The concept “mass media” is a collective term that stands for a broad variety of print media like newspapers, magazines, books and electronic media like radio, television, and the Internet. The concept “newspaper,” in turn, comprises daily and weekly newspapers, and “magazines” publications like news magazines, fashion magazines, sports magazines, etc. All mass media offer different features – news analysis, news, editorials, commercial messages, and so on – and deal with a broad variety of topics – including domestic and foreign politics, economy, arts, sports. Some mass media present only visual information (printed media), some only acoustic information (radio), some primarily moving images (TV, movies), while some present them all (Internet).
The concept of the effects of mass media is also a collective term for the effects of different signals like verbal messages and nonverbal signals; of the style of coverage; depicted behavior like violence, sex and pornography; and of coverage of different topics like politics, economy, and sports. Therefore, when we discuss the effects of mass media we would be interested in the specific effects, for example, of the news on specific topics, of entertainment programs, of visual images, and so on.
Types of Media Effects
The mass media can produce a broad spectrum of effects. Among them are physical effects, effects on beliefs (knowledge), effects on attitudes and values, effects on emotions, effects on social behavior, effects on public opinion, and effects on the reputation of people covered by the media. These effects may be the consequences of media use, but they may also be a result of interactions with people who have used the media and subsequently been influenced by prior media exposure. Alternately, the media may intensify or change existing beliefs, opinions, or attitudes. Media effects may also be manifest in the acquisition of knowledge or in the imitation of behavior depicted by the mass media. In some instances, the mass media may also influence the conclusions drawn by the recipients of media messages, which might actually exceed the information presented.
Media effects may be present after just a single contact with an article and may occur within a short period of time. Alternately, observable effects of media exposure may also be the result of ongoing multiple contacts with several related articles and may set in only after a certain lapse of time. The effects of mass media may erode quickly in some cases, or they may last for a long time. Furthermore, media effects are often proportional to the intensity of coverage or the frequency of media exposure. Thus, effects may already be very strong after only a few articles (contacts) or may set in only if the media have reported often on a subject.
The effects of mass media may be limited to singular aspects such as beliefs or emotions or there may be interdependencies between various effects. For example, media information may arouse emotions that contribute to an increase in the credibility of consistent information. However, because of the multitude of possible causes and effects, general statements on the effects of the media cannot be made. The type and strength of media effects depend on the respective production techniques and consumption conditions.
History of Media Effects Research
The effects of the mass media have been and are still being examined by a considerable number of researchers who have made substantial contributions. Some of them are outstanding and deserve special attention, among them Albert Bandura, Steven H. Chaffee, Leon Festinger, George Gerbner, Carl I. Hovland, Elihu Katz, and Elisabeth Noelle-Neumann. Following the work of these pioneers, the history of media effects research is often roughly divided into three phases: the phase of strong media effects (c.1910 –1945), that of weak media effects (c.1946 –1970), and that of moderate media effects (c.1971 onwards).
The first (strong effects) phase is characterized by case studies based on outstanding events. Two of the most notable are the panicky reactions to the Orson Welles radio play War of the Worlds (Cantril 1940) and Kate Smith’s radio campaign for the sale of war bonds that yielded US$39 million within 15 hours (Merton 1946). Even after the end of the first phase, strong media effects were documented most thoroughly by Hovland and his colleagues. Their results, however, were relativized because of the limited external validity of their laboratory experiments where there was no possibility of selective exposure and unnaturally high attention levels among subjects.
The second (weak effects) phase started with Lazarsfeld et al.’s (1944) study on the 1940 presidential election in the USA. The election was a recurring routine event and the coverage was less spectacular and the effects far less dramatic. The authors offered a range of explanations for what was at the time a surprising result. Most prominently, they introduced the two-step flow of communication model, the influence of opinion leaders, and the role of selective media use.
