Do More Professionalized Legislatures Discriminate Less? The Role of Staffers in Constituency Service

Abstract

Research suggests that organizational structure can influence the ability of actors to discriminate. In this research note, we examine whether the structure of state legislatures affects observed discrimination in correspondent audit studies. We find that increased legislative professionalization is associated with reduced discrimination against racial minorities. By analyzing thousands of emails collected in a prior study, we find that legislative professionalization is related to a higher likelihood that staffers respond to email contacts and staffers are less likely to discriminate against racial minorities across multiple measures of discrimination. Our findings emphasize the importance of substantively relevant heterogeneity in audit studies and identify a potential mitigator of discrimination—legislative professionalism. Our results also highlight the importance of staffers in representation and the legislative process.

Keywords

discrimination staffers constituency service state legislatures

Introduction

Legislative professionalization affects a host of different outcomes involving both legislative behavior (Berry et al., 2000; Maestas, 2000; Squire, 1992) and policy output (Kousser & Phillips, 2009). Despite the various effects of professionalization, scholars have not yet examined whether it might influence discrimination against constituents. Literature across a number of fields suggests that organizational professionalization could reduce discrimination (Andersen & Guul, 2019; Banerjee et al., 2018; Hirschman & Bosk, 2019; Lipsky, 1983; Mladenka, 1977; Thomas, 1986; Weber, 1922). Several features associated with professionalism, in particular the increased role staffers play in constituent communications (Herrick, 2011), could play a role in reducing discrimination against constituents. Given the plausibility that professionalization could reduce discrimination, we re-analyzed data from a prior field experiment about discrimination against hypothetical constituents, and we also coded for the first time approximately 3,000 emails received in the original study allowing us to examine who responded and some characteristics of their responses.

Our data analysis suggests that professionalization, and specifically an increase in staffer’s involvement in the legislative enterprise, reduces discrimination in both the rate of replies to hypothetical constituents and in the content of the emails that legislative offices send. This is the first evidence we are aware of suggesting that legislative professionalization can play a role in reducing discrimination against constituents. The substantive importance of these findings suggests scholars should further investigate the relationship between discrimination and legislative professionalization, including explicitly trying to understand the mechanisms that might lead to reduced discrimination. In addition, our results suggest that increases in professionalism and the increase in staffer support increases a legislatures’ capacity to better represent constituents.

We contribute to a growing literature on the importance of staffers in the legislative enterprise (Hertel-Fernandez et al., 2018; McCrain, 2018; Montgomery & Nyhan, 2017; Pertschuk, 2017) by studying staffers in state legislatures and their role in constituency service. Early research theorized the importance of staffers in increasing legislative capacity in Congress (DeGregorio, 1995; Fox & Hammond, 1977; Salisbury & Shepsle, 1981), but it is only recently that the field has studied the importance of staffers in state legislatures (Weissert & Weissert, 2000). State legislatures present an opportunity to advance the study of staffers because of the variation in their roles and utilization across states. Likewise, prior literature has discussed the role of staffers in constituency service (Ziniel, 2009), and we advance the conversation by showing that staffers have a substantive effect on constituency service by reducing discrimination.

Legislative Professionalization

Prior scholarship has not examined whether legislative professionalism can reduce discrimination. Although professionalization involves multiple factors (Squire, 2007), in this context we are most interested in its association with an increased likelihood that staffers respond to constituent communications. Our data and those of other scholars show that staffers are more likely to respond to constituents as professionalization increases (Herrick, 2011). This is a part of what Bowen and Greene (2014) refer to as the “support-intensity” dimension of professionalism; how much legislative work is supported by staffers.

There are multiple reasons why having staffers deal with constituent communications might reduce discrimination against constituents. First, hiring staffers may lead to the implementation of routinized work procedures that can reduce discrimination by creating norms against such behavior (Pager & Shepherd, 2008), and legislative actors are expected to adhere to norms if they wish to be successful (Matthews, 1959). In their study of legislative staffers, Romzek and Utter (1997) note that staffers are expected to protect their legislator’s reputation by upholding strict norms when interacting with constituents; “constituents are always to be treated with the utmost courtesy even when they are being obnoxious . . . .” This suggests that staffers in more professionalized legislatures are more likely to receive constituent service training. Staffers may also have fewer cognitive biases than legislators (Heß et al., 2018; Sheffer et al., 2018), which might also make staffers less likely to discriminate than legislators.

