Addition to the last post

[shkrobius]

I've always found the arguments for "enfranchisement" of this or that group (e.g., women) somewhat dubious. It sounds great, but it tacitly assumes that the voting patterns would not change, whereas the opposite is true.

...By 1840, voter participation levels reached 80% and remained high throughout the 19th century. By the end of the 19th century

Comments

[dmpogo]

So what ? 50% evenly distributed is more representative than 100 % drawn from preselected 1/2 of population

[shkrobius]

Do not we have representative democracy? If it is considered proper for some people represent the others, why is it improper for men to represent their households? If the principle of representation is openly admitted to fail at the level of family, it is difficult to make the consistent case for its general propriety. To put it differently: if we consider it repressive for husbands to represent the interests of their wives, why do we make exception for parents representing the interests of their children

[dmpogo]

Do not we have representative democracy? ....

Why stop at the level of family ? Afterall the king himself can well represent his people and call it representative democracy. Such retorical questions are neither deep nor interesting.

Whatever the history of the suffrage it's outcome is eliminating specific biases in the polls of today, which IS enfanchisement of the groups the
voting was biased against before. ( It is not measured in the number of participants BTW, every statistician will tell you that whether a sample is biased and sample size are different things.)

Sure biases still exisits in the current voting rights, e.g. against minors. But at least agains women they are largerly eliminated.

The fact that in US women vote came with antialcohol sentiment - well that does mean that male only sample was biased :). Tough luck, I guess men should have drank less to keep the other half of the population happy.

[shkrobius]

Family is the smallest unit of society. If the representation does not work for the smallest unit, why should it work for larger ones? I can see the case for direct democracy. I do not see the case for enfranchisement in representative democracy. You conflate the two issues.

Moreover, I do not see where your confidence of less bias comes from. Is it just an opinion or you can back it up?

[shkrobius]

I am not sure I expressed myself well. Let me give an example.

You say that sparse sampling of 100% is better than dense sampling of 50%. Let's consider families voting for two parties, R and D. They can be RR, RD, DR and DD. Now, consider these RD and DR couples. They know how they will vote. Because they cancel each other's votes, in all probability they will decide not to vote (and that might explain 50% turnout if the votes split almost evenely). Then your statistical sample is biased because it represents exclusively RR and DD couples as opposed to individual R and D voters. The RD/DR voters are disenfranchised by the very fact that they know each other's votes. So you trade one kind of disenfranchisement for another.