POLITICAL horserace polls are a suspect staple of campaign news coverage that help turn our elections into spectator sports instead of exercises in participatory democracy. Are polls enlightening
or just confusing?Horserace polls -- which measure how the races would go if the vote were held today -- spawn endless conjecture about who's winning, a meaningless question that has little to do with the issues that drive elections.
Elections aren't like baseball games where score is kept by the inning. Nobody is winning until the first returns come in. Trying to measure who's winning before Election Day is as pointless as speculating about who's winning a sumo match while the sumotori are throwing salt around before stepping into the ring.
Every election year, distracting controversies erupt over polls.
Voters say they're tired of polls and the results are increasingly unreliable. Hawaii polls have always shown an unusually high percentage of undecided voters, which makes it impossible to tell what's really going on in a close race. It's not that Hawaii voters are really less decided than mainland voters. We're just more circumspect about telling strangers on the telephone whom we intend to vote for.
Reports that special interest groups are posing as pollsters to spy on members who don't follow the party line will only harden the instinct to clam up.
So why do news media poll? We say it's to give voters the same information politicians have and keep candidates honest in their own polling. We say readers want the information. True, but let's admit it's also true that polls are a relatively inexpensive marketing tool to get our name out there with "exclusive" news.
The Star-Bulletin dropped horserace polling this year, focusing our polls with NBC Hawaii News 8 on issues other than who's winning phantom elections that won't be held today. Our TV partner continues to run horserace polls, in addition to the issue surveys we share.
Our one 1998 venture into horserace polling was a sorry example of how polls can obscure the truth we try to illuminate.
On our survey running today, we asked if respondents would vote "yes" or "no" on the constitutional amendment to allow the Legislature to ban same-sex marriage.
Our initial results showed voters rejecting the constitutional amendment by a ratio of 62 percent to 25 percent. We became concerned when even proponents of same-sex marriage told us our results were out of line with their own.
The Advertiser/KHON poll showed "yes" winning by a margin of 56 percent to 32 percent -- a 30-point difference from ours. The questions we asked were almost identical and the polls were taken about the same time.
WHAT a dilemma. If we published numbers that turned out to be terribly wrong, our credibility would take a beating. If we didn't publish the results, opponents of the constitutional amendment would accuse us of unfairly withholding information favorable to them.
We did the responsible thing and paid to poll the question again to double-check our results.
The new numbers came back opposite from the first poll, with 55 percent to 35 percent favoring "yes" this time. Who was to say which set of results was right?
Again, we tried to do the responsible thing by publishing both results, leaving readers to take the numbers for what they're worth -- which isn't much.
It's time to step back and ask ourselves if we're really performing a public service here. Are our polls fulfilling our high journalistic purpose of enlightening readers or are we just adding to the confusion?
David Shapiro is managing editor of the Star-Bulletin.
He can be reached by e-mail at editor@starbulletin.com.
Previous Volcanic Ash columns
