Literature Review

Source Credibility

Whether the goal of the communication is persuasion or the generation of understanding, contemporary research generally supports the idea that source credibility is an integral part of the communication process (McCroskey & Young, 1981). Playing such a vital role in nearly all communication, credibility's influence has far reaching effects that have only recently been explored by researchers. The credibility construct was unique to the field of communication, in that, it was not borrowed from other disciplines such as social psychology or sociology like so many early variables (Roloff & Miller, 1980). Early studies of source credibility focused on the basic characteristics, laying the foundation for future examination.

Early research on the source credibility construct within the last century focused primarily on message acceptance. Many of the basic characteristics were "understood" by communications experts, but very little testing of the validity of the characteristics was attempted. In an attempt to better understand message acceptance Hovland and Weiss (1951) conducted research that focused on communications from sources considered by subjects to be trustworthy and untrustworthy. When opinion change was measured immediately after communication, those messages from the trustworthy sources were found to be significantly more effective in changing opinion in the direction advocated by the communicator than were identical communications attributed to sources considered untrustworthy. Nearly a month later, however, there was a decrease in extent of agreement with trustworthy sources and an increase in agreement with untrustworthy sources. There were no measurable differences found in the amount of factual information retained from the two sources, immediately after or four weeks after communication.

The role of issue relevance and its impact on credibility was also analyzed as it applies to information retention. McDaniel and Vestal (1975) postulated that an individual readily accepts information that is similar to his own beliefs, values, and concepts. Results from the study indicate the high source credibility/high issue relevance group retained significantly more content than the high source credibility/low issue relevance group. The mean score for low source credibility/high issue relevance was also significantly lower, indicating the variables of source credibility and issue relevance interact to effect immediate and long-term retention of communication content.

As source credibility's influence in marketing and advertising concepts has received significant attention in recent years, the expansion of credibility issues in television has moved to the forefront. Pfau (1990) discussed source credibility issues as they relate to different mediums in his research of television's influence, noting that the factors that influence interpersonal communication are responsible for influence in television communication. Research indicated that source factors, rather than content, play a much larger role in interpersonal communication and television. Pfau (1990) discovered that television elevates person variables in the process of influence. Source credibility was the primary factor for influencing attitudes in television and interpersonal communications. Television was found to require limited involvement in message processing, consistent with the heuristic processing model. The study went on to report that the impact of source factors in persuasion was increased with television due to the use of low salience messages.

Yet another medium investigated to understand the effects of source credibility as it relates to persuasion is the newspaper industry. While television uses many of the same source factors as interpersonal communications, newspapers find themselves at the opposite end of the spectrum with regard to involvement. Kaufman, Stasson, and Hart (1999) conducted experiments to examine the influence of source credibility, among other variables, in need for cognition on perceptions of newspaper communication. Research indicated that a main effect of source credibility was found to increase perceived accuracy of the article, in that articles from the Washington Post were found to be credible, regardless of communication strength. The newspaper's high credibility rating influenced readers' opinions of the articles they read, whether the articles communicated a message effectively or not. Researchers pointed out that a source's perceived level of trustworthiness and expertise were more important than the true level of credibility. Many of the same concepts that scholars discovered in previous research turned up in the research of newspapers, specifically, articles attributed to low-credibility sources may motivate greater scrutiny among those low in need for cognition. The critical factor resulting from this study is the indication that people may dismiss factual and accurate information when it is presented by a source perceived as low in credibility. Individuals have prejudices and judgments concerning the media that can overshadow the actual quality or merit of the information the source conveys.

Source credibility's impact on the newspaper industry was also noted in research conducted by Koomen, Visser, and Stapel (2000). The study on credibility of newspapers and fear of crime found that readers of a newspaper article on street robberies, published in a credible paper, reported more fear of robbery, and fear of crime in general, than did readers who thought the article had been published in a less credible newspaper. Readers of the credible newspaper were also more concerned about robbery as a societal problem than the nonreaders. The article was only effective when the source of the news, the newspaper, was credible. Once again, researchers discovered if the source is less credible, information, although credible in itself, may be disqualified in its implications. People may believe it, but attach a tag to it, declaring it less relevant.

Through continued research and testing of the construct, source credibility characteristics provide the foundation for understanding credibility's impact in nearly all communication contexts. The progress made in the last decade by scholars who recognize the implications of source credibility in modern communication have enabled communicators to mold their messages to have greater impact on their target audience.

Heuristics

The study of heuristics and issue involvement seem to go hand-in-hand. Heuristics can be defined as simple rules that allow people to make unexamined decisions. People use heuristics because they do not want to put forth the effort required of conscious processing (Miller, 2002). Issue involvement refers to the level of importance a person places on a subject. If a person is highly involved in something, they will exert more effort to gain knowledge about it. If the topic is of little interest, they will spend little, if any, energy obtaining information (Chaiken, 1982).

