If you have "no place to go," come here!

Misogyny, Sexism, & the Gender Gap in the 2008 Election

[Welcome, readers!]

Misogyny, sexism, and the gender gap are alive and well in the American electorate.

Overall, men prefer McCain over either Democrat, while women prefer either Democrat over McCain. But the gender gap widens considerably when a woman is running. When McCain is matched with Obama, the gap is 13.9% (comprising 7% of voters), but when it is McCain versus Clinton, the gap nearly doubles to 26.9% (comprising 13.5% of voters.)

The expansion of the gender gap is due almost entirely to changes in how men vote. Only a few more women (1.6%, comprising 0.9% of all voters) prefer Clinton when matched with McCain than when Obama is matched with McCain—women pretty much stick with the Democrat regardless of whether its Clinton or Obama.

But far more men (11.3%, comprising 5.6% of all voters) prefer Obama when matched against McCain than when its Clinton versus McCain. As a result, when the margins in all 50 states are averaged, Clinton gets 4.3% fewer votes than McCain, while Obama get 0.4% more votes than McCain.

(This study is based on the SUSA National Poll released on 3/5/07 than polled voter on their preference for Clinton or Obama when matched against John McCain in all 50 states. Links to each state’s data will be provided in Part 2)

Table G1 (see below) shows the averages for all 50 states for the margins found in all states, for males, for females, and the gender gap – the difference between the male average and the female average (male minus female), for both each Democratic candidate when matched against McCain. (Rows 1 and 2). Row 3 shows the difference between support for Clinton and Obama (negative numbers mean less support for Clinton) relative to Obama in each category) .

Rows 4, 5, and 6 show the same data weighted based on the percentages of male and female voters in each state. While Rows 1, 2, and 3 compare margins within specific categories, rows 4, 5, and 6 show the percentage of the electorate that the margin comprises. (i.e. In Row 1 the “Male” percentage represents the margin among male voters – McCain received 18.1% more male voters than Clinton. In Row 4, the “Male” percentage represents the margin as compared to all voters – McCain’s advantage among males represented 8.9% of all voters.)

G1A--Unweighted Averages
Row		         All	Male	Female	Gender Gap
1	Clinton         -4.3%	-18.1%	8.8%	26.9%
2	Obama	         0.4%	-6.8%	7.1%	13.9%
3	HRC minus BHO	-4.6%	-11.3%	1.6%	13.0%
G1B--Weighted Averages
Row		          All	Male	Female	Gender Gap
4	Clinton	        -4.3%	-8.9%	4.7%	-13.5%
5	Obama	         0.4%	-3.3%	3.7%	-7.0%
6	HRC minus BHO	-4.6%	-5.6%	0.9%	-6.5%

Table G2 (see below) shows where these margins come from. The data shows total vote percentages (“All”), is broken down by gender (“Male” and “Female”, and the “Gender Gap” (“Male” percentage minus “Female” percentage) is shown.

In Table G2A, McCain’s vote is shown the average for each Democratic candidate matchup. Row 2 shows that in a McCain-Obama matchup, McCain gets 45.1% of the total vote, consisting of 49.3% of the male vote, and 41.2% of the female votem resulting in a gender gap of 8.1%. Row 1 shows that in a McCain-Clinton matchup, McCain gets 47.2% of the total vote (2.1% more than against Obama), which consists of 54.1% of the male vote (4.8% more than against Obama), and 40.8% of the female voter (0.4% less than against Obama) , resulting in a gender gap of 13.3% Clinton’s gender gap is 5.2 points higher than Obama’s, largely because of the 4.8% difference in the male vote for McCain.

In G2B, the average vote percentage for each Democratic candidate is shown in their own matchups with McCain. Row 2 shows that Obama gets 45.5% of the votes, comprise of 42.5% of the male vote and 48.3% of the femail vote, creating a gender gap of –5.8%. Row 1 shows that Clinton gets 42.9% of all voter (2.6% less than Obama), comprised of 36% of the male vote (6.5% less than Obama) and 49.5% of the female vote (1.2% more than Obama), creating a gender gap of -13.5%. Again, Clinton’s gender gap is greater than Obama’s (by -7.7 points), due primarily to Obama getting 6.5 more of the male vote than Clinton.

