In polling, the answer you get depends heavily on the question. Obvious enough, right?

Few should understand this point better than the prestigious, independent Field Institute, whose polls on California issues often contribute to the public debate.

So why is Field polluting the discussion around revenues in California with bad questions and bad data?

We can and must do better, and soon.

The most recent Field Poll scans the horizon for support levels on prospective special-election issues

(get the poll here, and the fascinating cross-tab results here).

It’s helpful to see, in this poll, where Californians would cut to help reduce the state budget deficit. Basically, most voters would cut nothing except prisons and, perhaps, the costs of environmental regulation. Republicans would cut a lot more, except K-12 public schools.

The broad consensus is against almost any cuts that would be needed to meet the state’s widening budget chasm.

So are people willing to support new revenues? (OK, let’s go ahead and say “tax increases.”)

Not according to Field. But look at how they asked the question. Registered voters were asked to respond to: “I would be willing to pay higher taxes to help the state balance its budget.”

Even that poor wording got 43% support overall. Democrats and independent voters gave it 53% support.

But look at what’s missing in the question. The wording speaks to who would pay the higher taxes – the respondent – but how much? And for what purpose?

There’s no limit stated, no type of tax enumerated. And the purpose is as grim as they get: “help[ing] the state balance its budget.” If you have ever seen California voters talk about the state budget, they view it as a black hole rife with waste.

It’s astonishing that there’s a core of 43% willing to shovel their own money into a ditch to help “the state” fix “its budget.”

Of course there are many better ways to ask this kind of question. For instance, you could ask voters if they’d pay more to protect the specific programs and services they like. (Which is basically everything but prisons.) Linking tax increases to specific purposes helps a lot.

Or you could ask about specific kinds of taxes that don’t mainly affect the individuals responding to the poll. An oil extraction tax gets big numbers in polling for lots of reasons, one of them being who it’s targeted at (highly profitable oil companies).

An increase in income taxes on the wealthiest Californians similarly scores well because – as much as we may try – most Californians don’t fit into the top 1%, 5%, or even 10% on the income scale. (Disclosure: I managed the campaign for Prop. 63, which added a 1% surcharge on annual income over $1 million to support mental health programs.)

And if you present the facts properly, I’d bet you’d see majorities supporting a split-roll property tax, or at least a fix to change-in-ownership rules for commercial property.

We need to educate voters about the degree to which we have a revenue problem that’s been papered over for 20 years and salvaged occasionally by bogus economic bubbles. There are lots of sensible ways to raise money and not have undue impact on the people who can least afford it. If there’s ever going to be a time to find billions of dollars in steady new revenue sources, we’re pretty near the “hair on fire” phase now where that will be both necessary and possible.

Don’t believe you’d be limited to 43% support for a raft of revenue measures. Smartly designed, longer-lasting revenue solutions will have an audience – an even bigger audience if we don’t see this special election happen, or if the tax extensions fail and we flop into devastating austerity. Let’s get to work now.  

One advantage the Legislature has had for years in referring ballot measures directly to the people is that lawmakers were not subject to any standard of impartiality. They could, and often did, write a sunny “official” description of their own referred measures for official ballot materials. The hope was to encourage some number of extra votes by casting such measures in a positive light.

That’s in the past tense now, though, due to a court ruling last week. And that change could jeopardize legislatively referred measures intended for the much-discussed June 2011 special election ballot.

Without the power to control the framing of the special-election measures, the Legislature and Gov. Brown could find passage just marginally more difficult in an election that could be extremely close.  

The Court of Appeals ruled late last week that the Legislature cannot supplant the Attorney General, who is normally charged with writing a fair and “impartial” description of statewide ballot measures. From now on, no more advocacy language from legislators can appear in the official title & summary.

The case came out of the successful Nov. 2008 high-speed rail bond measure, whose title and summary were pointedly upbeat. For instance, the title was the “Safe, Reliable High-Speed Passenger Train Bond Act,” and the first bullet read, “Provides long-distance commuters with a safe, convenient, affordable, and reliable alternative to driving and high gas prices.” (For more see here: http://bit.ly/gK7c05)

The measure won (52%-48%), and that victory won’t be revisited due to this ruling, but the appeals court found that the Legislature’s practice of writing such chipper language for its own ballot measures violated the law controlling initiative descriptions. The Legislature has routinely exempted itself from that law in the course of stipulating the language to appear as the title and summary. But those days are over.

Going forward, every measure, whether it’s put on the ballot by citizens or legislators, will now need to have a neutral ballot title, summary and label prepared by the AG. And that could make the much-discussed June 2011 special election harder for Gov. Brown and his slate of initiatives.

To see why this matters, look back only to the most recent statewide special election. In May 2009, the Legislature put a slate of special measures on the ballot, most related to the budget. I worked against two of those, Props. 1D & 1E, which sought to cut early-childhood programs and mental health services that had previously been created by voter enactments.

Suffice it to say that the title and summary descriptions written by the Legislature skipped over a lot of the details, like the fact that we were being asked to amend past voter initiatives. The summaries made these cuts sound painless and even, in some ways, made the proposals sound like efforts to ensure funding for important programs, not to cut programs.

Hoping to find votes, the authors of these summaries had inverted the meaning of the actual proposals to make them sound good. For those of us hoping to persuade voters to stop the cuts, the official language was maddening.

Opponents of 1D were not able to challenge the ballot description, but the 1E opponents did. They feared that the initial summary language would falsely convey to voters that 1E would protect and enhance children’s mental health services, instead of cutting significant mental health services. But it was unclear whether we could win a court fight. If the court were to find that the Legislature was free to exempt itself from the requirement for an impartial title and summary, 1E opponents would have been forced to accept the language as initially drafted

Before their case went to a judge, attorneys for the Legislature and the 1E opponents worked out a settlement. The revised summary was better. But settlements get forced when both sides are unsure of the outcome. The 1E opponents took half a loaf rather than risk going with the first draft.

There was more of a titanic battle around that year’s Prop. 1A ballot description, a fight over the nature of the tax increases proposed therein and how they were tied to a spending cap. Opponents of 1A won a couple of tweaks to the language, but mostly lost their effort to recast 1A in a more neutral/negative light. In their case, the court upheld the basic right of the Legislature to put almost anything they wanted into the title & summary.

Now, all of the budget measures in May 2009 went down to defeat by about 2-to-1, so you can argue that any puffery put into ballot materials by the Legislature was a nonfactor. But in a close race, everything matters. The June 2011 special election – if it comes together – was surely conceived with one big assumption: whatever the Legislature puts on the ballot, it will control the framing of those measures in the official ballot materials. No more.

The Court of Appeals decision in this case is remarkably simple to read and grasp. It’s online here:

http://www.courtinfo.ca.gov/op…

The gist of the ruling is this: The requirement for the AG to write an impartial summary of “each measure” on the statewide ballot was itself instated by a ballot measure – the Political Reform Act (PRA) of 1974. Even the Legislature can’t ignore the statutory requirements of a ballot measure enacted by the people. So the appeals court found that the Legislature could not exempt itself from those requirements. The fact that legislators were only doing it for measures they had drafted themselves was immaterial. The PRA says it’s the AG’s job – period.

This is a simpler Article II case than even many critics of this longstanding legislative practice seem to have imagined.

The bottom line for June 2011 now is that any legislatively referred measure must go to the AG for a description, and the content of that description will be subject to litigation on any claim that the language may be prejudicial.

That could be a big negative for some of the issues that have been rumored to be part of a new special-election package, including possible repeats of 2009’s 1D & 1E to cut children’s services or mental health care. The job of selling the special election just got harder.