In 1965, Krugman suggested that high involvement learning is absent from most consumer behavior. Dealing specifically with the effects of advertising in print media versus television, Krugman established that people could learn without cognition (1965). To test his theory, Krugman employed the use of peripheral messages or heuristics in an advertising campaign. Krugman's article, "The Impact of Television Advertising: Learning without Involvement," focused on the different cognitive approaches taken when consuming print media vice television. His research found that unlike print media, advertising messages on television were not fully comprehended by viewers, yet still appeared to impact purchases (Krugman, 1965), forming a concept or memory that was stored until it was needed (1977).

In 1982, Leippe and Elkin defined two distinct forms of involvement: response and issue. Issue involvement occurred when people were motivated to resolve issues of personal relevance. Conversely, response-involved recipients formed attitudes that were socially acceptable, or could "successfully undergo public opinion" (Leippe & Elkin, 1982, p. 270). This distinction was particularity useful when studying persuasion. If a person had issue involvement, then messages aimed at their personal goals would surely have an effect on them. Those with response-involvement will tend to pay closer attention to an actual message, as personal feelings are not a concern, therefore they would be harder to persuade (Leippe & Elkin, 1982). Also of interest were findings that people could be both issue- and response-involved, or as Leippe and Elkin called it, "multiply-involved" (1982, p. 271), on one issue. Leippe and Elkin wondered which of these two motives would override the other in the event that the motives were in conflict. They found that neither won. "Multiply-involved subjects seemed to have achieved a compromise any politician would admire" (Leippe & Elkin, 1982, p. 277). People with multiple-involvement employed skills from response- and issue-involvement, paying more attention to the message content but also aligning their attitudes near the middle ground, thereby appeasing the audience.

Zaichkowsky (1985) was interested in issue involvement as it pertained to consumer research. She concluded that issue involvement was a function characteristic of consumer behavior, yet research did not agree on a definition or measurement (Zaichkowsky, 1985). Zaichkowsky devised a Semantic Differential scale utilizing twenty items directly related to personal relevance. The scale was tested and found to have content and criterion-related validity and reliability; however, Zaichkowsky admits that no tests were conducted to verify convergent or discriminate validity (1985). Despite the lack of such information, Zaichkowsky's issue involvement scale, or some alteration of it, remains the most reliable and credible instrument of involvement measurement (Pfau, 2003).

Issue involvement came to the forefront of communication with the development of Chaiken's Heuristic-Systematic Model (HSM) in 1980. Primarily studied as it pertained to persuasion, HSM stated that people will use the most effortless modes of mental processing available to get a valid and accurate result (Chaiken, 1987). According to Chaiken, this can be accomplished in two ways, either systematically or with heuristics. As systematic processing requires more effort, people generally rely on it when they have high issue involvement. Like Krugman, Chaiken found that people will seek to minimize cognitive efforts and maximize confidence in their judgment. She called this the sufficiency principle (Chaiken, Ginger-Sorolla, & Chen, 1996). In her research, Chaiken (1987) identified several heuristic cues: attractiveness, liking/agreement heuristic, the expertise credo, the length-strength rule, and the consensus heuristic.

In 1991, Chaiken and Maheswaran re-examined research on HSM and found evidence that people may use heuristic and systematic thought processing simultaneously, what they called the additivity process. Equally important were their findings demonstrating this model could be applied not only to persuasion, but motivation as well. They concluded that people want valid and accurate attitudes and use both heuristic and systematic processing to develop these attitudes. Research also showed there were limitations to heuristic processing which could bias results. Heuristics had to be available in memory, accessible from memory and perceived as reliable (Chaiken & Maheswaran, 1991) for people to feel confident about their attitude. When heuristic processing resulted in insufficient evidence, the person would use systematic processing. However, their research concluded that heuristic processing is occurring all the time, regardless of what stage of systematic processing is being employed (Chaiken & Maheswaran, 1991).

In 2000, Pfau, Holbert, Zubric, Pasha and Lin explored the differences between print and television. Their research showed that "print places primary emphasis on the content of messages, whereas video brings into play the role and influence of sources of messages" (Pfau et. al, 2000, p. 25). Message source was one of many heuristic cues. Pfau (2003, March) returns to the study of issue involvement and heuristic processing in studying presidential debates. In this research, he found that most people are not active seekers of information but rely instead on heuristics to help guide their decisions. Pfau concludes that print messages and internet content have the greatest effect on those with high issue involvement, while television messages often cause involuntary exposure.

In spite of the fact that there is a lot of controversy over what specific heuristics people utilize and in what situations they are employed, most research points to the fact that they are utilized. Furthermore, the extent which people make use of heuristics relies heavily on the level of involvement given an issue.

