Appendix A The Sample
YouGov fielded our survey of potential candidates from November 18, 2021 to March 8, 2022.Footnote 1 They broke the sample into two parts. First, they compiled the “Four Profession Sample” portion of the candidate eligibility pool. This included interviews with 1,576 people who identified as lawyers, educators, or business professionals, as well as 500 politically active, college-educated women and men. The frame for the politically active sample was representative of respondents in the 2020 Cooperative Election Study who engaged in at least four of the following activities in the last year: (1) attended local political meetings (such as school board or city council); (2) put up a political sign (such as a lawn sign or bumper sticker); (3) worked for a candidate or campaign; (4) attended a political protest, march, or demonstration; (5) contacted a public official; and (6) donated money to a candidate, campaign, or political organization. The respondents were matched to a sampling frame on age, race, and education and evenly split on gender.
For the “Broader Sample,” YouGov surveyed 3,417 full-time employed, college-educated respondents who were then matched down to a sample of 3,000 to produce the final dataset. The frame for this sample was constructed by stratified sampling from the full 2019 American Community Survey (ACS) one-year sample with selection within strata by weighted sampling with replacements (using the person weights on the public use file).
In an effort to ensure that the “Four Professions Sample” matched our 2001 and 2011 eligibility pool samples as closely as possible, we included on the survey – and fielded to all respondents – a question that asked them to identify their profession as either (1) lawyer; (2) company executive or business owner; (3) teacher, principal, professor, or college administrator; (4) government or political party staff member; or (5) other. This allowed us to ensure that the law, business, educator, and political activist subsamples matched the specific roles included in the previous studies. It also allowed us to classify from the broader sample of college-educated respondents people who worked full time in one of the four eligibility pool professions, but whom YouGov did not screen as such. The analyses presented throughout the book, and the gender gap in political ambition depicted in Chapters 1 and 3, rely on respondents’ self-identified profession (in other words, how they answered the profession question we included on the survey).
Essential for our purposes, Table A1 illustrates that the women and men share similar demographic profiles. Aside from household income, the statistically significant differences that emerge are not substantively meaningful. Moreover, the demographic differences we uncover in race, party affiliation, and age, while statistically significant, are roughly the same magnitude as those among the potential candidates from the 2001 and 2011 samples, to whom the 2021 sample is being compared.Footnote 2 Moreover, we control for these demographic variables in the multivariate analyses.
1 See Appendix D for information about human subjects protections and IRB protocols.
2 See Lawless and Fox (Reference Lawless and Fox2005) for a description of the 2001 sample and Lawless and Fox (Reference Lawless and Fox2012) for a description of the 2011 sample.
Appendix B Survey Instrument
This appendix includes all the questions from the survey whose results are reported in the book, either as frequencies or in the bivariate and multivariate analyses.
Have you done any of the following things in the past two years? (Check all that apply)
Have you ever taken any of these steps that precede a campaign? (Check all that apply)
Have any of these people ever suggested that you run? (Check anyone who has suggested it)
Which statement best describes your attitude toward running for office in the future?
If you were to run for public office, how would you feel about the following aspects of a campaign?
Would any of the following interactions make you think that maybe you should consider running for office?
Answer options:
Definitely
Maybe
Probably not
Definitely not
A representative from the Democratic or Republican Party encourages you to run
A group of friends at a dinner party tells you that you’d be a great candidate
A coworker tells you that we need more people in politics like you
After a conversation about politics with an acquaintance, the person tells you to run for office
An elected official makes a speech encouraging everyday Americans to run for office
An email from a political organization informs you that someone submitted your name as an excellent potential candidate
What is your level of agreement with the following statements?
Answer options:
Strongly agree
Agree
Disagree
Strongly disagree
Women face more scrutiny and challenges when they run for office than men do.
Overturning Roe v. Wade would be a serious setback for women’s equal rights.
Ordinary people would be better than most elected officials at solving the country’s problems.
It would upset me if someone in my immediate family married a Trump supporter.
It would upset me if someone in my immediate family married a Biden supporter.
(For anyone who is married or living with a spouse/partner) Which of the following statements best describes the division of labor on household tasks, such as cleaning, laundry, and cooking?
Which statement best characterizes your childcare, or characterized it when your children lived at home?
For each pair, choose the statement that better describes you?
I have a lot of skeletons in my closet./I have a squeaky-clean background.
I don’t know enough about public policy to run for office./I am very knowledgeable about some areas of public policy.
My looks would be criticized if I ran for office./I have the look of a successful candidate.