However, in less-professionalized legislatures, there are multiple reasons why we might observe more discrimination. First, such legislatures are less likely to have staffers handle constituent communications and therefore the aforementioned attributes of staffers are missing. Furthermore, legislators in these states are less likely to receive ethics training (Rosenthal, 1996), and they have less experience with legislative norms due to term limits (Kousser, 2005). Legislators in less-professionalized legislatures are also more likely to violate ethics due to their inexperience (Herrick & Fisher, 2007).

Second, scandals in the late 20th century left trust in state legislatures at a historic low, which led some states to pass stringent ethics codes regulating legislators’ interactions with constituents and staffers (Rosenthal, 1996; Squire, 1993). The implementation of ethics codes reduced legislators’ control over staffers in favor of nonpartisan constituency service agencies, which feature more training than is otherwise common (Rosenthal, 2008). These reforms were more common in professionalized legislatures, providing a further reason to expect professionalism to reduce discrimination (Rosenthal, 1996).

Third, prior research suggests that constituent communications in professionalized state legislatures are conducted by a racially diverse group of staffers who might be expected to discriminate less against minority constituents. Furthermore, racial minority staffers are more likely to be assigned constituent communication work in legislatures (Jones, 2017) possibly to help the legislator garner a personal vote with racial minority constituents (Ziniel, 2009). This is particularly important in our case because the legislative audit literature has found that discrimination is mitigated by non-White legislative actors (Costa, 2017).

Furthermore, the substantive importance of the topic provides reason to investigate the relationship between legislative professionalization and discrimination in constituent contact. We turn now to a discussion of the research design and data we use to shed light on professionalization and discrimination.

Data and Research Design

The data we use comes from Butler and Broockman’s (2011) audit study that focused on whether hypothetical Black and White constituents were equally likely to receive a response to an email sent to state legislators requesting assistance with voter registration. Audit studies are field experiments in which researchers use aliases to request assistance from a service provider. These fictional aliases allow researchers to signal a characteristic of interest (i.e., race, religion, sexual orientation) without varying other factors that might affect behavior, which means that observed differences in service can be attributed to the manipulated characteristic (Gaddis, 2018). In the context of race and constituent communications, prior research has focused on the average difference in response rates between a minority and White constituent. We advance this literature by showing that more professionalized legislatures are less likely to discriminate than less-professionalized legislatures, and this heterogeneity in response rates can provide evidence about factors that may mitigate discrimination (Butler & Crabtree, 2017).

In October 2008, Butler and Broockman (2011) emailed state legislators for information about how to register to vote. “Dear [Representative], My name is [Jake Mueller/DeShawn Jackson] and I’m trying to figure out how to register to vote . . .”¹ Constituent race was signaled by varying the name of the email sender using purportedly White (Jake Mueller) or Black (DeShawn Jackson) constituent names. The key outcome is whether the emails from the Black and White constituents receive different response rates. The original study found that the Black constituent received about 5 percentage points fewer responses than the White constituent (Butler & Broockman 2011), which suggests legislative offices discriminated against the Black constituent.

We were given access to an anonymized version of the Butler and Broockman (2011) data. This means that we cannot examine whether district-level (e.g., percent Black in district) or individual-level (e.g., legislators’ party id, age, or gender) variables are associated with reply rates, but given our interest with a state-level factor—professionalization—this is not a major concern. The original article included professionalism as a covariate in some of their analyses, but they did not examine the moderating effect of professionalization at multiple relevant values as necessary for interaction effects (Brambor et al., 2005). To augment the data on response rates, we also use new data culled from almost 3000 email replies received as part of the original audit study. These emails allow us to identify who responded to the emails and characteristics of the responses.

