Media, issue information and the vote choice in a referendum campaign

Early theorists framed public opinion as an emergent product of broad discussion - emanating ideally from a debate open to wide popular participation, free-flowing and uncensored, and well-informed (Lasswell 1941; Price 1992; Price/Neijens 1997). However, early scientific analysts (e.g. Allport 1937) found the concept of public opinion as an "emergent product" of discussion difficult to grasp empirically and problematic in a number of respects. Over time, they came to accept mass survey data as the only workable empirical expression of public opinion (Key 1961; Converse 1987). The extent to which general population surveys provide valid measures of what has traditionally been defined as public opinion - grounded in public discussion and well-informed by debate - has been questioned by scholars of many stripes (Price & Neijens 1998; Crespi 1989; Saris/Sniderman 2004). Empirical evidence suggests that opinions given to pollsters and survey researchers are often unorganized, disconnected, individual responses that have not been influenced by public debate (Bishop/Oldendick/Tuchfarber/Bennet 1980).