My political views are out of sync with my community./My political views are representative of my community.
Someone like me would have a hard time running./Someone like me would have a leg up running for office.
I don’t have thick enough skin./I am pretty unflappable when criticized.
Appendix C Interviews with Potential Candidates
Throughout the book, we draw on more than 100 in-depth phone interviews with potential candidates from the 2021 sample and 200 interviews with respondents to the 2001 survey. These interviews provide a valuable addition to our main empirical approach, giving us potential candidates’ first-hand perspectives on their attitudes toward running for office, their perceptions of the political system, and the changes they’ve observed in women’s presence in the political arena.
This book draws primarily on the interview evidence gathered from the 2021 sample.Footnote 1 The last question of the survey potential candidates completed asked if they would be willing to participate in a follow-up phone interview. YouGov did not release these data to us, but in June 2023, they randomly selected 500 respondents – 250 women and 250 men – who expressed a willingness to be interviewed and sent us respondents’ first names and email addresses. We emailed the entire list and asked if they were still interested in participating in an interview. We heard from 148 respondents, 128 of whom were willing to be interviewed.
We scheduled the interviews throughout June and July 2023. After taking into account scheduling issues and unanswered calls, we ultimately conducted 101 semi-structured interviews in which we promised respondents anonymity. The following questions guided the interviews, which ranged from twenty-five to forty-five minutes in length (although not every respondent answered all of them):
1. Have you ever thought about running for office?
a. If they have, follow up on what office, when, why that was appealing to them. If it turns out that they have already run for office, find out which one, when, whether they won, and what that experience was like.
b. If they haven’t, find out why they think it has never occurred to them.
c. If it’s not clear that they have thought about running, ask if they are open to running in the future, what might make them do it, or why they wouldn’t.
2. When you were growing up, how political were your parents and friends? Did you talk about politics a lot? What’s your first political memory? Was politics a big part of your schooling? Did you watch the news regularly?
3. Were you ever encouraged as a child to think about running for office? What about by the time you were in college? (If they were, find out when and from whom, as well as what they thought about the suggestion – did they perceive it as serious?)
4. What about later in life? Has anyone ever suggested that you run for office as an adult? (If yes, find out when and from whom, as well as what they thought about the suggestion – did they perceive it as serious? Be sure to list all types of people who suggested they run.)
5. Do you think you’re qualified to run for office? Why or why not?
6. What qualifications do you think are most important for political candidates?
7. When you think about the last 10 or 20 years, have you noticed an increase in the number of women in politics? Regardless of their answer, ask why they think that.
8. We’ve been studying interest in running for office for many years and have consistently uncovered a big gender gap in political ambition. Men are much more interested than women are in running for office. This was true in our first study in 2001 and it is true in our most recent study. This was sort of surprising to us given all the changes to the political landscape since 2000. Does this surprise you? Any thoughts on why the gender gap hasn’t closed?
9. Do you think that it’s harder for women than men to succeed in politics? Why or why not?
10. Does seeing high-profile women in politics like Hillary Clinton, Nancy Pelosi, or Nikki Haley make you think the system is more open to women?
11. Is there anything you’d like to add that we haven’t discussed?
We took detailed notes and transcribed the interviews in real time as we spoke to potential candidates on the phone.
1 See Lawless and Fox (Reference Lawless and Fox2005) for the questionnaire that guided the 2003 interviews with potential candidates.
Appendix D Human Subjects Protocol
We commissioned YouGov to build our sample of potential candidates. Although our study received institutional review board (IRB) approval from the University of Virginia, YouGov also provided us with a twelve-page document, entitled “Survey Procedures: Information Related to the Treatment of Human Subjects,” that fully describes their sampling recruitment, compensation, and other protocols. The details that follow are taken directly from the March 2020 document YouGov provided (those were the practices in place when our survey was carried out).
In a nutshell, YouGov’s practices should mitigate standard concerns about engaging with human subjects: power, consent, deception, harm and trauma, confidentiality, and impact. The Citizen Political Ambition Study does not involve deception or the potential for harm or trauma as regularly conceived. In addition, we see no broad concerns about impact. While it’s true that the survey questions might prompt some citizens to think about running for office, civic engagement is a positive externality. Finally, the protocols put in place by YouGov – summarized in this appendix – should eliminate concerns regarding power, consent, and confidentiality.
YouGov Sample Procedures
According to YouGov, our sample is drawn from a proprietary opt-in survey panel, comprised of 1.2 million US residents who have agreed to participate in YouGov’s web surveys. At any given time, YouGov maintains numerous recruitment campaigns based on salient current events.
