DaytonDailyNews.com

By Russell Carollo and Elliot Jaspin
rcarollo@DaytonDailyNews.com and ejaspin@coxnews.com

For more than a decade, the Peace Corps has offered an incomplete picture of violence against its volunteers.

The agency, which didn’t collect crime data for analysis during its first 28 years, omits many victims from its published crime statistics and fails to track some serious crimes at all.

One problem is the way the agency calculates crime rates. Instead of counting victims of crimes, the agency counts "incidents" and then compares that number to the amount of time volunteers spend in a country — ignoring that many incidents have multiple victims. The effect is a rate that is artificially low.

In 1996, three female volunteers were raped and two male volunteers were assaulted by a group of armed men in El Salvador, but the event was recorded as a single rape "incident" in the agency's Assault Notification and Surveillance System database.

Barbara Daly, the Peace Corps' press secretary, said the database was not designed to measure crimes against volunteers.

"We count all people involved in an event as one event," she says in a written response to the Dayton Daily News. "We do this, not to

count crimes against individuals or even to count crime events, but to track trends for purposes of prevention. This is the purpose of the database. This was why it was set up and any other purpose for the database misuses and misrepresents the database."

Dr. A. Russell Gerber, the Peace Corps' chief of epidemiology, said the agency uses its system to identify when and where crimes occur. He said the system has helped the agency reduce the number of major sexual assaults in recent years.

An analysis of the agency's data shows sexual assaults dropped from 113 in 2001 to 94 in 2002. But the 2002 total was still more than double the number of reported sexual assaults in 1991. The data for all assaults shows a 125 percent increase during that span.

However, the actual number of crimes against volunteers is likely much higher. According to the newspaper's analysis, 21 percent of the "incidents" in the Peace Corps' assault database involved more than one volunteer. The database provides enough description to show that a number of those other volunteers were crime victims, too, but the database doesn't describe exactly what happened to all of them.

More than 17,000 law enforcement agencies in the United States participate in a data collection system coordinated through the FBI, which uses a method different from that of the Peace Corps. All victims of so-called "crimes against persons," such as murder, aggravated assault and rape, are included in the FBI's crime statistics. Like the Peace Corps, the FBI doesn't count people victimized of lessor crimes in the single incident, though a new system being developed by the FBI counts all victims in a single incident.

The 1996 incident in El Salvador, counted as one rape by the Peace Corps, would count as three rapes under the FBI system.

Another problem with the Peace Corps system is that it doesn't collect information on certain crimes, including kidnappings and abductions, nor does it track arrests or convictions.

Still another problem is underreporting. Volunteers have unique reasons for not reporting crime, including fear of being moved from their sites. The agency’s own auditors and Michael O'Neill, its former security chief, estimate underreporting as high as 50 percent. A 2002 report from the General Accounting Office, Congress' investigative arm, found that Peace Corps volunteers did not report 60 percent of rapes and 20 percent of other sexual assaults in 1998.

Even crimes that are reported are not necessarily counted in statistics. Mike Mulvaney and a second volunteer were abducted at gunpoint in Bolivia in January 2001. The country director at the time, Meredith Smith, confirmed that Mulvaney reported the incident to her, and O'Neill confirmed that he spoke to Mulvaney about the abduction. But the incident doesn't appear on the agency's assault database.

O'Neill said Mulvaney should have reported the incident to the medical office. A written response from Peace Corps says the agency has no record the assault was reported.

Peace Corps officials said they are considering changes to the way they track crime, including counting all victims and tracking arrests and prosecutions.

Staff Writer Mei-Ling Hopgood contributed to this report.

[From the Dayton Daily News: 10.26.2003]