Another modest data reliability problem has come up with the global warming issue.
The idea that global warming is leading to more hurricanes is getting pretty deeply embedded in the collective consciousness; although in their defense it should be noted that very few climatologists have actually made this claim.
Anyway, the data problem is that better measurement is allowing storm trackers to name more storms. This bumps up the number of named storms without actually providing evidence that they really are more frequent.
The details are that better satellite tracking is allowing them to spot storms far out to sea earlier, and to get wind speed data from them without sending out a plane. This is certainly an improvement, but it is inconsistent with the old method of sending out a plane to measure the central pressure of the storm (something that still can't be done with a satellite).
This is all analogous to the fact that cancer rates are rising, although no one is certain to what extent that is because we are looking for it more. Of course, in the cancer case this uncertainty hasn't prevented those broadly opposed to modern western society to blame it for cancer. The problem here then really isn't the hurricane data, but the lack of awareness of this pitfall, and the potential for abuse.
FWIW: you should also note that sciences that are more serious than climatology visit similar issues openly. As an example, there is broad awareness in economics of the Stock and Watson argument that recessions are less common than they used to be because of changes in measurement.
Read the whole thing in The Houston Chronicle.
You'd probably better - it isn't like the wire services are going to pick up something from a flyover state that doesn't support the politically correct hegemony.