Miscomprehension and Believability of Information Presented in Print Advertising

ABSTRACT - Previous research in this area has raised the question of whether or not conclusions based on the forced exposure paradigm are generalizable to more naturalistic settings. In the present investigation the miscomprehension and believability of information targeted toward a student audience are measured under as ecologically valid conditions as operationally feasible. Based upon the results obtained, it appears that some earlier results were indeed well "within the ballpark."

Citation:

Richard F. Beltramini and Steven P. Brown (1994) ,"Miscomprehension and Believability of Information Presented in Print Advertising", in NA - Advances in Consumer Research Volume 21, eds. Chris T. Allen and Deborah Roedder John, Provo, UT : Association for Consumer Research, Pages: 218-223.

Advances in Consumer Research Volume 21, 1994    Pages 218-223

MISCOMPREHENSION AND BELIEVABILITY OF INFORMATION PRESENTED IN PRINT ADVERTISING

Richard F. Beltramini, Arizona State University

Steven P. Brown, Southern Methodist University

ABSTRACT -

Previous research in this area has raised the question of whether or not conclusions based on the forced exposure paradigm are generalizable to more naturalistic settings. In the present investigation the miscomprehension and believability of information targeted toward a student audience are measured under as ecologically valid conditions as operationally feasible. Based upon the results obtained, it appears that some earlier results were indeed well "within the ballpark."

INTRODUCTION

Research on miscomprehension has relied on the "forced exposure paradigm" to estimate the proportion of advertising content that is misunderstood (e.g., Jacoby and Hoyer 1982a, 1989). In this paradigm, study participants are instructed to watch television commercials or read print ads and then answer factual or inferential questions regarding their content immediately afterward. Thus, these studies have assessed miscomprehension rates under conditions unlike those of ordinary advertising exposure, which makes generalization of miscomprehension rates to more naturalistic conditions inappropriate. The present study investigates miscomprehension rates for print advertising under naturalistic exposure conditions and compares these rates to those reported in forced exposure studies.

Also, relatively little attention has been given in previous research to the relationship between miscomprehension and measures of advertising effectiveness. Most existing research has studied deceptive advertising and elicitation of "pragmatic inferences" that mislead consumers regarding the attributes and potentialities of products (e.g., Harris 1977; Harris and Monaco 1978). This research has proceeded from the assumptions that some advertisers encourage miscomprehension and that miscomprehension is positively related to ad believability through the mediation of false inferences (Monaco and Kaiser 1983). The present study explores the relationship between miscomprehension of the factual content of print ads and their perceived believability.

A final limitation of previous research in this area concerns its lack of attention to ensuring the advertising information being assessed is specifically targeted toward an audience of interest to advertisers, and one which is sufficiently defined to permit practical application of the research results. The present study affords that focus by measuring miscomprehension and believability of advertising information among college students, an audience at which millions of dollars are currently directed and which is growing in importance to advertisers.

PREVIOUS RESEARCH

Miscomprehension Rates

The original work on miscomprehension of television commercials by Jacoby and Hoyer (1982a) reported a miscomprehension rate of 29.6 percent for advertising content. This work was criticized by Ford and Yalch (1982) on the basis that methodological characteristics of the study biased the miscomprehension findings upward. Among other things, Ford and Yalch criticized the study's laboratory setting, forced exposure to stimuli, instructions given to study participants, and lack of correspondence between the intended target markets of the advertisements and the study participants whose comprehension was assessed. Jacoby and Hoyer (1982b) responded that the laboratory setting and forced exposure procedure would have the effect of biasing miscomprehension findings downward rather than upward. No subsequent study has directly addressed the external validity questions raised by Ford and Yalch (1982) by assessing miscomprehension rates under ordinary exposure conditions and comparing them to those established in forced exposure studies.

A further criticism leveled against the original Jacoby and Hoyer study concerned inclusion of unequal numbers of true-false items for each ad (i.e., four false and two true). Ford and Yalch (1982) argued that "yea-saying" (i.e., respondents' tendency to answer "true" when uncertain of the correct response) resulted in an overstatement of miscomprehension. Schmittlein and Morrison (1983) reanalyzed the Jacoby and Hoyer data, concluding that if a single index were used to describe the proportion of copy correctly comprehended, that index would be 46 percent.

