Sage Journals: Discover world-class research

Abstract

Does descriptive representation lead people to evaluate elected officials and their institutions more favorably? Does it improve political efficacy and engagement? We report findings from a survey experiment that uses treatments drawn from respondents’ real political context—elected officials who make policy in respondents’ county of residence. Specifically, we present a sample of Cook County residents with a member of the Cook County Board of Commissioners—who may or may not “match” the respondents gender or ethnoracial identity—to assess whether signaling that the respondent is “descriptively represented” on the Board affects their assessments of the Board and other attitudinal outcomes. Our pre-registered design positions us to identify effects of roughly one-eighth of a standard deviation in our full sample, but the estimated effects of signals of ethnoracial- and gender-based descriptive representation are null across the five outcomes we consider. In pre-registered exploratory analysis re-estimating effects by subgroup, we find evidence that suggests that descriptive representation affects some attitudes among women and Black respondents. This said, the effects we find in these groups are modest in magnitude, scattered, and, in most cases, statistically indistinguishable from those that emerge in other groups.

Keywords

Descriptive representation local politics experiment pre-registered gender race

Introduction

Although scholars often conceive of and evaluate representation in substantive terms (e.g., Gilens and Page 2014; Miller and Stokes 1963; Stimson et al., 1995), the concept of representation is complex (Mansbridge 2003; Pitkin 1967; Rehfeld 2009) and a rich body of research has considered the potential importance of descriptive representation (Mansbridge 1999; Pitkin 1967). Representatives’ descriptive characteristics affect their behavior and, ultimately, substantive policy outcomes (e.g., Dittmar et al., 2018; Juenke and Preuhs 2012; Lowande et al., 2019; Reingold et al., 2020). They may also affect how people evaluate and engage with elected officials and the political system (Mansbridge 1999). In this paper, rather than proposing novel theory regarding the effects of descriptive representation, we offer a practical, empirical supplement to existing research on how descriptive representation affects attitudes about political institutions and elected officials. We make four contributions.

First, using a sample of residents of Cook County, IL¹ we combine the internal validity of an experimental design with the external validity benefits of presenting participants with real political figures who make policy on their behalf. We take advantage of the diversity of the Cook County Board of Commissioners—which includes both men and women who identify as White, Black, and Hispanic—to present respondents with a Commissioner who may or may not share their own ethnoracial or gender identity. This political body is also relatively obscure.² Few respondents are likely to have well-formed pre-existing opinions about the Board or its commissioners. Additionally, the relative obscurity of the Board and its role means that respondents’ substantive preferences regarding the Board’s actions are likely to be uncrystallized (Mansbridge 1999). Thus, having a group member with a “seat at the table” may be viewed as particularly valuable (Dittmar et al., 2018). In concert with claims that descriptive representation may be particularly important in local politics (Marschall and Ruhil 2007), this would seem to present a “most likely” case for identifying descriptive representation effects.

Our second contribution is that we consider a range of attitudinal outcomes: 1) evaluations of the board’s past performance, (2) trust in commissioners to make good choices in the future, (3) a summary rating of efficacy, (4) interest in the next Board election, and (5) inclination to support the commissioner presented. Third, we explore the possibility that the effects of descriptive affinity vary across groups: analysis facilitated by the ethnoracial diversity of Cook County’s population, which yields a sufficient number of White (non-Hispanic; 41%), Black (24%), and Hispanic (26%) respondents to allow us to detect effects smaller than one-third of a standard deviation in each group. Finally, to our knowledge, ours is one of the first studies of the effects of descriptive representation on public attitudes to pre-register an analysis plan (see SM A2).³ This practice mitigates concerns about the “file drawer” problem and researchers’ “degrees of freedom” (Franco et al., 2014; Gerber et al., 2010; Wicherts et al., 2016).

Descriptive representation and public attitudes

Observational research on the attitudinal consequences of descriptive representation has yielded important insights regarding how elected officials’ race and gender affect how people evaluate individual representatives, as well as the institutions in which they serve. Some findings suggest that racial and gender affinity can affect how voters evaluate candidates (Costa and Schaffner 2018). However, these effects are often strongly conditioned by—and in some cases largely explained by—partisanship and ideology (e.g., Ansolabehere and Fraga 2016; Badas and Stauffer 2018; Dolan 2008).

