Typical media story about outcomes - crime stats increase September 24, 2007
Posted by Paul Duignan in : Outcomes theory & the news , trackbackThe issues dealt with by outcomes theory - which looks at the ways we think, discuss and argue about outcomes - are central to many of the political debates which we read about daily in the media. For instance today (24 September 2007) the Washington Post carries a story with the headline FBI Report: Violent Crime on the Rise. It’s fairly typical of the type of story we see. From an outcomes perspective what are the important elements of these types of stories? Firstly, there’s a claim by the media that there’s some ‘news’ in what they’re reporting - that something is changing. The media need this, basically what they sell is news of change - if nothing changes there’s nothing to tell us about, there’s no ‘news’. In this case the subtitle is: ‘First significant increase in homicides and robberies since 1993 continues.’ Outcome measures can go up and down due to statistical variation, is this change actually significant from a statistical point of view? There’s no way to tell from the story, I presume they’re not referring to statistical significance in the title. If they were they would need to make this clear in the article if we are to be able to have any basis for deciding whether we should regard this as an actual change or not.
If there is an actual change, from an outcomes theory point of view what is being measured here is what is technically called a ‘not-necessarily attributable’ indicator. Its an indicator, a measure of an outcome, but its mere measurement does not say anything about what that change can be attributed to. However, as is usual in stories like this, this story includes an ‘attributional’ claim from a group of stakeholders. This is a claim that the change can be blamed on someone for doing or not doing something. In this case there’s the claim that the Bush Administration has faced ‘growing criticism from congressional Democrats, big-city mayors and police chiefs for presiding over cuts in federal assistance to local law enforcement agencies over the last six year.’ The implication is that these stakeholders are blaming the change in the outcome on the policies of the Bush Administration. Is this attributional claim accurate? Who knows? Without detailed independent analysis of what the changes were, their likely effect and linking these to the changes in the outcome over time, there’s no way of assessing the truth or falsity of such a claim. This is not to say that you could not do so if someone put the work in, just that the mere claim does not establish it.
Thirdly there’s the spin, at the end of the story it runs as follows: “Justice Department spokesman Brian Roehrkasse said in a statement that there is some ‘encouraging news’ in the FBI report but acknowledged that ‘violent crime remains a challenge for some communities’. He said that the overall crime rate - combining violent and property crimes - ‘was the lowest crime rate measured by the UCR [Uniform Crime Reporting Program] in more than 30 years.’” There is plenty of room for such spin because the relationship between the different indicators presented in the story is never made clear. For instance, there’s a paragraph which reads: ‘In addition to the overall number of violent crime reports, the violent crime rate - which measures the number of reported crimes based on population - rose by 1 percent, the FBI said.’ Now, is this population-adjusted rate a better measure of the crime rate that the violent crime reports (the figures on which the whole story of ’significant change’ is based) and if it is, is the implication that there has not been a significant change and if that is the case, why is the Washington Post going on about it?
However, the story certainly works for the various stakeholders. The Washington Post gets to sell its ‘news’, those who want to criticise the Bush Administration get a warm feeling that there’s one more thing that they can criticise it about. On the other hand, because of the confusion about the interpretation of the figures, those who support the Bush Adminisration can latch onto the spin that there was some ‘encouraging news’. But for the rest of us it was really a complete waste of the couple of minutes it took to read it. There are two news stories I would be interested in. The first one is ‘Crime trends worsen, no one really sure why’ which would be the story if the change were statistically significant. And the other one is ‘Independent analysis (and it would have to be independent - not something put together by a partisan think tank) points to Bush Administration policies leading to increasing crime trends’. Short of this, we’re mostly better off spending our time on the sports section.
Paul Duignan (outcomesblog.org)
Comments»
no comments yet - be the first?