May 11, 2026

What the Godbe poll actually says — and what it doesn't

In December 2025, Godbe Research presented Ross Valley School District trustees with the most concrete public read on where Measure H actually stands. The poll surveyed 418 likely June 2 primary voters by phone, text, and email between November 10–12, 2025, with a stated margin of error of 4–5%. It was reported by Keri Brenner in the Marin IJ on December 21, 2025.^[1]

There are three findings from the poll worth understanding in detail. Two of them are reasons for cautious optimism on Yes. The third is the single most strategically important data point the campaign has, and it directly answers the opposition’s core argument.

Finding 1: ~65% favorability, against a 66.67% threshold

The headline number was about 65% favorability for Measure H — within the 4–5% margin of error of the 66.67% supermajority threshold required for passage.

Godbe Research president Bryan Godbe described that as “within striking distance” of the threshold but “not enough to offer an overwhelming endorsement.”^[1]

The honest read of this number is that Measure H is closer to passing than failing, but not by enough to be safe. A measure polling at 65% with a 4–5% MOE could land anywhere from ~60% to ~70% on election day. The lower end is below Measure E’s 62.53% in May 2025; the upper end clears the threshold.

This is what the Yes campaign is working with. The persuasion work between now and June 2 is about closing the last 2–4 points and ensuring turnout doesn’t shift the electorate against the measure.

Finding 2: 73% approval of how RVSD manages money

The more interesting number — the one that actually defines the campaign’s strategy — is that 73% of respondents approved of how RVSD manages taxpayer dollars.^[1]

That’s meaningfully higher than the 65% favorability for the measure itself, and notably higher than what the district saw in pre-Measure-E polling.

Trustee Daniel Cassidy said: “This is a big difference. People are more in favor of how we’re managing the money. That, to me, is a sign that we’re telling the right story.”^[1]

The strategic implication: there’s an 8-point gap between voters who think the district manages money well (73%) and voters who currently support the measure (65%). That gap is the persuasion space. Closing some of it gets Measure H to two-thirds.

This suggests the persuasion job isn’t to convince skeptics that the district is competent (most already believe it is). It’s to convince voters who trust the district’s management to extend that trust to specifically supporting Measure H — which means making the consequences of No more concrete and the structure of the measure more legible.

Finding 3: Lower-cost variants didn’t move the needle

This is the most important strategic finding in the whole poll, and it directly answers the opposition’s most respectable argument.

Per Godbe, the survey tested lower-cost variants of the measure — a smaller increase, a lower annual escalator, etc. The reported result: “those did not move the needle of favorability.”^[1]

This is the data point that justified the board’s decision to land at $540 rather than CST’s suggested $149 or the parent’s suggested “in the $300s.”^[2] The polling implied that the price point isn’t the principal binding constraint on Yes support. Trimming the ask wouldn’t have unlocked the supermajority. Voters who oppose Measure H aren’t going to flip to Yes if the measure shrinks; voters who support it would have supported either version.

That’s a non-obvious finding, and it’s worth taking seriously. The intuition that “ask for less and you’ll pass” is everywhere in the opposition framing. But the actual data — gathered from a 418-voter sample by a polling firm Marin districts routinely use for measure feasibility — found that intuition to be wrong.

This doesn’t mean smaller asks can’t pass. It means the gap between 65% and 67% on Measure H is not principally a dollar-amount gap. It’s a salience-and-stakes gap.

What the poll didn’t test

A few things worth being honest about. The Godbe poll:

Was paid for by RVSD. Godbe is a respected nonpartisan firm Marin districts widely use, but the survey was commissioned by the district, not by an independent third party. The methodology is standard, but the choice of what to test was made by the client.
Was conducted in November 2025, before the contingency plan was presented in January 2026. The closure-and-receivership stakes that have been part of the Measure H argument since January didn’t exist as a public document when the polling was done. Voters’ reaction to those specific stakes wasn’t measured.
Measured favorability, not voting behavior. Favorability and votes correlate strongly but aren’t identical. Some respondents who say they “favor” a measure don’t return their ballot; some who say they “oppose” change their mind.
Was a single point in time. Poll-to-poll movement in measure favorability over the spring is unmeasured in this dataset.

The Measure E parallel is the unsettling part. Per the same December reporting, the poll’s results were similar to what Godbe presented to the board before Measure E went on the May 2025 ballot — and Measure E ultimately got 62.53%, a few points below the pre-poll number. Polls in the upper-65s for parcel tax measures have a track record of softening on election day. The 4–5% margin of error is doing real work here.

Mimi Willard’s pollster critique, briefly

CST president Mimi Willard told trustees that long polls produce voter fatigue (“they are annoyed that you keep asking questions, so they just keep saying ‘no’”) and that the survey didn’t ask about the cumulative tax burden across multiple 2026 ballot measures.^[1]

The fatigue critique is empirically testable post-election; it’s not a frivolous concern about long-form opinion surveys, but it cuts in unclear directions for a Yes campaign. The cumulative-burden critique is more substantive and is what we address in our post answering the cumulative-burden argument — short version: the actual June 2 ballot has 11 tax measures countywide, and as a Fairfax or San Anselmo voter your school decision is just Measure H.^[3]

Our read

The Godbe poll is the most credible public data about Measure H’s chances. It says:

The measure is plausibly within reach of two-thirds, but not safely above it.
There’s an 8-point persuasion gap between voters who trust district management and voters who currently support the measure.
The “ask for less” alternative would not have changed the outcome.

If you’re voting Yes, that translates concretely: the campaign’s success depends on converting some of the 8% who trust the district into specific Yes voters on Measure H. That’s a turnout-and-persuasion conversation with neighbors, not a structural change to the measure.

That’s why we recommend a Yes vote on Measure H — and why we keep coming back to the importance of two-or-three-neighbor conversations between now and June 2.

Sources

← All posts