Elaboration Likelihood Model

The Elaboration Likelihood Model (ELM) is a general theory of attitude change devised to explain how a person processes messages and how that processing will affect attitudes when exposed to persuasive messages. Practitioners are successfully reaching and impacting audiences with persuasive messages by using the model to give directional success to advertising campaigns, impact the prevention of alcohol abuse, and assist in education.

Attitudes, in the eyes of the researchers, are ideas and feelings people have about issues important to them. Attitudes are based on a number of life experiences and guide personal decision making. ELM is defined as a processing model with two mental routes to attitude change: the central processing route and peripheral processing route. Message processing follows the central route when a person is interested in the issue, has direct experience, or high involvement. The person may take active steps to seek out information and make an effort to digest the message before accepting or rejecting it. Peripheral route message processing is accomplished when the message or issue has very little or no relevance, the person is unmotivated to surrender to the cognition effort required, or involvement is low. In other words, they take a "mental short-cut" to accept or reject the persuasion message.

Petty and Cacioppo went further into defining the model by identifying key constructs: argument quality, peripheral cues, and elaboration. These variables affect processing direction. Argument quality refers to the strength and direction of the message. Developed messages are tested and defined. Strong messages are those developed to induce positive thoughts. Weak messages are those developed to induce negative thoughts. Peripheral cues are thought to change attitudes without the processing of a message. It is believed that they affect change by appealing to "primitive affective states" that have been associated with the object or issue. Elaboration determines the extent or direction the processing of the message with take. High elaboration leads to central processing and low elaboration leads to peripheral processing. The model explains how variables affect attitudes (Petty & Cacioppo, 1981a, 1981b, 1986, 1989) (Cacioppo & Petty 1987).

Drug information messaging is another arena for the ELM to affect persuasive messages. Christensen, Ascione, and Bagozzi (1997), used the model to test "direct to consumer" advertising messages directed at elderly patients. Their research results showed some proof that involvement level may affect whether the patients would process the messages centrally, thus if there is low involvement, the patient may not make drug-use decision based on high involvement information like drug facts and use warnings. They concluded that the model could be an effective tool for determining when information content or promotional aspects would persuade patients.

The use of the ELM as a consumer behavior model was found to be beneficial for determining when individuals are more likely to be influenced by information content or the promotional aspects of consumer advertisements for prescription drugs. ELM was used to explain how individual and situational differences among consumers determine when information content or promotional aspects of an advertisement are more likely to influence attitudes. For example, according to the ELM, a highly involved individual is more likely to pay attention to the product-relevant information presented in an advertisement. Attitudes can be determined, primarily by argument quality when elaboration likelihood is high (central route processing) but primarily by peripheral cues when elaboration likelihood is low.

ELM was applied to the field of employee recruitment and attraction to the organization by those recruits by Larson and Phillips (2002). The model brings together ability and motivation factors to explain how unrelated variables can affect the attractiveness a potential employee may have for a firm. Both ability and motivation are necessary for high elaboration. The authors found that the ELM could be used to predict motivational and ability factors related to peripheral cues and attraction. Research found that the recruiter had an effect on a job applicant only right after the interview stage; they concluded that the recruiter may only be a peripheral cue.

Research of late includes work by Hosman, Heubner and Siltanen (2002) which explores the areas that include argument strength and need for cognition among others. The strength of argument was found to have a direct relation to the processing of messages. Weak speakers appeared less credible and trustworthy, therefore drawing the audience's attention away from the persuasive messages and toward personal characteristics. The researchers failed to find a correlation for the need for cognition within their research, but recommended further research in the area.

Chaiken (1987) developed a different yet similar message processing model: The Heuristic Model of Persuasion. This alternative approach identifies two modes of processing: systematic processing and heuristic processing. Systematic processing is similar to the central processing route in ELM in that it has high involvement. Individuals are likely to give consideration to message content when processing systematically. Heuristic processing is similar to, although is not exactly like peripheral processing. Heuristic processing defines message processing with decision rules. These rules allow for decision making without effort or evaluation. It might be something long believed, and still believed, that has never been validated, and doesn't require validation by the processor. The major stray from ELM is in the belief that the two processes work together in parallel, unlike ELM in which the two processes are mutually exclusive.

The roots of the idea defined within the elaboration likelihood model were studied in the field of social psychology before being applied to the communication field and persuasion theory as it is taught today. The use of the model may be applied in the advertising and health education arenas of communication. Both arenas have a strong emphasis on the persuasion of others where practitioners want a predicted outcome from the message. Researchers have been continuously testing the model ever since it existed. This continuous research will continue to reshape the way the model can be used to affect attitude change. The model also has its criticisms. As defined by Petty and Cacioppo, ELM has only two mutually exclusive routes to processing. One of the major criticisms is that the model should adapt so that it encompasses processing in a third direction. The third option would have some level of processing within both the central and peripheral routes. An alternate processing model developed by Chaiken, the Heuristic Model of Persuasion. Maybe the elaboration likelihood model will prove to be only a foundation for the further use and acceptance of the heuristic model of persuasion.

