How reliable are the Justice Department’s stats on rape and sexual assault? According to a new report from the National Research Council (NRC), not very. The survey, released Tuesday, reveals that rape is vastly undercounted in America and suggests ways to improve the measurement tools.

The Justice Department uses the National Crime Victimization Survey (NCVS), which indicates that the rate of completed and attempted rape in the U.S. has gone down from a high of 5 percent of girls and women victimized in 1995 to a low of 2 percent from 2005 to now. That would be good news, if true. But the NRC found that a survey by the Centers for Disease Control and Prevention estimated 1.2 million victims of rape and attempted rape in 2010, which is significantly higher than the 188,000 counted by the NCVS.

What’s behind the large disparity? Slate’s Emily Bazelon writes that one cause might be that the NRC survey asks better, more targeted questions. The NCVS also doesn’t include questions about whether a person was unable to consent to sex because he or she was “drunk, high, drugged or passed out.” Having accurate numbers is critical to establishing effective approaches to rape prevention.

So what should be done to ensure more accuracy? Another complicating issue is that rape is one of the most underreported crimes in the world. So what more can be done to get victims to report crimes of sexual violence more consistently?

Guests:
Emily Bazelon, is a Senior Editor for Slate and the Truman Capote Fellow at Yale Law School; Author of Sticks and Stones: Defeating the Culture of Bullying and Rediscovering the Power of Character and Empathy (Random House 2013). She tweets @emilybazelon

Christopher Krebs, senior researcher at RTI International, an institute that provides research and technical services to governments and businesses