Good Statistics

January 29th, 2007

This morning shayna and i had a discussion on child abuse. I’m a bigger fan of statistics than she is, and it led to a brief tangential discussion of good statistics vs bad statistics. Yes, sometimes statistics are a bad thing. I define good statistics are ones that are fact based, complete, and useful (i should prolly work on that definition). Bad statistics are ones that are incomplete, contain subjective elements, and/or could be used to support either side of the same argument – you gotta take ’em with a grain of salt. For example, “4 out of 5 dentists recommend crest”. That is bad because it is incomplete – we don’t know how many and what type of dentists were asked – they could have asked only dentists that are employed by crest. We assume that alot of dentists were asked, and because we have to assume it is incomplete and therefor a bad statistic. And statisics aren’t always bad, but they could be better. Both of the following are from 2006 Statistics – Which is better and why?

  • 36.7% of all women in prison and 14.4% of all men in prison in the US were abused as children.
  • More than a third of women in the nation’s prisons and jails reported abuse as children, compared with 12% to 17% for women in the general population. About 14% of male inmates reported abuse as children, compared with 5% to 8% of men in the general population.

I would say the second, because it puts it in context. That is, if percentage in jails and out of jails were the same, it’s not as noteworthy. But it’s 2-3 times greater for those in prison. Makes sense. However, the second doesn’t mention US, which it should, and actually is the case. Here are some more interesting statistics from the US in 2004 .

  • An estimated 1,490 children died due to child abuse or neglect (src).
  • In the US, 12 out of every 1,000 children (age 0-17) were known victims of child abuse (1.2%). For children 3 and under, its 16 out of 1,000 (src).
  • There were 868,000 known victims total out of about 73 million children total (age 0-17). Florida had the higest rate of about 32  victims per 1,000 children (130,000 victims of 4 million florida children) (src).
  • When looking at each race, and comparing only child victims in that race to other children in that race: (src)
    • African-American – 19.9 victims per 1,000 children
    • Pacific Islander – 17.6 victims per 1,000 children
    • American Indian or Alaska Native – 15.5 victims per 1,000 children
    • White – 10.7 victims per 1,000 children
    • Hispanic – 10.4 victims per 1,000 children
    • Asian – 2.9 victims per 1,000 children
  • 83% of victims were abused by a parent acting alone or with another person (src g1 g2).
  • 56% of alleged child abuse reports were made by professionals (educators, law enforcement and legal, social services, medical, mental health, child daycare, and foster care). 44% of reports were submitted by nonprofessionals (including friends, neighbors, relatives, etc.) (src).

However, i sadly could not find info linking child abuse to economic status (but race was included), which my gut tells me might matter, but i could totally be wrong. I actually found a great webpage by Jim Hopper that discusses child abuse statistics (how i found the above), with an emphasis on taking statistics with a grain of salt, which is really the point of this blog. He also points out that many child abuse cases go unreported, its hard to tell how many. He suggests a few books to read regarding statistics:

“Again, widespread uncritical faith in statistics is historically fairly recent. And it causes significant confusion – among members of the media, politicians, judges, and advocates for various causes, not to mention average citizens. Therefore, having tools for thinking critically about statistical findings reported in the media (and on the web) will help you better understand a variety of important issues, not just child abuse. Two good, recently published books can help you cut through the confusion and hype that surround most presentations of statistical and scientific findings in the popular media … It Ain’t Necessarily So .. and Damned Lies and Statistics ..” – http://www.jimhopper.com/abstats/#s-statistics