Half-arsed reporting

Here is a stupid headline from the NZ Herald;

More US kids report abuse

Nearly half of United States children who've been victims of crime or abuse are stepping up to tell school officials, doctors and police, according to a 2008 telephone survey of more than 4500 US children and teens. That compares with 25 per cent in a similar study done in 1992.

That's it. Two sentences.

This leaves more questions than answers the central being, how many of the 4,500 were victims?

According to the National Incidence Study of Child Abuse and Neglect, overall abuse is down on the 1990s. The data reported on is gathered from various state Child Protection Services (CPS). NIS 4 was published this year. Excerpt:

The number of children who experienced Harm Standard abuse declined significantly, by 26%, from an estimated 743,200 in the NIS–3 to 553,300 in the NIS–4. This reflects a 32% decrease in the rate of Harm Standard abuse per 1,000 children in the nation. Moreover, the incidence of all specific categories of abuse decreased: The incidence of sexual abuse decreased significantly, while the declines in physical abuse and emotional abuse were both close-to-significant.

What the survey reported in the Herald shows is that of the children experiencing a crime or abuse, more are reporting it.

To illustrate my point, let's say 100 children were abused in 1992 and 25 (quarter) of them reported it. In 2008 only 68 children were abused but 34 (half) of them reported it.

More children reported abuse but fewer children were abused.

Yet the headline, without greater context, leaves the distinct impression that child abuse in the US is getting worse. The alternative headline may very well have been that

Fewer US children abused

(Although the half-arsed type would probably say, Less US children abused, if bothering).

And yes I could go hunting for the survey myself but that is irrelevant to the point I am making about half-arsed reporting.