What is the connection between this trend and opinion leaders’ role in the election campaign? In order to understand this, we must consider both the question of how media consumption has changed over the past decades and the nature of reporting on the flood in August 2002.

The opinion leader theory and the accompanying notion of the two-step flow of communication were devised in the United States in the 1940s, before the advent of television. At the time, researchers focused primarily on radio, a relatively new medium that was only beginning to have an impact on the formation of popular opinion. One basic assumption was that opinion leaders used the mass media—both newspapers and radio—more intensively than the remaining population did. This was believed to give them a head start in acquiring information, which they then passed on to third parties through personal communication.

As mentioned earlier, the only problem was that there was no sure way of identifying opinion leaders in surveys, although there was certainly evidence supporting the so-called “more-and-more rule,” which posited that people who were particularly intensive newspaper readers also used radio and other media, such as the cinema, more intensively than others did. This finding also confirmed one important prerequisite for the opinion leader theory: that there were apparently people who simultaneously used all kinds of informational media more intensively than others did.

The same media consumption pattern was also observed in the initial years after the Federal Republic of Germany was founded. In a 1952 Allensbach Institute survey, 66 percent of radio listeners said they also read the newspaper on a regular basis, as compared to only 45 percent of respondents who did not own a radio.

Since then, however—and particularly in the past twenty years—media consumption patterns have changed considerably in Germany. An investigation conducted by Noelle-Neumann and Rüdiger Schulz in the 1990s found that average television consumption of three hours or more per day by young people seriously reduced the likelihood of reading or subscribing to a local daily newspaper. Succinctly stated, the “more-and-more rule” no longer applied. Instead, two different groups of media consumers had emerged: those who read the newspaper intensively and those who watched a great amount of television.

At the same time, the nature of television consumption also changed dramatically. First of all, the amount of television watched steadily increased. According to Allensbach surveys, in 1967, West Germans watched an average of nine hours and forty minutes of television per week. By 1983, that figure had climbed to twelve hours and twelve minutes, and by 2002, it was thirteen hours and thirty-six minutes per week.

Meanwhile, the reach of TV news programs declined. In 1985, the four major news programs, Tagesschau, Tagesthemen, Heute, and Heute-Journal, reached a combined total of about 28 million viewers per day. In 2003, the total reach of the seven leading TV news programs broadcast by the five German channels was just under 23 million viewers. The increase in television consumption, then, primarily represented an increase in consumption of entertainment programming, instead of—or at least to a lesser degree than—programs providing political information.

Under these circumstances, the relationship between opinion leaders and the remaining population also changed with respect to television consumption. Especially interested in politics, opinion leaders presumably used the new medium of television more intensively than the remaining population during the Federal Republic’s early years. By 1985, however, there was no difference between opinion leaders and the remaining population in terms of the average amount of television consumed. And during the 2002 election year, the situation was the exact opposite of what it had been in the 1950s: Now, it was opinion leaders who watched television less frequently. In August 2002, 15 percent said they watched more than three hours per day, as compared to 25 percent of the remaining respondents who said the same (Table 2).

At the same time, opinion leaders displayed typical preferences when it came to the contents of the programs they generally watched. When asked which television shows they particularly enjoyed watching, an above-average share of opinion leaders chose news programs, political reports, and other informational shows, and were clearly less interested than the average population in entertainment programs (Table 3).