The researchers whose study Ted Cruz's campaign cited to back up his claim that "the overwhelming majority of violent criminals are Democrats" said Tuesday night the Texas senator is "misrepresenting" their work.
"We have not explored partisanship by type of crime, violent or otherwise, and so our work, as published, cannot support Senator Cruz’s claim," University of Pennsylvania professor Marc Meredith and Harvard doctoral student Michael Morse said in a response prepared to appear on The Washington Post's Monkey Cage blog.
Cruz made his comments Monday when asked by radio host Hugh Hewitt about the media's response to the shooting at a Colorado Planned Parenthood over the Thanksgiving weekend.
“You know, every time you have some sort of violent crime or mass killing, you can almost see the media salivating, hoping, hoping desperately that the murderer happens to be a Republican so they can use it to try to paint their political enemies,” Cruz said, before making his claim about Democrats representing the majority of violent criminals.
When pressed for evidence of his assertion, Cruz's campaign pointed CNN to research from Meredith and Morse suggesting that ex-felons register to vote as Democrats.
But Meredith and Morse said Cruz oversimplified their research.
For starters, the author's 2013 paper was about ex-felons in general, not violent criminals specifically. In other words, they studied the voter registration behavior of convicted felons who had been released from prison.
Politifact rated Cruz's claim "mostly false" Tuesday.
"Only a small share of ex-felons are convicted of violent crimes," Meredith and Morse said, surfacing data from Iowa in their response to Cruz showing that only about 12 percent of the population of discharged felons in the state had been convicted of committing violent crimes.
"Senator Cruz is correct that people convicted of a violent crime are more likely to identify as Democrats than as Republicans," the researchers said. "But people convicted of violent crimes have similar political orientations as nonviolent offenders."
"Senator Cruz was on stronger footing when, later in the interview, he dropped his focus on violent crimes and claimed that, instead, 'convicted felons tend to vote Democrat,"' the researchers continued.
But there's also statistical problem with Cruz's extrapolation of their research. Meredith and Morse's study investigates voting registration behavior in just three states — New Mexico, New York and North Carolina, which recently passed laws requiring ex-felons to be notified of their right to vote.
And although they did find that of the small percentage of ex-felons who registered to vote in those states, more registered as Democrats; in another paper in which they examined Iowa, Maine and Rhode Island, they found that a plurality of ex-felons registered as independents or with other parties.
They argue that ex-felons registering with the Democratic party might represent a demographic correlation more than it does any kind of causal link between being a felon and being a Democrat.
As Meredith and Morse point out, "African-Americans are both overrepresented in the ex-felon population and have some of the strongest allegiances to the Democratic Party."
Pointing to data in Florida, Morse and Meredith explain that blacks "overwhelmingly identify as Democrats whether or not they are ex-felons." And a majority of disenfranchised ex-felons in Florida are non-black, and a significant proportion of them register as Republicans.
The researchers' point, then, is that "in stark contrast to Senator Cruz’s comments," many Republicans are also disenfranchised by state laws that do not allow ex-felons to vote.