Panel members are recruited by various methods to help ensure diversity in the panel population. These include web advertising campaigns (public surveys), permission-based email campaigns, partner-sponsored solicitations, telephone-to-web recruitment (RDD-based sampling), and mail-to-web recruitment (voter registration-based sampling).
The primary method of recruitment for the YouGov panel – and the manner in which most of our respondents were solicited – is web advertising campaigns that target respondents through keyword searches. In the words of YouGov:
A search in Google may prompt an active YouGov advertisement inviting their opinion on the search topic. At the conclusion of the short survey respondents are invited to join the YouGov panel to directly receive and participate in additional surveys. All recruited members must pass through a double opt-in procedure, where respondents must confirm their consent again by responding to an email.
The database then checks to ensure that the newly recruited panelist is new and has a valid address.
YouGov augments the panel with difficult-to-recruit respondents by soliciting panelists in telephone and mail surveys. Respondents provide a working email address where they can receive an electronic invitation and confirm their consent and interest in receiving and participating in YouGov web surveys.
Participants are not paid to join the YouGov panel, but they do receive incentives through a loyalty program to take individual surveys.
YouGov Consent and Privacy Practices
Each respondent receives the following consent statement upon providing contact information and indicating an interest in receiving survey invitations from YouGov: “YouGov invites people to complete self-administered surveys via the web using a panel of respondents.” Panelists are provided the privacy policy when they voluntarily sign up and a link to this with each study request. Specifically, each invitation states that their participation is voluntary and confidential.
YouGov IRB Statement and Protection of Human Subjects
YouGov works with the Western IRB to ensure its research protocol and specific studies are consistent with Good Clinical Practices as defined under the US Food and Drug Administration (FDA) regulations and the International Conference on Harmonisation (ICH) guidelines. In addition, the YouGov research protocol has been reviewed and approved for Federalwide Assurance (FWA) by the Department of Health and Human Services (number FWA00010960). All members of the scientific research team have IRB training.
YouGov Incentives and Compensation
YouGov compensates participants through an incentive program in which suxrvey respondents receive “PollingPoints” they may redeem:
| Points | Reward |
|---|---|
| 25,000 | UNICEF Donation |
| 30,000 | $25 gift cards from AMC, Fandango, CVS, Regal, or Walgreens |
| 35,000 | $15 Amazon gift card or $25 Kmart gift card |
| 40,000 | $25 Foot Locker gift card |
| 45,000 | $25 Nike gift card |
| 57,500 | $50 FreshGift gift card |
| 60,000 | $50 gift card from Lowe’s, Chili’s, Applebee’s, GameStop, Groupon, iTunes, Sears, TJX, Best Buy, Bed, Bath & Beyond, Walmart, Foot Locker, Target, Macy’s, or Old Navy |
| 65,000 | $50 gift card from Nike, Amazon, Global Hotel, or Visa |
| 100,000 | $100 Visa Prepaid card, $100 UNICEF Donation, or $100 Amazon gift card |
Each panelist receives between 250 and 5,000 points to complete a survey. YouGov considers the survey reward policies and incentives “to serve as a genuine token of appreciation for YouGov panelists.”
