Proposition 50, the California-slaps-back initiative, is cruising to a comfortable victory on Nov. 4, a slam dunk for Gov. Gavin Newsom and efforts to get even with Texas.

Or not.

It's actually a highly competitive contest between those wanting to offset the GOP's shameless power grab and opponents of Democrats' retaliatory gerrymander — with many voters valuing California's independent redistricting commission and still making up their minds.

Obviously, both things can't be true, so which is it?

That depends on which of the polls you choose to believe.

Political junkies, and the news outlets that service their needs, abhor a vacuum. So there's no lack of soundings that purport to show just where Californians' heads are at a mere six weeks before election day — which, in truth, is not all that certain.

Newsom's pollster issued results showing Prop. 50 winning overwhelming approval. A UC Berkeley/L.A. Times survey showed a much closer contest, with support below the vital 50% mark. Others give the measure a solid lead.

Not all polls are created equal.

"It really matters how a poll is done," said Scott Keeter, a senior survey advisor at the Pew Research Center, one of the country's top-flight polling organizations. "That's especially true today, when response rates are so low [and] it's so difficult to reach people, especially by telephone. You really do have to consider how it's done, where it comes from, who did it, what their motivation is."

Longtime readers of this space, if any exist, know how your friendly columnist feels about horse-race polls. Our best advice remains the same it's always been: Ignore them.

Take a hike. Read a book. Bake a batch of muffins. Better still, take some time to educate yourself on the pros and cons of the question facing California, then make an informed decision.

Realizing, however, the sun will keep rising and setting, that tides will ebb and flow, that pollsters and pundits will continue issuing their prognostications to an eager and ardent audience, here are some suggestions for how to assay their output.

The most important thing to remember is that polls are not gospel truth, flawless forecasts or destiny carved in implacable stone. Even the best survey is nothing more than an educated guess at what's likely to happen.

That said, there are ways to evaluate the quality of surveys and determine which are best consumed with a healthy shaker of salt and which should be dismissed altogether.

Given the opportunity, take a look at the methodology — it's usually there in the fine print — which includes the number of people surveyed, the duration of the poll and whether interviews were done in more than one language.

Size matters.

"When you're trying to contact people at random, you're getting certain segments of the public, rather than the general population," said Mark DiCamillo, director of the nonpartisan Berkeley IGS Poll and a collaborator with The Times. "So what needs to happen in order for a survey to be representative of the overall population ... you need large samples."

Which are expensive and the reason some polls skimp on the number of people they interview.

The most conscientious pollsters invest considerable time and effort figuring out how to model their voter samples — that is, how to best reflect the eventual composition of the electorate. Once they finish their interviews, they weight the result to see that it includes the proper share of men and women, young and old, and other criteria based on census data.

Then pollsters might adjust those results to match the percentage of each group they believe will turn out for a given election.

The more people a pollster interviews, the greater the likelihood of achieving a representative sample.

That's why the duration of a survey is also something to consider. The longer a poll is conducted — or out in the field, as they say in the business — the greater the chances of reflecting the eventual turnout.

It's also important in a polyglot state like California that a poll is not conducted solely in English. To do so risks under-weighting an important part of the electorate; a lack of English fluency shouldn't be mistaken for a lack of political engagement.

"There's no requirement that a person be able to speak English in order to vote," said Keeter, of the Pew Research Center. "And in the case of some populations, particularly immigrant groups, that have been in the United States for a long time, they may be very well-established voters but still not be proficient in English to the level of being comfortable taking a survey."

It's also important to know how a poll question is phrased and, in the case of a ballot measure, how it describes the matter voters are being asked to decide. How closely does the survey track the ballot language? Are there any biases introduced into the poll? ("Would you support this measure knowing its proponents abuse small animals and promote gum disease?")

Something else to watch for: Was the poll conducted by a political party, or for a candidate or group pushing a particular agenda? If so, be very skeptical. They have every reason to issue selective or one-sided findings.

Transparency is key. A good pollster will show his or her work, as they used to say in the classroom. If they won't, there's good reason to question their findings, and well you should.

A sensible person wouldn't put something in their body without being 100% certain of its content. Treat your brain with the same care.

____

©2025 Los Angeles Times. Visit at latimes.com. Distributed by Tribune Content Agency, LLC.