Sen. Ted Cruz misrepresented an academic study when he claimed that the "overwhelming majority of violent criminals are Democrats."
The claim is based on research that found a majority of ex-felons in three states registered as Democrats. But that was a study of all ex-felons, not just violent criminals, as Cruz framed it. Also, a follow-up study of three additional states by the same authors found the majority of ex-felons in those states were neither Republican nor Democrat.
Another study estimated ex-felons are more likely to vote Democratic than Republican, but the authors caution that that doesn't mean criminality is associated with partisanship. Rather, it is a reflection that those in the criminal justice system are more likely to be black, Latino or poor whites.
Cruz's comment came during an interview with Hugh Hewitt, who referenced the recent Planned Parenthood shooting and said that despite frequently participating in "pro-life events" he had "never met, not once, a single pro-life activist who is in favor of violence of any sort. Have you, Senator Cruz?"
We reached out to the Cruz campaign for backup and did not hear back, but in response to a similar query from CNN, the Cruz campaign cited research published in January 2014 in The Annals of the American Academy of Political and Social Science and written by Marc Meredith, an associate professor in the University of Pennsylvania's Department of Political Science, and Michael Morse, a Ph.D. student in the Department of Government at Harvard University, who once worked as an undergraduate fellow at FactCheck.org.
Their research into whether voting rights notification laws increase ex-felons' voter turnout found that in three states — New York, New Mexico and North Carolina — a majority of ex-felons registered post-incarceration as Democrats. The breakdown was as follows: New York, 61.5 percent Democrat, 9 percent Republican; New Mexico: 51.9 percent Democrat, 18.9 percent Republican; and North Carolina, 54.6 percent Democrat, 10.2 percent Republican.
But that's far from the full story.
"Cruz is misinterpreting our research," Morse told us via email. "We only calculate the registration rates by party for discharged felons, and do not break this down by type of crime … So our work cannot speak to the partisanship of 'violent criminals.' "
Moreover, the two men published another paper in March 2014 with data from three other states — Iowa, Maine and Rhode Island — and the results were different. In all three of those states, a plurality was affiliated with no party (though in every case, there were far more registered as Democrats than Republicans).
"Ex-felons' partisan affiliations vary across states and I don't think there's enough evidence to claim that the national ex-felon population is 'overwhelmingly Democratic,' at least in terms of party registration," Morse said. "There is more evidence that ex-felons are not very supportive of the Republican Party."
Morse pointed to the book "Locked Out: Felon Disenfranchisement and American Democracy" — written by Jeff Manza of New York University and Christopher Uggen of the University of Minnesota — which he said estimated that about 73 percent of ex-felons would vote Democratic.
Manza told us via email that their research, based mainly on data from Oregon and Minnesota, was an estimate that relied on matching the felon population with "all the demographic characteristics we could to estimate the partisanship of similar people in the general population (that is, an average voter with the same education, race, gender, marital status, etc. of the felon population). It is an indirect method of trying to figure out how felons and ex-felons might have voted (and how many of them might have voted if they were eligible)."
Indeed, Manza told us that he is aware of no direct survey of the felon population that asked about partisanship, meaning Cruz "has no direct evidence for what he said."
More broadly, Uggen told us in a phone interview that Cruz's statement is "quite a stretch" of the research on the subject.
"I think the main things that are responsible for that correlation (between criminality and Democratic party affiliation) are race and poverty," Uggen said. "If you look closely, it isn't criminality that is so clearly associated with partisanship. We know very clearly that the criminal justice system falls more heavily on people of color in urban areas and people who are farther down the rungs of the socioeconomic ladder."
Uggen noted that their research focused mainly on Oregon and Minnesota and so, "I can't really speak to the national picture with confidence." However, he said, "it is certainly the case (among ex-felons) that there are likely more individuals who would identify as Democrats rather than would identify as Republicans."
Part of what their study is picking up is the race effect, he said. According to the Sentencing Project, 38 percent of people in state or federal prisons in 2011 were black, though black people make up just 13 percent of the overall U.S. population. According to the Pew Research Center, about 80 percent of the black population in the U.S. aligns with the Democratic Party.
Cruz also said that any time there is a violent crime or mass killing "you can almost see the media salivating, hoping, hoping desperately that the murderer happens to be a Republican." That's an opinion, but we would note that the media generally focuses on the political leanings of a mass shooter when the possibility of political motivation arises. For example, it was widely reported that Jared Lee Loughner, who shot and killed 6 people and seriously wounded Democratic Rep. Gabrielle Giffords, was registered as an independent.
In the Planned Parenthood shooting case, the political leanings of the alleged attacker, Robert L. Dear, received media scrutiny after reports from an unidentified senior law enforcement official that Dear told police "no more baby parts" after his arrest.
Dear's ex-wife told the New York Times that Dear was "generally conservative, but not obsessed with politics" and that "he believed that abortion was wrong, but it was not something that he spoke about much." Colorado voting records listed Dear as unaffiliated. (The voter registration indicated Dear was female, which led some in the blogosphere to speculate that he was a transgender person — reports repeated by Cruz — but that turned out to be simply a clerical error.)
Cruz criticized the media for refusing to report that "the overwhelming majority of violent criminals are Democrats." But the claim is unsupported by research.
The researchers behind the study cited by Cruz's campaign say their research doesn't look at violent criminals — they only looked at ex-felons generally, violent or not. Nor does their ex-felon population represent all criminals, just those who were arrested by police. And finally, they say, they're not even sure that with all those qualifiers it's even true that a majority of ex-felons go on to register as Democrats. That's true in some states studied, but not others, and no national data is available.
Cruz was on firmer ground when he went on to say that Democrats "go in and fight to give the right to vote to convicted felons. Why? Because the Democrats know convicted felons tend to vote Democrat." There is evidence from the studies that ex-felons are more likely to register as Democrats than Republicans when their voting rights are restored.
FactCheck.org is a nonpartisan, nonprofit "consumer advocate" for voters that aims to reduce the level of deception and confusion in U.S. politics. Based in Philadelphia, Factcheck monitors the factual accuracy of what is said by major U.S. political players in the form of TV ads, debates, speeches, interviews and news releases. Its goal is to apply the best practices of both journalism and scholarship, and to increase public knowledge and understanding. Find a list of Factcheck.org funders here.