Overestimating Salesperson Truthfulness: the Fundamental Attribution Error

Robert Baer, Bradley University
ABSTRACT - This study examined consumers' susceptibility to the fundamental attribution error--the tendency to overestimate the importance of internal causes while underestimating the significance of external causes. The results showed that a short product line caused subjects to overestimate the extent to which the salesperson's product claims reflected their true beliefs about the product. Yet, a sales commission prompted subjects to underestimate the extent to which the sales message reflected the salesperson's genuine beliefs.
[ to cite ]:
Robert Baer (1990) ,"Overestimating Salesperson Truthfulness: the Fundamental Attribution Error", in NA - Advances in Consumer Research Volume 17, eds. Marvin E. Goldberg, Gerald Gorn, and Richard W. Pollay, Provo, UT : Association for Consumer Research, Pages: 501-507.

Advances in Consumer Research Volume 17, 1990      Pages 501-507

OVERESTIMATING SALESPERSON TRUTHFULNESS: THE FUNDAMENTAL ATTRIBUTION ERROR

Robert Baer, Bradley University

[The author thanks Rustan Kosenko and four anonymous reviewers for their helpful comments. This research was partially funded by The Caterpillar Fellows Program.]

ABSTRACT -

This study examined consumers' susceptibility to the fundamental attribution error--the tendency to overestimate the importance of internal causes while underestimating the significance of external causes. The results showed that a short product line caused subjects to overestimate the extent to which the salesperson's product claims reflected their true beliefs about the product. Yet, a sales commission prompted subjects to underestimate the extent to which the sales message reflected the salesperson's genuine beliefs.

One of the more pervasive problems encountered in marketing is to establish the perceived trustworthiness of marketing messages. Trustworthiness refers to the extent to which the source is perceived as intending to communicate valid statements about the product. Trustworthy statements are perceived to be objective and furthering no vested interest of the source. The more trustworthy a source, the more believable and effective are his or her persuasive communications. In conjunction with expertise (perceived knowledge regarding the product category), trustworthiness creates the perception of credibility. An untrustworthy source, on the other hand, is one who is insincere, dishonest and is perceived as having a vested interest in the claim. When a source's self-interest can be obviously furthered by acquiescence to his or her's persuasive communication, a target should discount what is said.

Marketing trustworthiness and credibility can be considered from an attributional perspective (Settle and Golden, 1974; Hansen and Scott, 1976; Calder and Burnkrant, 1977). Attribution theory is a collection of theories (Bem 1965; Kelley 1967; Jones and Davis 1965) that seek to explain the cognitive processes involved when an individual infers the causes of another person's behavior. Within the past ten years there have been two major reviews of attribution theory and consumer research. Mizerski, Golden and Keenan (1979) provided an overview of four major attributional theories and reviewed the relevant consumer behavior research from 1971 to 1978. Folkes (1988) has summarized the relevant consumer behavior research since then and has concluded that "attribution theory is a rich and well-developed approach that has a great deal to say about a wide range of consumer behavior issues' (p. 548).

Studies of attribution have dwelt on the conditions which determine whether a behavior is attributed to inner traits, beliefs or dispositions (internal attribution) or to prevailing social and environmental pressures (external attribution). Consumers are typically confronted with just such an attributional problem. For example, given an opinion expressed by a salesperson, the customer must decide whether the act (stated opinion) was caused by (1) the salesperson's true beliefs about the product (internal attribution) or (2) variable situational contingencies such as the desire to make the sale, the reward for making the sale or the role requirements of the job (external attribution). An attribution to the external situation means that the customer believes that most salespeople would behave in the same way. By contrast, an internal attribution indicates that the customer is assigning traits and opinions t-o the salesperson that are more important than the situational pressure surrounding the salesperson's behavior.

However, it is not always easy to discern whether a salesperson's verbal statements were caused by internal or external reasons. The cause will be more ambiguous and therefore less revealing about his or her true beliefs to the extent that external pressure exists. For example, customers who encounter a friendly salesperson should not make an internal attribution (i.e., this is a friendly person) because friendliness is perceived to be an important part of an individual's role as a salesperson. In this case, the external constraint (job requirements) can quite easily account for the observed behavior (friendliness). Thus, under these circumstances, it is logical for consumers to regard the sales message with caution (Sparkman, 1982).

