Dissecting Dueling IGS & PPIC Polls on Prince Gavin


A few months ago, when the Coronavirus pandemic was just getting real, a majority of Californians thought Gov. Gavin Newsom was doing a good job of managing the crisis.

Not so much anymore.

Today, only 31% of California voters give Newsom good or excellent marks for handling the “coronavirus pandemic” in a survey by Berkeley’s Institute of Governmental Studies, compared to 43% who give him poor or very poor marks.

And in the poll by the Public Policy Institute of California, registered voters are split almost equally with 49% approving and 46% disapproving Newsom’s handling of the “coronavirus outbreak.”

Which is why Newsom’s overall approval rating has dipped to 46% among registered voters in the IGS Poll and 50% among RVs in the survey from PPIC.

Much has been made of the difference between the findings of the two polls in part because IGS gave results for registered voters (46%) while PPIC focused on results from likely voters (52%).

But if you compare apples to apples – registered voters – the two polls are only about 4 percentage points apart on Newsom’s overall performance.

Bottom line, voters are not particularly happy with Newsom these days and his handling of the coronavirus pandemic/outbreak is the main reason. (Also: rank partisanship abides).  

What it means for recall

All of which suggests the drive to recall Newsom is likely to get on the ballot because Republicans hate his guts.

However, Prince Gavin is likely to survive a recall because Republican partisans represent only a small portion of California voters.

Our old friend George Skelton at the LA Times, along with Ben Christopher at CalMatters and Jeremy White at Politico have already done some fine reporting and commentary on Newsom’s standing in the polls and his chances in a recall election. They rely somewhat on analyses from Mark DiCamillo and Co. at IGS and Mark Baldassare at PPIC.

IGS said, in part:

Fueling the decline is the public’s much more negative assessment of the way Newsom and state government are handling the pandemic. T

he latest poll finds fewer than one in three Californians (31%) rating Newsom as doing an excellent or good job in handling the pandemic overall,down from 49% last September.

Also, just 22% offer a positive rating of the job he and state government are doing in overseeing the distribution of the coronavirus vaccines to the public.

In addition, only about half (47%) have a great deal or some trust in the way the governor and state government are setting the rules when issuing stay-at-home orders or setting  guidelines for business to follow to slow the spread of the virus, with majorities describing them as inconsistent (62%), confusing (60%) and ineffective (53%).

While PPIC said:

Assuming there will be a governor’s recall election in 2021, the political wildcard is the status of COVID-19 in California.

In the January PPIC Survey, about half of likely voters say that COVID-19 is the most important issue for the governor and legislature to work on in 2021.

Currently, Governor Newsom has mixed reviews for his handling of this issue (50% approve, 47% disapprove). And less than three in ten give the state government an excellent or good rating for its handling of the COVID-19 vaccine distribution. In contrast, seven in ten approve of the way that the pandemic is being handled by Joe Biden in his early days of presidential leadership.

Deep in the weeds

So, what are the key differences between these two surveys, much deconstructed and dissected in political circles this week, both produced by very fine pollsters?

(For what it’s worth, the dispositive 538 pollster ratings show that PPIC has about the same high rating as the now defunct Field Poll had, but enjoys a much better rating than Berkeley.)

PPIC uses live interviews and random digit dialing, which gives every telephone exchange (land line and cell phone) an equal chance of being surveyed, and they use an established method to get someone from the household on the phone.

This is classic polling methodology, developed over decades to ensure that a random sample of the population is surveyed.

But it has become increasingly more difficult (and expensive) to get people to take phone surveys.

PPIC’s response rate in its latest survey was about 5% for landline calls and 3% for cellphones (although for those in the sample who had participated in a prior survey it was higher – 44% for land lines and 25% for cells).

In addition, pollsters determine who is and who is not a registered voter (or a likely voter) by asking questions that respondents may or may not answer truthfully like:

Are you registered to vote? What party are you registered in? How much are you following the news? Do you plan to vote the next election?

Every pollster has their own combination of questions to try to make sure they distinguish who is and who is not a voter, but it’s as much art as it is science.

In addition, some people are so suspicious of authorities, institutions and researchers they won’t ever participate in a survey, even when the call is from the Public Policy Institute of California and not some “partisan” or egghead caller like, say, the University of California Berkeley.

IGS, on the other hand, can’t afford the huge cost of live surveys (for the most part) and so has turned to online polling, with sophisticated abilities to target and engage voters in the actual voter file from the California Secretary of State.

But they have to invite about 190,000 voters to participate to get a sample of 10,000 voters. And they can only invite voters who have listed an email address, which is now about half the registered voters.

So, their effective response rate is also about 5%, and it’s a pre-screened group of people who have listed an email address and will go to the trouble of filling out the survey online.

They have the same problem that live calling has of Trumpistas, who don’t want to participate at all so their views can’t even be given weight to represent their share of the population.

Moreover, online surveys can’t claim to represent a random sample of voters because not every voter has an equal chance of being surveyed.

So, they have to construct a representative sample based on sophisticated use of variables like age, gender, location, education, etc.

What this method does know for sure is that they’re dealing with actual voters and even how often they have voted, all of which is in the voter file.

There are many more issues that confront pollsters using either method. Needless to say, polling has become a hugely difficult endeavor which the most recent presidential campaign revealed starkly.

But this much we know for certain: Gov. Gavin Newsom is facing voters who no longer thinks he’s the bee’s knees.

And the only way he can guarantee survival in office is to make people believe that California is doing a better job of containing the coronavirus, distributing vaccine, opening schools and businesses and slowing the death toll.

subscribe to comments RSS

There are no comments for this post

Please, feel free to post your own comment

You must be logged in to post a comment.