Despite criticisms of the Jacoby and Hoyer research, subsequent studies have generally found miscomprehension rates consistent with those of the original study. Jacoby, Hoyer, and Zimmer (1983), for example, found miscomprehension rates ranging from 12 to 30 percent, depending on whether the information was presented visually, aurally, or audiovisually (visual presentation resulted in the lowest miscomprehension). Gates (1986) reported miscomprehension rates ranging from 7 to 36 percent for 30-second television commercials. Edwardson, Grooms, and Proudlove's (1981) broadcast journalism study revealed a 37.7 percent miscomprehension rate for newscasts, whereas Hoyer and Jacoby (1985) found a 33.7 percent miscomprehension rate for public affairs programs. Chaiken and Eagley (1976) reported a 38 percent miscomprehension rate, and a 1980 study by Lipstein indicated a 32 percent miscomprehension rate for advertising content. Morris, et al. (1986) found miscomprehension rates as low as 9 percent and as high as 44 percent for true-false items related to print advertisements for blood pressure and arthritis medicines among members of the target market for these products (although only two ads related to a single product category were studied).

Jacoby and Hoyer (1989) concluded that miscomprehension of print ads was between 21 and 37 percent, but have subsequently revised these estimates stating:

This reanalysis showed that for TV advertising the median miscomprehension rate for individual meanings is 23.3 percent (or, when adjusted for guessing based on the analysis conducted by Schmittlein and Morrison 1983, 14.9 percent). For advertising communications appearing in mass-media magazines, the median miscomprehension rate for individual meanings is 11.8 percent (Jacoby and Hoyer 1990), p. 14).

All of these studies used forced exposure procedures, thus leaving external validity questions unexamined. Consistent with the argument of Jacoby and Hoyer (1982b), it seems reasonable to expect somewhat higher miscomprehension rates under naturalistic exposure conditions relative to those observed in forced exposure studies.

As this review of the miscomprehension literature reveals, varying rates of miscomprehension have been observed under varying study conditions. To integrate these findings as systematically and precisely as possible, and to provide a benchmark for comparison against the results of the present investigation, meta-analytic techniques were used to generalize across prior studies. First, a sample-size-weighted mean miscomprehension rates was computed across all studies that assessed the miscomprehension of advertising messages. Studies of miscomprehension of message types other than advertising were excluded from the analysis. Results are reported in Table 1.

TABLE 1

RESULTS OF A META-ANALYSIS OF MISCOMPREHENSION RESEARCH

The overall weighted mean miscomprehension percentage across all studies using advertising stimuli was 24.3 percent, with a standard deviation of 5 percent. Sampling error (computed according to formula derived in Hunter and Schmidt 1990, p. 105) accounted for 44 percent of the observed variance in miscomprehension proportions. Because this was less than the 75 percent that Hunter and Schmidt regard as sufficient evidence of effect homogeneity, the studies were divided into groups of print advertisement and television advertisement studies, and separate analyses were conducted for each group. The weighted mean miscomprehension proportion for print advertisements was 20.6 percent, whereas the proportion for television advertisements was 26.7 percent. Thus, the weighted mean proportion for studies of print advertisements was selected to provide the most appropriate benchmark for comparison against the results of the present research.

Miscomprehension and Believability

Believability is a key attribute for most advertising, since consumers are not likely to respond to advertising in the desired manner if they do not believe what it says. Previous research has indicated that individuals tend to place more credence in two-sided than in one-sided ads (e.g., Settle and Golden 1974; Folkes 1988), that moderate claims are believed more than extreme claims (Beltramini and Evans (1985), and that familiar statements are believed more than unfamiliar statements (Beltramini 1988). Relevant to the investigation at hand, however, Hoyer, Srivastava, and Jacoby (1984) found familiarity to be unrelated to miscomprehension.

A substantial amount of previous research (e.g., Harris 1977; Harris and Monaco 1978; Preston and Richards 1986) has associated miscomprehension with deceptive advertising and attempts by advertisers to foster false beliefs favorable to the advertised product. Jacoby and Hoyer (1982a) maintained that a certain amount of miscomprehension is natural and unavoidable, and that only miscomprehension in excess of that baseline amount should be taken as evidence of deceptive advertising. They pegged that normative baseline at approximately 30 percent. Preston and Richards (1986), however, argued that Jacoby and Hoyer considered only cases in which the literal message of the ad and the meaning that is conveyed to the consumer are identical. They advance the concept of "induced miscomprehension," which they describe as the conveyance of false messages by inference or false implication.