Appendix E Variable Description and Coding in the Multivariate Analyses
| Variable | Range | Mean | Std. Dev. | Coding |
|---|---|---|---|---|
| Dependent variables | ||||
| Considered running for office | 0, 1 | 0.39 | 0.49 | Indicates whether respondent ever considered running for office (1) or not (0). |
| Took a concrete step | 0, 1 | 0.24 | 0.43 | Indicates whether respondent took at least one of the following steps that often precede a campaign – spoke to party leaders; discussed running with friends or family; discussed financial contributions with potential supporters; investigated how to get on the ballot; spoke to candidates about their experiences; attended a candidate training – (1) or not (0). |
| Recruited by political actor | 0, 1 | 0.18 | 0.38 | Indicates whether respondent ever received the suggestion to run for office from a party leader, elected official, or political activist (1) or not (0). |
| Ran for office | 0, 1 | 0.06 | 0.23 | Indicates whether respondent ever ran for office (1) or not (0). |
| Independent variables – Baseline indicators | ||||
|---|---|---|---|---|
| Gender (female) | 0, 1 | 0.51 | 0.50 | Indicates whether respondent identifies as a woman (1) or a man (0). |
| Education | 1–2 | 1.50 | 0.50 | Indicates whether respondent has a bachelor’s degree (1) or a postgraduate degree (2). |
| Income | 1–16 | 9.34 | 3.01 | Indicates respondent’s annual family income. Ranges from less than $10,000 (1) to $500,000 or more (16). |
| Black | 0, 1 | 0.12 | 0.32 | Indicates whether respondent identifies as Black (1) or not (0). |
| Latino | 0, 1 | 0.12 | 0.32 | Indicates whether respondent identifies as Latino (1) or not (0). |
| Asian | 0, 1 | 0.07 | 0.26 | Indicates whether respondent identifies as Asian (1) or not (0). |
| Married | 0, 1 | 0.64 | 0.48 | Indicates whether respondent is married or living with a partner (1) or not (0). |
| Children under 18 at home | 0, 1 | 0.34 | 0.47 | Indicates whether respondent has children under the age of 18 living at home (1) or not (0). |
| Birth year | 1936–1999 | 1974 | 13.31 | Indicates respondent’s year of birth. |
| Democrat | 0, 1 | 0.58 | 0.49 | Indicates whether respondent self-identifies as strong Democrat, Democrat, leaning Democrat (1) or not (0). |
| Republican | 0, 1 | 0.27 | 0.49 | Indicates whether respondent self-identifies as strong Republican, Republican, leaning Republican (1) or not (0). |
| Political efficacy | 1–5 | 3.30 | 1.16 | Indicates whether respondent thinks that “ordinary people would be better than most elected officials at solving the country’s problems.” Ranges from “strongly disagree” (1) to “strongly agree” (5). |
| Political activity | 0–7 | 2.67 | 1.82 | Indicates the number of the following activities respondent has engaged in during the last two years: voted in the presidential election; donated money to a political campaign; volunteered for a political campaign; attended a rally or protest; contacted an elected official; wrote or commented about politics on social media; served on a nonprofit board. |
| Political interest | 1–4 | 3.47 | 0.78 | Indicates how closely respondent follows politics. Ranges from “hardly at all” (1) to “most of the time” (4). |
| Independent variables – Traditional family roles | ||||
|---|---|---|---|---|
| Political discussions with parents | 1–4 | 2.55 | 1.08 | Indicates how frequently respondent discussed politics with parents when growing up. Ranges from “never” (1) to “frequently” (4). |
| Ran for office as a student | 0, 1 | 0.36 | 0.48 | Indicates whether respondent ran for office in high school and/or college (1) or not (0). |
| Majority of household tasks | 0, 1 | 0.21 | 0.41 | Indicates whether respondent is responsible for the majority of the household tasks (1) or not (0). |
| Majority of childcare | 0, 1 | 0.37 | 0.48 | Indicates whether respondent is responsible for the majority of the childcare (1) or not (0). |
| Encouraged by personal source | 0, 1 | 0.37 | 0.48 | Indicates whether respondent ever received the suggestion to run for office from a spouse, family member, colleague, or someone from a church, synagogue, or mosque (1) or not (0). |
| Independent variables – Masculinized ethos | ||||
|---|---|---|---|---|
| “Strong” Democrat or Republican | 0, 1 | 0.45 | 0.50 | Indicates whether respondent identifies as a “strong” Democrat or Republican (1) or not (0). |
| Recruited by political actor | 0, 1 | 0.18 | 0.38 | Indicates whether respondent ever received the suggestion to run for office from a party leader, elected official, or political activist (1) or not (0). |
| Independent variables – Gendered psyche | ||||
|---|---|---|---|---|
| Self-perceived qualifications | 1–4 | 2.52 | 1.08 | Indicates how qualified respondent feels to run for office. Ranges from “not at all qualified” (1) to “very qualified” (4). |
Table A1 Demographics of the candidate eligibility pool
| Women | Men | |
|---|---|---|
| Race | ||
| White | 63%Footnote * | 66% |
| Black | 14Footnote * | 10 |
| Latino | 13 | 11 |
| Asian | 7 | 8 |
| Other | 3 | 5 |
| Party Affiliation | ||
| Democrat (including leaners) | 64Footnote * | 53 |
| Independent | 12Footnote * | 16 |
| Republican (including leaners) | 24Footnote * | 31 |
| Household Income | ||
| Less than $70,000 | 32Footnote * | 21 |
| $70,000–$99,999 | 24 | 22 |
| $100,000–$149,999 | 24Footnote * | 28 |
| $150,000 and above | 21Footnote * | 29 |
| Profession | ||
| Lawyer | 11 | 12 |
| Business executive/owner | 9Footnote * | 15 |
| Educator | 22Footnote * | 18 |
| Other | 58 | 55 |
| Education | ||
| Bachelor’s degree | 50 | 49 |
| Post-graduate degree | 50 | 51 |
| Mean age | 45.5Footnote * | 48.8 |
| N | 2,580 | 2,496 |
Note:
* indicates that the gender gap is significant at p < 0.05.
