While the election hasn’t been officially certified yet, it looks likely Measure S will end up failing. It didn’t miss by much — the latest count shows it getting 65.5% of the vote — but there are probably so few votes left to count that it’d be a stunning turnaround for it to end up passing.
Which naturally raises the question “How did this happen?”
But before we get into that, it’s worth looking at how votes were cast. I’ve looked at the current, near-final results in a number of different ways, and the most intuitive one I’ve come up with is to look at what I call the Net Yes vote.
The Net Yes vote is defined this way:
Net Yes = Yes Vote – (2 x No Vote)
The multiplicative factor two is used because parcel taxes must garner a 2/3 supermajority to pass. In other words, each No vote has twice the effect of each Yes vote on the outcome of the election.
At this point, Measure S has a Net Yes vote of (469), negative 469. That means that Measure S fell short by 469 Yes votes (i.e., if there had been 469 more Yes votes somewhere in the District the initiative would have passed). Here’s how that 469 Yes vote deficit was generated within the District, by precinct:
Net Yes votes for Measure S
You can click on the image for a larger version (please note it’s a pretty large file). Areas tinted red produced Net Yes deficits, while those tinted green produced Net Yes surpluses. Color intensity corresponds to how big a deficit or surplus was produced. The more intense the color, the bigger the deficit or surplus. The golden borders are the approximate boundaries of the elementary school service areas.
There are two striking things about this map. First, there are more intense red areas than intense green areas. While Measure S pulled ahead in a number of precincts, it didn’t do so by much. Conversely, there were a good number of areas where it lost pretty heavily.
The other interesting thing I see is the way Measure S lost heavily in the hills behind Arundel. While both of my kids are now in high school, they both attended Arundel and Tierra Linda, and I know more people from that area of town. I thought I knew the area pretty well, and I thought it was a pretty strong supporter of the District. It was surprising to see it go so heavily against Measure S.
I was less surprised to see what I might loosely call the southwest hill country of San Carlos go against Measure S. That’s because I believe there are relatively fewer households in that area with school age children, which would tend to reduce overall awareness of District issues.
A takeaway from this analysis is that future campaigns ought to think about how to do a better job of outreach into the “hill country”. The density of kids attending District schools is probably lower in those areas, which means less “inherent” understanding of District issues. An organized effort to hold neighborhood outreach efforts might pay big dividends.
It was also interesting to see that Measure S lost, although not heavily, some of the most “kid friendly” areas of the District, down on the “flats” (if someone would like to suggest a better term for that part of the District please email me at mark@arcabama.com).
It’s also worth looking at how turnout varied across the District. Areas that generated large Net Yes vote surpluses but had low turnout could be targets for get-out-the-vote efforts the next time a tax initiative is put on the ballot. Here’s how turnout varied across the District:
Once again, you can click on the image for a larger version (it’s a pretty large file). The intensity of the tint corresponds to the relative turnout. The more intense the color, the higher the turnout.
What leaps out at me from this map is that the precinct that generated the most Net Yes votes also had one of the lower turnout rates. Assuming the campaign didn’t get every single Yes voter to go to the polls from that precinct, it could be an attractive place on which to focus. It also looks like we could do a better job of getting voters out from the eastern parts of the “flats”.
One last observation about turnout: precinct 3626 — the one on the leftmost edge, with the lightest tint — had the lowest turnout for Measure S. In fact, it was so low that if I hadn’t compressed the way the tints were calculated almost all the rest of the District would have been uniformly dark blue. There may be an interesting story in that precinct, which I’ll come back to in a later post.
Very interesting analysis.
Maybe you can take this to a next step of guestimating what the best selling points for the measure are in the various districts? My wife volunteered on the pro Measure S call campaign which attempted to educate voters on the measure, but the campaign seemed to lack consistent convincing arguments. This could be due to a poorly ran campaign, taking voter interest for granted, or possibly the failure to make a connection to voter interests. With San Carlos being a community with such a diverse population (baby boomers and retirees, blue color workers and young affluent families) as seen through your district analysis, it would seem that a potential key to getting this measure passed may lie in more targeted campaigning. I believe your analysis can help focus our efforts by identifying which message/sales pitch should be made to which area…
With your further analysis, we may just get this passed.
FWIW, the new principal of Brittan Acres, John Triska, is very well regarded and highly intelligent making me think he may be one of the better people to help promote your analysis. He writes regular articles to school parents which are always well written and very well received. Perhaps he can help get a well written article in support of the measure published to all district parents.
Please keep me posted and I appreciate your efforts.
Nice analysis. How do you know which districts have school age children? I was surprised by the bright green in 3608. Other than the areas by Prospect and Cedar, it seems that stretch goes right down through most of the retirement communities in the area which I would assume do not have school age children. Were you surprised by that green spot?
Adam,
I’m going to write one or two posts about why I think Measure S failed over the next day or so.
I wouldn’t characterize the campaign as poorly run, although I’m probably biased because I was one of the people on the campaign committee and did a lot of work to try and get it passed :).
For now, I’ll leave you with the observation that there are roughly twice as many reasons being offered for why it failed as there are people offering them :). In other words, there are a LOT of hypotheses out there. There’s a lack real data on why people voted the way they did, however, and so figuring out the “real” reasons for the failure is difficult.
Mike,
Good point about the demographics of 3608. Yes, I was surprised, but initially chalked it up to 3608 encompassing an area north of San Carlos Avenue which has a lot of homes and, I think, a lot of school age children.
But actually 3608 is south of San Carlos Avenue.
It’s possible that retired voters were aware that anyone aged 65 or older could exempt themselves from Measure S. The campaign did to to some trouble to try to get that message out there.
We are in the midst of a worldwide recession, which is not the fault of anyone in SCSD, but may have affected many voters’ willingness to tax themselves.
Years ago, a SD my daughter attended did a pre-election survey, which resulted in asking for less money in order to get it passed. At the time school parents were seven percent of the voters there.
Flatlands is East San Carlos, Bayside San Carlos, Bayshore San Carlos?
Mark – Great information. One interesting comment from a “No” voter I spoke with (paraphrasing): “If only 50% of San Carlos school district parents felt compelled enough to make a contribution benefitting their own kids via the San Carlos Ed Fund (not even $5), why should 100% of property owners be mandated to contibute even more from their property taxes.” A rather simplified viewpoint, but I think I do understand the fairness perspective she presented.
Steve,
Wow, I hadn’t heard that before! FWIW, 50% participation is actually pretty high.
I suspect you heard yet another application of what I call the “Rationalization Rule of Politics”: people make up their mind to vote one way or another, and then look for a “reason” to justify their decision.