The third (moderate effects) phase started with McCombs & Shaw’s study on “The agenda-setting function of mass media” (1972) and Noelle-Neumann’s programmatic article on the “Return to the concept of powerful mass media” (1973). McCombs & Shaw shifted the attention of researchers from media effects on attitudes and opinions to their effects on beliefs. Noelle-Neumann introduced three ground-breaking concepts: the consonance of the media, their cumulative effects, and their influence on the climate of opinion. These constructs became the basis of the spiral of silence theory. The theoretical reorientation of the field was accompanied by a growing number of long-term studies based on content analyses of media coverage and representative surveys. Thus the development of media coverage and public opinion could be categorized and statistically analyzed.
From today’s point of view, the differences in effects in the three phases were smaller than they have historically appeared. As far as such differences actually existed, they can be explained by the specifics of the events covered as well as by theoretical and methodical innovations in research. In cases of outstanding events, the effects today are at least as strong as the effects that were detected during the first phase of media effects research. For example, two weeks after the attacks of September 11, 2001 on the World Trade Center, American celebrities collected US$116 million for the victims within two hours by holding a benefit concert that was broadcast on TV and radio. For routine events, the effects correspond approximately to what was actually determined in the second phase of media effects research. That is, in the case of long-term thematization of facts and opinions, the beliefs of a remarkable part of the population follow the media’s tenor.
Theories of Media Effects
The effects of mass media can be described by models and explained by theories. Theories on the effects of the mass media can be roughly categorized into two general types: learning theories and cognitive theories.
Despite numerous similarities between the two theoretical bases, there is one important difference. Specifically, learning theory approaches address the correct reproduction of information, and therefore divergences between beliefs and information provided by media are considered learning deficits that may also be interpreted as a lack of media effects. In contrast, cognitive theories do not necessarily understand deficits as a lack of media effects, but rather as the result of predictable contributions of recipients who process the information provided. Thus, because of this essential difference the two theories are treated separately even though they share certain similarities.
Learning Theory Approaches
According to learning theory, learning is based on the association (conditioning) of stimuli and reactions. In the present context, operant conditioning is most significant. This describes the accidental co-occurrence of a stimulus and a reaction that is established as a unit if the reaction is rewarded (reinforcement). In the area of media effects the reward could, for example, be the social recognition awarded by friends and peers for knowledge about political incidents a viewer gathered from a news program watched by chance. Presumably, such processes are important motivators for individuals to become more vigorous media consumers.
A variation of learning theory is the theory of observational learning, which is relied upon above all to explain the acquisition of aggressive behavior. Accordingly, observers learn patterns of aggressive behavior in mass media that they later apply under certain conditions of action. In most effects studies oriented to learning theories, the key variables – objective contents and subjective understanding, attentiveness and effects, cognitive and emotional aspects of the effect, and others – are not differentiated and recorded separately. This is because, in existing studies available for secondary analyses, the required data are often missing, and measuring them all simultaneously in primary studies would be too difficult and expensive.
Remembering Media Content
No one can remember all the media reports they have read, heard, or seen. This raises two related questions. First, what percentage of media information on current events is retained? Second, what percentage of knowledge about current events is based on media information? Research has been shown that immediately after watching television news, viewers remember an average of two to three news stories out of 15. From this and similar results it can be concluded that a single report about an event normally leaves nearly inconsequential effects on the memory. Only repeated reports are effective, and the number of repetitions has to exceed a critical threshold.
From the poor recollection of single reports, it may be assumed that media coverage generally has a small influence on the beliefs a population holds about current events. This assumption, however, is wrong. Evidence has shown that about two-thirds of what ordinary people know about current events derives from media reports (Früh 1992). The reason for the apparent contradiction is the large number and frequency of media reports on all important events, which therefore structure the dominant beliefs of the population.
Influence On the Perceived Relevance of Issues
Media coverage on current affairs has an influence on the public’s assessment of the significance of social problems and the urgency for solving those problems. The theoretical basis of the agenda-setting hypothesis is the notion that the media have no strong influence on what the recipients think, but a considerable influence on what they think about. Two main research designs have been established in agenda setting research. The first constructs a comparison of all issues on the media’s agenda with the population’s agenda over a short period of time such as a few weeks. The second examines a comparison of the development of media coverage on single issues with the development of the population’s beliefs over a longer period of time of several years or more.
