proposition 1.  It’s a weak relationship, but one that shows the voters exercised independent judgement on the ballot measures.  This is significant because the “Yes” vote supporters used television advertising of Romney speaking out about teacher unions and the need for school reform.  This campaign tactic was not effective because the Romney-supporting counties displayed significant cross over to vote No on the ballot measures.

A subspecies of the Idaho voter, long thought extinct and not seen in 22 years made a brief reappearance last Tuesday.

Voters gave Republican Presidential candidate Mitt Romney 64 percent of the vote and then down the ballot supported the “no” votes on Proposition One and Two with about 58 percent of the vote.  This is the first case of ticket-splitting since the 1990 election, a practice where voters give landslide support (more than 55 percent of the vote) to one candidates or ballot measures for opposing political parties.  It can be traced back to as far as 1936 as the chart below shows.

Since the ballot measures were inspired by Idaho Education Association opposition of so-called school reform legislation, and the no vote strongly supported by Idaho Democratic voters and office holders, it follows that the vote on Propositions 1 & 2 qualify as a surrogate for a cross-over.  And that’s what the voters did.  This happened in opposition to the Republican Establishment which was lined up in favor of all three ballot measures, and the “no” votes winning was no mean feat given the strong role the Establishment has in calling the shots on initiatives and referenda.

Similarly, in 1978 Democratic Governor John Evans was elected on the same night when voters adopted the 1% Initiative, both with about 57 percent of the vote, and Governor Evans campaigned in opposition to the initiative.

Proposition One limited collective bargaining for teachers in their contracts with school districts, putting the thumb on one side of the scale in labor negotiations.  What’s more, the supporters of a “yes” vote ran a television ad featuring Mitt Romney and his opinions against teacher unions in an attempt to persuade the Romney supporters to also vote yes on the ballot measures.  It didn’t work.

But it’s the candidate races with four examples shown above where the ticket-splitting Idaho voter would elect candidate from opposing parties with landslide votes in the same election.  It goes back to 1936 when President Franklin Roosevelt and Senator William Borah both got 63% of the vote.  The most recent example where two candidates won in landslides was 1990 when Governor Cecil Andrus was reelected and Larry Craig was elected to the US Senate.

Republican Presidential Candidate Mitt Romney won Idaho’s four electoral votes with some 64 percent of the vote.  This amount was an improvement over John McCain’s 61 percent in the 2008 election.  But both of those percentages fall short of that of George W. Bush who racked up 67.2% and 68.4% in 2000 and 2004 respectively.

Idaho maintained 4th place for the percentage of Republican ballots, behind Oklahoma, Utah and Wyoming.  When Bush was on the ballot Idaho ranked 2nd.

One would think with the Romney hype in Idaho during the past year, in large part for his LDS religious membership, the $6 million raised in Idaho to support the national campaign, and having all the endorsements of the establishment, that there would have been an even larger percentage of vote than he achieved.  You know, like do better than Bush in both percent of vote as well as raw vote (in part because there are more voters in the state than eight and twelve years ago).

But statewide turnout as measured by votes cast for President was down 3,500 from 2008.  Taking votes cast from 2008 and 2012 for President Obama and John McCain/Mitt Romney and creating a percentage change from one election to the next give you this comparison:

President Obama shows a percentage decrease in 43 of 44 counties.  That means he got fewer votes in all but one county in 2012 v. 2008.  The one county where his total went up was the anomaly of Clark County, Idaho’s smallest, where he got 66 votes this year compared to 64 in 2008.

Romney’s performance compared to McCain shows a mixed result.  There is an increase in 23 counties and a decrease in 21 counties.  Romney seemed so much more of a hometown favorite than McCain for the reasons stated above it is surprising to see the numbers did not move all that much.  A tide did not rise across the state.

The change in aggragate voter turnout was also very uneven across the counties:

Only thirteen counties showed increased voter turnout while in 31 counties voter turnout went down.  So much for Idahoans being fired up to turn out big to vote for Romney.  Or at least really big.  The counties with the largest percentage increases in turnout are also those with large percentage LDS membership.  We’ll dig in to the Romney vote and LDS membership in a future analysis.

The unofficial vote for President in Idaho totals 651,661, with Mitt Romney getting 420,390 ballots or 64.5 percent of the vote.  He beats the 61.5% that John McCain earned, but McCain garnered his 403,012 votes out of a total 655,032 ballots.

Raw voter turnout is down about 3,500 votes.

Numbers are not yet available for total ballots cast and numbers of registered voters. But predictions of Romney support driving an 80 percent turnout or Propositions 1, 2 and 3 driving a 78 percent turnout could be too optimistic (unless voter registration did not improve).  Assuming an estimated total turnout of 664,000 voters a 78 percent turnout would need be based on 851,000 registered voters and an 80 percent turnout would equate to 830,000 voters.  Since there were nearly 862,000 registered voters in 2008 and every four years the number of registered voters has increased, its not likely the 78 or 80 percent turnout rates will be achieved.

That data will come when the votes are canvassed later in the month.  For now the available numbers are the votes for President which can be compared to the US Census Bureau’s Voting Age Population (VAP) estimate as well as the Voting Eligible Population (VEP) numbers developed by the United States Elections Project at George Mason University.  

So the graph above shows a voter turnout that flatlines for 2012 while VAP and VEP continue an upward trend.

For those wondering, the VEP figure subtracts from VAP those who are not eligible due to disqualifications such as a felony conviction and still on probation and/or incarceration, those residents who are not US citizens, and adjusted by those currently overseas who are eligible voters.  Over time there is a growing difference in the numbers between VAP and VEP.

The Chart below gives you the long-term look at Idaho ballots cast for President as a percent of VAP, and since 1980 VEP.  In 1960 Idaho led the nation in turnout.

Idaho also led the nation in 1980 in total votes cast but was second to Minnesota in Presidential voting because more Idahoans skipped the Presidential election race of Ronald Reagan v. Jimmy Carter v. John Anderson and more voted in the US Senate race between Steve Symms and Frank Church.

The 57 percent No vote on Proposition 1 bears a relationship to the 2006 school supporters’ sales tax initiative that fell short at 45 percent.  With some exceptions many counties showed an increase of 13 percent or more in the vote supporting teachers’ contentions that the law should be overturned.  Plot the percent of vote in the 44 counties from these two ballot measures six years apart and here’s what it looks like:

Above the regression line I colored three points in orange to show three counties where the NO vote in 2012 approached 65 percent, between ten and 20 percent better than the sales tax vote in 2006.  These three counties, from left to right Shoshone, Nez Perce and Bannock respectively, have a history of significant private sector union members in the workforce.  Many residents include sons and daughters of union households and some retirees among the ranks.  They probably helped push a higher NO vote percentage on Proposition One out of a sensibility that teachers should be able to negotiate on more than just wages and benefits.  Until the mid 1990s these three counties were among the most Democratic voting areas.  And given the rare opportunity to vote on a labor issue voters in these counties responded.

The green data points are for Latah and Blaine counties, a couple of liberal strongholds which voted in strong support for both the sales tax initiative in 2006 and marked the NO vote on the Proposition 1 ballot.

All five counties well above the regression line, coming out on the winning side of a vote, they wake up the echoes of a time when a candidate like a Cecil Andrus could put together a winning coalition in Idaho.