Table A2 The gender gap in political ambition, by demographic group
| Considered Running for Office | |||
|---|---|---|---|
| Women | Men | Gap | |
| Race | |||
| White | 36% | 52% | 16 points |
| Black | 22 | 34 | 12 |
| Latino | 26 | 44 | 18 |
| Asian | 15 | 21 | 6 |
| Party Affiliation | |||
| Democrat | 34 | 47 | 13 |
| Republican | 32 | 51 | 19 |
| Independent | 22 | 41 | 19 |
| Household Income | |||
| Less than $70,000 | 24 | 42 | 18 |
| $70,000–$99,999 | 31 | 48 | 17 |
| $100,000–$149,999 | 36 | 49 | 13 |
| $150,000 and above | 39 | 51 | 12 |
| Age | |||
| Under 40 | 29 | 47 | 18 |
| Ages 40–59 | 33 | 47 | 14 |
| 60 and over | 33 | 48 | 15 |
| Parental Status | |||
| No children at home | 32 | 46 | 14 |
| Children at home | 31 | 49 | 18 |
| Marital Status | |||
| Married | 35 | 49 | 15 |
| Not married | 27 | 44 | 17 |
| N | 2,580 | 2,496 | 15 points |
Notes: Entries represent the percentage of women and men within each demographic group who have considered running for office. The final column presents the size of the gender gap in political ambition in each demographic category. The gender gap is significant at p < 0.05 in every comparison.
Table A3 Baseline models of candidate emergence
| Considered Running | Took a Concrete Step | Ran for Office | |
|---|---|---|---|
| Gender (female) | ˗0.649Footnote * (0.070) | ˗0.664Footnote * (0.081) | ˗0.536Footnote * (0.145) |
| Education | 0.478Footnote * (0.071) | 0.465Footnote * (0.082) | 0.067 (0.144) |
| Income | ˗0.005 (0.013) | 0.001 (0.015) | 0.022 (0.026) |
| Black | ˗0.376Footnote * (0.117) | 0.228 (0.132) | 0.597Footnote * (0.245) |
| Latino | ˗0.086 (0.110) | 0.439Footnote * (0.121) | 0.324 (0.226) |
| Asian | ˗0.998Footnote * (0.158) | ˗0.642Footnote * (0.192) | 0.281 (0.402) |
| Married | 0.197Footnote * (0.082) | 0.232Footnote * (0.095) | 0.099 (0.171) |
| Children at home | 0.040 (0.080) | 0.241Footnote * (0.091) | 0.397Footnote * (0.162) |
| Birth year | 0.015Footnote * (0.003) | 0.015Footnote * (0.003) | ˗0.034Footnote * (0.006) |
| Democrat | ˗0.040 (0.107) | ˗0.049 (0.128) | 0.811Footnote * (0.285) |
| Republican | 0.203 (0.114) | 0.339Footnote * (0.135) | 0.935Footnote * (0.292) |
| Political efficacy | 0.125Footnote * (0.031) | 0.172Footnote * (0.036) | 0.026 (0.062) |
| Political activity | 0.454Footnote * (0.022) | 0.487Footnote * (0.025) | 0.221Footnote * (0.041) |
| Political interest | 0.236Footnote * (0.055) | 0.109 (0.066) | ˗0.360Footnote * (0.120) |
| Constant | ˗33.060Footnote * (5.807) | ˗34.046Footnote * (6.699) | 63.844Footnote * (12.125) |
| Adjusted R2 | 0.271 | 0.246 | 0.088 |
| N | 4,594 | 4,417 | 1,838 |
Notes: Entries represent logistic regression coefficients (and standard errors) predicting whether a respondent ever considered running for office, took a concrete step that often precedes a candidacy, or actually ran for office. The analysis predicting an actual candidacy is restricted to the subsample of respondents who considered running. Children at home include only those who are under the age of 18.
* p < 0.05.