The original formulation of the agenda-setting hypothesis includes the unspoken assumption that public opinion follows the ups and downs of the emphasis placed on issues by the media. This assumption, however, does not apply to all cases. For example, the population often remains worried after massive reports on serious disasters even after the coverage has already decreased. In these cases, there is no linear relation between the media coverage and public opinion. If correlations that assume linearity between variables are calculated, the results paradoxically demonstrate weak effects. To avoid this error several nonlinear models have been developed.
Influence on Assessment of Risks
Individuals generally have good judgment concerning the relative frequency of causes of death. They are aware of causes of death that are more likely and those that are less frequent. They typically make two mistakes, however, by overestimating the occurrence of rare fatalities and underestimating the occurrence of frequent causes of death. The concept of availability heuristic explains how this is related to media coverage. Since exceptional events have high news value the media report intensively on rare causes of death such as tornados, poisoning, lightning strikes, and so on, but relatively seldom on frequent causes of death including heart attacks, lung cancer, diabetes, and other common life-threatening diseases.
People focus less on the real frequency of causes of death and more on the emphasis in current media coverage. Therefore, the beliefs of the population on chances and risks follow the representation of unlikely but possible damages rather than the development of the actual number and size of damages.
Influence on Beliefs about the Distribution of Opinions and Behaviors
Media reports contain two types of information on the distribution of opinions and the relative frequency of behaviors. One type common in media coverage is quantitative information, such as used in reporting polls where a certain percentage of respondents share a position. The other type is reports on individuals who represent different behaviors and opinions. The presentation of individuals has a stronger influence on the beliefs about majority opinions than quantitative statements. If, for example, a report specifically over-represents members of the minority, many recipients consider the minority to be the majority even if the same report contains quantitative statements on the actual majority opinion.
Influence on Suppositions about Effects
Most people estimate unwanted (negative) media effects to be higher on others (alter) than on themselves (ego). The concept of third-person effects was introduced by Davison (1983). The occurrence of the effect was confirmed by several surveys and experiments, which also found that the power of third-person effects increases with the social distance between alter and ego. Third-person effects are also enhanced when the sizes of the comparison groups are greater.
The source of the third-person effect may be an overestimation of media effects on other individuals, an underestimation of the effects on oneself, or a combination of both. The third-person effect presumably has important consequences, but these remain relatively unexamined. There are indications that it reduces the disposition to present one’s own opinions in public controversies. Also, third-person reasoning has been shown to promote the demand for banning certain media content that is believed to have undesirable effects, such as pornography.
Influence on the Distribution of Knowledge in Society
In the 1940s it was already well known that there was a connection between education and the use of information presented by the media. In simple terms, more educated individuals make better use of information offered by the media. Nevertheless, social scientists in the 1950s continued to assume that the distribution of the media would improve the knowledge of the population in general. This optimistic assumption was called into question by Tichenor et al. (1970) and their discovery of the knowledge gap. This framework identified the growing gap between the knowledge of higher-educated and lower-educated groups that develops over the course of coverage on singular events or issues.
Influence on the Perception of the State of Society
The cultivation hypothesis developed by Gerbner and his colleagues is based on the assumption that the media – and above all TV – are an important factor of cultural and political socialization. Through both information and entertainment TV conveys ideas of the state of society in which people live. The more frequently and intensely people watch TV, the stronger the influence of its presentation of reality. The influences of information and entertainment presented on TV co-mingle and therefore have to be regarded in conjunction.
On the individual level, intensive use of TV leads to inaccurate beliefs in the frequency of violent acts in society. On the societal level, widespread intensive use has been linked to the homogenization of beliefs and opinions on current issues. In contrast to the significantly more limited knowledge-gap hypothesis, which examined shorter periods of time and singular issues, the cultivation hypothesis deals with longer periods of time and general beliefs, opinions, and attitudes.