Other studies find that descriptive representation affects levels of trust and political efficacy (Abney and Hutcheson Jr. 1981; Atkeson and Carrillo 2007; Banducci et al., 2004; Gay 2002; Merolla et al., 2013; Pantoja and Segura 2003; Sanchez and Morin 2011; Stauffer 2021; Torres 2006; Ulbig 2007; West 2017). For example, Bowen and Clark (2014) find that constituents whose representatives in the House share their race or ethnicity perceive those legislators to be more responsive—patterns that tend to be more pronounced among non-White Americans. However, we note that other studies fail to find a link between descriptive representation and efficacy (Snagovsky et al., 2020).

Descriptive representation may also enhance political engagement, participation, and knowledge (Atkeson 2003; Campbell and Wolbrecht 2006; Gleason and Stout 2014; Hansen 1997; Marschall and Ruhil 2007; Reingold and Harrell 2010; Rocha et al., 2010; Tate 2004). However, some of the effects reported across these studies are conditional. For example, the relationship between gender-based descriptive representation and political knowledge and participation appears to be stronger earlier in the life cycle (Dassonneville and McAllister 2018; Wolbrecht and Campbell 2007). Fairdosi and Rogowski (2015) and Washington (2006) find that the presence of Black candidates on the ballot can increase turnout, but only when those candidates are Democrats. Griffin and Keane (2006) find that liberal Black voters are more likely to turn out when they are descriptively represented, but that descriptive representation decreases turnout among moderate and conservative Black Americans. Hansen (1997) finds a relationship between the presence of women candidates and political efficacy and engagement in some years, but not others. Other observational studies simply fail to find a relationship between descriptive representation and this class of outcomes (e.g., Costa and Schaffner 2018; Lawless 2004).

One limitation of these studies is that they face threats to internal validity. Perhaps the thorniest is that being descriptively represented by elected officials is the product of candidates’ and voters’ strategic choices (Espírito-Santo and Verge 2017). For example, Bobo and Gilliam (1990) find that Black respondents living in a city with a Black mayor report higher levels of participation. This may reflect a dynamic where descriptive representation encourages participation, or one where politically active Black voters are more likely to elect a Black mayor. These challenges have led other scholars to study the effects of descriptive representation by taking advantage of discontinuities (e.g., Broockman 2014) and experimental designs that obviate concerns about omitted variables or reverse causality.

Some experimental studies use treatments that present participants with information on the descriptive characteristics of candidates running for office (Costa and Wallace 2021; Wolak 2015), the composition of hypothetical or actual decision-making bodies (Clayton et al., 2019; Hayes and Hibbing 2017; Scherer and Curry 2010), or highlight communications from a particular candidate to make descriptive representation salient (Bonneau and Kristin 2020). Others ask participants to evaluate hypothetical candidates (Jones 2016) or judicial nominees (Kaslovsky et al., 2021). These studies have offered novel insights into the consequences of descriptive representations that, like their observational counterparts, are often conditional. For example, Costa and Wallace (2021) report that information about increasing numbers of women running for office increased ambition among women, but had more nuanced effects on vote choice and reports of political efficacy. Bonneau and Kristin (2020) found that treating participants with ads from Hillary Clinton’s campaign increased ambition among some women who supported her, but decreased ambition among other groups. Wolak (2015) found that a story that varied the gender composition of candidates in Pennsylvania failed to affect women’s political interest or engagement.

These studies address concerns about internal validity. However, they typically either use hypothetical or distant political elites or treatments that focus on broad trends in political elites’ characteristics. An additional concern is that, in some cases, patterns of apparent conditionality may be the product of exploratory analysis where null interactions go unreported. Here we build on the strengths of existing experimental work with an eye to enhancing the external validity of our findings. The stimuli we use are drawn from a pool of real political elites who most respondents are unlikely to know much about, but who are actively making policies in the county where respondents reside. All analyses we report are pre-registered.

