Wisconsin judges can weigh a risk assessment test that considers gender when sentencing offenders but can't use it as the sole factor in determining a sentence's severity, the state Supreme Court ruled Wednesday.
The court said COMPAS reports can be used in Wisconsin but warned the assessment comes with limitations. Judges can't use it to determine a sentence's severity and should explain the report is only one of many factors they consider during sentencing, the court said.
The ruling stems from a case involving Eric Loomis, who was charged in 2013 with multiple counts in connection with a drive-by shooting in La Crosse. He ultimately pleaded no contest to stealing a car and guilty to fleeing an officer.
A pre-sentence investigation into Loomis' background included a COMPAS report, a recidivism risk assessment developed by Northpointe, Inc. The report is based on information taken from a defendant's criminal file and an interview and predicts the recidivism risk by comparing the defendant to a similar data group.
Loomis' COMPAS report found he presented a high risk of recidivism. Judge Scott Horne referenced the report when he sentenced him to six years in prison.
Loomis argued defendants can't challenge the reports' accuracy because Northpointe considers its methodology a trade secret. He also contended that since the reports consider a defendant's gender they violate the due process right not to be sentenced based on sex.
The 4th District Court of Appeals kicked Loomis' case directly to the state Supreme Court. The high court ruled unanimously that the report is a valid tool in sentencing.
The court noted, however, that studies on COMPAS accuracy in other states show shortcomings. The justices ordered that any pre-sentence report containing a COMPAS report must come with caveats. The master reports must inform the judge that some studies have raised questions about whether COMPAS disproportionately classifies minorities as higher recidivism risks and no study on COMPAS has been in done in Wisconsin yet, the court said.
As for the gender argument, the court said both Loomis and state attorneys agree that men on average re-offend more often. Therefore, considering a defendant's sex promotes accuracy, the court found. What's more, Loomis failed to show that the judge relied on his gender as a sentencing factor, the court said. Horne explained that Loomis' sentence would have been the same without the COMPAS report, the court added.
Loomis' attorney didn't immediately respond to a voicemail seeking comment.