How Trump's false claim about African American support happened
Excerpts to illustrate some statistical ideas

Original story published in CNNMoney by Brian Stelter May 6, 2018.

On May 2, The Daily Caller published a story titled "Black Male Approval For Trump Doubles In One Week."

The Daily Caller cited Reuters/Ipsos polling. On April 22, the weekly tracking poll had "Trump's approval rating among black men at 11 percent, while the same poll on April 29, 2018, pegged the approval rating at 22 percent."

The story pointed out that "Reuters only sampled slightly under 200 black males each week." But it didn't explain why that would matter when interpreting the results. And in any case, that nuance was quickly lost as the headline was shared tens of thousands of times on social media.

Charlie Kirk of Turning Point USA copy and pasted part of The Daily Caller story and tweeted, "The defection from the Democrat party is happening and they are moving to Trump."

The buzz continued on May 3. Liz Wheeler, a host on a small conservative cable channel called OANN, asked if the supposed bump in support was "the Kanye effect."

Normally you wouldn't be reading about the poll here, because the Reuters/Ipsos tracking poll does not meet CNN's standards for reporting.

CNN's director of polling and election analytics Jennifer Agiesta explained why: "It was conducted using a non-probability online sample, meaning that those who participated signed up to take the poll rather than being randomly selected." Bottom line: This means "there could be bias in the sample."

"The sample sizes for those two measurements were too small to reliably suggest any shift in public opinion," Kahn told CNNMoney through a spokeswoman.

This is common sense: The more narrowly you slice the data, the less reliable it becomes. Reuters and Ipsos survey about 350 people a day, 11,000 a month, on the internet. The outlets call the relative precision of the results as a "credibility interval."

In this case, the president's approval rating among black men, "the credibility interval was more than +/- 9 percentage points for each measurement, which leaves open the possibility that his approval also could have dropped in this time frame," Kahn said. It could have dropped!