Reporting crime

The percentages weren’t wrong. But they needed more explanation.

When the Bangor Daily News reported recently that Washington County had the best crime-solving record in the state and Penobscot County had the worst, we could have done a better job providing context.

That’s because percentages are easily altered by small data pools, and there is room for inconsistency between police departments in declaring whether a case is cleared or not. It’s also more accurate to look at trends over time, not a single year.

The Uniform Crime Reporting statistics show that the Washington County Sheriff’s Office resolved 66 percent of crimes in 2010, and the Penobscot County Sheriff’s Office resolved only 9 percent. That’s a jaw-dropping statistic until you look closer.

First, the Uniform Crime Reporting Program does not track everything reported to police, as not all crimes occur often enough to be meaningful in an index. It tracks the most serious, common crimes: murder, rape, robbery, aggravated assault, burglary, theft, motor vehicle theft and arson.

The Penobscot sheriff’s office got about 17,477 calls in 2010, but the index only tracked about 740 cases. Still, the 740 indexed Penobscot crimes represent nearly four times the number in Washington County.

And Penobscot deputies encountered a greater percentage of crimes that are difficult to solve, regardless of what police force you work for. (A case is typically cleared when an offender is charged and taken into custody.) Those difficult-to-solve crimes include burglary, theft and motor vehicle theft.

Consider this: If a teenager opens 17 unlocked car doors and steals extra change at night when no one is around, and police don’t find the culprit, that goes on the department’s record as 17 unsolved crimes. But if police solve a rape case, that counts as only one solved crime.

These factors played into the Penobscot sheriff’s office clearance numbers, according to Sheriff Glenn Ross. But it turns out the department also needs to correct its recording process.

For example, let’s say a man is arrested on charges of burglary and theft, but he reaches a court agreement to plead guilty to the burglary charge and have the theft charge dismissed. The police department might then claim the burglary but not mark the theft as cleared.

Or, if someone breaks into a camp but doesn’t steal anything, it should be marked as criminal trespassing. But the crime might stay in the record system as a burglary because that’s what the property owner first reported.

It’s similar to another example of when people call police to tell them they had their narcotics stolen, but they’re really saying it to get more pills; there was no crime. The Penobscot sheriff’s office marks the call as inactive, which works against it because it means no one found out who committed the act. The code should be changed to unfounded, Ross said, because the crime didn’t occur.

It comes down to this: Many cases are solved or connected throughout the year, but the department doesn’t go back and change the codes.

And there’s a reason why it doesn’t. The department switched to an electronic system a few years ago that does not streamline the process.

Before, deputies wrote out manually on a form whether the case was cleared by arrest. Now that they submit the records electronically they don’t have that option. Ross said the department is examining the system to see how to correct it, and he will look at every case handled in 2011 to make sure its status is updated.

Anyone can look and see: from 1995 to 2002, the Penobscot sheriff’s office had yearly clearance rates of 16 to 30 percent, while the Washington sheriff’s office had similar yearly rates of between 16 and 33 percent (The national average is around 22 percent).

In 2003, though, Penobscot’s rate plummeted to about 10 percent and stayed constant through 2010. Were deputies doing a poor job of making arrests? No. Their reporting could have been more thorough. As could ours.