Hawaii voters made a number of important decisions Tuesday. Was there a common thread between any of them? Back by popular demand, it’s Civil Beat’s correlation coefficients!

After the primary election in September, we crunched the numbers and told you, for example, that there was a strong mathematical relationship between gubernatorial candidate James “Duke” Aiona and mayoral hopeful Panos Prevedouros.

This time around, we can tell you that supporters of Neil Abercrombie tended to support the constitutional amendment allowing the state to keep excess tax revenue instead of returning it to you. Read to the end for a countdown of the five most interesting correlations we found.

First, let’s review the rules. Here’s what we wrote six weeks ago:

What is a correlation coefficient, you ask? The Pearson product-moment correlation coefficient is a measure of the linear dependence between two variables and can be determined by using this formula:

Source: Screen capture of Wikipedia

Don’t worry, it’s all Greek to us, too. Luckily, Microsoft Excel has a handy PEARSON function that lets us quickly input two arrays of data — for example, the district-by-district results for a gubernatorial candidate and a mayoral candidate.

The formula’s output is a number somewhere between 1.0 (for perfect correlation, when the data go up and down in tandem) and negative 1.0 (for when they mirror each other perfectly, with one going up when the other goes down). A result of zero means the two sets of data are mathematically independent of each other.

To be clear, the above table only means that where Hannemann performed relatively well, Caldwell was somewhat likely to do the same. Just because the coefficient representing their relationship was .770 does not mean that Caldwell got 77 percent of the votes that Hannemann did, and doesn’t mean the likelihood of Caldwell winning a district was 77 percent if Hannemann won it.

It’s impossible to draw a straight line between candidates in different races. The correlation coefficient is not about causality, and to attribute a causal relationship between two candidates from this data would be a bridge too far.

It’s up to us humans to interpret the data and figure out why things went the way they did. The relationships we’re describing are in no way personal but entirely mathematical, like love between two robots.

Civil Beat used the formula to investigate the relationships between the relative successes and failures of candidates in different races across Hawaii’s 51 House districts. As you saw above, some of the findings pretty much support conventional thinking, and some are counterintuitive.

For the general election, we threw the following district-by-district percentages into the blender to make our cocktail: voter turnout; Abercrombie; Aiona; Djou; Colleen Hanabusa; Daniel K. Inouye; Mazie Hirono; Appointed Board of Education amendment; independent Honolulu transit authority amendment; tax rebate amendment; and per capita income. Below you’ll find tables describing the relationships between those ingredients.

All tables are sorted from strongest relationship to weakest, regardless of whether the relationship was positive or negative.

Takeaway 5: Voting Down Party Lines

This table holds the fewest surprises. There were positive relationships between members of the same party and negative relationships between members of opposite parties. It’s interesting to note that both gubernatorial candidates had their weakest relationship with Sen. Inouye. He performed strongly with voters of all stripes, making him a less polarizing figure even as he’s the “godfather” of Democratic politics in Hawaii.

Candidate 1 Candidate 2 Correlation
Abercrombie Hirono 0.829
Abercrombie Hanabusa 0.823
Abercrombie Djou -0.823
Aiona Djou 0.817
Aiona Hanabusa -0.817
Aiona Hirono -0.819
Abercrombie Inouye 0.717
Aiona Inouye -0.701

Source: Civil Beat analysis

Takeaway 4: Rich People Vote

We see here that there’s a strong positive relationship between income and turnout, which we reported on yesterday. It’s also apparent that regions with lower turnout tended to move toward Republican Aiona but away from Republican Djou, if only slightly. Democrats Hirono and Inouye performed stronger when turnout was high.

Voter Turnout Other Factor Correlation
Voter Turnout Income 0.596
Voter Turnout Honolulu Transit Authority “Yes” 0.459
Voter Turnout Appointed Board of Education “Yes” 0.401
Voter Turnout Hirono % 0.394
Voter Turnout Inouye % 0.326
Voter Turnout State can keep excess revenue “Yes” -0.216
Voter Turnout Abercrombie % 0.156
Voter Turnout Aiona % -0.133
Voter Turnout Djou % 0.075
Voter Turnout Hanabusa % -0.075

Source: Civil Beat analysis

Takeaway 3: Wealthy Backed Djou, Appointed BOE

Wealthy districts backed GOP candidates in Hawaii this year, even as Democrats dominated the election. Hirono, Hanabusa, Inouye and Abercrombie all had negative relationships with per capita income, while Republicans Djou and Aiona had positive relationships.

Also of note is the relationship between high income and support for an appointed Board of Education.

Per Capita Income Other Factor Correlation
Income Hirono -0.412
Income Djou 0.361
Income Hanabusa -0.361
Income Appointed Board of Education “Yes” 0.339
Income Inouye -0.287
Income Aiona 0.193
Income Honolulu Transit Authority “Yes” 0.192
Income Abercrombie -0.179
Income State can keep excess revenue “Yes” 0.022

Source: Civil Beat analysis

Takeaway 2: No Correlation on Tax Rebate

One number in the above table that we struggled to explain is the lack of a connection between income and the tax-related constitutional amendment. If wealthier residents are more fiscally conservative and wanted the government to return the revenue, you’d see a negative figure there. If less affluent voters need the money more urgently, you’d also see a negative number.

Maybe those theories are both off, or maybe the ballot question was just confusing and people didn’t know what they were voting for, but the figure turned out to be 0.022, pretty darn close to zero.

Takeaway 1: Abercrombie Says ‘Yes,’ Supporters Say ‘No’

It’s hardly shocking that Republicans backing Aiona’s candidacy would tend to oppose any change that would allow the government to hang onto tax dollars any longer than previously permitted.

The interesting takeaway here, though, is that areas supportive of Abercrombie tended to oppose the appointed Board of Education amendment, even though he changed his mind during the campaign and cast a vote in support. That’s likely due to the fact that the Hawaii State Teachers Association endorsed Abercrombie and simultaneously ran a campaign opposing the amendment.

Meanwhile, Republican-friendlier areas supporting Aiona voted to give more power to the government. Not particularly conservative, right?

Candidate Initiative Correlation
Aiona State can keep excess revenue “Yes” -0.565
Abercrombie State can keep excess revenue “Yes” 0.558
Aiona Appointed Board of Education “Yes” 0.309
Abercrombie Appointed Board of Education “Yes” -0.280
Abercrombie Honolulu Transit Authority “Yes” 0.193
Aiona Honolulu Transit Authority “Yes” -0.181

Source: Civil Beat analysis

Now that we’ve all had our fill of these correlation smoothies, it’s time to put the blender away for two years — unless you can find some more numbers for us to crunch.

About the Authors