For whom does descriptive representation matter?

Existing work suggests that descriptive representation may not matter equally for all groups in society and there is reason to expect it to be particularly important for those who identify with groups that have historically faced political disadvantages or subordination. For example, Mansbridge (1999) notes that “[a] history of dominance and subordination typically breeds inattention, even arrogance, on the part of the dominant group and distrust on the part of the subordinate group.” Thus, demographic groups like women and Black Americans may view representation by those who share their characteristics as especially beneficial.

Some existing empirical evidence supports this expectation. For example, Costa and Schaffner (2018) find that having a woman representative affects attitudes among women, but not men (though, surprisingly, they find that women are less likely to report having contacted their representative when that representative is a woman). Similarly, in their experiment, Hayes and Hibbing (2017) find that descriptive representation affects attitudes among Black respondents as expected, but not among White participants. This said, other studies find evidence that representatives’ descriptive characteristics can also affect those who identify as men or White (Jones 2016). Per our pre-analysis plan, below we estimate treatment effects in our full sample, as well as separately by gender and racial subgroups.

Research design

The experiment was embedded in a survey of Cook County residents, fielded January 12–17, 2023. As noted above, this context is ideal in that both the Cook County population and Cook County Board are demographically diverse, allowing us to estimate the effects of descriptive representation tied to gender and ethnoracial identities for men, women, and three ethnoracial groups. Respondents were recruited by Dynata to be demographically representative of the Cook County population. Following our pre-analysis plan, we excluded three of the 1202 respondents who completed the survey because they did not provide responses to all demographic items (see Table SM A2 for sample characteristics) and the experimental outcome measures. The median respondent encountered the experiment 3.5 minutes into the survey (median total survey completion time = 8.4 minutes).

Respondents were assigned to a condition that highlighted one of 12 Commissioners serving on the Cook County Board.4 The experiment leverages the descriptive diversity of the Board—these 12 commissioners fill six cells defined by crossing ethnoracial identity (White, Black, Hispanic) with gender (Man, Woman), with some cells including multiple Commissioners (photos are presented in Table SM A3). Respondents were assigned to one of the six cells with equal probability, and presented with one Commissioner from that cell.

The experiment began with the following text, which was accompanied by the photo of the randomly assigned Commissioner:

[COMMISSIONER NAME] currently serves on the Cook County Board of Commissioners. The Board manages how public health services are provided in Cook County, allocates funding for the Sheriff’s department, and oversees the maintenance of Cook County’s highways and forest preserves. The board is made up of 17 Commissioners like [COMMISSIONER NAME].

A series of seven questions asked respondents to: (1) rate the Board’s performance of the past year (Board Job Rating), (2) indicate their level of trust in “Commissioners like [COMMISSIONER NAME]” to make good choices in the coming year (Trust Commissioner), (3) indicate how likely they would be to “contact a Commissioner like [COMMISSIONER NAME],” as well as how likely they thought it was that they’d be listened to (averaged to create Efficacy measure), (4) indicate their interest in following the next Board election (Interest in Election), and (5) indicate how likely they would be to contribute to the Commissioner’s campaign and vote for them if they ran again (averaged to create Intent to Support measure). Images of the randomly assigned commissioner accompanied all questions except (4) and (5). We illustrate what this would look like for a respondent completing the survey on their phone in Figure SM A2. Full question wording is reported in Figure SM A3. All outcomes were rescaled to have a mean of 0 and standard deviation of 1.⁵

Analysis

Per our pre-analysis plan, we estimate a series of regression models predicting each of the five outcomes with two indicators, one for (A) whether the Commissioner’s ethnoracial identity matched an identity reported by the respondent⁶ and one for (B) whether the Commissioner’s gender matched the respondent’s gender. We control for respondent age, education, income, income refusal indicator, gender, and ethnoracial ID (indicators for each option that could be selected).7 We account for the fact that our design uses real elected officials who respondents may broadly view as more competent/attractive/etc. by including indicators for each Commissioner in our models (we report the average outcomes for each Commissioner in Table SM A4).