Cognitive Theory Approaches
Cognitive theories generally inquire as to how people perceive their environment and themselves, how they interpret it in a reasonable way, and how they experience it emotionally. These theories are centered on human information processing and its triggers, which include media presentations. Beliefs, opinions, attitudes, emotions, and actions are thereby considered interrelated parts of a system. Cognitive theories are based primarily on two assumptions that were first outlined by Fritz Heider in the 1940s, but initially received scant attention in research.
The first assumption, which became the basis of consistency theories was that people do not perceive reality surrounding them as single elements, but as prestructured units. The second assumption, which later became the foundation of attribution theories is that people intuitively attribute causes to occurrences. According to cognitive theories, beliefs and opinions are not merely copies of media presentations; in fact, most clearly go beyond the presentations but are based on them. Differences between media presentations and public beliefs do not indicate a lack of effects, but rather signify information processing predictably triggered by media reports.
Influence on Schema-Conducted Processing of Information
Schema theory is based on the assumption that media recipients do not take up individual pieces of information independently of one another and derive meaning from them, but interpret them consistently according to a predetermined schema. Schema theory was introduced into effects research by Graber (1984). Schema-induced information processing can be controlled by media reports that present events from a certain perspective.
For example, differently framed crime news (possibilities of fighting against crime, suffering of victims, or the responsibility of government, for example) stimulates public interpretations that correspond to the respective frame. Thus, personal schematic information consistent with the frame of news is retained better than information that contradicts it. In addition, readers, listeners, or viewers tend to believe they have received information consistent with the frame even when such information was not actually included in the reports.
Influence on the Relevance of Individual Characteristics of Politicians
Priming theory is based on the assumption that certain emotions, thoughts, and memories can be seen as knots in a network that are more or less closely connected to other parts of the network. Information activates parts of the network and sensitizes them for similar information. Thus, such information that follows will be consumed more attentively and processed more intensively.
The theory was introduced into media effects research by Iyengar and Kinder in 1986 and included several crucial assumptions. Citizen assessments about politicians and voting intentions are based in part on beliefs about politicians’ competence. Repeated coverage of issues sensitizes recipients to some issues and makes solutions to the issues seem especially urgent. Thus the presumed ability of politicians to deal with the issues becomes more significant, contributing to a positive or negative image of them. General opinions as well as voting intentions may therefore change, although judgments about the individual characteristics of politicians may remain constant.
Influence on the Interaction of Cognitions and Emotions
Appraisal theory (Nerb et al. 1998) combines elements of attribution theory and emotional arousal theory. Appraisal theory operates under conditions where media reports on great damages trigger high attention and evoke great excitement.
Descriptions of events trigger predictable emotions. If the damage is attributed to uncontrollable natural forces, the event evokes sadness, but if it is attributed to a person acting in a controlled way, it evokes anger. The extent of reactions is enforced or diminished by the interaction of emotions and cognitions. Therefore, emotions are not only effects of media coverage but also causes of beliefs and vice versa.
Influence on Public Opinion
According to Aristotle, people depend on the society of others and therefore fear social isolation (zoon politikon). The basic ideas of this notion were introduced into media effects research by Noelle-Neumann (1984), who suggested that people constantly monitor their environment to stay abreast of popular majority opinion in order to avoid social isolation.
Therefore, individuals often draw on their interactions with other people and personal observation as well as media presentations. Each of these resources can incidentally stimulate correct or incorrect ideas about the distribution of opinions. Nevertheless, in the long term most people will be influenced more strongly by media presentations than by their personal observations. People who consider themselves in the minority tend to withhold their opinions in public. In so doing, the presumed majority opinion is artificially inflated, which in turn increases the pressure on the actual or alleged minority.
Influence on Opinion Formation in Public Controversies
The theory of cognitive affective media effects incorporates elements of Rosenberg’s (1956) affective cognitive consistency theory and applies them to the analysis of public controversies (Kepplinger et al. 1991). According to the authors of cognitive affective media effects theory, all conflicts comprise events or factors that favor one side or the other (issue ownership). In this sense, the events have an instrumental quality where opponents and unsympathetic media coverage emphasize information that marginalizes competing claims.