However, one of the more significant consequences that has emerged in the investigation of attributional processes is the discovery that people do not always make attributions in this logical fashion. Rather, the attributional process is plagued-by a number of systematic errors and biases that distort the judgmental process (Nisbett and Ross, 1980). One of the more common errors is that people are generally biased toward making internal attributions. That is, they tend to see the dispositions of a person as causing that person's behavior. Several investigators (Jones, 1979; Miller, 1976; Ross, 1977) report that people tend to infer that an individual's actions are congruent with their beliefs even when the individual clearly had no choice in his or her actions. Logically, observers should discount the diagnosticity of behavior when the person has little or no control over their behavior. Their failure to do so is widely regarded as attributional error in the direction of personal causation.

Although it has been the exploration of attribution error which has shaped much of the attribution research in other disciplines, attributional error has had little impact on the field of consumer behavior. The present study examined the extent to which consumers were prone to make internal attributions. A bias toward internal attributions would cause consumers to overestimate the extent to which salespeople believe their own product claims. Since consumers are cognizant of the fact that salespeople operate in a highly constrained environment and that they are not often free to express their true beliefs about the product, one might expect that consumers would be exempt from this error. However, attributional research has shown that the error is a pervasive tendency in the perception of other people. In fact, the error is so impermeable to change that it is often referred to as the "fundamental attribution error" (Ross, 1977).

The purpose of the present study was to determine whether consumers in a sales situation were vulnerable to the fundamental attribution error. Attribution research leads one to speculate that since the error is so robust, consumers too should commit the error, i.e., they should underestimate the extent to which the salesperson's word is a reflection of their job requirements and should be somewhat inclined to take the salesperson's word at face value. This is a rather startling prediction that runs counter to prevailing thought which depicts consumers as highly sensitive to the salesperson's vested interest in the sale and cautious about his or her motives. In this study, a sales situation was chosen to examine the fundamental attribution error because it is not a neutral environment. Rather, consumers are likely to question the salesperson's motives and are not likely to automatically assume that the salesperson is trustworthy. Thus, the present study created a stringent test of the fundamental attribution error.

The present study investigated two different variables (length of product line and sales incentive) that could mediate the perception of salesperson trustworthiness. Salespersons who have the luxury of a long product line or who are known to not make a commission from the sale, may be perceived as having more freedom to convey their genuine beliefs about the product than salespersons with a short product line or who make a sales commission. Those with the long product line or no commission may be perceived as having less stake in any single product item, and, as a result, their word may be considered more trustworthy. In the case of a short product line or a sales incentive, the external constraint can quite easily account for the salesperson's behavior. Under these circumstances, it would be logical for the customer to be cautious about trusting the salesperson's expressed opinion.

In this study, it was hypothesized that: (1) When the message was not constrained by external factors (no sales commission or a long product line), consumers would assume that the salesperson's true beliefs about the product were congruent with their verbally expressed beliefs. Thus, a salesperson who speaks highly about the product should be perceived as genuinely believing in that product more than one who speaks unfavorably about the product. (2) Subjects would commit the fundamental attribution error in the no choice conditions (short product line or sales incentive) by attributing an attitude toward the salesperson that was consistent with their publicly expressed beliefs. A significant difference between the beliefs attributed to a salesperson who speaks favorably versus unfavorably about the product in the no choice conditions would be indicative of attributional error. (3) The same pattern of results was expected when the salesperson's constraint was manipulated by either the length of the product line or by the sales incentive. However, it was predicted that attributional error would be diminished in the sales incentive compared to the product line condition. Knowledge about a sales incentive for sales personnel heightens consumer sensitivity about the salesperson's vested interest which, in turn, diminishes perceptions of salesperson trustworthiness.

METHOD

Overview

The experimental-conditions provided for two levels of the salesperson's expressed beliefs toward the product (favorable versus unfavorable), two levels of product line length (long-choice versus short-no choice) and two levels of sales incentive (no incentive-choice versus incentive-no choice). The design was not fully crossed. Subjects were assigned to only one of the levels of either the product line or sales incentive conditions.

Participants

The participants were 49 male and 61 female undergraduate business students who volunteered to participate in the study. Participants were run in small groups ranging in size from eight to fourteen. Participants were randomly assigned to one of the eight cells in the experimental design as they appeared for their experimental session. Sixty-two subjects were assigned to one of the four cells in the product line length condition while 48 subjects were assigned to one in four cells of the sales incentive condition.