We report the direct effects of the treatments in Figure 1 (regression models reported in Table SM A5). Although our design positions us to identify effects of roughly one-eighth of a standard deviation, none of the 10 treatment effects reaches conventional thresholds of statistical significance. The estimated effects range from −0.117 to 0.082 standard deviations, with an average of 0.016; associated p-values range from 0.082 to 0.920, with an average of 0.496.8 Expressed in units of the unstandardized outcome scales, the largest estimated effect in the expected direction is the 0.08 unit effect of am “ethnoracial ID match” on reported efficacy (p = .211).

Figure 1.

Direct Treatment Effects. Markers indicate effects of treatment commissioner’s characteristic “matching” respondent characteristic. Outcomes scaled to have means of 0 and standard deviations of 1. Whiskers are 95% confidence intervals.

Effects by subgroups

As discussed above, it is possible that the effects of descriptive representation vary across groups. Per our pre-analysis plan, in Figure 2, we report effects from models estimated separately for respondents who identified as women, men, White, Black, and Hispanic.9 To ensure comparability across groups, we exclude respondents who identified as a gender other than a man or woman, as well as those who did not identify as White, Black, or Hispanic. We emphasize that this analysis was preregistered as exploratory.10 This decision was rooted in our recognition that identifying interaction effects—particularly modest conditionality—requires substantial statistical power that our design simply does not have. We encourage readers to keep the exploratory nature of this analysis (as well as the large number of potential hypothesis tests in play) in mind as they consider the evidence and tests of statistical significance reported in this section.

Figure 2.

Treatment Effects by Subgroup. Markers indicate effects of treatment commissioner’s characteristic “matching” respondent characteristic. Outcomes scaled to have means of 0 and standard deviations of 1. Whiskers are 95% confidence intervals.

Signals of each type of descriptive representation appear to affect some outcomes among women respondents. Women respondents whose reported ethnoracial identity matched the commissioner’s rated the board more favorably (p = .040) and reported higher efficacy (p = .061). Each effect is approximately .15 standard deviations. Women respondents presented with a woman commissioner also rated the board’s performance (p = .003) and their trust in the board (p = .014) going forward approximately .20 standard deviations more favorably than those presented with a commissioner who was a man. These four point estimates are substantively small, amounting to approximately 3–5 percent of the ranges of the unstandardized outcome measures. Formal tests pooling the data and interacting the treatment indicators with indicators for respondents’ race and gender indicate that the “gender match” effects among women are statistically distinguishable from those among respondents who identified as men, but the “ethnoracial identity match” effects are not (see Table SM A7). Figure 2 also suggests a statistically significant, negative effect of a “race match” on interest in the next election among respondents who identified as men. This effect is statistically significant (p = .011; p = .059 for test of equality of effects between men and women respondents).

None of the 10 estimated treatment effects reaches conventional thresholds of statistical significance among the White or Hispanic subgroups. However, among Black respondents, the coefficient on the “race match” indicator is positive and yields p-values less than 0.05 in the Rating and Efficacy models (estimated effects of just over 0.27 standard deviations in each case—approximately 6–7 percent of the ranges of the raw outcome measures). We also find a negative, statistically significant coefficient in the Interest model. However, all formal tests of the equality of treatment estimates across ethnoracial groups (Table SM A7) fall short of conventional thresholds of statistical significance.¹¹

Discussion

Our experiment leveraged Cook County’s diversity to present participants with real elected policy-makers with varied descriptive characteristics, whose decisions have consequences for residents. In our pooled analysis, the estimated effects of our treatments are small and fall short of conventional thresholds for statistical significance. However, our pre-registered exploratory subgroup analysis suggests that descriptive representation may be consequential for those who identify as either women or Black. These patterns are consistent with a dynamic where individuals who identify with groups that have been—and continue to be—under-represented in elected office attach particular value to being descriptively represented. However, these effects only emerge for a subset of our five outcomes, and the estimated effect of a “race match” on interest among Black respondents is unexpectedly negative.

