EXIT POLLS….I talked to a magazine writer today who asked me who I thought was right: the networks for not publicizing early exit poll numbers or blogs for publicizing them. I told her I thought they were both right.
Basically, I think exit poll numbers are interesting data, and anyone who’s interested should have access to them. That’s especially true since the media has access to them all day long, and it annoys me when their coverage is so obviously driven by something they won’t tell us about. So if blogs make exit polls accessible to political junkies and other people who want to see them, that’s great.
At the same time, television and radio are mass mediums, and it seems likely that early exit poll results can affect election turnout if large numbers of people see them. What’s more, if the networks decide to air exit poll results, you’re pretty much forced to hear about them unless you choose to simply not watch the news at all ? hardly a likely proposition on election day.
Roughly speaking, then, I think the best solution is for TV and radio to refrain from publicizing early exit poll results, while the internet should make them available to people who actively want to see them. That’s about what happens now, and the only improvement would be for sites like CNN to post the official results ? along with enough data to judge how seriously to take them ? instead of having blogs publish whatever rumors and reports of rumors they happen to get via email. I haven’t really heard any credible arguments against doing this.
UPDATE: And speaking of exit polls, I’m still curious about my question from last night: why is everyone complaining about how inaccurate they were? In fact, the final exit poll results were almost spot on.
It’s true that many of the early results were off, but those were based on small samples and were skewed toward early voters. I don’t know if we’ll ever know for sure, but it’s quite plausible that Kerry supporters jammed the polls in the morning, and the early exit results accurately reflected this.
Or am I missing something here?