Procedure

Stimulus materials were presented in booklet form. Independent variables were manipulated by asking subjects to take the role of observers while reading-written scenarios portraying an automobile salesperson's behavior. The cover page of the booklet described the experiment as an investigation of the way in which people make judgments about products and salespeople with limited information. Subjects were asked to imagine a potential customer who had come to an automobile dealership inquiring about the "Austin automobile or, perhaps, some other car of medium price."

Sales Message

Persuasive messages pertaining to the sale of an automobile were composed. The messages were adopted from sets of persuasive messages reported by Pratkanis, Greenwald, Ronis, Leippe and Baumgardner (1983). The messages were used in 11 previous studies published by those authors and have been found to produce strong, immediate persuasion effects that tend to dissipate quickly.

The sales messages consisted of a brand name and a description of the product attributes. For the brand name, a fictitious name (Austin) was randomly selected from a list of the 72 most common six-letter names in the United States (U.S. Social Security, 1964). The same brand name was used in all experimental conditions. To create a description of the product attributes, paragraphs based on articles appearing in Consumer Reports were composed. Three attributes (durability, safety and handling) were chosen because different automobiles should vary among these characteristics. Each sales message contained a separate statement about durability, safety and handling. These product attribute statements provided general information about the meaning and importance of the attribute. The product attribute statements were identical for all experimental conditions.

Independent Variables

Salesperson's Expressed Beliefs. Salespersons expressed a belief about the Austin that was either favorable or unfavorable. The belief consisted of one sentence for each of the three product attributes that assigned a value of either "poor" (unfavorable) or "excellent" (favorable) to each of the three attributes. In addition, salespersons in the favorable condition began their sales message with an overall excellent assessment of the Austin and concluded by recommending its sale. Salespersons in the unfavorable condition began with a poor evaluation and concluded by not recommending its sale.

Length of Product Line. Subjects in the short product line (no choice) condition were told that there was a short product line of medium-priced cars. In fact, these subjects were told that the Austin was the only medium- priced car on the lot. Because of the short product line, it was emphasized that salespersons were not free to sell the models in which they most believed. If these salespersons thought that the Austin was not a good car, they were not free to express their opinion to the customer.

Subjects in the long product line (choice) condition were told that there was a long product line, i.e., that there were 11 medium-priced cars in addition to the Austin. Subjects in this condition were told that salespersons were permitted by their supervisor to sell any medium-priced car that they wished to sell. These sales persons were described as free to choose models to sell that they saw fit. If the salespersons thought that some models were better than others, they were free to say so to the customer.

Sales Incentive. Subjects in the sales incentive condition (no choice) were told that salespersons received special incentives to sell the Austin. That is, these subjects were told that salespersons would receive a much larger commission for selling the Austin than would result from selling other cars in its price range. These subjects were told that salespersons were pushed by their supervisor and if they did not think that the Austin was a good car, they were not free to express that opinion.

Subjects in the no sales incentive condition (choice) were told that salespersons were given no special incentives to sell the Austin, i.e., that salespersons received the same commission for selling any medium-priced car. Thus, these salespersons were described as free to choose models to sell as they saw fit. If salespersons thought that some models were better than others, they were free to say so to the customer.

Manipulation Check

In order to assess the effectiveness of the constraint (choice) manipulation, subjects were asked, "How much free choice did the salesperson have to express his or her true opinions about the Austin?" Subjects responded on a 9-point rating scale, 1 = very little choice, 9 = very much choice.

Dependent Measures

As the major dependent measure, subjects attributed an attitude to the salesperson. Subjects were asked, "What do you think is the salesperson's deepest and most genuine belief about the Austin?" (1 = extremely poor car, 9 = extremely good car). The degree of confidence in their attitude attribution was also assessed. Perceptions of the salesperson were measured on 9- point bipolar scales (1 = very little, 9 = very much). Finally, subjects were asked to estimate the monetary worth of the Austin.

RESULTS

Results were analyzed by two 2x2 ANOVAs. Manipulated variables were length of the product line (short versus long), sales incentive (incentive versus no incentive) and the salesperson's expressed beliefs toward the product (favorable versus unfavorable). Since subjects were assigned to one of the levels of either the product line length or sales incentive conditions, separate 2x2 ANOVAs were performed on the product line length and sales incentive data.