We use two different measures to capture the extent to which staffers are likely to be involved in constituent communications. One measure, the Squire index, was measured at approximately the same time as the original audit study and which was used in the original analysis of the data (Butler & Broockman, 2011; Squire, 2007). The Squire index is composed of data on legislature salary and benefits, number of days in session, and staff per legislator. Each of the three components is divided by the corresponding value for Congress, and then, the three values are summed together to achieve the final score. Scores for state legislatures range from a low of 0.027 for New Hampshire to 0.626 for California. In addition to the Squire index, we use the 2008 measure of “support intensity” developed by Bowen and Greene (2014).² This measure captures the degree to which staffers support legislative work, which is a key aspect of how we expect professionalization to affect responses to constituents. Our results are consistent whether we use the Squire index or the Bowen–Greene support-intensity index. While others have studied heterogeneity in legislative discrimination, they have not examined whether professionalism and/or staff usage affects discrimination (Costa, 2017).

In Table 1, we present the regression results from multiple estimates of the relationship between professionalization and discrimination in reply rates.³ The outcome variable is a binary indicator of whether a reply is received or not. The unit of analysis is a state legislative office, and offices were randomly assigned to receive an email from either the Jake or DeShawn hypothetical constituent (Vuolo et al., 2018). We re-scaled the Bowen and Greene (2014) and Squire (2007) measures of professionalization into z-scores for each state (i.e., each state’s standard deviation from the mean) to facilitate comparison and interpretation. The relationship between professionalization and replies to constituent emails is estimated as a linear probability model (LPM), as is the norm in legislative audit studies (Costa, 2017).⁴ The moderating measure, state professionalization, is not randomly assigned. As such, we adjust for possible state-level factors that may be correlated with both our moderating measure (professionalization) and our outcome (reply rates). We account for such factors as the state’s Black population (% of total), median household income, total state population (in millions), percent vote for Obama in 2008, term limits, and voting ID laws (Kousser, 2005). Demographic covariates come from the 2008 American Community Survey, election data come from states’ secretary of state records, and state legislative institutional factors (i.e., voting id and term limits) come from the National Conference of State Legislatures (NCSL).⁵

Table 1.

Relationship Between Professionalism and DeShawn Treatment.

Variables	(1)	(2)
Variables	Squire Index	Support-Intensive Index
DeShawn treatment	−.054** (.024)	−.047* (.025)
Professionalism (Squire), standardized	−.033 (.021)
DeShawn treatment × professionalism (standardized)	.030 (.024)
Percent state Pop. Black	−.303** (.125)	.050 (.154)
State population (millions)	.001 (.003)	−.003 (.002)
Household income, US$10,000s	.080*** (.014)	.078*** (.015)
Voter ID	.043 (.033)	.023 (.034)
Term limits	.047 (.031)	.079** (.033)
Obama vote share percent, 2008	.255 (.175)	.222 (.174)
Support-intensive (Bowen and Greene), standardized		−.012 (.019)
DeShawn treatment × support-intensive (standardized)		.046* (.025)
Constant	−.070 (.098)	−.044 (.100)
Observations	1,618	1,496
R ²	.050	.037

Note. Standard errors in parentheses.

p < .1. **p < .05. ***p < .01.

Table 1, Column (1) presents the interaction between the DeShawn treatment and the Squire (2007) professionalism index after adjusting for the aforementioned variables. Table 1, Column (2) presents the interaction between the DeShawn treatment and the support-intensive index of Bowen and Greene (2014). Both the Squire and support-intensive indices produce near-identical results. The small reduction in sample size when using the support-intensive index is because it was not calculated for some states in 2008. We stress that the statistical significance of the interaction cannot be interpreted solely from examining the coefficients in Table 1. The standard errors reported for interaction terms in regression tables are incorrect, and standard errors must be calculated at multiple levels of the moderating term and/or the interaction should be visually presented to allow for meaningful interpretation (Brambor et al., 2005).