(Table G2D -- which is the same as table G1, above --is created by subtracting the values in Table G2A from the the values in Table G2B.)

Table G2C looks at the “Undecided” vote. In the average McCain-Clinton matchup, there are 9.9% “undecided” voters, just 0.6% more than in a McCain-Obama matchup. But there are 2.5% more undecided voters in a McCain-Obama matchup than a McCain-Clinton contest. And while 1.8% more men are undecided when it’s the Clinton matchup, slightly fewer women (-0.7%) are undecided when Clinton is the nominee. (And people say that women can’t make up their minds!)

G2A--Average McCain percentage
Row		        All	Male	Female	Gender Gap
1	vs Clinton	47.2%	54.1%	40.8%	13.3%
2	vs Obama	45.1%	49.3%	41.2%	8.1%
3	HRC minus BHO	2.1%	4.8%	-0.4%	5.2%
G2B--Average Democratic percentage versus McCain
Row		        All	Male	Female	Gender Gap
4	Clinton 	42.9%	36.0%	49.5%	-13.5%
5	Obama   	45.5%	42.5%	48.3%	-5.8%
6	HRC minus BHO	-2.6%	-6.5%	1.2%	-7.7%
G2C--Average Undecided Percentage versus McCain
Row		        All	Male	Female	Gender Gap
7	vs Clinton	9.9%	10.0%	9.8%	0.2%
8	vs Obama	9.3%	8.2%	10.5%	-2.3%
9	HRC minus BHO	0.6%	1.8%	-0.7%	2.5%
G2D--Average Democratic margin versus McCain
		     All	Male	Female	Gender Gap
10	Clinton	     -4.3%	-18.1%	8.8%	-26.9%
11	Obama	      0.4%	-6.8%	7.1%	-13.9%
12	HRC-BHO	     -4.6%	-11.3%	1.6%	-13.0%

Table G3 presents a “weighted” version of Table G2. (see above explanation of Table G1B.) It should be noted each state was weighted according to its own demographic profile; Row O represents the average democgraphic profile of all 50 states.

Demographic Break down by Gender
row		All 	Male	Female	Gender Gap
0	Average	100.0%	48.4%	51.6%	-3.2%
G3A--Average McCain percentage
row		         All 	Male	Female	Gender Gap
1	vs Clinton	47.2%	26.2%	21.0%	5.3%
2	vs Obama	45.1%	23.9%	21.2%	2.7%
3	HRC minus BHO	2.1%	2.4%	-0.3%	2.6%
G3B--Average Democratic percentage versus McCain
row		         All	Male	Female	Gender Gap
4	Clinton	        42.9%	17.4%	25.6%	-8.3%
5	Obama	        45.5%	20.6%	24.9%	-4.4%
6	HRC minus BHO	-2.6%	-3.2%	0.7%	-3.9%
G3C--Average Undecided Percentage versus McCain
row		        All	Male	Female	Gender Gap
7	vs Clinton	9.9%	4.9%	5.1%	-0.2%
8	vs Obama	9.3%	4.0%	5.4%	-1.4%
9	HRC minus BHO	0.6%	0.9%	-0.4%	1.2%
G3D--Average Democratic margin versus McCain
row		         All 	Male	Female	Gender Gap
10	Clinton	        -4.3%	-8.9%	4.7%	-13.5%
11	Obama	         0.4%	-3.3%	3.7%	-7.0%
12	HRC minus BHO	-4.6%	-5.6%	0.9%	-6.5%

There can be little question that it is not simply “Hillary hatred” or “Clinton Derangement Syndrome” that is at work here. It is sexism and misogyny. Women are voting for the Democrat, regardless of who is running. It is men who switch their votes in droves to McCain when it’s a woman running.

But there is a great deal of variation in the extent to which this misogyny and sexism shows itself in various states. That will be the subject of Part 2. Part 3 will look at group of states where the “gender factor” plays a significant role in individual states.

No votes yet


Becki Jayne's picture
Submitted by Becki Jayne on

Sexism and misogyny is alive and thriving in America. However, women who support Hillary are getting very, very angry. So I wouldn't be surprised if there isn't a huge backlash come November if Obama is the nominee based on what I've read (and I read a lot of blogs and their comments). It will be interesting to see how the SUSA results change (if they repeat the survey) a few months from now.

