On Fri, 17 Nov 2000 15:53:58 GMT, "Robert Chung" <[EMAIL PROTECTED]>
wrote:
>In today's edition of the NY Times was this article on the accuracy of
>machine vs. hand counts as described by the makers of vote-counting
>machines.
>
>http://www.nytimes.com/2000/11/17/politics/17MACH.html
>
>In summary, under ideal conditions the machines can be up to
>99.99% accurate. Ideal conditions do not often apply, and
>in a 1975 Federal Election Commission study, it was found that
>99.5% of the ballots were read accurately.
>
>The article continues, "Ultimately, industry officials said, the
>most precise way to count ballots is by hand."
This assumes the standards for acceptable ballots have been set a
priori. Would leaving it up to vote counters to "divine" what the
intent of the voter was on election night be more precise? Would
judgment calls on "indentations" be more precise? Stuffing the ballot
box is as American as apple pie. Rascals, of course, can "stuff" a
machine too I suppose. When a vote like this is too close to call and
either human or mechanical errors can be operant, it seems the toss of
a coin or winning a poker hand might suffice. Unfortunately, in this
particular instance, the stakes are awfully high for simplistic
tie-breakers.
>
>"'The important thing here is that there may be no way to get
>a 100 percent accurate count by a machine,' said Mr. Swartz,
>whose card readers are approved by the Federal Election
>Commission for use in punch-card voting systems. 'It is totally
>reasonable that the most accurate way to do it is a carefully
>run recount.'"
>
>--Robert Chung
>
>
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