Table A4 Considering a candidacy: the impact of a politicized upbringing and external support
| (1) | (2) | |
|---|---|---|
| Baseline Indicators | ||
| Gender (female) | ˗0.625 (0.072)Footnote * | ˗0.570 (0.078)Footnote * |
| Education | 0.411 (0.073)Footnote * | 0.293 (0.079)Footnote * |
| Income | ˗0.018 (0.013) | ˗0.036 (0.014)Footnote * |
| Black | ˗0.503 (0.121)Footnote * | ˗0.758 (0.133)Footnote * |
| Latino | ˗0.203 (0.113) | ˗0.375 (0.124)Footnote * |
| Asian | ˗1.018 (0.161)Footnote * | ˗0.904 (0.169)Footnote * |
| Married | 0.202 (0.084)Footnote * | 0.203 (0.091)Footnote * |
| Children under 18 at home | 0.038 (0.082) | ˗0.006 (0.089) |
| Birth year | 0.014 (0.003)Footnote * | 0.010 (0.003)Footnote * |
| Democrat | ˗0.062 (0.109) | 0.030 (0.118) |
| Republican | 0.166 (0.116) | 0.162 (0.126) |
| Political efficacy | 0.113 (0.032)Footnote * | 0.073 (0.035)Footnote * |
| Political activity | 0.411 (0.023)Footnote * | 0.292 (0.025)Footnote * |
| Political interest | 0.193 (0.057)Footnote * | 0.162 (0.062)Footnote * |
| Politicized Upbringing | ||
| Political discussions with parents | 0.169 (0.034)Footnote * | 0.109 (0.037)Footnote * |
| Ran for office as a student | 0.826 (0.072)Footnote * | 0.595 (0.079)Footnote * |
| External Support | ||
| Encouraged by personal source | 1.885 (0.079) | |
| Constant | ˗31.173 (5.958)Footnote * | ˗23.473 (6.450)Footnote * |
| Pseudo-R2 | 0.310 | 0.440 |
| N | 4,592 | 4,592 |
Note: Entries represent logistic regression coefficients (and standard errors) predicting whether a respondent ever considered running for office.
* p < 0.05.
Table A5 Considering a candidacy: the impact of family responsibilities
| (1) | (2) | (3) | |
|---|---|---|---|
| Baseline Indicators | |||
| Gender (female) | ˗0.493 (0.092)Footnote * | ˗0.548 (0.103)Footnote * | ˗0.458 (0.116)Footnote * |
| Education | 0.450 (0.091)Footnote * | 0.435 (0.093)Footnote * | 0.468 (0.105)Footnote * |
| Income | ˗0.014 (0.017) | ˗0.016 (0.018) | ˗0.021 (0.020) |
| Black | ˗0.679 (0.157)Footnote * | ˗0.790 (0.150)Footnote * | ˗0.874 (0.193)Footnote * |
| Latino | ˗0.352 (0.139)Footnote * | ˗0.415 (0.143)Footnote * | ˗0.507 (0.161)Footnote * |
| Asian | ˗1.350 (0.216)Footnote * | ˗1.182 (0.236)Footnote * | ˗1.214 (0.251)Footnote * |
| Married | – | 0.066 (0.122) | – |
| Children at home | 0.092 (0.099) | ˗0.039 (0.125) | 0.203 (0.148) |
| Birth year | 0.015 (0.004)Footnote * | 0.016 (0.005)Footnote * | 0.009 (0.006) |
| Democrat | ˗0.028 (0.135) | ˗0.076 (0.141) | ˗0.076 0(.157) |
| Republican | 0.099 (0.142) | 0.036 (0.144) | 0.004 (0.160) |
| Political efficacy | 0.106 (0.040)Footnote * | 0.109 (0.041)Footnote * | 0.118 (0.046)Footnote * |
| Political activity | 0.387 (0.028)Footnote * | 0.399 (0.029)Footnote * | 0.388 (0.032)Footnote * |
| Political interest | 0.177 (0.071)Footnote * | 0.149 (0.073)Footnote * | 0.195 (0.083)Footnote * |
| Politicized Upbringing | |||
| Political discussions with parents | 0.194 (0.042)Footnote * | 0.200 (0.043)Footnote * | 0.178 (0.048)Footnote * |
| Ran for office as a student | 0.957 (0.090)Footnote * | 0.967 (0.092)Footnote * | 0.999 (0.104)Footnote * |
| Family Responsibilities | |||
| Majority of household tasks | ˗0.104 (0.095) | – | ˗0.139 (0.120) |
| Majority of childcare | – | 0.083 (0.106) | 0.210 (0.128) |
| Constant | ˗32.853 (8.119)Footnote * | ˗34.702 (10.119)Footnote * | ˗22.125 (11.652)Footnote * |
| Pseudo-R2 | 0.313 | 0.314 | 0.314 |
| N | 2,932 | 2,807 | 2,209 |
Notes: Entries represent logistic regression coefficients (and standard errors) predicting whether a respondent ever considered running for office. The analysis of household responsibilities – equation 1 – is restricted to respondents who are married or living with a partner. The analysis of childcare responsibilities – equation 2 – is restricted to respondents with children. The analysis of the joint effect of household and childcare responsibilities – equation 3 – is restricted to respondents who are married or living with a partner and who have children.