Readers, listeners, and viewers prefer information supporting their own view, but are confronted with a considerable amount of contradictory information. The knowledge of information that favors one side, or is against the other side, influences opinions about the opponents. This in turn increases or diminishes the range of legitimate behavior and thereby minimizes the chances of success of the opponents.
De Facto Effects
In studies based on learning and cognitive theories, the information provided by mass media is identified and the processing strategies analyzed. This procedure is based on a significant reduction of the complexity of coverage, since in reality the relevant information is not isolated but presented as part of a mixture of correct and incorrect assertions of facts, presumptions, and opinions. Furthermore, there are many ways in which recipients of mediated reports might be influenced. For example, individuals can develop and form their opinions by perusing statements of opinion from journalists and other sources, but they can also derive their opinions from neutral information. Accordingly, the beliefs and opinions of recipients may be influenced by a broad variety of sources that can hardly be isolated in the flow of media information and that converge to create a more or less consistent picture of reality. Such effects are called “de facto effects,” or the trap effect.
De facto effects may always be assumed when the formation of opinion follows the media tenor. In these cases, effects of media coverage have to be assumed, despite the fact that they cannot be traced back to single sources and explained by specific theories. Examples of de facto effects include the development of opinions about the state of political parties or about the chances of parties and candidates in election campaigns. Other examples involve opinion development about public officials, such as the American President George Bush or the German Chancellor Helmut Kohl, or about issues like the development of the economy and ecological scandals. In this context, studies on cultivation theory must also be considered.
Most studies in the effects of mass media are based on three doubtful, mostly unspoken, axioms. These axioms have a considerable influence on the design of effects studies, the relevance of theories for the explanation of media effects, and the interpretation of the social importance of the results of effects research.
The first axiom is: Events happen, media cover. According to this axiom, current events on which the media report happen independently of the media. The validity of this axiom, however, is doubtful because a number of events on which the media report are the result of previous coverage. Furthermore, in conflicts and crises, people and organizations on which the media report change their behavior as a result of media coverage. For example, if coverage by influential media is critical, politicians may change their strategy, parties may withdraw parts of their agenda, governments may change their mind about the implementation of changes in laws, or companies may phase out or otherwise modify their products. When the media follow up and report on these decisions, strictly coverage often speaks about the effects of previous coverage. Another reason why the first axiom is untenable is because many of the events on which the media report would not take place or happen in the way they do if the actors did not expect media coverage. This applies to all press conferences, but also to party congresses, sporting events, and other planned media-centric events.
There are three different types of events that can be distinguished. First, some events happen independently of the media, such as natural disasters, accidents, or terrorist attacks (genuine events). Second, certain events would happen without media coverage, but their character is modified by media coverage (mediated events). Third, there are events that happen only in order to generate media coverage (staged events), such as press conferences. After spectacular events, a large proportion of the press reports is not devoted to the key events that had originally triggered media coverage. Instead, media reports tend to focus on staged and mediated events that are thematically related (Kepplinger & Habermeier 1995), which is partly due to the reciprocal effects of the mass media. That is, the subjects of media reports frequently adapt their behavior to meet the conditions of successful portrayal in media coverage. The subjects of coverage are a comparatively small number of people, most of whom wield considerable influence. Thus, decisions influenced by media reports may affect many more people indirectly.
The second axiom is: No effect without change. This suggests that the reinforcement of existing beliefs, opinions, attitudes, and behaviors cannot be regarded as an effect of media coverage. The axiom holds true only under two conditions. First, even if the media did not support the existing beliefs, opinions, attitudes, and behaviors of its audience, these characteristics and attributes would still exist. At most this assumption applies to attitudes and habits, but it is not true of most beliefs and opinions. An example is public assessments of the most important problems facing the nation: if the media report less frequently than they used to on a particular issue, fewer people consider it to be a problem.
