26
10
For men, the key factor was whether their job was unionized or not. The union vote moved
back to the NDP in 2004 and it stayed there is 2006. Being a union member increased men’s
probability of voting NDP by 19 points and decreased their probability of voting Conservative
by 10 points. Being a union member also enhanced women’s likelihood of voting NDP, but the
effect was more modest (11 points). The implication is that the votes of male union members
helped to keep the gender gaps in vote choice from being bigger.
11
We do not have a good explanation for this effect. It is not an artifact of over-control: 31 per
cent of women with a mortgage on their home voted NDP, compared with 20 per cent of those
who did not. Conversely, even in the absence of controls, being a renter makes little or no
difference to the probability of voting NDP. This is a change from 2004 when the NDP did
almost as well as the Liberals and Conservatives among renters (Gidengil et al. 2006).
12
Togeby (1994) has suggested that the impact of parenthood varies depending on the number of
children a woman has. Her study of Danish women suggested that having a single child made
for more left-wing views, but having more than one child encouraged more conservative views
.
13
Being a Protestant fundamentalist remains a highly significant correlate of Conservative voting
even when value orientations are added to the vote model.
14
For example, almost two-thirds of the women interviewed in 2006 favored a publicly funded
day care system; only 30 per cent preferred the Conservative policy of paying money directly to
parents to spend as they please. Only 30 per cent of women were opposed to same-sex marriage.
Convention