It's election year, we gotta please the public! On a more serious note-- I agree with John, in that 2 years is insufficient to set a trend...however 24 months (which is 24 points on the graph instead of 2) is enough to make future projections/ or strategies to curb those projections. We can't take those statistics at face value unless all of the variables stayed the same (which they didn't). If they calculate drug possession into 'crime rates' for 2 years, and then remove it from the list 2 years later-- logic tells us???? On that same token, legislation could have equal impact on the statistics, regardless of controlled variables. Diversion programs for certain crimes can raise or drop those figures significantly. So then we're back to square one... statistics are meaningless, absurd, and can be swayed to convince the public to vote a certain way on certain issues.... Reminiscent of "we is in Iraq gettin' Al-Kayda [sic]"

I remember when the results came out for each of those two years (05 and 06), when those upward slopes were a reversal of a long-term (several years) trend downwards. Many were attributing that to Meth use at the time. It may have been those two years that were the flukes because if you do look back further, the rates have continued on a long downward trend. Despite that, most Americans think crime rates are going up, mostly because media reports of crime and drug use have continued to increase during the same time period. The media tends to sensationalize and focus (or maybe fixate is a better word)its attention on crime and devastation.

Reporting two year trends in crime or drug use rates is absurd. After all, we only have three choices -- up, down, or the same -- and none of them will ever be a trend when compared to only two prior years. Methinks the Feds have too many public relations (i.e., propaganda) staff.