Key points are not available for this paper at this time.
In a polity that relies on representation to achieve a semblance of democracy, the composition of a non-elected arm of government such as the bureaucracy is important. In recent years, public administration scholars have addressed the topic in a number of ways. The well-substantiated premise undergirding this work in each case is that bureaucrats exercise discretion that makes a difference in the bureaucracy's outputs (e.g., Skolnick, 1966; Hedge, Menzel, and Williams, 1988; Meier and Stewart, 1992). This conclusion has held in examinations of higher level bureaucrats and lower street level, bureaucrats, although there is some debate about at which level representativeness is more important (Meier and Nigro, 1976; Thompson, 1976). The extant literature can be characterized as supply-side in approach because the focus is either on (1) the degree to which a specified bureaucracy is representative of a larger population in terms of various demographic variables (e.g., Subramanian, 1967; Meier, 1975; Grabosky and Rosenbloom, 1976; Hall and Saltzstein, 1977; Cayer and Sigelman, 1980; Dometrius and Sigelman, 1984; and Henderson and Preston, 1984), (2) the forces which make bureaucracies more or less representative (e.g., Saltzstein, 1986; Stein, 1986; Mladenka, 1989; Kellough, 1990a, 1990b), (3) the extent to which the policy-relevant attitudes of bureaucrats are representative of a clientele or of the public at large (e.g., Meier and Nigro, 1976; Rosenbloom and Featherstonehaugh, 1977; Thompson, 1978; Lewis, 1990; Garand, Parkhurst, and Seoud, 1991); or, most recently, (4) the links between the representativeness of bureaucracy and policy outputs and outcomes (e.g., Meier and Stewart, 1992; Hindera, 1993). Regardless of the focus, the assumption is that a representative bureaucracy is a good to be provided, the attainment of which will produce a better-served, more-satisfied clientele. Whether this assumption is true or not is particularly important for individuals for whom interaction with a bureaucracy is literally a life-or-death matter, such as those infected with HIV. After all, who should care less about who provides them with services than individuals whose very lives can be extended and whose quality of life can be improved by those services? Survival time with HIV is correlated with use of services (Rothenberg et al., 1987) and with demographic factors such as race (Seage et al., 1993; Bergner, 1993), on which bureaucracies can vary significantly. If clients of HIV service delivery agencies care about the characteristics of those who run and staff: the service delivery bureaucracy, and consequently are more or less likely to use the services depending on its representativeness, they exhibit a demand for a representative bureaucracy. Thus, this issue area provides a most stringent test of the importance of representative bureaucracy. Do those in need of bureaucratically delivered services care whether or not the bureaucracy is representative? If they do, are they more concerned about representativeness at the higher or lower reaches of the organization? The analysis presented here often an answer to these questions drawing on a survey of persons living with AIDS (PLWAs) in Dallas, Texas. These answers are important because AIDS services delivery impacts the quality of life of the PLWAs by facilitating acceptance of the reality of the disease and providing the basic preventative services that enhance both quantity and quality of life. Recent evidence indicates that basic services such as nutritional counseling (AIDS Treatment News, 1987, 1989) and counseling to aid in the client's psychological adjustment (Keet et al., 1994) are tremendously important in increasing the longevity of survival. AIDS: The Initial Response The initial organized and funded response to AIDS came not from the public sector but from the private sector. Public administrators and even the gay community failed to heed the early warnings of front-line health workers (Shilts, 1987). …
Thielemann et al. (Fri,) studied this question.