We present the estimated relationship between legislative professionalism and the difference in reply rates in Figure 1 from ±2 SDs of each measure of professionalism. If professionalism reduces discrimination, then we expect to see that as professionalism increases, the difference in reply rates between Jake and DeShawn should disappear (i.e., the estimated difference should disappear, and we should fail to reject the null hypothesis of equal rates of response to DeShawn and Jake). In Figure 1, this is precisely the pattern that appears regardless of which professionalism measure we use. In the less-professionalized legislatures, there are significantly fewer responses to the DeShawn (Black) name compared to the Jake (White) name; however, as evidenced by the upward slope in Figure 1, increases in professionalization are associated with reduced discrimination against DeShawn. In the less-professionalized legislatures, the experimental data are indicative of discrimination, but in the most professional legislatures, we do not find evidence of discrimination.⁶ The 95%confidence intervals around the point estimates indicate no significant difference in response rates (i.e., no discrimination) by about the midpoint of the distribution of the Squire index.⁷ Importantly, this is not just a result of increased uncertainty as the point estimate for the difference in reply rates also moves toward no difference and eventually is even suggestive of more responses to DeShawn than Jake. Although the number of states observed at the higher levels of professionalization is smaller than at the lower levels, our estimates of discrimination are still based on the behavior of hundreds of individual legislative offices in these professionalized legislatures.

Figure 1.

Legislative professionalism reduces differences in reply rate between Jake and DeShawn.

Responding to Constituent Communications

In the previous section, we show that more professional legislatures are less likely to discriminate against a Black constituent. In this section, we present results based on content analysis of emails received by the hypothetical constituents, which have not been previously used in research. Prior analyses have not identified the likely respondents of constituency communications because of data limitations, and this information can help us to understand whether observed discrimination is driven by legislators or staffers responding to constituents. Staffers are agents of their legislators but, like any actor, control over them is imperfect. In several states, staffers are managed through a nonpartisan service agency (Rosenthal, 2008) and do not work directly for a legislator. Even if legislators and staffers have identical interests, the two may act differently because of asymmetric information about whether the legislator has a reason to discriminate against minority constituents. Staffers, especially in term-limited legislatures, may also have more experience with legislative norms than legislators (Kousser, 2005). Racial minority staffers are more likely to be assigned constituent communication work (Grose et al., 2007; Jones, 2017; Ziniel, 2009), and although the placement of racial minority staffers in these positions may limit their ability to influence policy, it may influence whether vulnerable constituents receive equitable service. These reasons suggest that legislators and staffers may behave differently, but existing studies have not investigated this possibility.

We coded emails based on whether a staffer or legislator responded.⁸ As shown in Figure 2, we find that more professionalized legislatures, as measured by the Squire index, are more likely to have their communications handled by staffers based on the emails received as part of this audit study.⁹ The figure demonstrates that the difference is quite significant and in many states there did not appear to be any (or very, very few) emails handled by staffers.

Figure 2.

Percentage of staffer respondents by legislative professionalism.

We also coded the contents of emails to determine whether legislators and staffers differ in how they responded to hypothetical constituents. We coded the emails responses for two issues that seem related to discrimination—whether the respondent mentioned voter ID requirements or felon disqualification laws. If a respondent discriminates in their responses, then we would expect them to be more likely to bring up ID requirements or felon disqualification laws when responding to a Black constituent, because these issues disproportionally affect Black voters (Barreto et al., 2018; Biggers & Hanmer, 2017; Cottrell et al., 2019). We cannot infer whether this form of discrimination is malicious or well-meaning. It is possible that respondents bring these issues up to suppress votes or a friendly reminder, so they are not denied voting at the poll.

An important issue regarding analysis of email responses in audit studies is that they suffer from post-treatment bias. The initial emails are randomized across legislators, but we can only analyze the content in the observed responses and who responds is not random and may be affected by the treatment (Coppock, 2018). One method to address the possibility of post-treatment bias that has been used in prior studies is to assume that all nonresponses do not mention the outcome of interest (Kalla et al., 2018; Loewen & MacKenzie, 2018). We also must make an assumption about who would have sent the emails that were not actually sent, and we assume that the proportion of non-sent responses would have been equivalent to the division we observed among actual responses for each state. These are both strong assumptions, but they remove the possibility of post-treatment bias. The appendices include additional analyses where we assume that all missing replies would have been sent by either staffers or legislators, and these alternatives do not affect the substantive conclusions. In Table 2, we present the bias-corrected results regarding how staffers and legislators respond to the two different constituents.

