Is There A War on Women? Attitudes About Women in the Workplace and in Politics
A Report from the 2012 Blair Center-Clinton School Poll
By
Angie Maxwell
Pearl Ford-Dowe
Rafael Jimeno
Todd Shields
Despite the progress that women have made in their struggle for equality in both the workplace and in political life, there are substantial obstacles that women face in both arenas. Scholars of public opinion have investigated attitudes about women in work and public life as well as the changing nature of how women are perceived and evaluated.1 The Blair Center-Clinton School (BCCS) Poll, conducted by the Diane D. Blair Center of Southern Politics and Culture and the Clinton School of Public Service, adds to the understanding of these attitudes.
I. Work Life
While many scholars have demonstrated that although traditional sexism has declined substantially since the 1950s, contemporary attitudes toward women have adapted to the increasing presence of women in politics and leadership roles in the workforce. The Modern Sexism scale is intended to uncover negative attitudes regarding changes in traditional gender roles over the past several decades. Five of the scale’s 23 questions relate to women in the workforce. Specifically, respondents are asked to read each of the following statements and then express the extent to which they agree or disagree.
- Many women are actually seeking special favors, such as hiring policies that favor them over men, under the guise of asking for “equality.”
- Most women interpret innocent remarks or acts as being sexist.
- Feminists are seeking for women to have more power than men.
- When women lose to men in a fair competition, they typically complain about being discriminated against.
- Discrimination against women is no longer a problem in the United States
Combined, these five questions form a unique scale that allows us to examine the ongoing resistance toward women holding equal positions or, in many cases, positions of authority.2 Since respondents are asked to agree or disagree with each statement on a Likert scale ranging from 1=strongly disagree to 5=strongly agree, the scale ranges from 5 to 25 with higher scores indicating increasingly negative attitudes concerning women in the workplace.
While the mean for all participants is 14.29, the average for men is 15.0 compared to women who have an average of 13.6. If we look at the percentage of respondents who report attitudes that are one standard deviation above the mean (standard deviation=3.89) we find that 24.3% of men reported “extremely negative” attitudes.
The percentage of respondents expressing these “extremely negative” attitudes varies significantly by gender, race and region. For example, 25.7% of white men nationally would be classified as expressing sexist attitudes, but only 19.2% of African-American men and 22.0% of Latino men fall into this category.
Figure 1: Percent of Men Expressing “Sexist” Attitudes by Race
When we compare the South to the non-South, we see even more variation in the percentage of respondents who reported negative attitudes toward women. Southern men are more likely to report negative attitudes that are one standard deviation above the mean (25.5%) compared to men that live outside of the South (23.8%). Further, a greater percentage of southern women also expressed negative attitudes that were one standard deviation above the mean, 16.9%, compared to 12.5% of women living outside of the South.
Figure 2: Percent Expressing “Sexist” Attitudes by Gender and Region
Surprisingly, and perhaps why gender discrimination remains a persistent issue in American politics, a slightly larger percentage of 18-29 year-old-males (25.4%) hold highly negative attitudes toward women compared to 24.7% of men over 60.
Figure 3: Percent Expressing “Sexist” Attitudes by Age Group
Educational attainment appears to be related to these attitudes, with the lowest levels being reported by respondents with the highest levels of education. Perhaps reflecting the competition in the workplace, the greatest percentage of respondents reporting negative attitudes toward women that were above the average were those respondents who had “some college” education.
Figure 4: Percent Expressing “Sexist” Attitudes by Education Group
There were also substantial differences in these attitudes across party identification. Again, looking at the percentage of respondents who reported negative attitudes that were at least one standard deviation above the mean, 15.1% of male Democrats, 22.6% of male Independents and 36.7% of male Republicans expressed these attitudes.
II. Political Life, 2012 Election
In addition to negative attitudes toward women in the workplace, we are also interested in the extent to which women continue to face obstacles in the political.
Looking first at voting in the 2012 presidential election, we see that there was a large gender gap that clearly benefited president Obama. Specifically, we find a greater percentage of women, compared to men, supported the Democratic candidate. There were substantial regional differences as well. As shown in the figure below, the gender gap in the non-South was quite substantial with 60% of non-southern women voting for Obama while only 45% of non-southern men voted for Obama. In the South, however, the gender gap was almost non-existent with 49.6% of southern women voting for Obama and 48.3% of southern men voting for President Obama. Despite the many reports about Obama’s lead among women voters, the advantage appears to have not been true in the South.
