A PAC known as "Fight Back California" commissioned a new poll of the 49th district for the period March 4-7. Change Research was the pollster. The results are very questionable, as we'll see below.
Like FM3, Change Research did an initial ballot test, putting the names of the potential candidates in front of likely voters and asking them which candidate would get their vote. The results from the simple ballot test are mostly consistent with what we've seen before. We say "mostly consistent" because relative to the mid-February polls, Chavez is polling a bit lower, Harkey is lower, Kerr has picked up steam (from low single digits to high single digits), and Gaspar has faded:
So far, so good, right? Well, things go seriously sideways from here in a way that casts serious doubt on the whole poll.
What got reported out to the world by "Fight Back California" from this poll was a table of results that bears no recognition to the chart above. The reason for the difference? The pollsters fed likely voters "information" about the candidates and asked them again how they would vote. You can do this sort of thing the right way... or the wrong way. The right way is how FM3 did their testing (see the FM3 poll results in the post before this one to get the full picture). The wrong way is to bias the information for or against a candidate or candidates, manipulating the results to get a certain outcome.
Just like the FM3 poll, Change Research followed up the simple ballot test by asking likely voters how they'd vote after they heard some information about the candidates.
However, the "Fight Back California"/Change Research results shifted in a really big way after "information" was given. This didn't happen with the FM3 poll, where support moved by a few points at most (either way) depending on the candidate.
What makes us quite skeptical of the results is that ALL candidates in the Change poll benefited from the "information" given to the likely voters except for Doug Applegate. After "information", Sara Jacobs is suddenly at 15% from 7%. Mike Levin is suddenly at 16% from 11%. Kerr got a two point lift. Chavez and Harkey got moderate lifts as well, four points for Chavez and five for Harkey. Poor Doug Applegate saw his support cut almost in half after the "information" was given, from 23% to 13%. Here is how the results look after this "information" was presented to the likely voters:
These results don't look credible to us. We're agnostic about individual candidates but we don't believe there is actually any new information available anywhere that would cut Doug Applegate's support in half. We think FM3 did a fair job of putting positive and negative information in front of the voters for their poll. We can't say the same here at all. We have to question the methodology and the intentions of the pollsters and the PAC that paid for the poll.
UPDATE: We've been alerted to the fact that Change Research isn't exactly a top-shelf pollster and the poll likely has serious methodology issues. We don't know the demographics of the respondents or their party affiliation (so we don't know if the sample is a true proxy of the 49th district). We don't know what "information" was communicated or how. Doubts have been raised to the point where we believe it's quite possible the poll may have been set up to make one candidate look worse than others. We're going to significantly discount the results. This is troubling because these poll results were included in a San Francisco Chronicle article about Democrats in jungle primaries all over California.
For reference, here is how "Fight Back California" announced their results:
In CA-49, when voters are shown the list of all potential candidates with information about the candidates, Republican Assemblymember Rocky Chavez is in the lead with 19% followed by Democrat Mike Levin at 16% and Republican Board of Equalization member Diane Harkey and Democrat Sara Jacobs tied at 15%. In a narrowed field of candidates, Chavez maintains the lead with 23% followed by Democrat Levin at 17%, Republican Harkey at 16%, and Doug Applegate at 15%.
Here is a link to the post on their web site announcing the results:
We've asked the "Fight Back California" PAC to share their methodology and the information they gave the likely voters to understand what generated the huge swings. We hope they oblige. We suspect they won't.
Voices of the 49th is committed to making sure that voters in the 49th congressional district have a meaningful choice in November... which means engaging, educating, informing, and influencing voters through the jungle primary in June so that a Democrat makes it on to the final ballot.