Table 2.

Email Content by Respondent and Name (Jake vs DeShawn), Corrected for Nonresponse Bias.

Respondent	ID mention		Felon mention
Respondent	Jake	DeShawn	Jake	DeShawn
All	8.30%	8.37%	0.36%	0.96%
n = 4,472^a	(.9333)		(.0118)
Legislator	8.48%	8.76%	0.34%	0.86%
n = 3,149	(.7863)		(.0547)
Staffer	6.23%	5.41%	0.58%	0.90%
n = 957	(.5958)		(.5638)

Note. p values of two-sided t-test between Jake and DeShawn in parentheses.

The all category includes emails whose likely respondent, staffer or legislator, is unknown.

We find that legislators are more likely to mention felon disqualification laws (p = .05) in their responses to DeShawn compared to Jake despite assuming that all of the nonresponses would have failed to mention felon disqualification laws. Staffers do not differ in mentioning felon disqualification to the two constituents (p = .564). We do not find evidence of discrimination in mentions of ID requirements by either legislators or staffers in Table 2. Our analyses of the emails provide suggestive evidence that legislators and staffers may differ in how they interact with constituents.

Our combined results suggest that observed discrimination by legislative offices is lower (or nonexistent) in more professionalized legislatures and that one reason is that in these contexts staffers are more likely to handle constituent communications. These are the first results of their kind and demonstrate the importance of delving further into legislative discrimination to determine its source(s) and what might moderate it.

Conclusion

In this research note, we examine the moderating effect of professionalization on discrimination against minority constituents, because there are multiple reasons to expect that more professional legislatures will be less likely to discriminate than less-professional legislatures. In particular, prior research suggests that the larger role played by staffers in professional legislatures and the characteristics and training of staffers should make discrimination less likely where they handle constituent communication. We find that legislatures with high professionalism do not engage in racial discrimination in responses to emails from constituents. Professionalized legislatures are more likely to have staffers who respond to constituents, and our analyses of the content emails from an audit study suggests evidence of behavioral differences between legislators and staffers, which is consistent with recent literature on legislators’ cognitive biases. Our results advance the representation literature by showcasing a legislative activity, constituent communications, in which an increased role for staffers increases legislators’ capacity to better represent constituents. Our results also advance the legislative literature by exploiting interstate differences in staffer utilization and by showing the importance of staffers in constituency service.

The Squire index that we use to measure professionalization includes factors other than staffing levels, and therefore, we cannot make a strong claim that it is difference in staffing alone that drive the heterogeneity in discrimination across values of the Squire index. However, we have good reasons to believe that differences in the prevalence and behavior of staffers explain the reduction in discrimination as professionalization increases. The measure developed by Bowen and Greene (2014) captures the degree of staffer involvement in the legislative process and produces near-identical results as the Squire index. In the online appendix (Supplemental material), we present a variety of results using different measures of professionalization/staffing that also reinforce the importance of staffers in reducing discrimination.

Our research advances knowledge about discrimination in political contexts (i.e., constituent interactions with legislators and bureaucrats), and especially about whether and how legislative professionalization may mitigate discrimination, highlighting the importance of staffers in the legislative process. Further research should continue to investigate whether the observed discrimination against constituents (and other types) is driven by legislators or others and use this knowledge to enhance our understanding of the conditions under which discrimination occurs and how it might be ameliorated. Although beyond the scope of our present data, future research should also attempt to examine the importance of racial minority staffers in the representation process.

Supplemental Material

APR_OnlineAppendices.2_4_2020 – Supplemental material for Do More Professionalized Legislatures Discriminate Less? The Role of Staffers in Constituency Service

Supplemental material, APR_OnlineAppendices.2_4_2020 for Do More Professionalized Legislatures Discriminate Less? The Role of Staffers in Constituency Service by Michelangelo Landgrave and Nicholas Weller in American Politics Research

Footnotes

Acknowledgements

Thanks to Dan Butler and David Broockman for graciously sharing both the original data reported in their paper and all the emails they received from legislative offices, which have not been analyzed previously. Thanks to Rebekah Herrick for sharing her 2008 state legislature survey data. Additional thanks to S. Michael Gaddis, Daniel Biggers, and participants of the UCR Political Behavior Workshop for helpful comments.