Another thing, based on decades of marketing and research employment experience, there's what people say they will do vs. what they actually do. I have a tendency to take all surveys with a big shaker of salt. I'm not disputing your excellent analysis, Paul. I'm just saying that it remains to be seen whether women will vote for the Democrat no matter who the nominee is. If the misogyny continues in the media, and if Democratic leaders continue to remain silent about it, oh, boy. Not good.

Submitted by Paul_Lukasiak on

as to the general unreliabily of polls this far out when it comes to predicting what will happen in November.

What this is about is the remarkable difference in the attitudes of men toward Obama and Clinton. Less than 1/50 of women voters switch from the male republican to the female democrat. 1/7 of male votes switch from the male democrat to the male republican when the democratic candidate is a female. Men are thus 7 times more likely to switch their vote based on gender than women.

intranets's picture
Submitted by intranets on

I'm not sure about the "7 times more likely statement"... you are comparing women voters switching from R to D, and male voters from D to R. What if it is just easier to voter for the R? And I'm not 100% sure it isn't a statement about the candidate (HRC) and not women candidates in general, but it probably would be similar for any women candidates.

I don't think there will be a lot of women who won't vote because Hillary isn't on the ballot. Of course I do think that would happen with black voters if Obama wasn't on the ballot, so I already hold an impossible double standard.

I just can't imagine Hillary supporters would let McCain win, rather than vote for Obama. (I do think black voters would stay home for a different effect of the my-vote-doesn't-count factor, which is what Ohio 2004 did by making long lines for first time black voters)

Submitted by Paul_Lukasiak on

the survey allowed for "undecided" voters, and also varied its demographic composition for each matchup. (For instance, the male/female demographic breakdown is different for McCain v Obama and McCain v Clinton in New Jersey -- see see SUSA New Jersey data

The key here is that women stick with the democrat regardless of who the candidate is. Almost half (0.7%) of the greater support for Clinton among women (1.6%) can be attributed to women who were "undecided" in McCain v Obama.

Men who support Obama abandon in droves the Democrat when its McCain v Clinton -- either switching to McCain (4.8%) or becoming "undecided" (1.8%)

BDBlue's picture
Submitted by BDBlue on

According to CNN's exit polls, yesterday in Mississippi 28% of Democratic voters said that gender was either the most important consideration (7%) or one factor among many (21%). You might think Clinton won those votes because of women's enthusiasm for her. You'd be wrong. Obama won those who said gender was the most important 64-36% and those who said it was a factor 70-27%. In fact, Obama won by more among people who said gender was important (69-29) than he did among those who said it wasn't (56-42).

Of course in Ohio, when Clinton won among voters who said "race" was important, Big Media Matt and others pronounced her win due to the racist vote. I await the prounouncement that Obama won the sexist vote in Mississippi. Of course, I won't be holding my breath.

And on a related note, let me say that posts like this one only make me appreciate and love all of the Democratic men in my life who either support Hillary or would support Hillary if she is the nominee. I know it's somewhat sad to be grateful for men who aren't misogynists, but given what I've learned about my culture during this election, I'm grateful nevertheless.

Submitted by Paul_Lukasiak on

I was thinking about doing something similar to this study using exit polling data (in general, the gender gap is wider among democrats than among republicans).

on the gender matters question, Mississippi seems to be either an anomoly, or evidence of an emerging pattern -- in most cases, Obama's percentage in that category was lower than his general election percentage. In Mississippi, he got 69% of the "gender matters vote, and only 61% of the overall vote. (In Louisiana, it was 54% of "gender matters" and 57% overall; in GA it was 54% of "gender matters" and 67% overall -- I've picked demographically similar states here, but a spot check of other states shows the same pattern. However, my SUSA data suggests that in the vast emptimess states that held caucuses -- and thus have no exit polling data--the pattern might not hold.)

Becki Jayne's picture
Submitted by Becki Jayne on

For your reply and seems we both agree. Your statement, "Men are thus 7 times more likely to switch their vote based on gender than women" is a sad reality but thanks for keeping it real and daring to speak the truth, Paul. Props to you.

Looking forward to reading more of your analysis.