Source Credibility Effects: a Test of Behavioral Persistence

Ruby Roy Dholakia, University of Rhode Island
ABSTRACT - The effect of source credibility on behavioral persistence is tested in an experimental setting The results support the behavioral hypothesis based on self-perception theory the moderately credible source is found to facilitate greater behavioral persistence than a more credible source
[ to cite ]:
Ruby Roy Dholakia (1987) ,"Source Credibility Effects: a Test of Behavioral Persistence", in NA - Advances in Consumer Research Volume 14, eds. Melanie Wallendorf and Paul Anderson, Provo, UT : Association for Consumer Research, Pages: 426-430.

Advances in Consumer Research Volume 14, 1987      Pages 426-430


Ruby Roy Dholakia, University of Rhode Island


The effect of source credibility on behavioral persistence is tested in an experimental setting The results support the behavioral hypothesis based on self-perception theory the moderately credible source is found to facilitate greater behavioral persistence than a more credible source


There appears to be some acceptance of the belief that the characteristics of a communicator influence an audience's reception of a message This belief affects not only the use of testimonials for commercial advertising but also the choice of endorsers for political candidates and social causes

Experimental research dealing with the persuasive effect of source credibility has created some support for this premise (cf Sternthal, Phillips, and Dholakia 1978) Researchers have found that sources having more of the credibility dimensions induce greater attitude change immediately than do sources having less of these dimensions (Hovland and Weiss 1951; Johnson, Torcivia, and Poprick 1968; Kelman and Hovland 1953; Miller and Baseheart 1969; Schulman and Worrall 1970; Warren 1969; Watts and McGuire 1964; Whittaker and Meade 1968; but not Hovland and Mandell 1952) In more commercial settings, research indicates similar effect of source credibility on readership scores (Freeman 1957) and product and company attitudes (Fireworker and Friedman 1977; Friedman and Friedman 1979; Kamer, Azhari, and Kragh 1975; Mowen and Brown 1981)

However, studies that have looked at the interactions of source credibility with other variables have not supported the greater effectiveness of higher credibility sources The conditions that have been reported to interact with source credibility include (a) message conditions such as message congruence with source's self-interest (Eagley and Chaiken 1975; Walster, Aronson, and Abrahams 1966), and the novelty of message arguments (McCroskey 1969); (b) audience characteristics such as level of authoritarianism (Johnson, Torcivia, and Poprick 1968), issue-involvement (Johnson and Izzet 1972; Johnson and Scileppi 1969; Rhine and Sevarance 1970) or locus of control (Ritchie and Phares 1969)

Furthermore, when researchers have focused on behavioral change and not only on attitudes, less credible sources have been found to be more effective than higher credibility sources (Dholakia and Sternthal 1977; Hill, Smith, and Mann 1986; Tybout 1978; Powell 1965; except Brock 1965) If one's behavior is available as a cue, it appears that attitude change is greater among those exposed to a less credible source (Dholakia and Sternthal 1977; Tybout 1978) If the behavioral manipulation was perceived to be a choice situation, then it was observed that the less credible source was more persuasive when compliance was perceived to be voluntary (Eagley and Chaiken 1975; Himmelfarb and Arazi 1974; Jones and Brehm 1967)

It appears from a review of the experimental literature that the use of source credibility for persuasion must consider the situational context and rely on different models to explain the outcomes (Sternthal, Phillips, and Dholakia 1978) Specifically, cognitive response analysis (Greenwald 1968) may be employed to predict that highly credible sources will be more persuasive than low credibility sources when only the source and communication- related cues are available On the other hand, when an individual's own behavior is also available as a cue, then self-perception theory (Bem 1972) can be employed to predict that low credibility sources will be more persuasive than highly credible ones

If the implications of these models are to have any practical and theoretical significance, it must be determined whether the reported persuasive effectiveness of a less credible source, when behavior is available as a cue, results in repeated behavior. In other words, can a less credible source elicit great behavioral response in the future? This is particularly relevant to consumer behavior where repeat behaviors are consistently sought through marketing actions


The purpose of the investigation reported in this paper is to test the persistence of the source credibility effect While there are some studies that have examined the effect of source credibility on behavior (Brock 1965; Dholakia, and Sternthal 1977), there are no studies that have looked at the impact on behavior over time The study examines a situation where an individual has available one's own behavior before responding to a subsequent request Second, it examines the impact of source credibility on behavioral responses over time

Research Hypothesis

If the subjective causal inference process implied by self-perception theory is valid, then one can hypothesize greater behavioral response from individuals exposed to a low credibility source than those exposed to a highly credible source This relationship is expected in only those communication situations where initial behavior is Available as a cue to assess one's attitudes.

