When is a poll not really a poll?
When a "poll" asks questions in such a way as to get a certain answer or if the number of people questioned is too small to draw any conclusions, it isn't really a poll.
WHAS TV ran as a story on the air (and on the net) that 38% of Kentuckians thought Governor Ernie Fletcher should resign office because AG Greg Stumbo is investigating his administration's approach to the merit system.
Pretty sensational numbers, right? Well, not so fast.
The alert staff at Kentucky Progress called the WHAS studio to question some red flags that were evident in the reporting of the "poll."
The response from WHAS to this questioning was to amend its story to say that only 224 Kentuckians were asked "What should happen to the governor now?"
Why would they poll 224 people and then report it as news? Well, the claim is that 500 people were called and asked if they were familiar with the investigation. The original reporting claimed that 45% said they were familiar. Interestingly, the original reporting also claimed that 59% of the 500 said the hiring practices were corrupt. So how was it, we wondered, that more people had an opinion than had a familiarity? We wondered what "information" was given by pollsters that might explain this discrepancy. That is when the story changed to indicate that only those respondents who claimed an awareness of the investigation were asked follow up questions. That would be 224 people. So which is it: was this merely a push poll to create a story, or are we really being asked to draw conclusions about a straw poll of 224 people?
The far left is already getting worked up about this. Sure hate to rain on that parade.