* p < 0.05.
Table A6 Considering a candidacy: the impact of party
| Baseline Indicators | |
| Gender (female) | ˗0.649Footnote * (0.071) |
| Education | 0.478Footnote * (0.071) |
| Income | ˗0.005 (0.013) |
| Black | ˗0.376Footnote * (0.117) |
| Latino | ˗0.086 (0.110) |
| Asian | ˗0.998Footnote * (0.158) |
| Married | 0.197Footnote * (0.082) |
| Children under 18 at home | 0.040 (0.080) |
| Birth year | 0.015Footnote * (0.003) |
| Democrat | ˗0.041 (0.114) |
| Republican | 0.203 (0.118) |
| Political efficacy | 0.125Footnote * (0.031) |
| Political activity | 0.454Footnote * (0.022) |
| Political interest | 0.236Footnote * (0.055) |
| Strength of Partisanship | |
| “Strong” Democrat or Republican | 0.000 (0.076) |
| Constant | ˗33.060Footnote * (5.809) |
| Pseudo-R2 | 0.271 |
| N | 4,594 |
Note: Entries represent logistic regression coefficients (and standard errors) predicting whether a respondent ever considered running for office.
* p < 0.05.
Table A7 Political recruitment by a political actor
| Gender (female) | ˗0.532Footnote * (0.090) |
| Education | 0.358Footnote * (0.091) |
| Income | 0.029 (0.017) |
| Black | 0.322Footnote * (0.150) |
| Latino | 0.241 (0.145) |
| Asian | ˗0.497Footnote * (0.225) |
| Married | ˗0.197 (0.104) |
| Children under 18 at home | 0.221Footnote * (0.102) |
| Birth year | ˗0.007 (0.004) |
| Democrat | ˗0.220 (0.141) |
| Republican | 0.072 (0.150) |
| Political efficacy | 0.090Footnote * (0.040) |
| Political activity | 0.592Footnote * (0.027) |
| Political interest | 0.122 (0.082) |
| Constant | 8.424Footnote * (7.331) |
| Pseudo-R2 | 0.276 |
| N | 4,594 |
Note: Entries represent logistic regression coefficients (and standard errors) predicting whether a respondent reported ever being recruited to run for office by an elected official, party leader, or nonelected political activist.
* p < 0.05.
Table A8 Considering a candidacy: the impact of political recruitment
| (1) | (2) | |
|---|---|---|
| Baseline Indicators | ||
| Gender (female) | ˗0.521 (0.080)Footnote * | ˗0.478 (0.085)Footnote * |
| Education | 0.286 (0.081)Footnote * | 0.282 (0.081)Footnote * |
| Income | ˗0.040 (0.015)Footnote * | ˗0.040 (0.015)Footnote * |
| Black | ˗0.820 (0.138)Footnote * | ˗0.821 (0.138)Footnote * |
| Latino | ˗0.418 (0.127) | ˗0.420 (0.127) |
| Asian | ˗0.893 (0.172)Footnote * | ˗0.883 (0.172)Footnote * |
| Married | 0.256 (0.094)Footnote * | 0.256 (0.094)Footnote * |
| Children under 18 at home | ˗0.045 (0.091) | ˗0.044 (0.091) |
| Birth year | 0.013 (0.003)Footnote * | 0.013 (0.003)Footnote * |
| Democrat | 0.079 (0.121) | 0.076 (0.121) |
| Republican | 0.175 (0.129) | 0.171 (0.129) |
| Political efficacy | 0.074 (0.036)Footnote * | 0.073 (0.036)Footnote * |
| Political activity | 0.215 (0.026)Footnote * | 0.217 (0.026)Footnote * |
| Political interest | 0.179 (0.063)Footnote * | 0.181 (0.063)Footnote * |
| Politicized Upbringing | ||
| Political discussions with parents | 0.089 (0.038)Footnote * | 0.090 (0.038)Footnote * |
| Ran for office as a student | 0.504 (0.081)Footnote * | 0.503 (0.081)Footnote * |
| External Support/Recruitment | ||
| Encouraged by personal source | 1.721 (0.081)Footnote * | 1.723 (0.081)Footnote * |
| Encouraged by political actor | 1.508 (0.118)Footnote * | 1.660 (0.161)Footnote * |
| Gender Footnote * Encouraged by political actor | – | ˗0.331 (0.231) |
| Constant | ˗29.354 (6.646)Footnote * | ˗29.201 (6.651)Footnote * |
| Pseudo-R2 | 0.475 | 0.475 |
| N | 4,592 | 4,592 |
Note: Entries represent logistic regression coefficients (and standard errors) predicting whether a respondent ever considered running for office.