Figure 5: 2012 Gender Vote Gap by Region
Looking next at the 2012 Presidential vote by race and region, we again see substantial variations. Not surprisingly, we see that both non-southern and southern African-Americans overwhelmingly supported President Obama. Non-southern whites, however, split their votes almost equally between President Obama and Mitt Romney. Southern whites, however, reported substantially more support for Mitt Romney (66% for Romney and 32% for Obama). Hispanics in the South and non-South supported Obama, but southern Hispanics reported greater support for Romney (29%) than did non-southern Hispanics (25%).
Figure 6: 2012 Presidential Vote by Race and Region
There were also differences in support for the candidates across men and women within a single race, as shown in the following figure. Since there were few differences between male and female African-Americans, in either the South or the non-South, we omit their responses in the figure below.
Figure 7: 2012 Presidential Vote by Race, Gender and Region
As shown in the figure above, white men gave more support to Romney, but the support was greater in the South. While 52% of non-southern white men supported Romney, 65% of southern white men supported Romney. Non-southern white women supported Obama (53.8%), although southern white women did not (66.6% for Romeny and only 32% for Obama). This pattern is attenuated, but similar among Hispanic women. For example, while 73% of non-southern Hispanic women voted for Obama, only 55% of southern Hispanic women voted for Obama (and 33% of southern Hispanic women voted for Romney compared to only 19% of non-southern Hispanic women). Clearly, President Obama benefited from a gender gap among non-southern white women, as well as overwhelming support from all African-Americans, but the gender gap did not extend to Caucasian or Hispanic women who live in the South.
III. Political Life: Will you see a female president in your life time?
In addition to asking respondents about vote choice, the BCCS poll asked respondents whether or not they believed they would see a female president in their lifetime. Among men, 77.2% indicated yes, compared to 22.8% who said no. Women, of course, reported equally high numbers, with 79.9% saying yes, and only 20.1% saying no. Again, however, these views are not shared equally among racial sub-groups. While 83.2% of African-American men and 80.1% of Latino men agree, only 75.6% of white males answer yes to this statement.
Figure 8: Percent Who Believe They Will See a Female President in their Lifetime
In this situation, regional differences do not appear to be as large. Southern men are almost equally inclined to agree with the statement (76.7%) as non-southern men (76.7%). However, a larger gap exists between southern women (77.9%) and non-southern women (80.9%). As shown in the figure below, southern Hispanic women and non-southern African-American men had the greatest percentage of respondents indicating that they would see a female president in their lifetime (84%), and southern white men had the lowest percentage of respondents (74%).
Figure 9: Percent Believing They Will See a Female President: Race, Gender and Region
Overall, the above results suggest that substantial percentages of respondents not only express negative views of women that are higher than the average, but substantial percentages of respondents also do not believe they will see a female president in their lifetime.
IV. Political Life: 2016 Presidential Match Ups
The 2012 BCCS poll also asked respondents to think about the 2016 presidential election. In the context of the above findings, it is very interesting to see how respondents would react to the Democratic front-runner, Hillary Clinton. In looking toward the 2016 elections, we paired potential Democratic candidate Hillary Clinton against Republican early front-runners, Chris Christie, Marco Rubio and Jeb Bush.
Almost three in five respondents said they would support Hillary Clinton over Jeb Bush (61% to 32%), Marco Rubio (61% to 30%), or Chris Christie (58% to 34%).
Figure 10: Hypothetic 2016 Presidential Contests
While Clinton enjoys fairly wide popularity nationwide, gaps in her popularity exist by gender, race and region.
Women are more likely than men to support a hypothetical President Hillary Clinton, and they support the Secretary of State by a 2:1 margin. Men are more likely than women to support the Republican candidate in all three match-ups, although a majority of them state they would vote for Clinton.
Figure 11: Hypothetic 2016 Presidential Contests by Gender
The gender gap is the most striking among Republicans. While a majority of Republican men and women would prefer the Republican candidate to the former Secretary of State, Republican women are more likely than men to say they would vote for her against all three potential Republican challengers. Among Independent respondents, the difference between genders is smaller, though women are still more likely than men to support Clinton over Marco Rubio or Chris Christie. Among self-identified Democrats, however, men and women are equally likely to say they would vote for Hillary Clinton.