The support for this hypothesis is drawn from attribution theory (Kelley 1975) This theory can be used to specify the process by which people evaluate communications from others, and it identifies the attitudinal effects of this evaluation (Kaplan 1976) For individuals who enact a behavior in response to a communication, observing that the behavioral compliance occurred despite message advocacy by a low credibility source should augment the attribution of behavior to internal causes Having made this causal attribution, subsequent behavior becomes more likely from those exposed to a low credibility source Conversely, if the behavioral compliance can be attributed to an external cause like a high credibility source, then individuals are likely to be less favorable about the issue and also less certain about the attitudinal inference Subsequent behavior, therefore, 18 likely to be more uncertain from those exposed to a high credibility source.


Research Setting The experiment was conducted in India while the author was visiting one of the Institutes of Management Consumer activism was in its nascent stages Development and marketing trends highlighted weaknesses and gaps in the delivery of customer satisfactions, and the stage was being set for the emergence of Nadar-style consumer activism. It provided an opportunity to test research implications of prior source studies conducted by the author in a different socio-economic context while staying within a comparable issue/topic

Subjects Sixty-one participants in the study were recruited from an introductory graduate course in management During one of the regular class hours, the students were asked to participate in a study on consumer protection The sponsors of the study were described as an external group who had sought the assistance of the course instructors to administer the study Participation was made voluntary, and all chose to cooperate The participants were randomly assigned to the two-source credibility conditions

Procedure Experimental subjects were given booklets containing the independent and dependent variables The experimental task involved reading a one-page message dealing with the need for a consumer-protection organization to meet the growing problems of consumer rights in India After reading the message, the individuals were asked to volunteer for a consumer-protection organization Then, they were assessed on their attitudes toward consumer protection and their perceptions of the source One day after reading of the one-page message in class, all individuals were contacted through an internal mall delivery system with a second request Finally, three months after the initial contact, a meeting was scheduled with a prominent consumer activist as the featured speaker Notice of the meeting was again delivered to individuals through the internal mail system

Independent Variables. Source credibility was the only independent variable manipulated in the study The source description was presented at the beginning of the message In the high credibility condition, the source was described as a successful lawyer who was a consumer activist and volunteer and held the non-salaried position of president of a fictitious consumer federation The less credible source, on the other hand, was portrayed as a state government official who spoke on various topics before public gatherings as part of his public relations job

Dependent Variables The major dependent variable of interest is behavioral persistence This has been operationalized as a categorical response (compliance-no compliance) to three behavioral requests a) request to be a volunteer; b) intention to attend a meeting, and c) actual attendance The second and third requests took place 1 day and 3 months, respectively, after the initial request

In addition, a measure of individual attitudes toward consumer protection in India and perception of the source was obtained following the first behavioral request Individual attitudes toward consumer protection were assessed on a seven point Likert-type scale composed of three items An overall measure of attitude was obtained by summing each individual 18 response on the three-item scale Finally, the source credibility manipulation was checked by a semantic differential scale comprising three bipolar adjectives The three adjectives related to the expert, trustworthy and attractive dimensions of source credibility The attitude and source perception measures were included in the booklet along with the message and source description and administered in the classroom setting In all cases, the measure of source credibility followed the behavioral request and attitude variables


Manipulation Check

A manipulation check was performed to determine whether or not the source credibility induction was successful After receiving the persuasive appeal from either a high or low credibility source, respondents evaluated the communicator's credibility on three items The analysis shows that the manipulation was highly successful (F=14.76, df=1/59, p <.001) (See Table 1)



Since the manipulation of the source was based on only the expertise and trustworthy dimensions of credibility, the source perceptions were analyzed at the item level It showed that the treatment groups differed in their source perceptions on the "expert" and "good" item but : not on the "attractive" item The finding that the two f sources were similarly perceived on the attractive dimension rules out the possibility of demand character as a possible explanation of the result (Orne 1969; Sternthal and Dholakia 1978)

Initial Behavioral Compliance

Compliance to the request for volunteers was quite high over two-thirds of the respondents agreed to be listed as volunteers to promote the cause of consumer protection in India Compliance is greater among those exposed to a message from the less credible source (76.7%) than those exposed to the more credible source (64.5%) The difference in compliance is, however, not significant (z=1.06, p=.15).


