mass media effects: empiricist approach

Mass media: effects research - empiricism

The 'empiricist' tradition

The empiricist researchers were concerned to find out as much as possible about media audiences, in much the same terms as advertisers today would seek information from, say the NRS: number of people, age, sex, social status, occupation, leisure and so on.

By and large these data tended to be used to support studies into the effectiveness of communication, rules for mounting effective campaigns and so on.

Contemporary commentators on media research are frequently dismissive of the 'scientific', experimental methods often employed in early empiricist 'Effects Research'. Whilst there is much to criticize in this approach, the critics often unfairly overstate their case, disregarding the methodological diversity which did exist at the time. Such diversity was often forced upon the researchers by the realization that their 'scientific', 'positivistic' approach was based on a transmission model of communication which conceives of a message being sent from sender to receiver, disregarding institutional, psychological, cultural and other factors which contribute to any possible effects the media may have.

Hovland

Very important amongst these researchers was Carl Hovland of Yale whose carefully controlled experiments were designed to test the separate variables in the communication process. The main focus of his research was persuasion. Many of the principles he established are generally accepted today - one finds them being repeated, in one form or another, by, for example, political spin doctors, PR people, advertisers. However, it's worth bearing in mind that such people are trying to sell their services and so may be making greater claims for Hovland's principles than they deserve. Certainly, as mentioned above, many contemporary critics would criticize the unashamedly positivist approach adopted by Hovland, an approach which implies that it is possible to discern general 'rules' for effective and persuasive communication.

Please click here for more details of the Hovland approach:

Lazarsfeld

Paul Lazarsfeld was also a very important researcher who contributed much to the development of empirical methods in the social sciences during his work at the Columbia Bureau of Applied Social Research. The most famous of the studies he conducted was that into voting behaviour carried out in the 1940s and which led him to develop the highly influential Two Step Flow Model of mass communication.

As a result of his research, Lazarsfeld concluded that the media actually have quite limited effects on their audiences. This view of the media is common to many of the researchers in the US. Hovland, for example, whilst showing what variables can be altered to make a communication more or less effective, also places considerable emphasis on those factors, especially social factors such as group membership, which limit the persuasiveness of the message. Consequently, this view of the media is often referred to as the 'limited effects' paradigm or tradition.

Limited effects

In Towards a Sociology of Mass Communication (1971), McQuail summarises some of the main findings of the research which confirms this 'limited effects' view:

'persuasive mass communication is in general more likely to reinforce the existing opinions of its audience than it is to change its opinion' (from Klapper (1960))
'people tend to see and hear communications that are favourable or congenial to their predispositions' (from Berelson & Steiner (1964))
'people respond to persuasive communication in line with their predispositions and change or resist change accordingly' (from Berelson & Steiner (1964))

Consequently:

'political campaigns tend to reach the politically interested and converted', as shown for example in Lazarsfeld's research
'mass media campaigns against racial prejudice tend to be unsuccessful', as demonstrated in Kendall and Woolf's analysis of reactions to anti-racist cartoons. The cartoons featured Mr Biggott whose absurdly racist ideas were intended to discredit bigotry. In fact 31% failed to recognise that Mr Biggott was racially prejudiced or that the cartoons were intended to be anti-racist (Kendall & Wolff (1949) in Curran (1990)).
'effects vary according to the prestige or evaluations attaching to the communication source', as demonstrated by Hovland
'the more complete the monopoly of mass communication, the more likely it is that opinion change in the desired direction will be achieved' - as in totalitarian societies, such as Nazi Germany, for example
'the salience to the audience of the issues or subject matter will affect the likelihood of influence: "mass communication can be effective in producing a shift on unfamiliar, lightly felt, peripheral issues - those that do not much or are not tied to audience predispositions"' (from Berelson and Steiner (1964)). This is also supported by the recent research of Hügel et al, who confirm other studies' findings that media agenda-setting effects are limited to unobtrusive issues. (Hügel et al (1989))
'the selection and interpretation of content by the audience is influenced by existing opinions and interests and by group norms', as suggested by Hovland's research
'the structure of interpersonal relations in the audience mediates the flow of communication content and limits and determines whatever effects occur', as suggested by Katz and Lazarsfeld's research.
(For more comment on limited effects, see the conclusions of the more recent research conducted on behalf of the BBFC)

Powerful effects

Schramm (1982) points to three powerful effects which the media can exert and which are pointed to by the research of the Columbia Bureau:

the media can confer status on organisations, persons and policies. As Schramm suggests, we probably work on the assumption that if something really matters then it will be featured in the media; so, if it is featured in the media, it must really matter;
the media can enforce social norms to an extent. The media can reaffirm social norms by exposing deviation from the norms to public view - this connects with British research by Cohen into folk devils and moral panics;
the media can act as social narcotics; sometimes known as the narcotising dysfunction, this means that because of the enormous amount of information in the media, media consumers tend not to be energised into social action, but rather drugged or narcotised into inaction.

Violence and Delinquency

As mentioned above, the empiricist vein of research in the US was funded to a large extent by major corporations concerned to investigate the influence of their advertising and public relations and by political parties which wished to devise the most effective campaigns. Another important impetus came from the government which responded to widespread public concern about media (especially film and then, later, television) portrayals of violence and their possible link with juvenile delinquency. The nature of the assumed links was then and continues to be unclear and confused. Klapper (1960) reduced the assumptions to six basic forms:
mass media messages containing the portrayal of crimes and acts of violence can

be generally damaging
be directly imitated
serve as a school of crime
in specific circumstances cause otherwise normal people to engage in criminal acts
devalue human life
serve as a safety valve for aggressive impulses

In essence, it is these assumptions which continue to underlie public concern over the media's possible harmful effects, notably on children. This concern has been reflected in the government funding of research into media violence and delinquency, both here and abroad. It is also reflected in the very extensive legislation in the UK (see the sections on regulation), and in frequently stated media assumptions that violent media messages cause violence. Because it is a matter of such widespread concern, there is a separate section on research into violence. Please click here to go to that section:

For an overview of each 'tradition', please click below:

Hypodermic needle model

Cultural effects

Uses and gratifications

Recent developments