Like all research, our design has limitations. Although our design positioned us to identify modest treatment effects in each subgroup, confidently identifying differences in effect sizes across groups requires substantially more statistical power (Gelman 2018). Thus, although we find statistically significant treatment effects in some groups and point estimates across groups suggest that groups responded differently to the treatments, in many cases we cannot reject the null hypothesis of “no difference in treatment effects across groups.” There are two exceptions: the effects of shared gender on women respondents’ ratings of the Board and trust in the Commissioner are distinguishable from the effects of a “gender match” among men. We fall even further short of having sufficient statistical power to investigate the possibility that the effects of descriptive representation are conditioned by intersections between gender, ethnoracial, and other identities (Reingold et al., 2020).

Other limitations of our design are tied to the inevitable tradeoffs researchers must make when crafting experiments. Our focus on Cook County has the substantial advantage of positioning us to present an ethnically diverse pool of respondents with stimuli drawn from a diverse pool of real elected officials. This said, we cannot fully “unbundle” Commissioners’ ethnoracial and gender identities from other attributes respondents may have inferred from the photos we used. It is also important to note that Cook County Commissioners are elected from single-member districts. Thus, although respondents were presented with a real official who makes policy for their county, in most cases that elected official was not their representative and we did not present respondents with a justification regarding why they saw one particular Commissioner. This may have attenuated our estimated treatment effects. Future research could identify contexts where a political body is comprised of a diverse pool of at-large members. This would open the door to randomly varying the characteristics of officials that have a more direct representation relationship with respondents. An additional limitation that stems from our experimental context is that Cook County is overwhelmingly Democratic. This said, existing work indicates that descriptive representation should be particularly consequential in this context.

These limitations aside, our study offers important new evidence regarding the extent to which descriptive representation shapes public attitudes. Most published work finds that descriptive representation affects attitudes. Our findings—drawn from an experimental design that followed a pre-registered analysis plan—suggest that, even in the controlled context of a survey experiment, the effect of descriptive representation on political attitudes is limited. The effects we identify in our pooled analysis are null. Although some statistically significant effects emerge among historically under-represented subgroups, these effects are modest in size, only emerge for some outcomes and, for the most part, are statistically indistinguishable from the estimated effects that emerge in other groups.

More broadly, the present research underscores the value of both pre-registration and replication. Existing work often posits reasonable but nuanced theories regarding the conditions under which descriptive representation “matters.” Patterns indicating that the effects of the presence of Black candidates on the ballot depends on those candidates’ party affiliation or that the effects of descriptive representation on Black voters depends on those voters’ ideological leanings may reflect the true state of the world. However, identifying this type of conditionality requires a level of statistical power that is rarely available to researchers. This opens the door to a risk of false positives. The more nuanced theories regarding when and how descriptive representation matters become, the more crucial it will be for researchers to pre-register their analysis and resist the temptation to engage in post-hoc exploration of possible moderators.

Supplemental Material

Supplemental Material - Descriptive representation and attitudes about local government: An experimental test using real-world stimuli

Supplemental Material for Descriptive representation and attitudes about local government: An experimental test using real-world stimuli by David Doherty, Madeline Schade, and Dana Garbarski in Research & Politics.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Disclaimer

This publication was made possible (in part) by a grant from the Carnegie Corporation of New York. The statements made and views expressed are solely the responsibility of the author.

Supplemental Material

Supplemental material for this article is available online.

The replication files are available at:

Notes

References

Abney

Hutcheson

Jr (1981) Race, representation, and trust: changes in attitudes after the election of a black mayor. Public Opinion Quarterly 45(1): 91–101.

Ansolabehere

Fraga

(2016) Do Americans prefer coethnic representation? The impact of race on house incumbent evaluations. Stanford Law Review 68(6): 1553.

Atkeson

(2003) Not all cues are created equal: the conditional impact of female candidates on political engagement. The Journal of Politics 65(4): 1040–1061.

Atkeson

Carrillo

(2007) More is better: the influence of collective female descriptive representation on external efficacy. Politics and Gender 3(1): 79–101.

