One of the ways in which I've seen statistics disingenuously used by the government is to characterize incidental happenstance as a success attributable to government intervention. This is routinely done, especially with regard to Iraq.
Here's an example of what I mean:
Let's suppose you're running a government health-care program for a million people. Within the first year, 10,000 people die needlessly because of shortages, lack of access to certain drugs kept off the market, waiting lines, etc. This means that there are 10,000 less people to support the following year. As a result, fewer needless deaths are likely to occur even if no changes are made to the system. The government spins this and claims that things are improving based solely on numbers on a spreadsheet.
This is exactly what has been happening in Iraq.