The second condition under which the axiom holds true is that the relevant beliefs, opinions, attitudes, and behaviors have developed independently from previous media use. This assumption, however, also applies only in part. One example is the processing of TV news, where there are two kinds of effect. First, where viewers don’t know anything about the events or individuals covered, new knowledge and opinions are at least partly the effects of TV coverage. Second, where viewers already know the events or individuals covered and have opinions on them, their knowledge and opinions are mostly based on previous media reports. Thus, it is primarily the mass media that establish the context for the interpretation of the news on current events, which in turn reinforce and supplement the context (Kepplinger & Daschmann 1997).
The exclusive interpretation of reinforcement as evidence for a lack of media effects is a result of a static and thereby unhistorical approach. Such an interpretation does not do justice to the actual role of the media in the development of individuals and societies. In reality, the status quo is often a result of previous media effects. From these theoretical notions one has to conclude that existing beliefs, opinions, attitudes, and behaviors can also be regarded as effects of previous media coverage. Thus, insofar as these beliefs, opinions, attitudes, and behaviors would vanish without ongoing media coverage, reinforcement must be regarded as a central and important effect of media coverage.
The third axiom is: No effect without contact. This axiom is acceptable if at least one of the two following conditions is fulfilled: first, existing attitudes and peer bonds largely prevent the reception of dissonant information; second, dissonant information that is received from the mass media will be reinterpreted in such a way as to correspond to existing attitudes. If one or both conditions are fulfilled, the opinions and attitudes of media users are an efficient protection against the transmission of dissonant information to others. In this case, effects implied by the conveyance of information from the media must be attributed to the conveyors. Only then do conveyors work as filters that marginalize the effects of media on third parties considerably. However, when conveyors or opinion leaders also use and pass on dissonant information, the effects should be attributed at least partly to the media.
There are several reasons why the assumption that people exclusively use consonant media information and exclusively pass on consonant media information to third parties is incompatible with the findings of empirical research. One is that attitudes have only a small influence on the use of most mass media, especially TV newscasts, which are the most important and frequently utilized source of information. Another, related reason is that media recipients do not use negative information about people and organizations that they respect in a selective way. Additionally, the tendency toward the selective use of positive information can be overcome by journalistic means of design. Among these are big or spectacular headlines, long and well-placed articles, and ostentatiously illustrated reports (Donsbach 1991).
For these reasons, the preference for consonant information does not present an impermeable barrier for dissonant information on any level of selection. Even though media effects may be reduced by such preferences, the effects of the mass media are certainly not eliminated. Accordingly, predispositions and selective exposure do not represent an efficient protection against the reception of dissonant information. Therefore, the no effect without contact axiom is untenable. Opinion leaders and other interlocutors do not restrain the influences of media reports, but rather extend them in unique ways to those who lack direct contact with the same media coverage.
The model of indirect media effects has consequences for the interpretation of multilevel effects processes. This concerns, first, studies that have shown an influence of interlocutors on the beliefs and opinions of other people. Provided that no theoretical reasons or empirical facts suggest that the interlocutor relied on sources independent of the media, the effects must be assigned to the media. Second, it also concerns studies in the diffusion of information and innovation and in the change of social values and norms. Opinion forming mass media with a low distribution level may alter the perceptions of minority groups, which may then be adopted by media with high distribution levels and thereby shape the beliefs of the majority. Third, the consequences of the model of indirect media effects concern contemporary studies of the role of the mass media in historical developments. This addresses mainly questions as to whether and how the opinion-forming media shape public opinion, restrict the scope of decision-makers, or predetermine which kind of decisions are deemed legitimate and appropriate.