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.

ORCID iD

Michelangelo Landgrave

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Michelangelo Landgrave is a doctoral candidate in the Department of Political Science at the University of California, Riverside.

Nicholas Weller is an associate Professor in the Department of Political Science at the University of California, Riverside.

References

Andersen

S. C.

Guul

T. S.

(2019). Reducing minority discrimination at the front line—Combined survey and field experimental evidence. Journal of Public Administration Research and Theory, 29(3), 429–444.

Banerjee

Reitz

J. G.

Oreopoulos

(2018). Do large employers treat racial minorities more fairly? An analysis of Canadian field experiment data. Canadian Public Policy, 44(1), 1–12.

Barreto

M. A.

Nuño

S. A.

Sanchez

G. R.

Walker

H. L.

(2018). The racial implications of voter identification laws in America. American Politics Research, 47(2), 238–249.

Berry

W. D.

Berkman

M. B.

Schneiderman

(2000). Legislative professionalism and incumbent reelection: The development of institutional boundaries. American Political Science Review, 94(4), 859–874.

Biggers

D. R.

Hanmer

M. J.

(2017). Understanding the adoption of voter identification laws in the American states. American Politics Research, 45(4), 560–588.

Bowen

D. C.

Greene

(2014). Should we measure professionalism with an index? A note on theory and practice in state legislative professionalism research. State Politics & Policy Quarterly, 14(3), 277–296.

Brambor

Clark

W. R.

Golder

(2005). Understanding interaction models: Improving empirical analyses. Political Analysis, 14(1), 63–82.

Butler

D. M.

Broockman

D. E.

(2011). Do politicians racially discriminate against constituents? A field experiment on state legislators. American Journal of Political Science, 55(3), 463–477.

Butler

D. M.

Crabtree

(2017). Moving beyond measurement: Adapting audit studies to test bias-reducing interventions. Journal of Experimental Political Science, 4, 57–67.

10.

Coppock

(2018). Avoiding post-treatment bias in audit experiments. Journal of Experimental Political Science, 6(1), 1–4.

11.

Costa

(2017). How responsive are political elites? A meta-analysis of experiments on public officials. Journal of Experimental Political Science, 4(1), 241–254.

12.

Cottrell

Herron

M. C.

Rodriguez

J. M.

Smith

D. A.

(2019). Mortality, incarceration, and African American disenfranchisement in the contemporary United States. American Politics Research, 47(2), 195–237.

13.

DeGregorio

(1995). Staff utilization in the US Congress: Committee chairs and senior aides. Polity, 28(2), 261–275.

14.

Fox

H. W.

Hammond

S. W.

(1977). Congressional staffs: The invisible force in American lawmaking. Free Press.

15.

Gaddis

S. M.

(2018). An introduction to audit studies in the social sciences. In Gaddis

S. M.

(Ed.), Audit studies: Behind the scenes with theory, method, and nuance (pp. 3–44). Springer.

16.

Grose

C. R.

Mangum

Martin

(2007). Race, political empowerment, and constituency service: Descriptive representation and the hiring of African-American Congressional staff. Polity, 39(4), 449–478.

17.

Heß

Von Scheve

Schupp

Wagner

G. G.

(2018). Are political representatives more risk-loving than the electorate? Evidence from German federal and state parliaments. Palgrave Communications, 4(1), 1–7.

18.

Herrick

(2011). Representation and institutional design. Lexington Books.

19.

Herrick

Fisher

S. H.

(2007). Are citizen legislators less likely to violate the official ethics standards? In Representing America (pp. 74–80). Lexington Books.

20.

Hertel-Fernandez

Mildenberger

Stokes

L. C.

(2018). Legislative staff and representation in Congress. American Political Science Review, 113, 1–18.

21.

Hirschman

Bosk

E.A.

(2019). Standardizing biases: Selection devices and the quantification of Race. Sociology of Race and Ethnicity. Advance online publication. https://doi.org/10.1177/2332649219844797.

22.