Attitude Attribution

First, consider the results for the subjects in the product line length conditions. A significant main effect for expressed beliefs, F(1,58) = 24.60, p < .01 was obtained. These results indicated that salespersons who spoke favorably about the product were perceived as having more favorable attitudes (M = 5.09) towards the product than those who spoke unfavorably (M = 3.83). In terms of the central questions of this research, however, the data are best understood in terms of the product line length (constraint) x expressed beliefs interaction, F(1,58) = 104.84, p < .01. The means are shown in the left-hand panel of Table 1. These results indicated that in the choice condition (long product line), attitudes were attributed in line with the direction of sales message, i.e., salespersons who spoke favorably about the product were perceived as having more favorable attitudes than those who spoke unfavorably. These results support hypothesis 1. Although this difference was diminished in the no choice condition (short product line), subjects still tended to infer attitudes that were consistent with the direction of the sales message. That is, salespersons who gave favorable messages were perceived as more favorable toward the product (M = 5.47) than those who spoke unfavorably (M = 4.13), F(1,29) = 13.94, p < .01. This significant difference in attitudes attributed to targets who spoke favorably or unfavorably in the no choice condition is widely regarded as attributional error in the direction of personal causation and supports hypothesis 2.

TABLE 1

ATTITUDE ATTRIBUTION MEAN SCORES AND (STANDARD DEVIATIONS)

The right-hand panel of Table 1 shows the results for the sales incentive conditions. A main effect for expressed beliefs, F(1,44) = 46.02, p < .01 was obtained. Salespersons who spoke favorably about the product were perceived as being more favorable (M = 5.42) about the product than those who spoke unfavorably (M = 3.46). However, a constraint (sales incentive) x expressed beliefs interaction, F(1,44) = 82,69, p < .01, indicated that this was true only in the choice condition (no sales incentive) thus supporting hypothesis 1. In the no choice condition (sales incentive), a reversal occurred. Salespersons in those conditions were attributed with attitudes in the opposite direction for the sales message. Salespersons who spoke favorably about the product were perceived as less in favor (M = 4.25) than salespersons who spoke unfavorably (M = 4.92), F(1,22) = 3.67, p < .05. No other effects were significant.

The constraint x expressed beliefs interaction provided only partial support for hypothesis 2 and 3. As predicted, subjects exposed to the sales incentive conditions were less likely than product line length subjects to commit the fundamental attribution error. However, unexpectedly, a different pattern of results was obtained in the sales incentive condition. Attitudes correspondent to the direction of the sales message were not attributed to these salespersons. In fact, sales incentive subjects attributed beliefs toward the product opposed to the direction of the sales message. This effect has been observed in previous studies (e.g., Jones, Worchel, Goethals and Grumet, 1971 and Miller and Rorer, 1981) when the extremity, quality and persuasiveness of the constrained behavior has been weak. When a person acting under pressure does a poor job, observers infer that the person holds an attitude in the opposite direction. Weak or ambivalent behavior under no choice conditions is perceived as "foot-dragging" and it signifies that the person is acting under pressure. As a result, perceivers do not believe the spokesperson and they make attributions that are less consistent with the expressed beliefs.

Confidence

Subjects rated their degree of confidence in their attitude attribution (1 = very little confidence, 9 = very much confidence). There was no difference in the amount of confidence that subjects in the different product line length conditions expressed (all means ranged from 4.40 to 5.13). However, for subjects in the sales incentive conditions, a main effect for sales incentive, F(1,44) = 32.49, p c .01, indicated that subjects were more confident of their attitude attribution when no sales incentive was available to the salesperson (M = 6.54) than it was not available (M = 3.46).

Salesperson Characteristics

Subjects rated the salesperson's trustworthiness, credibility, believability, sincerity and eagerness to help on 9-point rating scales (1 = not at all, 9 = very much). Results for the product line length and sales incentive conditions were very similar. For the product line length case, subjects in the choice compared to the no choice condition rated the salesperson as more trustworthy (M = 4.93 vs. M = 2.97), F(1,57) = 21.50, p < .01, more credible (M = 4.50 vs. 3.29), F(1,57) = 7.38, p < .01, more believable (M = 4.87 vs. M = 3.84), F(1,57) = 4.35, p < .05, more sincere (M = 5.27 vs. M = 3.13), F(1,57) = 20.03, p < .01 and more eager to help (M = 4.32 vs. M = 2.08), F(1,57) = 6.53, p < .01. No other effects were significant.