Preston and Richards (1986) also suggested that miscomprehension tends to work in favor of the advertiser. They argued that because consumer expect ads to contain information favorable to the advertiser, they tend to interpret miscomprehended information in a manner consistent with this expectation. This perspective is consistent with theorizing regarding cognitive schemes, which holds that perceptions of, and inferences about, incoming stimuli are influenced by their relation to information contained in the activated schema (Fiske and Taylor 1984). From this perspective, an advertising schema would include the information that advertising messages are favorable to the advertiser. Information that is miscomprehended will be interpreted in a manner consistent with this information, resulting in cognition favorable to the advertiser. Preston and Richards (1986) reanalyzed data from an earlier study (Preston 1967), and found that a far greater proportion of miscomprehended meanings were favorable rather than unfavorable to the advertiser.

These findings related to the meaning consumers ascribe to miscomprehended informationCto what they thought the ad said. They do not indicate whether consumers believed what they thought the ad said. Other psychological theories would suggest that consumers are likely to discount the believability of meanings they ascribe to miscomprehended advertising information. Kelley's (1967) attribution theory, for example, posited that individuals tend to discount the value of information when its source has a potential ulterior motive. Recognizing advertisers' self-interest in conveying positive information, consumers may be skeptical about the believability of advertising messages (Mizerski, Golden, and Kernan 1979; Folkes 1988). It is not clear from this perspective, however, whether miscomprehended material is subject to any greater or lesser discounting than the rest of the ad.

Miscomprehended information may be less believable than correctly comprehended information if it results from cognitive distortion of otherwise plausible facts presented in the ad (e.g., through transpositions of letters or words resulting in altered meanings). Message segments that are interpreted in a manner inconsistent with the context in which they occur are likely to be less believable than correctly comprehended information (Neisser 1976). On the other hand, if miscomprehension results from a mismatch between information contained in the receiver's cognitive schemata and the information contained in the actual message, the receiver may interpret the message as consistent with his/her beliefs and prior experience regardless of its actual meaning (Fiske and Taylor 1984), and thus as more believable than correctly comprehended information. The first type of error results from non-veridical perceptual encoding of verbal elements in the message, and hence might be described as message-related miscomprehension. The second type results from imposing semantic elements on the message that are not actually contained in it, and thus might be described as schema-related miscomprehension. The present research undertakes an exploratory investigation of the relationship between miscomprehension of advertisements and their perceived believability under ecologically valid study conditions.

Hypotheses

In accordance with the literature reviewed and the research objectives, the following hypotheses were developed for testing:

H1: Exposure to ads under naturalistic conditions will result in higher miscomprehension rates than those typical of forced exposure studies of print advertising (over 20 percent).

H2: There will be no relationship between miscomprehension and perceived believability of advertising information.

METHODOLOGY

To address the criticisms noted of previous research and test the hypotheses under ecologically valid conditions, a mock-up version of a campus newspaper containing a representative mixture of editorial and advertising content was developed. The mock-up was professionally printed and in every way resembled the actual student newspaper. All of the editorial material was actually taken from recent editions of the campus newspaper and reflected current events on campus. Additionally, ten actual advertisements targeted toward students for nationally distributed products were selected for testing based upon similar layouts and amount of information, and were included.

The mock-up newspaper was distributed to two classes of junior level business students entering a large lecture hall before class at a large university. It is typical for students at this university to pass the time before class reading the campus newspaper, so the mock-up newspaper was made available to the students without instructions regarding what to read or attend to. Since free distribution bins are typically located just outside of these classrooms, it was felt that minimal sensitization occurred with distributing the newspapers. In this respect, providing students with the mock-up newspaper resulted in exposure to advertising under as natural conditions as practically possible.

After the class period had begun, the mock-up newspaper was collected and a questionnaire booklet distributed containing four true-false questions relating to the information contained in each ad, and by a ten-item perceived believability scale well validated in other studies (Beltramini 1982, 1988; Beltraminin and Evans 1985; Gould 1988). The true-false questions assessing miscomprehension included two true and two false correct responses for each ad, with the pattern of correct true and false answers randomized across the ten ads. The four items utilized for each advertisement were chosen by the researchers from a larger pool of potential items derived in pretest to achieve consistency of item difficulty and to assure that the items fairly represented information contained in the ads. The items selected were intended to relate to factual material contained in the ads and consist of restatements of that material, more so than merely playing back advertising executions.

