How important is it, for democracy, that women can vote?
Well, I'm sure we'd all agree that this was a big step forward for democracy. In fact, many would say that when women were not allowed to vote, that wasn't democracy.
But let's make a more quantitative assessment. How many election results have actually been directly changed by women's votes?
Women got the vote in the USA a few months before the 1920 election. In the 22 presidential elections 1920-2004, women's votes apparently have altered the winner in exactly one case – Clinton's election in 1996 – and maybe not even then. (According to exit polls, the men supposedly preferred Dole by 44% to 43%, but this is a small enough margin that it is possible that men actually preferred Clinton, in which case the women did not change anything.) Also, as an asterisk, Gore "won" the popular-vote "election" in 2000 and this was definitely due to women. But Bush won the presidency so that had no effect.
So in total we'll give women credit for changing one presidential election out of 22.
Women have only voted especially detectably differently from men starting in about 1980 – and also, women began to exceed men in total number of votes also in 1980. (Since 1980, women have voted more Democratic and men more Republican – Eleanor Smeal dubbed this the "gender gap." Women now exceed men both in numbers and in turnout percentage.)
In the 2004 elections, there were 33 senate and 36 state-governor races. According to CNN exit polls, women altered the winners in 4 cases.
In the 2006 elections, there were 34 senate and 11 state-governor races. According to CNN exit polls, again women altered the winners in 4 cases.
| Senate 2004 | Governors 2004 |
|---|---|
| Colorado(Salazar) | New Hampshire(Lynch), Washington(Gregoire), Delaware(Minner) |
| Senate 2006 | Governors 2006 |
| Montana(Tester), Virginia(Webb), Missouri(McCaskill) | Maryland(O'Malley) |
You can find more information about women and US politics at CAWP and in the work of professors Barbara Norrander and Karen M. Kaufmann.
In 2004 & 2006 women altered 4+4=8 winners out of 11+34+33+36=114 races (sen&gov), i.e. altering 7.02% of the winners. During 1920-2004, women altered 1 president out of 22 races, i.e. altering 4.55% of the winners.
The 2004-2006 count probably overestimates the effect of women, since 1980-2006 were years with substantially greater "gender gap" than the historical norm during the 3-times-longer period 1920-2006. So I am going to give women credit not for 7.02%, but only half that – 3.5% – historically averaged over senate and governor races. (I have no idea what will happen to the gender gap in the future. I am reckoning historical average effects.) Combining all these figures and taking account of these historical trends and putting in a 1σ statistical error bar, our bottom line is...
Women have historically altered (3.7±1.2)% of US presidential, senate, and state-governor race winners, but at the current time (1985-2007) they are altering about (7.4±2.5)%.
[If you want to phrase it in a more feminist (?) way: essentially the same statements ought to be true about men's votes.]