TABLE 2

PERCEIVED PRODUCT VALUE MEANS

Similarly, for the sales incentive treatments, subjects in the choice compared to the no choice condition rated the salesperson more trustworthy (M = 6.04 vs. M = 3.42), F (1,44) = 43.04, p < .01, more credible (M = 6.0 vs. 3.29), F(1,44) = 33.36, p < .01, more believable (M = 5.60 vs. 3.05), F(1,44) = 47.24, p < .01 and more sincere (M = 5.68 vs. M = 3.21), F(1 44) = 28.80, p < .01. No other effects were significant.

Estimated Product Value

The results for estimated product value are shown in Table 2. A constraint x expressed beliefs interaction was obtained for the product line treatment, F(1,57) = 3.96, p < .05. The results indicated that in the choice conditions, the estimated value of the product was consistent with the direction of the sales message, i.e., the product was seen as more valuable when the salesperson spoke favorably (M = $10,175) than unfavorably (M = $8,871). Although this tendency was diminished in the no choice condition, subjects in the no choice condition still tended to infer that the product was of greater value when the salesperson spoke favorably (M = $10,726) than unfavorably (M = $9,662).

For subjects in the sales incentive treatment, a constraint x expressed beliefs interaction, F(1,44) = 4.21, p < .05, indicated that in the choice condition (no sales incentive), the product was perceived as more valuable when the salesperson spoke favorably (M = $8,950) rather than unfavorably (M = $8,686). However, a reversal effect occurred under no choice conditions (sales incentive). The product was seen as more valuable when the salesperson expressed unfavorable beliefs (M = $8,866) than favorable beliefs (M = $7,790).

DISCUSSION

The present study was designed to assess consumers' tendencies to commit the fundamental attribution error The error is a well-documented phenomenon whereby observers attribute attitudes or dispositions to another person, even when that person's behavior appears to have been facilitated by strong situational pressures. Researchers manipulating the salience of situational constraints have had to go to extraordinary lengths to decrease this bias (Miller, 1976; Snyder and Jones, 1974). Yet, even when subjects are aware of powerful situational constraint operating, they are still prone to make internal attributions.

Prior to this study, the fundamental attribution error has never been investigated in a marketing context. A sales environment is one in which the powerful constraints on the salesperson seem quite apparent. Is fact, it is reasonable to assume that customers approach this situation with a bias opposed to the fundamental attribution error. They expect to doubt the salesperson's word and expect to regard the sales message as caused by the external situation rather than by the salesperson's genuine beliefs about the product.

These considerations notwithstanding, the present study found evidence for the fundamental attribution error in the product line conditions. Although attributions were consistent with the direction of the sales message in the choice conditions, subjects tended to infer attitudes consistent with the sales message even when the salesperson's lack of choice was emphasized in the experimental instructions. This finding is even more surprising given subjects' perceptions of the salesperson. In all cases, subjects in the no choice conditions perceived the salesperson as less trustworthy, credible, believable, sincere and eager to please than subjects in the choice conditions. In the no choice conditions, subjects explicitly acknowledged that the salesperson's word should be regarded with caution. Yet, an attitude consistent with the salesperson's expressed beliefs were attributed to them by subjects.

Having committed the fundamental attribution error, subjects perceptions of the product value were compatible with the error. Even when the salesperson clearly had no choice in the views that he or she expressed, subjects inferred that the product was more valuable when the salesperson spoke favorably compared to unfavorably.

These results suggest that consumers are not exempt from the fundamental attribution error. Although they are more conservative in the attitudes that they attribute to salespersons under no choice conditions, there is still a tendency to think that the salesperson believes what he or she says. Marketers can capitalize on this bias with manipulations that emphasize that the salesperson's behavior is voluntary.

In the sales incentive conditions, salespersons who expressed beliefs in favor of the product were actually perceived as less in favor than those who expressed beliefs against the product. As noted above, this effect has been found in previous studies when the behavior is weak or ambivalent. In the present study, however, the sales message in all experimental conditions was identical in terms of wording, length, quality, extremity and persuasiveness. Only the brand evaluations at the end of the sales message differed. Subjects perceptions of the message quality (M = 3.88), extremity (M = 5.27) arid persuasiveness (M = 4.94, all on 9-point rating scales) were weak to moderate and did not differ from condition to condition. Thus, the reversal effect can not be explained by assuming that the sales message in the sales incentive condition was weaker than the message in the product line condition.