The effect of source credibility on attitudes was measured by three 7-point items An analysis of variance was performed, and it was found that source credibility had no significant effect on individual attitudes toward consumer protection (F> 1 df - 1/59; p>.8).

Comparison of mean attitude ratings reveals that the group as a whole had quite favorable attitudes toward the cause of consumer protection in India Both groups hat a mean rating of over 17 0 on a sum scale with a maximum value of 21 0 The very nascent stage of consumer protection appears to attract considerable sympathy and support, even from such a "management-oriented" audience This can be seen also in the very high willingness to become a volunteer in a consumer protection organization. This may be compared to behavioral compliance among management students in a society with more advanced consumer protection issues where intention to sign a petition in favor of a consumer protection bill was much lower (Dholakia and Sternthal 1977)

Behavioral Persistence

The major question of interest in this paper is the effect of source credibility on behavior over time Behavioral compliance was sought one day and three months after the communication Generally, and as to be expected, there is a lower level of behavioral compliance to the delayed requests than the initial request following the communication only about 26% of the group expressed a willingness to attend a meeting, and 31% actually attended a meeting held three months after the request

There appears to be no effect of source credibility on behavior one day after the communication (z-0 51, p > 8) However, when actual attendance at the meeting is observed, there is a marginally significant effect of source credibility on behavioral compliance (z-l 32, p- 09) More respondents exposed to the low credibility source actually attended the meeting than those exposed to the high credibility source This is despite holding the meeting three months after the initial communication during which a large number of external variables affected the lives of the participants

The effect of source credibility on behavioral persistence can be seen more clearly when the behavioral responses are aggregated (Table 2) Combining the behavioral compliance measures provides a stronger measure of response consistency and broadens the generalizability of the findings (Epstein 1980)



When we examine responses to the two delayed behavioral requests, the low credibility source is able to elicit a higher level of compliance (z-1 57, p- 06) On the other hand, there is a larger number of respondents who express an intention to attend a meeting but do not follow through with the intention among those exposed to a highly credible source than the group exposed to a less credible source (z-2 24, p- 01) Similarly, there is a higher degree of noncompliance to all three requests among respondents exposed to a high credibility source (z-1 71, p- 04)


The finding that emerges from the study is that a low credibility source is more effective in eliciting behavioral compliance over time than a more credible source The implications of the study relate to research and practical issues This study was conducted in a very different economic environment and found support for the behavioral predictions based on self-perception theory and extends the findings obtained from the Dholakia and Sternthal (1977) experiment The differences in behavioral compliance between the two source credibility conditions become more significant as we aggregate the responses As the number of behavioral requests increase, increasing the commitment required to the cause, we observe a greater rate of compliance and persistent behavioral response among those exposed to the low credibility source Or to put it another way, it appears that attrition in compliance from the initial request to the subsequent requests is forestalled by the less credible source.

To empirically test the persistence effect, it is important and necessary to operationalize the first behavioral request such that it leads to high degrees of compliance Then, given comparable responses to the initial request, one can attribute differences in motivations to the source conditions under which initial behavior occur ed Hence, one would expect weak external justifications for the initial behavior in the less credible source condition and a strong external justification for the highly credible source condition This, in turn, is predicted to lead to differences in the repetition of behavior under the two source conditions

The series of behavioral requests may be seen as a form of the foot-in-the-door technique (Freedman and Fraser 1986; Pliner et al 1974) starting with an initially small request which elicits a high degree of compliance However, compliance to subsequent requests is influenced by circumstances under which the first behavioral compliance occurs It appears that the lover credibility condition facilitates greater attribution to personal reasons, and this is enhanced as the time interval between the requests increases According to Sternthal and Craig (1982), one way of increasing the salience of thoughts about one's behavior is to allow enough time between requests (p 129); this appears to help the less credible source more than the highly credible source

