Raynella Dossett-Leath’s recent trial for murder of her second husband (which ended in a mistrial) prompted numerous comments on the frequency of spouse-murderers. TV commentators, in particular, like to point out how often husbands kill wives and wives kill husbands. But is it true that most murders are committed by spouses and significant others?

Well, not according to the Bureau of Justice Statistics, as of July 11, 2007. From 1976 to 2005, only 7% of murders were committed by the victim’s spouse. Only 3.8% were committed by “significant others.” I’m sure you can do the math: that’s slightly less than 11% of all murders. Let’s put this another way: almost 90% of all murders are NOT committed by the spouse. Clearly, if police studied statistics they would NOT suspect the spouse or significant other first.

As an example, take a look at this story from Beaver Dam, WI: a husband who has been under suspicion of murdering his wife since 1980 was just exonerated. His story is pathetic.

Eight percent of all murders are committed by other family members (other than spouses). That means that parents, siblings, cousins, and in-laws are slightly more likely than spouses to kill a family member, and more than twice as likely as significant others to kill a family member. In the Dossett-Leath trial, if police had used statistics to identify a near relative other than the spouse who might have killed Mr. Leath, they would have looked first at family members who had a motive to kill for the inheritance, for example.

Unfortunately for the pundits who point the finger first at intimate partners, 32.1% (almost a full third) of murders are committed by mere acquaintances. I suppose acquaintances might include friends of a murder victim’s spouse, such as jealous girlfriends. That would mean that statistically Amber Frey would have been a far more likely prime suspect in the murder of Laci Peterson than her husband Scott. Statistically, Ms. Frey is four to five times as likely to have been the murderer as the spouse.

Strangers commit 13.9% of all murders. That’s twice as many stranger murders as spouse murders.

And 35.2% of murders are committed by unknown persons. In other words, over a third of all murders go unsolved.

Dare I suggest that these unsolved murders are more likely to be stranger murders or mere acquaintance murders (including one-night stands) than anything else? Since the cops seem to look first to spouses and family members, it seems to me that such guilty family members would be easy to catch. The more distant the relationship, the more difficult it is to find a motive or connection between a victim and a murderer. In addition, I suspect that drugs and prostitution are implicated in many unsolved murders. And, I’m convinced that psychologists grossly underestimate the number of psychopathic serial killers out there. (I have some research data on this topic which I’ll use for a post soon.)

But remember, statistics like these cannot be used in court as evidence. Statistics aren’t evidence that any given person committed murder. Statistics are collected for other purposes. It might help, though, if investigators would study statistics when they’re trying to solve murders. If they didn’t always erroneously jump to the conclusion that the spouse did “it,” maybe they would explore other possibilities more aggressively in the early phases of an investigation when the “trail” is still “hot.”

I can think of one especially tragic case in which the cops might have saved a kidnapped child’s life if they had not assumed the child’s father was whodunit: the pathetic case of Jessica Lundsford.

My point: statistics are misunderstood by most lawyers and law enforcement officials. And I can’t figure out why. The Bureau of Justice Statistics has an easy-to-use website. I recommend it highly.