Badas

Stauffer

(2018) Someone like me: descriptive representation and support for supreme court nominees. Political Research Quarterly 71(1): 127–142.

Banducci

Todd

Karp

(2004) Minority representation, empower- ment, and participation. The Journal of Politics 66(2): 534–556.

Bobo

Gilliam

(1990) Race, sociopolitical participation, and black empowerment. American Political Science Review 84(2): 377–393.

Bonneau

Kristin

(2020) Stronger together: political ambition and the presentation of women running for office. Politics, Groups, and Identities 8(3): 576–594.

Bowen

Clark

(2014) Revisiting descriptive representation in Congress: assessing the effect of race on the constituent–legislator relationship. Political Research Quarterly 67(3): 695–707.

10.

Broockman

(2014) Do female politicians empower women to vote or run for office? A regression discontinuity approach. Electoral Studies 34: 190–204.

11.

Campbell

Wolbrecht

(2006) See Jane run: women politicians as role models for adolescents. The Journal of Politics 68(2): 233–247.

12.

Clayton

O’Brien

Piscopo

(2019) All male panels? Representation and democratic legitimacy. American Journal of Political Science 63(1): 113–129.

13.

Costa

Schaffner

(2018) How gender conditions the way citizens evaluate and engage with their representatives. Political Research Quarterly 71(1): 46–58.

14.

Costa

Wallace

(2021) More women candidates: the effects of increased women’s presence on political ambition, efficacy, and vote choice. American Politics Research 49(4): 368–380.

15.

Dassonneville

McAllister

(2018) Gender, political knowledge, and descriptive representation: the impact of long-term socialization. American Journal of Political Science 62(2): 249–265.

16.

Dittmar

Sanbonmatsu

Carroll

(2018) A Seat at the Table: Congresswomen’s Perspectives on Why Their Presence Matters. Oxford University Press.

17.

Dolan

(2008) Is there a “gender affinity effect” in American politics? Information, affect, and candidate sex in US House elections. Political Research Quarterly 61(1): 79–89.

18.

Espírito-Santo

Verge

(2017) The elusive measurement of symbolic effects on citizens’ political attitudes: survey experiments as alternative avenues. Politics, Groups, and Identities 5(3): 494–499.

19.

Fairdosi

Rogowski

(2015) Candidate race, partisanship, and political participation: when do black candidates increase black turnout? Political Research Quarterly 68(2): 337–349.

20.

Franco

Malhotra

Simonovits

(2014) Publication bias in the social sciences: unlocking the file drawer. Science 345(6203): 1502–1505.

21.

Gay

(2002) Spirals of trust? The effect of descriptive representation on the relationship between citizens and their government. American Journal of Political Science 46(4): 717–732.

22.

Gelman

. (2018) “You need 16 times the sample size to estimate an interaction than to estimate a main effect”. https://statmodeling.stat.columbia.edu/2018/03/15/need-16-times-sample-size-estimate-interaction-estimate-main-effect/

23.

Gerber

Malhotra

Dowling

, et al. (2010) Publication bias in two political behavior literatures. American Politics Research 38(4): 591–613.

24.

Gilens

Page

(2014) Testing theories of American politics: elites, interest groups, and average citizens. Perspectives on Politics 12: 564–581.

25.

Gleason

Stout

(2014) Who is empowering who: exploring the causal relationship between descriptive representation and Black empowerment. Journal of Black Studies 45(7): 635–659.

26.

Griffin

Keane

(2006) Descriptive representation and the composition of African American turnout. American Journal of Political Science 50(4): 998–1012.

27.

Hansen

(1997) Talking about politics: gender and contextual effects on political proselytizing. The Journal of Politics 59(1): 73–103.

28.

Hayes

Hibbing

(2017) The symbolic benefits of descriptive and substan- tive representation. Political Behavior 39: 31–50.

29.

Jones

(2016) Constituents’ responses to descriptive and substantive representa- tion in congress. Social Science Quarterly 97(3): 682–698.

30.

Juenke

Preuhs

(2012) Irreplaceable legislators? Rethinking minority representatives in the new century. American Journal of Political Science 56(3): 705–715.