The source credibility effect on behavioral compliance occurs despite the similar attitudinal assessments by the respondents exposed to the different source conditions It is likely that the premessage disposition towards the consumer protection issue was favorable for the group as a whole While pretest measures are not available, premessage disposition may be inferred from two observations the relatively high willingness to volunteer for the cause (initial behavior) and the very high mean attitude scores; even noncompliers had a mean (sum) score over 15 0 on a 3-21 point scale Research evidence indicates that a moderately credible source is more effective than a highly credible source when the message recipients are favorably disposed to an issue (Sternthal, Dholakia and Leavitt 1978)


While the study supports the behavioral predictions based on self-perception theory, it does not provide any insights on the underlying mechanisms To have greater confidence in the behavioral results, further research is needed to provide direct support for the process mediating the behavioral responses of individuals exposed to different source credibility conditions

The predicted effect on attitudes following initial behavioral compliance was not supported; future research must be able to support predictions regarding attitudes as well as behavior The failure to find the predicted effect on attitudes has been related to an imposed "ceiling" in the attitude measure It highlights problems in cross-cultural research Based on prior research in the United States, business and management students had exhibited a much more negative attitude towards consume r is t issues and organizations This was mistakenly assumed to characterize the Indian students enrolled in a business program modeled after U S institutions While pretests were conducted to select appropriate question-items, the pretests were not adequately used to anticipate the "ceiling" problem This problem has to be taken care of in future research


Bem, D (1972), "Self-Perception Theory," in L Berkowitz (Ed ), Advances in Experimental Social Psychology, New York Academic Press.

Brock, Timothy (1965), "Communicator-Recipient Similarity and Decision Change," Journal of Personality and Social Psychology, 1, 650-654.

Dholakia, R.R.and B Sternthal (1977), "Highly Credible Sources Persuasive Facilitators or Persuasive Liabilities," Journal of Consumer Research, 3, 223-32.

Eagly, A and S Chaiken (1975), "An Attribution Analysis of the Effect of Communicator Characteristics on Opinion Change The Case of Communicator Attractiveness," Journal of Personality and Social Psychology, 32, 1367-144.

Epstein, S (1980), "The Stability of Behavior II Implications for Psychological Research," American Psychologist, 35, 9 (September), 790-806.

Fireworker, R B and H H Friedman (1977), "The Effects of Endorsements on Product Evaluation," Decision Sciences, 8, 576-583.

Freedman, J J and S Fraser (1966), "Compliance without Pressure The Foot-in-the-Door Technique," Journal of Personality and Social Psychology, 4, 195-202.

Freeman, W (1957), The BiB Name, New York Printer's Ink.

Friedman, H H and L Friedman (1979), "Endorser Effectiveness by Product Type," Journal of Advertising Re- search, 19, 63-71.

Greenwald, A (1968), "Cognitive Learning, Cognitive Response to Persuasion, and Attitude Change," in A Greenwald, T Brock, and T Ostrom (Eds ), Psychological Foundation of Attitudes, New York Academic Press, 167-170.

Hill, T , N D Smith, and M F Mann (1986), "Communicating Innovations Convincing Computer Phobics to Adopt Innovative Technologies," Advances in Consumer Research, Vol 13, ed R J Lutz, Provo, UT Association for Consumer Research.

Himmelfarb, S and D Arazi (1974), "Choice and Source Attractiveness in Exposure to Discrepant Messages," Journal of Experimental Social Psychology, 10, 516-527.

Hovland, C and W Mandell (1952), "An Experimental Comparison of Conclusion-Drawing by the Communicator and by the Audience, Journal of Abnormal and Social Psychology, 47, 581-588.

Hovland, C and W Weiss (1951), "The Influence of Source Credibility on Communication Effectiveness," Public Opinion Quarterly, 15, 635-650.

Johnson, Homer and Richard Izzett (1972), "The Influence of Source Identification on Attitude Change as a Function of the Type of Communication," Journal of Social Psycho- logy, 86, 81-87.

