>So, people, how convincing is this research?
Andrew Beveridge, a demographer at Queens College who consults for the NY Times, says it should be taken very seriously. Beveridge is not an excitable partisan.
Doug
----
November 19, 2004 By itself, switching these votes still wouldn't make Kerry the winner. But it's two presidential elections in a row that appear to have been messed up in Florida. Can the press help avoid a trifecta?
By Barry Sussman Editor at niemanwatchdog.org
Q. Berkeley sociologists, including one who is a member of the National Academy of Sciences, have issued a report saying the odds are 999 to 1 that flawed electronic voting machines in Florida gave Bush about 130,000 votes more than he actually got. The first question here: Are the Berkeley group's findings credible?
Q. Putting it politely, that's an awful lot of votes for machines to mangle. So a second question is, should we be sure of the vote count in other states that used electronic voting machines?
Q. An expert we talked to lauds the quality of the Berkeley study. But let's suppose others say it's not credible. What then: In the future are we supposed to have blind faith that voting machines are accurate, can't be tampered with, aren't rigged? With no recount possible?
Q. It is obvious that recounts must be part of the electoral system, as they always have been. But this also was obvious long before the 2004 election. Who blocked recounts?
Q. One would think there is time between now and the 2006 elections to see that there is a paper trail to enable recounts. But is there any guarantee that a paper trail alone ensures accurate vote tallies? Are elections officials in your area working on this?
Bush is credited with about 350,000 more votes than John Kerry in Florida. If the Berkeley study is correct and all 130,000 extra Bush votes were really cast for Kerry, that would be a swing of 260,000. A big number but not enough to change the result.
The report was done by doctoral students in sociology at the University of California at Berkeley, in collaboration with Michael Hout of the UC Berkeley Survey Research Center, a member of the National Academy of Sciences.
Its theme - one that has been kicking around in blogs since right after the election - is that Bush couldn't have amassed the vote totals shown for him in some Florida counties where touch-screen voting was used. (This Watchdog Web site made mention of the issue some days ago, noting questions raised by Keith Olbermann on MSNBC.)
The Berkeley report singles out heavily Democratic counties of Broward, Palm Beach and Miami-Dade. It says that through the use of multiple regression analysis, a standard but high-powered statistical technique, "irregularities associated with electronic voting machines may have awarded 130,000 excess votes or more to President George W. Bush in Florida.We can be 99.9 percent sure that these effects are not attributable to chance."
In an interview, Andrew Beveridge (andy at troll.soc.qc.edu), a sociology professor at Queens College and a consultant to the New York Times, spoke highly of Michael Hout, saying he was someone whose work has to be taken seriously. He described the Berkeley study as highly professional and said "the paper is quite disturbing." He said it "really raises the issue - we don't know what's going on" in electronic voting machines. "This kind of analysis is exactly why we need more analysis and a paper trail."
"Say it was some sort of fraud," Beveridge noted. "Whoever was doing it didn't know that it wouldn't matter."
The Berkeley group also studied Ohio voting results but found no such irregularities there.