Jones

J. R.

(2017). Racing through the halls of Congress: The “black nod” as an adaptive strategy for surviving in a raced institution. Du Bois Review: Social Science Research on Race, 14(1), 165–187.

23.

Kalla

Rosenbluth

Teele

D. L.

(2018). Are you my mentor? A field experiment on gender, ethnicity, and political self-starters. The Journal of Politics, 80(1), 337–341.

24.

Kousser

(2005). Term limits and the dismantling of state legislative professionalism. Cambridge University Press.

25.

Kousser

Phillips

J. H.

(2009). Who blinks first? Legislative patience and bargaining with governors. Legislative Studies Quarterly, 34(1), 55–86.

26.

Lipsky

(1983). Street-level bureaucracy (30th Anniversary Edition): Dilemmas of the individual in public service. Russell Sage Foundation.

27.

Loewen

P. J.

MacKenzie

M. K.

(2018). Service representation in a federal system: A field experiment. Journal of Experimental Political Science, 6(2), 93–107.

28.

Maestas

(2000). Professional legislatures and ambitious politicians: Policy responsiveness of state institutions. Legislative Studies Quarterly, 25, 663–690.

29.

Matthews

D. R.

(1959). The folkways of the United States Senate: Conformity to group norms and legislative effectiveness. American Political Science Review, 53(4), 1064–1089.

30.

McCrain

(2018). Revolving door lobbyists and the value of Congressional staff connections. The Journal of Politics, 80(4), 1369–1383.

31.

Mladenka

K. R.

(1977). Citizen demand and bureaucratic response: Direct dialing democracy in a major American city. Urban Affairs Quarterly, 12(3), 273–290.

32.

Montgomery

J. M.

Nyhan

(2017). The effects of Congressional staff networks in the US House of Representatives. The Journal of Politics, 79(3), 745–761.

33.

Pager

Shepherd

(2008). The sociology of discrimination: Racial discrimination in employment, housing, credit, and consumer markets. Annual Review of Sociology, 34, 181–209.

34.

Pertschuk

(2017). When the senate worked for US: The invisible role of staffers in countering corporate lobbies. Vanderbilt University Press.

35.

Romzek

B. S.

Utter

J. A.

(1997). Congressional legislative staff: Political professionals or clerks? American Journal of Political Science, 41, 1251–1279.

36.

Rosenthal

(1996). Drawing the line: Legislative ethics in the states. University of Nebraska Press.

37.

Rosenthal

(2008). Engines of democracy: Politics and policymaking in state legislatures. SAGE.

38.

Salisbury

R. H.

Shepsle

K. A.

(1981). US Congressman as enterprise. Legislative Studies Quarterly, 6(4), 559–576.

39.

Sheffer

Loewen

P. J.

Soroka

Walgrave

Sheafer

(2018). Nonrepresentative representatives: An experimental study of the decision making of elected politicians. American Political Science Review, 112(2), 302–321.

40.

Squire

(1992). Legislative professionalization and membership diversity in state legislatures. Legislative Studies Quarterly, 17, 69–79.

41.

Squire

(1993). Professionalization and public opinion of state legislatures. The Journal of Politics, 55(2), 479–491.

42.

Squire

(2007). Measuring state legislative professionalism: The squire index revisited. State Politics & Policy Quarterly, 7(2), 211–227.

43.

Thomas

J. C.

(1986). The personal side of street-level bureaucracy: Discrimination or neutral competence? Urban Affairs Quarterly, 22(1), 84–100.

44.

Vuolo

Uggen

Lageson

(2018). To match or not to match? Statistical and substantive considerations in audit design and analysis. In Gaddis

S. M.

(Ed.), Audit studies: Behind the scenes with theory, method, and nuance (pp. 119–140). Springer.

45.

Weber

(1922). Wirtschaft Und Gesellschaft [Economy and Society] University of California Press.

46.

Weissert

C. S.

Weissert

W. G.

(2000). State legislative staff influence in health policy making. Journal of Health Politics, Policy and Law, 25(6), 1121–1148.

47.

Ziniel

C. E.

(2009). Faces in the office: Racial employment segregation among Congressional staff. https://ssrn.com/abstract=1434290.

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

0.10 MB