Johnson, Homer and John Scileppi (1969), "Effects of Ego-Involvement Conditions on Attitude Change to High and Low Credibility Communicators," Journal of Personality and Social Psychology, 13, 31-36.

Johnson, H., J Torcivia, and M Poprick (1968), "Effects of Source Credibility on the Relationship between Authoritarianism and Attitude Change," Journal of Personality and Social Psychology, 9, 179-183.

Jones, R and J Brehm (1967), "Attitudinal Effects of Communicator Attractiveness When One Chooses to Listen," Journal of Personality and Social Psychology, 6, 64-70.

Kamen, J M , A C Azhari, and J R Kragh (1975), "What a Spokesman Does For a Sponsor," Journal of Advertising Research, 15, 17-24.

Kelley, H (1973), "The Processes of Causal Attribution," American Psychologist, 28, 107-128.

Kelman, H and C Hovland (1953), "'Reinstatement' of the Communicator in Delayed Measurement of Opinion Change," Journal of Abnormal and Social Psychology, 48, 327-335.

McCroskey, James (1969), "A Summary of Experimental Research on the Effects of Evidence in Persuasive Communication," Quarterly Journal of Speech, 55, 169-176.

Miller, G and J Baseheart (1969), "Source Trustworthiness, Opinionated Statements, and Response to Persuasive Communication," Speech Monographs, 36, 1-7.

Mowen, J C and S W Brown (1981), "On Explaining and Predicting the Effectiveness of Celebrity Endorsers," Advances in Consumer Research, Vol, 8 ed Rent B Monroe, Chicago Association for Consumer Research.

Orne, Martin (1969), "Demand Characteristics and the Concept of Quasi-Controls," in R Rosenthal and R Kosnow (eds ), Artifact in Behavioral Research New York Academic Press.

Pliner, P , H Hart, J Kohl, and D Saabi (1974), "Compliance without Pressure Some Further Data on the Foot-in-the-Door Technique," Journal of Experimental Social Psychology 10, 17-22.

Powell, F (1965), "Source Credibility and Behavioral Compliance as Determinants of Attitude Change," Journal of Personality and Social Psychology, 2, 669-676.

Rhine, Ramon and Laurence Severance (1970), "Ego-Involvement, Discrepancy, Source Credibility, and Attitude Change," Journal of Personality and Social Psychology, 16, 175-190.

Ritchie, E and E Phares (1969), "Attitude Change as a Function of Internal-External Control and Communicator Status," Journal of Personality and Social Psychology, 37 429-443.

Schulman, G and C Worrall (1970), "Salience Patterns, Source Credibility, and the Sleeper Effect," Public Opinion Quarterly, 34, 371-382.

Sternthal, B and R R Dholakia (1978), "Rejoinder," Journal of Consumer Research, 4, 67-69.

Sternthal, B , R R Dholakia, and C Leavitt (1978), "The Persuasive Effect of Source Credibility Tests of Cognitive Response," Journal of Consumer Research, 4, 252-260.

Sternthal, B , L W Phillips, and R R Dholakia (1978), "The Persuasive Effect of Source Credibility A Situational Analysis," Public Opinion Quarterly, 285-314.

Sternthal, B and C S Craig (1982), Consumer Behavior An Information Processing Perspective, Englewood Cliffs, N J Prentice-Hall.

Tybout, A (1978), "Relative Effectiveness of Three Behavioral Influence Strategies as Supplements to Persuasion in a Marketing Context," Journal of Marketing Research, 15, 229-42.

Walster, E , E Aronson, and D Abrahams (1966), "On Increasing the Persuasiveness of a Low Prestige Communicator," Journal of Experimental Social Psychology, 2, 325-342.

Warren, I (1969), "The Effect of Credibility in Sources of Testimony on Audience Attitudes Toward Speaker and Message," Speech Monographs, 36, 456-458.

Watts, W and W McGuire, "Persistence of Induced Opinion Change and Retention of the Inducing Message Contents," Journal of Abnormal and Social Psychology, 68, 233-241.

Whittaker, J and R Meade (1968), "Retention of Opinion Change as a Function of Differential Source Credibility," International Journal of Psychology, 3? 103-108.