Last week I seriously pissed off at least one local journalist by posting two competing ledes side by side, and pointing out how they guide readers in two different directions. Apparently, because the facts in the articles were mostly correct, I was entirely “wrong-headed” and “fatuous” (and perhaps drunk) to suggest that reporters might suffer from the same inherent bias that afflicts the rest of the species.
Of course the larger point missed in all the personal offense taken where none was intended, was the impact that ledes have on the way readers interpret the news, regardless of whether the actual reports are truthful or accurate. Take for example this lede from an AP story that hit the wire today:
OLYMPIA — More than 176,000 names were removed from the state’s voting rolls last year under a new statewide voter database that was developed to help counties find duplicate registrations and dead voters, Secretary of State Sam Reed said today.
The purge of illegal registrations is the result of the new system that has consolidated all 39 separate county systems into one database in January 2006.
Oy.
What will readers take away from this story? That 176,000 “illegal registrations” were purged from the voter roles. That the body of the article tells a different story comes too late — a large number of readers will only remember their first impression, and an even larger number won’t bother to read beyond the opening paragraphs. You put the most important information near the top of the article; that’s Journalism 101.
As the reporter makes clear a few column inches further down, the vast majority of these 176,000 purged registrations were not “illegal.” 39,814 were duplicate registrations, a common occurrence when voters move and fail to notify election officials. 40,105 were deceased voters who had never been removed from the roles. And 91,954 were inactive registrations and voters who requested cancellation or moved to other states.
To say that these were “illegal registrations” would imply that there was some crime committed by the registrants, which couldn’t be further from the truth. Only the 4,500 canceled felon voter registrations could arguably be considered “illegal”, and even then we’ve seen absolutely no evidence that a single felon registered knowing he was violating the law.
These would be more correctly described as “invalid” and “inactive” registrations, and yes, the choice of words is important, as it shapes the way readers perceive the integrity of our elections. A lede like this only reinforces the popular misperception that our elections are corrupt and mismanaged, but as the Spokesman-Review’s Jim Camden points out on his blog, there was virtually zero evidence of voter fraud in Washington state in 2006:
So at most, we had one case of double voting out of 2,107,370 ballots cast. Which is a .00004 percent rate of possible fraud.
Secretary of State Sam Reed said he was “pleasantly surprised” with the results. The state is doing a better job of cleaning up its voter records, but added “we really don’t have a history of voter fraud here.”
Which will come as a huge shock to some of his Republican brethren, who still hope to run Dino Rossi in a gubernatorial grudge match against Chris Gregoire to win the seat that was “stolen” from him. It also might give pause to some of their pollsters, who seem to delight in reporting that people don’t have confidence that the problems of the 2004 elections have been cleared up.
On the eve of the November election, one polling firm said its survey showed that 71 percent of the people lacked confidence there’d be no problems in 2006.
Of course, if people keep insisting there are problems, even if they don’t provide any proof, some folks might just conclude those problems exist.
And some of those folks just might include AP reporters who get the facts right, but the lede wrong.