31.

Kaslovsky

Rogowski

Stone

(2021) Descriptive representation and public support for Supreme Court nominees. Political Science Research and Methods 9(3): 583–598.

32.

Lawless

(2004) Politics of presence? Congresswomen and symbolic representation. Political Research Quarterly 57(1): 81–99.

33.

Lowande

Ritchie

Lauterbach

(2019) Descriptive and substantive representation in congress: evidence from 80,000 congressional inquiries. American Journal of Political Science 63(3): 644–659.

34.

Mansbridge

(1999) Should blacks represent blacks and women represent women? A contingent yes. The Journal of Politics 61(3): 628–657.

35.

Mansbridge

(2003) Rethinking representation. American Political Science Review 97(04): 515–528.

36.

Marschall

Ruhil

AVS

(2007) Substantive symbols: the attitudinal dimen- sion of black political incorporation in local government. American Journal of Political Science 51(1): 17–33.

37.

Merolla

Sellers

Fowler

(2013) Descriptive representation, political efficacy, and A frican A mericans in the 2008 presidential election. Political Psychology 34(6): 863–875.

38.

Miller

Stokes

(1963) Constituency influence in congress. American Political Science Review 57: 45–56.

39.

Pantoja

Segura

(2003) Does ethnicity matter? Descriptive representation in legislatures and political alienation among Latinos. Social Science Quarterly 84(2): 441–460.

40.

Pitkin

(1967) The Concept of Representation. Berkley, CA: University of California Press.

41.

Rehfeld

(2009) Representation rethought: on trustees, delegates, and gyroscopes in the study of political representation and democracy. American Political Science Review 103(02): 214–230.

42.

Reingold

Harrell

(2010) The impact of descriptive representation on women’s political engagement: does party matter? Political Research Quarterly 63(2): 280–294.

43.

Reingold

Widner

Harmon

(2020) Legislating at the intersections: race, gender, and representation. Political Research Quarterly 73(4): 819–833.

44.

Rocha

Tolbert

Bowen

, et al. (2010) Race and turnout: does descriptive representation in state legislatures increase minority voting? Political Research Quarterly 63(4): 890–907.

45.

Sanchez

Morin

(2011) The effect of descriptive representation on Latinos’ views of government and of themselves. Social Science Quarterly 92(2): 483–508.

46.

Scherer

Curry

(2010) Does descriptive race representation enhance institutional legitimacy? The case of the US courts. The Journal of Politics 72(1): 90–104.

47.

Snagovsky

Kang

Sheppard

, et al. (2020) Does descriptive representation increase perceptions of legitimacy? Evidence from Australia. Australian Journal of Political Science 55(4): 378–398.

48.

Stauffer

(2021) Public perceptions of women’s inclusion and feelings of political efficacy. American Political Science Review 115(4): 1226–1241.

49.

Stimson

Mackuen

Erikson

(1995) Dynamic representation. American Political Science Review 89: 543–565.

50.

Tate

(2004) Black Faces in the Mirror: African Americans and Their Representatives in the US Congress. Princeton University Press.

51.

Torres

(2006) Latinos in New England. Temple University Press.

52.

Ulbig

(2007) Gendering municipal government: female descriptive representation and feelings of political trust. Social Science Quarterly 88(5): 1106–1123.

53.

Washington

(2006) How black candidates affect voter turnout. Quarterly Journal of Economics 121(3): 973–998.

54.

West

(2017) Descriptive representation and political efficacy: evidence from obama and Clinton. The Journal of Politics 79(1): 351–355.

55.

Wicherts

Coosje

LSV

Hilde

EMA

, et al. (2016) Degrees of freedom in planning, running, analyzing, and reporting psychological studies: a checklist to avoid p-hacking. Frontiers in Psychology 7: 1832.

56.

Wolak

(2015) Candidate gender and the political engagement of women and men. American Politics Research 43(5): 872–896.

57.

Wolbrecht

Campbell

(2007) Leading by example: female members of parliament as political role models. American Journal of Political Science 51(4): 921–939.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.86 MB