The model of indirect media effects outlines a basis for discussing the role of the media as a catalyst, amplifier, and moderator of dynamic complex processes that hitherto lay outside the interests of recipient-fixed effect research. At the same time, it points out connections to questions and approaches of other disciplines such as political science (studies in decision-making), sociology (studies in conflicts and crises), and economics (studies in consumer behavior, development of supply and prices). The more general the questions asked, the further removed the causes (i.e., media coverage) become from the effects (e.g., decisions made by politicians as a consequence of media coverage), and the greater the number of intervening variables that have to be taken into account (such as opinion leaders who derive their opinions from the media and might influence decision-makers). The disadvantages of such studies are in the inevitable reduction of the level of evidence, but what is important is that the advantages include a realistic estimation of the media’s role in modern societies.
- Bryant, J., & Zillmann, D. (2002). Media effects: Advances in theory and research, 2nd edn. Mahwah, NJ, and London: Lawrence Erlbaum.
- Cantril, H. (1940). The invasion from Mars: A study in the psychology of panic. Princeton, NJ: Princeton University Press.
- Davison, W. P. (1983). The third-person effect in communication. Public Opinion Quarterly, 47(1), 1–15.
- Donsbach, W. (1991). Exposure to political content in newspapers: The impact of cognitive dissonance on readers’ selectivity. European Journal of Communication, 6(2), 155 –186.
- Früh, W. (1992). Realitätsvermittlung durch Massenmedien: Abbild oder Konstruktion? [Pictures of reality caused by the mass media]. In W. Schulz (ed.), Medienwirkungen: Einflüsse von Presse, Radio und Fernsehen auf Individuum und Gesellschaft [Effects of mass media: The impact of press, radio and TV on individuals and society]. Weinheim: VCH, Acta Humaniora, pp. 71–107.
- Graber, D. (1984). Processing the news: How people tame the information tide. New York: Longman.
- Iyengar, S., & Kinder, D. R. (1986). More than meets the eye: TV news, priming, and public evaluations of the president. In G. A. Comstock (ed.), Public Communication and Behavior, 1, 135 –171.
- Kepplinger, H. M., & Daschmann, G. (1997). Today’s news – Tomorrow’s context: A dynamic model of news processing. Journal of Broadcasting and Electronic Media, 41(3), 548 –563.
- Kepplinger, H. M., & Habermeier, J. (1995). The impact of key events on the presentation of reality. European Journal of Communication, 10(3), 371–390.
- Kepplinger, H. M., Brosius, H. B., & Staab, J. F. (1991). Instrumental actualization: A theory of mediated conflicts. European Journal of Communication, 6(3), 263 –290.
- Lazarsfeld, P., Berelson, B., & Gaudet, H. (1944). The people’s choice: How the voter makes up his mind in a presidential election. New York: Columbia University Press.
- McCombs, M. E., & Shaw, D. L. (1972). The agenda-setting function of mass media. Public Opinion Quarterly, 36(2), 176 –187.
- Merton, R. K. (1946). Mass persuasion: The social psychology of a war bond drive. New York: Harper.
- Nerb, J., Spada, H., & Wahl, S. (1998). Kognition und Emotion bei der Bewertung von Umweltschadensfällen: Modellierung und Empirie [Cognition and emotion in the evaluation of environmental accidents: Modeling and empirical studies]. Zeitschrift für Experimentelle Psychologie, 45(4), 251–269.
- Noelle-Neumann, E. (1973). Return to the concept of powerful mass media. Studies of Broadcasting, 9, 67–112.
- Noelle-Neumann, E. (1984). The spiral of silence. Chicago: University of Chicago Press.
- Perloff, R. M. (2003). The dynamics of persuasion: Communication and attitudes in the twenty-first century, 2nd edn. Mahwah, NJ: Lawrence Erlbaum.
- Perse, E. M. (2001). Media effects and society. Mahwah, NJ: Lawrence Erlbaum.
- Rosenberg, M. J. (1956). Cognitive structure and attitudinal effect. Journal of Abnormal and Social Psychology, 53(3), 367–372.
- Schenk, M. (2002). Medienwirkungsforschung [The effects of mass media], 2nd edn. Tübingen: Mohr.
- Tichenor, P. J., Donohue, G. A., & Olien, C. A. (1970). Mass media flow and differential growth in Public Opinion Quarterly, 34(2), 159 –170.