A guide to polling on California’s redistricting measure

Proposal 50, the California-Slaps-Bar initiative, is in the process of comfortable victory on November 4, a Slam Dunk for Governor Gavin Newsom and efforts to do even with Texas.

Or not.

It is in fact a highly competitive competition between those who wish to compensate for the shameless takeover of the GOP and the opponents of the Gerrymander of the reprisals of the Democrats – with many voters evaluating the independent redistribution commission of California and still deciding.

Obviously, the two things cannot be true, so what is it?

It depends on which surveys you choose to believe.

Political drug addicts and the media that meet their needs hate a void. There is therefore no shortage of survey which claims to show where the Californian heads are just six weeks before election day – which, in truth, is not so certain.

The Newsom sounder has published results showing the prop. 50 winning overwhelming approval. A UC Berkeley / La Times survey has shown a much closer competition, with support below the 50%mark. Others give the measurement a solid lead.

Not all polls are created equal.

“It is really important to see how a survey is done,” said Scott Keeter, main investigation advisor to the Pew Research Center, one of the country’s high -flying organizations. “This is particularly true today, when the response rates are so low [and] It is so difficult to reach people, especially by phone. You really have to think about how it is done, where it comes from, which has done, what is their motivation. »»

The long -standing readers of this space, if applicable, know what your friendly columnist feels breeding. Our best advice remains the same as he has always been: ignore them.

Take a hike. Read a book. Cook a lot of muffins. Better yet, take the time to inform yourself about the advantages and disadvantages of the question that California is confronted, then make an informed decision.

Realizing, however, the sun will continue to get up and go to bed, that the tides will refuse and flow, that the probers and the experts will continue to deliver their forecasts to an impatient and ardent audience, here are some suggestions on how to test their production.

The most important thing to remember is that polls are not the truth of the Gospel, impeccable forecasts or the destiny carved in an implacable stone. Even the best investigation is nothing more than an enlightened supposition on what is likely to happen.

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

Given the opportunity, take a look at the methodology – it is generally there in small characters – which includes the number of people questioned, the duration of the survey and if interviews have been carried out in more than one language.

The size counts.

“When you try to contact people at random, you get certain public segments rather than the general population,” said Mark Dicamillo, director of the non -partisan survey of Berkeley Igs and collaborator with Times. “So what must happen so that a survey is representative of the global population … You need significant samples.”

Which are expensive and the reason why certain polls are skimp on the number of people they question.

The most conscientious pollsters invest a considerable time and efforts to find how to model their voter samples-that is to say how to best reflect the possible composition of the electorate. Once they have completed their interviews, they lay the result to see that it includes the appropriate part of men and women, young and old, and other criteria based on census data.

Then, the pollsters could adjust these results to correspond to the percentage of each group which, according to them, will take place for a given election.

More people interviews polls, the greater the probability of carrying out a representative sample.

This is why the duration of an investigation is also something to consider. The longer a survey is carried out – or in the field, as we say in the company – the more the chances of reflecting the possible participation rate.

It is also important in a polyglot state like California that a survey is not carried out only in English. To do this, he risks a significant part of the electorate to subtitia; A lack of mastery of English should not be confused with a lack of political commitment.

“There is no requirement that a person can speak English to vote,” said Keeter, from Pew Research Center. “And in the case of certain populations, in particular immigrant groups, who have been in the United States for a long time, they can be very well established voters but are still not competent in English in terms of the management of an investigation.”

It is also important to know how a question of survey is formulated and, in the case of a voting measure, how it describes the question of voters who are invited to decide. How much does the investigation follow the language of ballots? Are there any biases introduced into the survey? (“Do you support this measure knowing that his supporters abuse small animals and promoted gum disease?”)

Something else to monitor: has the survey led by a political party, or for a candidate or a group pushing a particular program? If so, be very skeptical. They have every reason to issue selective or unilateral results.

Transparency is the key. A good sounder will show his work, as they said in class. If they don’t, there are good reasons to question their conclusions, well you should.

A sensible person would not put anything in their body without being 100% certain of its content. Treat your brain with the same care.