We don’t know if you have seen the following headline that is making the mostly conservative digital media, but a new Zogby poll published this on Tuesday: “Trump approval at 45%; Support among Hispanics, western and union voters jumps double digits.”
“Hispanics” in the headline caught our attention, so we read what Zogby said in the post about tis finding:
Trump saw one the biggest improvements among any sub-group with Hispanics; his approval increased 11% to 42% approval versus 55% who disapprove of Trump’s job as president.
Our initial question, as with most of these polls that very likely don’t include a large sample of U.S. Latinos, was: how many “Hispanics” got polled? So we emailed Jonathan Zogby, the company’s CEO, about the sample size. This is what we asked Zogby: “Good evening, this is the editorial team at LatinoRebels.com. We are doing a story about Trump’s approval ratings going up with Hispanics and we are wondering if you can let us know how many Hispanics were included in the survey? Thank you.”
Here is what Zagby told us:
(Email sent to Latino Rebels)
Following Zogby’s confirmation that 143 Latino respondents were included in the poll, it would mean that 60 respondents would represent the 42% of the 143 who gave Trump a favorable rating in this latest poll. We don’t think that this is a representative sample of anything and we are hoping that Zagby sends us more demographic information about these 60 respondents (we asked, but we haven’t heard yet.)
In addition, the sample is so small that the Latino portion of the poll has a high margin of error. For instance, if you use the following tool and enter 58,600,000 (the estimated number of Latinos living in the U.S. in 2017) along with the 143 sample size from Zogby’s poll, you would get an 8.2% margin or error. In other words, Trump’s approval rating could be as high as 50% with U.S. Latinos (we doubt that) or as low as 32% (more likely).
From American Research Group’s public calculator for margin of error.