A total of 829 usable questionnaires were collected. Comparison of the two classes surveyed showed no significant differences in basic demographic profiles, so the data were aggregated for further analyses.

RESULTS

Miscomprehension rates for the ten ads ranged from a low of 13.2 to a high of 45.5 percent. These observed rates are summarized in Table 2, along with mean believability ratings and correlations between miscomprehension and believability for each advertisement. Although the variability in miscomprehension across test advertisements is substantial, it is consistent with variability shown in previous studies (e.g., Jacoby and Hoyer 1989). Averaging across test advertisements, the overall rate of miscomprehension was 31.5 percent, which is slightly above the 29.6 percent reported by Jacoby and Hoyer (1982b) for forced exposure to television ads and within the range of 21.5 to 36.9 percent reported by Jacoby and Hoyer (1989) for forced exposure to print ads. An item-level analysis similar to that of Jacoby and Hoyer (1990) resulted in precisely the same average miscomprehension figure (31.5 percent) as in the advertisement-level analysis.

In an attempt to reduce confusion regarding whether or not to include "non-comprehension" in calculating miscomprehension, the "don't know" response alternative was not included in the instrument. However, the missing data across items and test advertisements ranged from 3.0 to 31.7 percent, averaging 14.4 percent. If missing data are interpreted as analogous to non-comprehension, the 14.4 percent figure for missing data does in fact approximate Jacoby and Hoyer's (1989) 15.5 percent non-comprehension rate, suggesting a resultant miscomprehension rate of between 31.5 and 45.9 percent (depending on how the missing data are interpreted). This range is approximately 9 to 10 points higher on each end than that of the forced exposure study reported by Jacoby and Hoyer (1989), and higher still than the more conservative item-level results of their 1990 reanalysis. In either case, these results support H1, which predicted that miscomprehension of naturally exposed advertisements would exceed miscomprehension rates reported in a forced exposure studies of print advertising (i.e., 20 percent).

To estimate the effects of guessing on these results, a statistical procedure to correct observed miscomprehension rates for the effects of guessing and yea-saying developed by Schmittlein and Morrison (1983), and subsequently used on the Jacoby and Hoyer (1982) data, was applied. The procedure estimates a parameter representing the proportion of items answered correctly from true knowledge after correcting for guessing and yea-saying. (The procedure is rigorously derived and explained in detail in Schmittlein and Morrison [1983].)

TABLE 2

MISCOMPREHENSION AND BELIEVABILITY RESULTS

The analysis indicated that the overall proportion of items answered correctly from true comprehension was approximately 37 percent. Although this figure is seemingly low, it compares with the 46 percent true comprehension figure estimated by Schmittlein and Morrison from the Jacoby and Hoyer forced exposure to television data. Thus, the true comprehension rate for advertising under naturalistic exposure conditions is lower than the forced exposure data.

Table 2 also illustrates an overall average believability rating of 4.57 (where 1=unbelievable and 7=believable), moderately believable. Internal consistencies of the believability scale across the ten test advertisements, measured by Cronback's alpha, ranged from .93 to .96, evidencing unidimensionality. Table 2 also contains the correlation coefficients between miscomprehension and believability, where the number of each study participant's incorrect answers to the true-false items (out of four) were related to their believability ratings for each advertisement. The average correlation was -.08 across the test advertisements, indicating a weak relationship between increasing miscomprehension and decreasing perceived believability of the information contained in the advertisements tested, despite statistically significant correlations in six of the ten cases. Although the relationship is weak, its negative valence suggests a slight tendency to disbelieve miscomprehended information. It should also be noted that the most likely effect of guessing on questions about ad content would be to increase random error, and thus attenuate the strength of the miscomprehension-believability association. Thus, the strength of the "true" relationship is likely to be understated by these results.

DISCUSSION AND IMPLICATIONS

The results of this study suggest that exposure to print ads under naturalistic conditions leads to slightly higher miscomprehension rates than those observed in forced exposure studies. The fact that relaxing experimental controls during ad exposure resulted in moderately higher levels of miscomprehension appears to validate Jacoby and Hoyer's (1982b) assertion that forced exposure in laboratory settings would tend to bias miscomprehension rates downward, rather than upward as implied by Ford an Yalch (1982).

