[E-voting] Any statisticians subscribed? Looking for a tipping point.

bobyrne at statesoft.ie bobyrne at statesoft.ie
Fri Sep 9 08:31:57 IST 2005


I meant the 'good' votes are correctly recorded and accurately reflect the
voters' intent. 'Good' votes will include spoiled votes. If a voter spoils
their vote it is right that the system should record that.

The 'bad' votes look to be correctly recorded but are in fact not a true
reflection of the voters' intent, whether through coercion, vote selling,
ballot stuffing, malicious or incompetent programming or any other failure of
the voting channel to ensure the integrity, authenticity and secrecy of the

The ratio of 'good' to 'bad' votes is largely a function of the 
integrity of the
voting channel.
The best we have available (at least according to common wisdom) is a paper
ballot recorded in a polling booth. This ensures an immutable, secret and
well-authenticated vote.
Postal voting is immutable but neither secret nor well authenticated.
DRE in a polling booth is well authenticated but not entirely secret and
definitely not immutable.
Remote DRE is poorly authenticated, not secret and subject to changes between
the voter and the count.

I'm certainly not making any comments about voters, just about how well the
system is recording their intent during the election.


Quoting Catherine Ansbro <cansbro at eircom.net>:

> Tim's comment  about "good voters" brings a smile to my face.  And it 
> raises some interesting questions.
> Brian, did you mean "good votes" as in ones that were accurately 
> recorded so as to reflect the voter's intent? These would be randomly 
> distributed only if you meant "bad votes" to be ones that were spoilt 
> (accidentally or deliberately--but not with fraudulent intent).  
> Though it's possible that because of well-disciplined vote management 
> FF (and possibly FG) votes would get statistically more than their 
> share of "bad votes" through uninformed or intimated voters putting 2 
> or 3 #1 votes for the same party and thereby spoiling their vote 
> accidentally (or through being intimidated and not wanting to express 
> a preference among their promised #1 votes).
> One could also debate whether deliberately spoiled votes are more 
> likely to show up for one party or parties more than another.  I 
> think it is unlikely that deliberately spoiled votes would be 
> distributed randomly, since it's more likely that a spoiled vote be a 
> reflection on whoever happens to be in power (locally? nationally?) 
> at a particular point in time or in a particular local area.
> Catherine
> Timothy Murphy wrote:
>> On Thu 08 Sep 2005 15:07, Brian O'Byrne wrote:
>>> Assume first that the 'bad' votes are randomly distributed. Does the
>>> reduction in accuracy affect the result?
>> In a democracy there is no such thing as good voters and bad voters.
>> There are just voters.
> _______________________________________________
> E-voting mailing list
> E-voting at lists.stdlib.net
> http://lists.stdlib.net/mailman/listinfo/e-voting
> http://evoting.cs.may.ie/

More information about the E-voting mailing list