Furthermore, a manipulation check that asked subjects "How much free choice did the salesperson have to express his or her true opinions about the product?" yielded similar perceptions of constraint in the product line (M = 4.26) and sales incentive (M = 4.65), F < 1, conditions. Further subjects in the no choice-product line length condition perceived a similar amount of salesperson free choice in the favorable (M = 1.80) and unfavorable condition (M = 1.40), F < 1. Yet, there was a significant difference between the attitude attributions in these two conditions. Similarly, subjects in the no choice - sales incentive condition perceived the same amount of salesperson free choice in both the favorable (M = 2.25) and unfavorable condition (M = 2.0), F < 1. Yet, significant differences in attitude attribution occurred.

The reversal effect cannot therefore be accounted for by assuming greater constraint in the sales incentive condition. Apparently, then it took a mild sales message combined with subjects' awareness of the salesperson's incentive that yielded perceptions of "foot-dragging." These results indicate that an ineffective salesperson runs the risk of being perceived as untrustworthy. A less than convincing performance under constraining circumstances prompt consumers to think that the salesperson does not believe his or her verbal statements. This perceived "foot-dragging" phenomenon elicits attitude attributions opposed to the verbal statements.

REFERENCES

Bem, Daryl (1972), "Self Perception Theory' Advances in Experimental Social Psychology, Vol. 6 ed. Leonard Berkowitz, New York: Academic Press.

Calder, Bobby J., and Burnkrant, Robert E. (1977), "Interpersonal Influence on Consumer Behavior: An Attribution Theory Approach," Journal of Consumer Research, 4, 29-38.

Folkes, Valerie, S., (1988), "Recent Attribution Research in Consumer Behavior: A Review and New Directions," Journal of Consumer Research, XX, 548-565.

Hansen, Robert A., and Scott, Carol A. (1976), "Comments on 'Attribution Theory and Advertiser Credibility'," Journal of Marketing Research, 13, 193-199.

Jones, Edward E., (1979), "The Rocky Road from Acts to Dispositions," American Psychologist, 34, 107-117.

Jones, Edward E., and Davis, Ketch E. (1965). "From Acts to Dispositions: The Attribution Process in Person Perception," in Advances in Experimental Social Psychology Vol. 2 ed. L. Berkowitz, New York: Academic Press.

Jones, Edward E., Stephen Worchel, G.R. Goethals, and J.F. Grumet (1971), "Prior Expectancy and Behavioral Extremity as Determinants of Attitude Attribution," Journal of Experimental Social Psychology, 7, 1-24.

Kelley, Harold (1967), "Attribution Theory in Social Psychology," in Nebraska Symposium on Motivation, ed. David Levine, Lincoln, NB: University of Nebraska Press.

Miller, Arthur G., (1976), "Constraint and Target Effects in the Attribution of Attitudes," Journal of Experimental Social Psychology, 12, 325339.

Miller, Arthur G. and Len Rorer (1982), "Toward an Understanding of the Fundamental Attribution Error: Essay Diagnosticity in the Attitude Attribution Paradigm," Journal of Research in Personality, 16, 41-59.

Mizerski, Richard, Linda L. Golden and Jerome B. Kernan (1979), 'The Attribution Process in Consumer Decision-Making," Journal of Consumer Research, 6, 123-140.

Nisbett, Richard and Lee Ross (1980). Human Inference. Strategies and Shortcomings of Social Judgment, Englewood Cliff, NY: Prentice-Hall.

Pratkanis, Anthony, R., Anthony G. Greenwald, David L. Ronis, Michael R. Leippe and Michael H. Baumgardner, (1987) "Consumer-Product and Socio- Political Messages for Use in Studies of Persuasion," unpublished paper, Graduate School of Industrial Administration, Carnegie-Mellon University, Pittsburgh, PA 15213.

Ross, Lee (1977), "The Intuitive Psychologists and His Shortcomings: Distortion in the Attribution Process," in ed. Advances in Experimental Social Psychology, Vol. 10, New York: Academic Press.

Settle, Robert B., and Golden, Linda L. (1974), "Attribution Theory and Advertiser Credibility," Journal of Marketing Research, 11, 181-185.

Snyder, Melvin and Edward E. Jones, (1974), "Attitude Attribution When Behavior is Constrained," Journal of Personality and Social Psychology, 10, 585-600.

Sparkman, Richard M., Jr. (1982), 'The Discounting Principle in the Perception of Advertising," in Advances in Consumer Research, Vol. 9, ed. Andrew Mitchell, An I Arbor, MI: Association for Consumer Research. 277-280.

----------------------------------------