* p < 0.05.
Table A9 Considering a candidacy: the impact of self-perceived qualifications
| (1) | (2) | |
|---|---|---|
| Baseline Indicators | ||
| Gender (female) | ˗0.415 (0.082)Footnote * | ˗0.559 (0.232)Footnote * |
| Education | 0.180 (0.083)Footnote * | 0.181 (0.083)Footnote * |
| Income | ˗0.046 (0.015)Footnote * | ˗0.046 (0.015)Footnote * |
| Black | ˗0.873 (0.140)Footnote * | ˗0.873 (0.140)Footnote * |
| Latino | ˗0.497 (0.129)Footnote * | ˗0.497 (0.129)Footnote * |
| Asian | ˗0.880 (0.177)Footnote * | ˗0.884 (0.177)Footnote * |
| Married | 0.236 (0.095)Footnote * | 0.238 (0.095)Footnote * |
| Children under 18 at home | ˗0.106 (0.093) | ˗0.106 (0.093) |
| Birth year | 0.018 (0.003)Footnote * | 0.018 (0.003)Footnote * |
| Democrat | 0.118 (0.122) | 0.118 (0.122) |
| Republican | 0.173 (0.130) | 0.176 (0.130) |
| Political efficacy | 0.028 (0.037) | 0.029 (0.037) |
| Political activity | 0.199 (0.026)Footnote * | 0.198 (0.026)Footnote * |
| Political interest | 0.114 (0.064) | 0.114 (0.064) |
| Politicized Upbringing | ||
| Political discussions with parents | 0.071 (0.038) | 0.070 (0.038) |
| Ran for office as a student | 0.415 (0.083)Footnote * | 0.416 (0.083)Footnote * |
| External Support/Recruitment | ||
| Encouraged by personal source | 1.605 (0.083)Footnote * | 1.603 (0.083)Footnote * |
| Encouraged by political actor | 1.396 (0.120)Footnote * | 1.398 (0.120)Footnote * |
| Qualifications | ||
| Self-perceived qualifications to run | 0.397 (0.044)Footnote * | 0.371 (0.059)Footnote * |
| Gender Footnote * Self-perceived qualifications | – | 0.054 (0.081) |
| Constant | ˗38.090 (6.788)Footnote * | ˗38.070 (6.788)Footnote * |
| Pseudo-R2 | 0.491 | 0.491 |
| N | 4,589 | 4,589 |
Note: Entries represent logistic regression coefficients (and standard errors) predicting whether a respondent ever considered running for office.
* p < 0.05.
Table A10 Who runs for office?
| Baseline Indicators | |
| Gender (female) | ˗0.160 (0.155) |
| Education | ˗0.138 (0.157) |
| Income | 0.002 (0.027) |
| Black | 0.322 (0.265) |
| Latino | 0.128 (0.244) |
| Asian | 0.514 (0.432) |
| Married | 0.248 (0.182) |
| Children under 18 at home | 0.286 (0.176) |
| Birth year | ˗0.026 (0.007)Footnote * |
| Democrat | 0.860 (0.298)Footnote * |
| Republican | 0.896 (0.004)Footnote * |
| Political efficacy | ˗0.033 (0.064) |
| Political activity | 0.042 (0.046) |
| Political interest | ˗0.491 (0.130)Footnote * |
| Politicized Upbringing | |
| Political discussions with parents | 0.015 (0.071) |
| Ran for office as a student | ˗0.027 (0.152) |
| External Support/Recruitment | |
| Encouraged by personal source | ˗0.359 (0.171)Footnote * |
| Encouraged by political actor | 1.802 (171)Footnote * |
| Qualifications | |
| Self-perceived qualifications to run | 0.603 (0.107)Footnote * |
| Constant | 47.014 (13.009)Footnote * |
| Pseudo-R2 | 0.251 |
| N | 1,835 |
Notes: Sample is restricted to respondents who considered running for office. Entries represent logistic regression coefficients (and standard errors) predicting whether a respondent ever ran for office.
* p < 0.05.
