Studies of representation are widespread in the congressional literature, with recent work focusing on both descriptive and substantive representation of various minority groups. The authors focus on religious representation in Congress, examining the extent to which senators' own religious affiliation and the relative size of their religious constituencies affect substantive representation of Religious Right interests. The authors find that Christian senators (evangelical Protestants, conservative Catholics, and Mormons alike) provide better representation of the Religious Right, as do senators from states with larger Christian constituencies. The findings demonstrate that substantive representation is the product of characteristics of both the representatives and the represented.
Keywords: Religious representation; Congress; U.S. Senate; roll-call voting; Religious Right; Family Research Council
In her classic treatment of representation, Hanna Pitkin (1967, 9) observed that in democratic polities, "a man [sic] is represented if he feels that he is, and not if he does not." Because the quality of any democracy depends on meaningful elections, questions of representation are vital in any functioning democracy. As a result, a great deal of the scholarly literature-in particular, the congressional literature- has examined the concept of representation. Scholars have investigated the extent to which various groups actually obtain meaningful representation, with much of the recent literature focusing on the antecedents of representation of racial and ethnic minorities in Congress (e.g., Hero and Tolbert 1995; Lublin 1997; Tate 2003; Whitby 1997). Regardless of the group in question, the literature largely has established that the extent and quality of a group's representation is dependent on two primary factors: characteristics of the representatives themselves (e.g., Lublin 1997; Mansbridge 1999; Schwindt-Bayer and Mishler 2005; Swain 2006; Tate 2003) and characteristics of those being represented (e.g., Bullock and MacManus 1981; Hero and Tolbert 1995; Hutchings, McClerking, and Charles 2004). Studies of the former center on whether descriptive representation leads to better substantive representation. In other words, this literature considers whether members of Congress with certain shared characteristics (such as race and gender) are more likely to behave in ways that are preferable to corresponding demographic groups in the population. Scholars also have considered whether the size (or proportion) of particular groups within a representative's geographic or reelection constituency might affect how well elected officials provide substantive representation (e.g., Combs, Hibbing, and Welch 1984; Griffin and Newman 2005; Hill, Leighley, and Hinton-Andersson 1995; Hutchings, McClerking, and Charles 2004; Kerr and Miller 1997). The present study furthers our understanding of representation by examining how characteristics of both elected officials and the people they represent might combine to yield substantive representation.
Studies that focus on the characteristics of representatives have their intellectual roots in Pitkin's (1967) early discussion of descriptive representation. According to Pitkin, when elected officials and their constituencies share key personal characteristics, better representation ought to result. Moreover, elected officials with certain personal characteristics likely also share similar backgrounds and experiences-and therefore might be expected to be like-minded in their desire and ability to represent citizens with the same characteristics. Recent research has tested this relationship empirically, with mixed results. Some evidence suggests that descriptive representation leads to better outcomes for racial and ethnic minorities (e.g., Canon 1999; Tate 2003) and women (e.g., Mansbridge 1999)-that is, minority elected officials better represent minority constituents and women better represent the interests of women. …