“What this data reveals,” contends Angie Maxwell, “is that the gender gap, perhaps influenced by the perceived ‘war on women’ which made headlines during the 2012 campaign cycle, could have implications not only in contests between the two parties, but within the Republican party itself.”
Figure 12: Gender Gap in Support for Clinton by Party Identification
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as women – men. Refusals not shown. N = 3606
Regional Gaps
Support for Clinton is slightly lower in the South than in other regions of the United States. Between 66% and 69% of respondents outside the South would support Clinton over the three Republican challengers, while support for Clinton hovers around 64% among southern respondents.
Former Florida Governor Jeb Bush is supported by a larger proportion of southerner s than among respondents who do not live in the South. Support for Bush is 7 percentage points higher among southern respondents than among non-southerner s. For Marco Rubio and Chris Christie, support is about 4 percentage points higher among respondents in the South than those who live above the Mason-Dixon line, although neither difference is statistically significant.
Figure 13: Hypothetic 2016 Presidential Contests by Region
In the 2016 Presidential Race, who would you vote for in the following pairings?
Refusals not shown. N = 3606
The regional gap varies by partisanship. Republicans outside the South are more likely to support Hillary Clinton over Jeb Bush or Marco Rubio than are southern Republicans. Southern Democrats, however, support Clinton against Rubio more than do Democrats outside the South. Among Independents, support for Clinton does not significantly vary by region.
Figure 14: Regional Gap in Support for Clinton by Party Identification
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as non-South – South. Refusals not shown. N = 3606
Racial Gaps
Although a majority of respondents of all races and ethnicities would vote for Hillary Clinton rather than Jeb Bush, Marco Rubio, or Chris Christie, larger proportions of African-Americans and Latinos/Latinas say they support Clinton. Between 87% and 90% of African-Americans and between 67% and 71% of Latinos/Latinas would vote for Clinton over these Republican challengers, while Clinton’s support among whites ranges from 50% to 55%.
This finding is significant for two reasons. First, it demonstrates that among African-Americans, there is little if any residual effect of Clinton’s heated contest against President Obama for the 2008 Democratic nomination. Second, it suggests that shared ethnicity is not enough to encourage Latinos/Latinas to support Rubio. Only one in four Latinos/Latinas would vote for him over Clinton, which is statistically similar to the proportion that support Bush or Christie over the former First Lady.
Rafael Jimeno explains, “Part of the large amount of support Hillary Clinton has among Latinos is a function of name recognition. In the 2008 Democratic Primary, Latinos also preferred Hillary Clinton to then-candidate Barack Obama, but this did not decrease their support for him once he became the nominee. If nominated, Marco Rubio could erode some of Hillary Clinton’s support once he, and his policy stances, become better known. Clinton, however, would still benefit from the strong attachment Latinos have with the Democratic Party.”
Figure 15: Hypothetical 2016 Presidential Contests by Race/Ethnicity
In the 2016 Presidential Race, who would you vote for in the following pairings?
Refusals not shown. N = 3606
As was the case with the gender gap, racial differences emerge even after controlling for political party identification. Support for Hillary Clinton is between 33 and 50 percentage points higher among African-American Republicans than among white Republicans. The difference between African-Americans and whites who self-identify as Independents or Democrats is much lower–from less than a percentage point to 10 points.
Figure 16: African-American/White Gap in Support for Clinton by Party Identification
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as African-American – white. Refusals not shown. N = 3606
A different pattern emerges when Latinos/Latinas are compared to whites. More Latino/Latina Republicans than white Republicans would support Hillary Clinton. Among Independents, more Latino/as support Clinton over Bush or Christie. Support for Clinton over Rubio is higher among white Democrats than among Democratic Latinos/Latinas, although the difference between these two groups is inconsequential in contests between Clinton and Bush or Christie.
Figure 17: Latino(a)/White Gap in Support for Clinton by Party Identification
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as Latino/Latina – white. Refusals not shown. N = 3606
A gap also exists between African-Americans and Latinos/Latinas, even after controlling for party identification. African-American Democrats are more likely than Latino/Latina Democrats to support Clinton over Bush, Rubio or Christie. Among Independents, African-Americans are slightly more likely to support Clinton over Bush and Rubio, while Latinos/Latinas are more likely than African-Americans to support Christie over Clinton. African-American Republicans are much more likely than Latino/Latina Republicans to support Clinton over Bush or Rubio.