Communication Media

In "The Myth of Massive Media Impact: Savagings and Salvagings," McGuire discusses the hows and whys for people's persistant belief that "television and other mass media have sizable impacts on the public's thought, feelings and actions even though most empirical studies indicate small to negligible effects" (McGuire, 1986, p. 174). Time usage studies done in this era do show that the average American is exposed to media twice as much as he interacts with other people. This fact leads for research on media impact and/or influence to try to get to a true reality. McGuire (1986, p. 233) evaluated the twelve most researched media effects from two perspectives, intended and unintended, and found any proposed effects from these other studies were unexpectedly minor. He suggests that the believers in this mass communication impact by television and other media need to look at how little evidence supports this impact and revamp their research programs to try to find true scientific evidence that supports their beliefs (McGuire, 1986, p. 234).

"A Channel Approach to Television Influence" by Michael Pfau (1990) examines the ways television, radio, print, interpersonal, and public address communication effect how these mass media modalities influence or persuade people. The study's results confirm that television influence is more interpersonal than the other mediums. These results also show that content drives the use of print and radio communication. Source credibility ranks number one for effecting television viewers, while content continued to be the most important effect on users of print and radio (Pfau, 1990, p. 209).

"Access, Use and Preferences for Online Newspapers" written by Hsiang Iris Chyi and Dominic Lasorsa (1999) presents research on who reads newspapers in print versus online and what newspapers are read in which modality. The authors believe that the general public will decide the future of online and print newspapers. Chyi and Lassorsa look at the public's perspective to determine the status of online newspapers based on access, use, and preference for national and local newspapers in both formats --- online and print (Chyi & Lasorsa, 1999, 3-4). Three characteristics that affect adoption rates of online newspapers are: compatibility with existing values, easy to try on short term basis, and produce easily observable results. The characteristic that discourages online newspaper adoption is complexity of use or understandability (Chyi & Larousa, 1999, 33).

Peng, Tham and Xiaoming (1999) wrote "Trends in Online Newspapers: A look at the U.S. Web" using survey and content analysis to look at the direction newspaper publishers are going particularly in the areas of advertising, readership, content, and services. Gutenberg's movable printing press made print the mass communication dominant medium and led to newspapers as the main mass medium until radio and television came into existence.
Declining readership since the 1960's, a desire to find out what causes readers to choose print (particularly newspapers) and the advent of the World Wide Web online capabilities all influence the industry and the authors' research. Internet online readers have short concentration spans requiring short text and lots of graphics, plus they prefer to be entertained, as well as informed. The study concludes that online readers are different from average hardcopy newspaper reader, because they read their local newspaper(s) but want access to more than the services that are available in hardcopy.

"Patterns of Internet and Traditional News Media Use in a Networked Community" by Scott L. Althaus and David Tewksbury (2000) looks at Internet and traditional news media use in a college networked community. After studying the use patterns and reasons people chose television, newspapers, and the World Wide Web to get information, the authors concluded that people that who like reading the newspaper will go to the Web as an additional news source, but the same effect does not seem to apply to television viewers (Althaus & Tewsbury, 2000, 62). Despite the narrow sample that is not typical of the American adult population, the study's participants used the Web primarily for entertainment and then to get news. Althaus and Tewksbury (2000, 67) study did not show much correlation between Internet use and decline in the use of traditional news media (newspapers, television, or radio).

Harlan Lebo (2003) writes a very comprehensive overview of "Year Three of the University of California-Los Angeles Internet Report" which is the formal over all study title, with "Surveying the Digital Future" as the publicity-friendly title for the 2003 report. The report's purpose is to look at the impact of the Internet on behavior and views of a national sample of 2,000 Internet users and non-users. The study looks at who is and is not online, media use and trust, consumer behavior, communication patterns, and social and psychological effects and bumps the resulting data against the previous two years' UCLA Internet Reports (Lebo, 2003, p. 13-14). This report's results found that the Internet ranks number one in importance when compared to other major media with books, newspapers, television, radio, and magazines following in order of importance. However, television ranks number one as the most important entertainment source with books, radio, magazines, the Internet, and newspapers following in rank order. Television viewing time continues to decline as Internet use rises, according to Lebo (2003 p. 13). Internet time spent online rose for the third year and the users surveyed said that it is the "catalyst for creating and maintaining friendships and family relationships. (Lebo, 2003, p.55). Although the Internet is seen as the most important information source by the sample population, it is also seen as an unreliable information source demonstrated by the first decline in credibility in the three years history of this study (Lebo, 2003, p. 82).