November 28, 2010 07:01 AM UTC

Amendment 62 A First Look at Official Vote Totals

•  
• by: Gypsy Chief

The amendment failed with 29.498% yes,

70.502% no statewide. This much we knew on election night. To take a closer look I waited until official results were presented on the Secretary of State website on Wednesday, November 24, 2010.

Did Amendment 62 carry any county? No. The high

water mark was Conejos County with 45.037% yes, the low was San Miguel County with 15.387% yes.

The top ten counties in terms of percentage yes

on 62 are:

Conejos 45.037%

Yuma 40.910%

Archuleta 40.762%

El Paso 40.665%

Cheyenne 40.591%

Rio Grande 40.567%

Fremont 40.179%

Delta 39.950%

Las Animas 39.416%

Custer 39.289%

The bottom ten counties are:

Clear Creek 21.323%

San Juan 20.957%

Gunnison 19.839%

Jackson 19.777%

Summit 16.963%

Routt 18.699%

Denver 17.067%

Boulder 17.029%

Pitkin 16.963%

San Miguel 15.387%

Where were the votes cast?

There were 1,722,548 total votes.

I ranked the counties by total votes and ran a cumulative vote total:

Here are the top ten.

COUNTY Percent of State Cumulative

Jefferson 12.751% 219,643

Arapahoe 11.022% 409,510

El Paso 10.971% 598,494

Denver 10.639% 781,755

Boulder 7.037% 902,972

Larimer 7.027% 1,024,023

Douglas 6.526% 1,136,431

Adams 6.273% 1,244,489

Weld 4.501% 1,322,020

Mesa    3.119% 1,375,748

These top ten counties cast 79.867% of

all votes on the measure.

I ranked all counties by percent yes and calculated the effect their

yes votes had on the total yes vote. For example

El Paso County voted 40.665% yes and those votes contributed 15.125% to all of the yes vote on the measure. Jefferson County voted 26.749% yes and those votes contributed 11.563% to all of the yes vote. Arapahoe County voted 27.962% yes and those votes contributed 10.449% to all of the yes vote. There was no other county that contributed as much as 10.000% to the total yes vote. Clearly El Paso County is an outlier. If there were

more places like El Paso County in Colorado the margin of difference would have been closer.

The measure would have failed – just not as badly. I considered taking a standard deviation

on percent of support and doing z-scores on all counties. Decided not to because I don’t

think it would add much to the analysis.

It took 861,275 no votes statewide to sink the measure. I ranked counties by total no votes and ran a cumulative no. The no votes from eight counties sank 62:

COUNTY Total No Cumulative No

Jefferson 160,891 160,891

Denver 151,984 312,875

Arapahoe 136,777 449,652

El Paso 112,134 561,786

Boulder 100,575 662,361

Larimer 84,998 747,359

Douglas 74,361 821,720

Adams 73,851 895,571

There is our friend El Paso County again. The yes votes from this county contributed significantly to the total yes vote. However the no votes also contributed significantly to defeat the measure.



Method and Limitations

This analysis is done on a county by county basis from official election returns.

Colorado counties vary so much in population, economic activity, education, culture

and so forth that the idea of a typical county is meaningless. Nonetheless I wanted to see if there were wide variations in percent of support for the measure as between counties.

It appears that there is based on a range of Conejos County (45.037) to San Miguel County (15.387).

Certainly a next step would be to break down the vote by gender, income, education, political

party and so forth. Since the measure failed so badly some may wonder if further analysis

along these lines is warranted. Then again there is a steady stream of political science

majors (such as yours truly eons ago) who may want to look deeper. Staying with county by

county analysis for a moment a scatter diagram might plot percent support against total votes.

Perhaps counties could be grouped by geographic region or main economic activity.

My own interest lies in another direction. I took on this project as a warmup for a look at the 4th Congressional District race. What can election returns tell us about that? An analysis of election strategy posted by user LDOP before the election got me thinking back to the days when my friends and I were doing real estate appraisals on typewriters. ‘Cut and paste’ meant do arithmetic on a calculator, type it up, photocopy, use scissors and cellophane tape to assemble a spreadsheet then photocopy that to make the final presentation. If there was a late-breaking comparable you got to do it over. Someone found an unused Everex 286-12 with 640k of ram. It had a mystery program which turned out to be SuperCalc 4. Love at first sight for an appraiser.

When I get to the 4th Congressional District I want to take a look specifically at nine

counties that I arbitrarily label as southeast. They are Baca, Bent, Cheyenne, Crowley, Kiowa, Kit Carson, Lincoln, Otero, and Prowers. The statewide average yes vote on 62 was 29.498; the average for these nine counties was 36.919%. There were not enough votes there to make a material difference in the outcome for 62. Was that true as well in the Markey-Gardner race?