Figure 18: African-American/Latino(a) Gap in Support for Clinton by Party Identification
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as African-American – Latino/Latina. Refusals not shown. N = 3606
Intersection of Gender and Region
Difference between men and women are more pronounced outside the South than within it. While approximately equal proportions of men and women who live in the South say they would support Clinton over Bush, Rubio or Christie, women outside the South are more likely than their male counterparts to indicate support for the former First Lady and Secretary of State, a difference of 11 to 14 percentage points.
Figure 19: Gender Gap in Support for Clinton by Region
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as women – men. Refusals not shown. N = 3606
Intersection of Race and Gender
The gender gap is more pronounced among whites than among African-Americans or Latinos/Latinas. Support for Clinton among white women is between 9 and 13 percentage points higher than among white men. The Secretary of State has more support among African-American women than among African-American men, a difference between 5 and 9 percentage points. Latinas and Latinos support the former First Lady in approximately even numbers.
Figure 20: Gender Gap in Support for Clinton by Race/Ethnicity
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as women – men. Refusals not shown. N = 3606
Intersection of Race and Region
The racial gaps seen nationwide are larger within southern states than the rest of the nation. Among southerner s, support for Clinton is about 45 percentage points higher among African-Americans than among whites; while African-Americans outside the South are still more likely than whites to vote for Clinton, the difference hovers around 30 percentage points.
Figure 21: African-American/White Gap in Support for Clinton by Region
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as African-American – white. Refusals not shown. N = 3606
Similarly, the gap between whites and Latinos/Latinas is larger in the South than outside the South. In non-southern states, Latinos/Latinas are more likely than whites to say they would vote for Hillary Clinton by about 10 to 20 percentage points. In southern states, however, the difference between Latinos/Latinas and whites is approximately 24 percentage points.
Figure 22: Latino(a)/White Racial Gap in Support for Clinton by Region
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as Latino/Latina – white. Refusals not shown. N = 3606
The difference between African-Americans and Latinos/Latinas is also larger in the South than in non-southern states. Support for Clinton is between 14 and 19 percentage points higher among African-Americans than among Latinos/Latinas outside the South. In southern states, however, the difference between African-Americans and Latinos/Latinas is between 19 and 27 percentage points.
Figure 23: African-American/Latino(a) Racial Gap in Support for Clinton by Region
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as African-American – Latino/a. Refusals not shown. N = 3606
Similarly, regional gaps vary by racial group. Whites who live outside the South are more likely to support Clinton than are white southerner s. Among Latinos/Latinas, respondents who live outside the South are more likely to support Clinton over Rubio; however, there is no regional difference when Clinton is paired against Bush or Christie. African-Americans who live in the South are more likely than those who live outside the South to express support for the Secretary of State.
Figure 24: Racial Gap in Support for Clinton by Region
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as reflected in legend. Refusals not shown. N = 3606
Intersection of Race, Region, and Gender
The racial differences in the gender gap are quite different within and outside of the American South. For almost every racial/regional combination, women are more likely to support Clinton than are men. White men and women in the South are equally likely to support Clinton over Bush. African-American men and women who do not live in the South are equally likely to support Clinton over Rubio.
Among Latinos/Latinas in the South, however, the gender gap is reversed. Men are more likely than women to support Clinton against all three candidates.
“This is the kind of interesting relationship that our data will allow us to explore in-depth in the coming weeks and months,” noted Pearl Ford-Dowe. “Regional differences between African-Americans hints at contrasts in political socialization.”
Figure 25: Regional Gap in Support for Clinton by Race/Ethnicity
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as Non-South – South. Refusals not shown. N = 3606
Figure 26: Gender Gap in Support for Clinton by Race/Ethnicity and Region
In the 2016 Presidential Race, who would you vote for in the following pairings? Difference calculated as women – men. Refusals not shown. N = 3606
1. “Sexism and racism: Old-fashioned and modern prejudices.” Swim, Janet K.; Aikin, Kathryn J.; Hall, Wayne S.; Hunter, Barbara A. Journal of Personality and Social Psychology, Vol 68(2), Feb 1995, 199-214.
2. “Ambivalent Sexism Revisited.” Glick, Peter and Susan T. Fiske. Psychology of Women Quarterly September 2011 35: 530-535