Moreover, the results give a preliminary indication of the magnitude of miscomprehension under one set of naturalistic exposure conditions. The generalizability of the findings is limited by the nature of the sample, setting, and stimuli, but the results do suggest that forced exposure studies slightly understate actual miscomprehension rates for print ads. The point is significant in light of the vigorous criticisms of Jacoby and Hoyer's (1982a) estimates as upwardly biased (Ford and Yalch 1982; Mizerski 1982).

What is perhaps more remarkable about the results is the fact that exposing study participants to test advertisements under naturalistic conditions without instructions regarding what to read or attend to resulted in only slightly greater observed levels of miscomprehension than forcing them to read and attend to advertisements. The results suggest that casual exposure to print advertising can, at least under some circumstances, be effective in conveying intended messages.

The relatively slight increase in miscomprehension rates under more naturalistic exposure conditions provides an additional indication of the amount of miscomprehension that is "normal" or "natural." It would appear that the amount of miscomprehension that should be expected depends to some extent on context and exposure conditions. This is not surprising. However, the relatively slight increase in miscomprehension, given a dramatic change in exposure conditions, lends additional credence to earlier estimates and suggests that the "normative" or "natural" rate of miscomprehension falls within a fairly narrow range.

The low negative correlations observed between miscomprehension and believability suggest that there is a slight tendency to place less credence in meanings that are miscomprehended. Some attenuation in the strength of association between miscomprehension and believability may well have resulted from guessing. However, the negative valence of the relationship may suggest that miscomprehension has the effect of potentially decreasing the believability of the ad overall.

Several caveats must be noted. Exposing study participants to test advertisements under the most naturalistic conditions possible necessarily entailed some relaxation of internal controls. Study participants were free to attend to any of the test advertisements as much or a little as they desired (within the time provided), as they are when naturally exposed to advertising. Thus, without specifically measuring attention to each ad tested there was no assurances that they attended equally to the various test advertisements. This tradeoff was made to preserve the naturalistic conditions, rather than calling undue attention to specific components of each ad, or imposing additional measures to lengthen the completion time for subjects. Still, averaging across study participants and advertisements, the results are not drastically different from those obtained in more tightly controlled studies in which subjects' attention is directed to and focused on each test advertisement for a fixed length of time.

Although the study design provided more ecologically valid conditions for studying miscomprehension, the generalizability of the results is limited by the context. The relative homogeneity, education level, and cognitive abilities of the study participants may not be representative of the larger population, although they do constitute an identifiable and important target market and were, in fact, the intended target market for the test advertisements. Also, even though no instructions were given at the time the newspaper containing the test advertisements was distributed, the fact that they were handed out at all may have somehow influenced how study participants read the contents. Across a total of ten test advertisements and over 800 study participants, however, any such effect is not likely to be large or highly systematic.

Despite these limitations, this exploratory investigation provides a number of implications to both researchers and advertisers. The study has shown that exposure to print advertisements under naturalistic conditions resulted in slightly higher miscomprehension rates than have been reported in previous forced exposure studies. The fact that relaxation of controls on exposure to test advertisements led to only slight increases in observed miscomprehension rates suggests that previously reported "normative" or "benchmark" rates are well "within the ballpark." Contrary to previous critiques, they appear to understate somewhat the amount of miscomprehension that occurs under naturalistic conditions. Statistical integration of prior research findings indicated that information presented in print advertising is miscomprehended less than information presented in television advertising. Correction of the results of studies using both media for the effects of guessing show true comprehension rates to be low regardless of medium or exposure conditions. Estimated true comprehension of print advertisements under naturalistic conditions was 9 percent lower than the estimate for forced exposure to television advertising. However, future research efforts are needed to further explore the relationship between miscomprehension and believability among other media and other samples utilizing naturalistic (versus forced exposure) approaches to enhance the generalizability of research in this area.

REFERENCES

Beltramini, Richard F. (1982), "Advertising Perceived Believability Scale," Proceedings of the Southwestern Marketing Association, 1-3.

Beltramini, Richard F. (1988), "Perceived Believability of Warning Information Presented in Cigarette Advertising," Journal of Advertising, 17, 26-32.

Beltramini, Richard F. and Kenneth R. Evans (1985), "Perceived Believability of Research Results Information in Advertising," Journal of Advertising, 14, 18-24.

