Jesse Richman, a professor at Old Dominion, retook the stand to begin day six of Fish v. Kobach. Richman and two rebuttal witnesses comprised an extended day in court.
Richman’s expert report provided four estimates ranging from one percent to 28 percent of the total number of noncitizens in Kansas who attempted to register or registered in the state. While Richman based each estimate on a different set of data, he testified that he was most confident about his 1 percent estimate. Richman also mentioned, but did not rely upon, a fifth data set from a survey he conducted with 500 Kansans that produced a zero percent estimate.
Moreover, major questions over the reliability and significance of Richman’s findings were called into question during cross-examination and from the rebuttal witnesses.
The cross-examination revealed Richman’s use of the Cooperative Congressional Election Survey (CCES) to calculate noncitizen registration rates, which had been roundly criticized in an open letter signed by over 200 political scientists. Richman then used this same widely criticized data set to reach his highest range estimate of noncitizens attempting or successfully registering to vote in Kansas.
Secretary Kobach had also previously used Richman’s data to support President Trump’s claim that noncitizens voting entirely explained his loss the popular vote. A video of Kobach making this claim was played in court, followed by testimony from Richman that even his contested analysis does not support that claim.
The creator of the CCES and Harvard professor, Stephen Ansolabehere, also testified in court today to refute Richman’s use of the data. Ansolabehere said the 14 respondents that Richman identified from the CCES as noncitizens attempting or successfully registered to vote in Kansas should be understood as a response error. Ansolabehere also responded to Richman’s other findings saying they provided, “no information on non-citizenship registration in Kansas.” He concluded that all of Richman’s estimates were “not statistically different than zero.”
Eitan Hersh, a professor at Tufts University, also testified today as a rebuttal expert witness to Richman. The Plaintiffs asked Hersh to confirm the matching data provided by the state that identified 127 cases of alleged noncitizens who attempted or succeeded in registering to vote. Hersh was able to confirm 122 who appeared on the database the state uses but only 48 who were successfully registered to vote.
Hersh went on to testify that about 400 people were currently on the Kansas rolls with a birthdate after they supposedly registered to vote; using this example Hersh concluded that the 122 alleged noncitizens should be considered likely administrative errors. Though data cannot extrapolate intent or accident on the part of the alleged noncitizens, Hersh testified that the voter rate of these 122 is at zero or one percent, thus concluding that these instances were likely accidents. To intentionally register as a noncitizen but not vote is like “holding up a bank and not taking any money,” said Hersh.
The trial will resume for its intended last day in court on Monday, March 19.