Chaiken, Shelley and Alice Eagley (1976), "Communication Modality as a Determinant of Message Persuasiveness and Message Comprehensibility," Journal of Personality and Social Psychology, 34, 605-614.

Edwardson, M., D. Grooms, and S. Proudlove (1981), "Television News Information Gain from Interesting Video vs. Talking Heads," Journal of Broadcasting, 25, 15-24.

Fiske, Susan T. and Shelley E. Taylor (1984), Social Cognition, New York: Random House.

Folkes, Valerie S. (1988), "Recent Attribution Research in Consumer Behavior: A Review and New Directions," Journal of Consumer Research, 14, 548-565.

Ford, Gary T. and Richard Yalch (1982), "Viewer Miscomprehension of Televised CommunicationCA Comment," Journal of Marketing, 46, 27-31.

Gates, Fliece R. (1986), "Further Comments on the Miscomprehension of Televised Advertisements," Journal of Advertising, 15, 4-9.

Gould, S.J. (1988), "Physician Professional Opinion Leadership and Physician Advertising: A Consumer View," Journal of Health Care Marketing, 8, 47-57.

Harris, R.J. (1977), "Comprehension of Pragmatic Implications in Advertising," Journal of Applied Psychology, 62, 603-608.

Harris, R.J. and G.E. Monaco (1978), "Psychology of Pragmatic Implication: Information Processing Between the Lines," Journal of Experimental Psychology: General, 107, 1-22.

Hoyer, Wayne D. and Jacob Jacoby (1985), "The Public's Miscomprehension of Public Affairs Programming," Journal of Broadcasting and Electronic Media, 29, (4), 437-443.

Hoyer, Wayne D., Rajendra K. Srivastava, and Jacob Jacoby (1984), "Sources of Miscomprehension in Television Advertising," Journal of Advertising, 13, (2), 17-25.

Hunter, John H. and Frank L. Schmidt (1990), Methods of Meta-Analysis: Correcting Error and Bias in Research Findings, Newbury Park, CA: Sage Publications.

Jacoby, Jacob and Wayne D. Hoyer (1982a), "Viewer Miscomprehension of Televised Communication: Selected Findings," Journal of Marketing, 46, 12-26.

Jacoby, Jacob and Wayne D. Hoyer (1982b), "On Miscomprehending Televised Communication: A Rejoinder," Journal of Marketing, 46, 35-43.

Jacoby, Jacob and Wayne D. Hoyer (1989), "The Comprehension/Miscomprehension of Print Communication: Selected findings," Journal of Consumer Research 15, 434-443.

Jacoby, Jacob and Wayne D. Hoyer (1990), "The Miscomprehension of Mass-Media Advertising Claims: A Re-Analysis of Benchmark Data," Journal of Advertising Research, 29, (2), 9-16.

Jacoby, Jacob, Wayne D. Hoyer and M.R. Zimmer (1983), "To Read, View, Or Listen? A Cross-Media Comparison of Comprehension," Current Issues and Research in Advertising, 6, 201-217.

Lipstein, B. (1980), "Theories of Advertising and Measurement Systems," Attitude Research Enters the 80's, Chicago: American Marketing Association.

Mizerski, Richard W. (1982), "Viewer Miscomprhension Findings are Measurement Bound," Journal of Marketing, 46, 32-34.

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

Monaco, G.E. and D. Kaiser (1983), "Effects of Prior Preference, Inferences, and Believability in Consumer Advertising," in Information Processing Research in Advertising, R.J. Harris, ed., Hillsdale, N.J.: Erlbaum.

Morris, L.A., D. Brinberg, R. Klimberg, C. Rivera, and L.G. Millstein (1986), "Miscomprehension Rates for Prescription Drug Advertisements," Current Issues and Research in Advertising, 9, 93-117.

Neisser, Ulrich (1976), Cognitive and Reality, San Francisco: Freeman.

Preston, Ivan L. (1967), "Logic and Illogic in the Advertising Process, "Journalism Quarterly, 44, 231-239.

Preston, Ivan L. and J.I. Richards (1986), "The Relationship of Miscomprehension to Deceptiveness in FTC Cases," in Advances in Consumer Research, R.J. Lutz, ed., 138-142.

Schmittlein, David C. and Donald G. Morrison (1983), "Measuring Miscomprehension for Televised Communications Using True-False Questions," Journal of Consumer Research, 10, 147-156.

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

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