How Comparative Product Information Affects Consumers and Competition: the Effects of the Business Week and U.S. News &Amp; World Report Ratings

Paul N. Bloom, University of North Carolina at Chapel Hill
Lisa R. Szykman, University of North Carolina at Chapel Hill
ABSTRACT - Comparative product information programs are fast becoming a popular way to assist consumers in making product decisions. Despite the explosion of such programs, little has been done to study the "impact" that these programs have on consumer behavior and choices. This paper reports on an initial study that attempts to study these effects. Using the Business Week and U. S. News and World Report rankings of MBA programs and national universities, respectively, it was found that rankings do seem to have an effect on how consumers behave. Specifically, these rankings have been shown to be significantly related to the number of applications received by the program/university, the amount of tuition charged, the salary recruiters are willing to pay graduates, and the amount of expenditures spent per student. Future research is discussed.
[ to cite ]:
Paul N. Bloom and Lisa R. Szykman (1998) ,"How Comparative Product Information Affects Consumers and Competition: the Effects of the Business Week and U.S. News &Amp; World Report Ratings", in NA - Advances in Consumer Research Volume 25, eds. Joseph W. Alba & J. Wesley Hutchinson, Provo, UT : Association for Consumer Research, Pages: 433-439.

Advances in Consumer Research Volume 25, 1998      Pages 433-439

HOW COMPARATIVE PRODUCT INFORMATION AFFECTS CONSUMERS AND COMPETITION: THE EFFECTS OF THE BUSINESS WEEK AND U.S. NEWS & WORLD REPORT RATINGS

Paul N. Bloom, University of North Carolina at Chapel Hill

Lisa R. Szykman, University of North Carolina at Chapel Hill

ABSTRACT -

Comparative product information programs are fast becoming a popular way to assist consumers in making product decisions. Despite the explosion of such programs, little has been done to study the "impact" that these programs have on consumer behavior and choices. This paper reports on an initial study that attempts to study these effects. Using the Business Week and U. S. News and World Report rankings of MBA programs and national universities, respectively, it was found that rankings do seem to have an effect on how consumers behave. Specifically, these rankings have been shown to be significantly related to the number of applications received by the program/university, the amount of tuition charged, the salary recruiters are willing to pay graduates, and the amount of expenditures spent per student. Future research is discussed.

INTRODUCTION

Comparative product information programs have usually been viewed as positive contributors to a free-market economy. It is generally believed that consumers make better decisions when they are provided with larger amounts of low-cost, trustworthy, easy-to-access, easy-to-process information about the purchase options they are considering in the marketplace (Mazis et al. 1981, Tellis and Wernerfelt 1987, Hjorth-Anderson 1984, Moorman 1997). When inexpensive and reliable information is present, consumers should be able to find and buy the options that come closest to matching their prferred or ideal bundles of attributes. Such information should also help reduce the likelihood that consumers will be misled or deceived into choosing options that serve them poorly, since they will be able to do a better job of comparing and evaluating the product choices available to them. Furthermore, over time, manufacturers and resellers should begin to compete vigorously for the favor of increasingly better-informed consumers. This would lead to the development of products that come progressively closer to matching consumers’ preferred or ideal bundles of attributes. This logic has motivated the introduction of a host of government-mandated information programs or "remedies," including those that have created truth-in-lending disclosures, assorted safety warnings, nutritional labeling, and energy-consumption labeling. Many private efforts to supply more information to consumers, such as the publication of magazines like Consumer Reports and Consumers’ Checkbook, have also been motivated by this logic.

Criticism of comparative product information programs typically revolves around two related issues. First, some have questioned whether the costs of certain information programs are worth the benefits. This has been a concern expressed about many government-mandated programs rather than private ones (which are typically paid for by consumers who want to use the information). Critics of government-mandated programs argue that the costs to private companies of complying with the programsCplus the costs to governments of developing and policing the programsCexceed the benefits consumers obtain from the programs, primarily because consumers tend to prefer information from sellers instead of objective sources. Second, others express concern about the potential for the information providers to mislead or bias consumers. Critics worry that either (a) the measures reported (e.g., APR, mileage) are not valid and reliable, or (b) the format in which the information is presented can lead consumers to weight attributes differently and less appropriately (for them) than they might otherwise if the information was not available. For example, they might focus on calories when they might be better off focusing on saturated fats - or vice versa. In other words, the information provider may be able to manipulate consumer choice by influencing consumers’ product attribute preferences. Moreover, attempts to police these programsClimit the use of misleading or biasing measures and formatsC could add substantially to the costs of these programs (Beales et al. 1981, Wright 1979, Bettman 1986).

Although a host of new comparative product information programs have emerged over the last thirty years, very little research has been done that can provide insights into the true impact of these programs. Most of the research that has been done on these programs could be labeled as "developmental evaluations," in that the studies were done to help determine what measures or formats should be required before the programs were put in place (Andrews 1995, Bettman et al 1986, Levy and Derby 1995, Levy et al 1996, Magat and Viscusi 1992, Stewart and Martin 1994, Vankatesan et al 1986). Very few studies can be identified as "impact evaluations," examining how the programs actually affected consumer behavior, competition, or consumer welfare (exceptions include Hoffer et al 1992, Moorman 1996, Rudd and Glanz 1991, and Wholey et al 1992, and Moorman 1997). Research like this could be helpful in reaching decisions about whether to discontinue or modify existing programs. Furthermore, it could be helpful in guiding the development of emerging comparative product information programs on offerings such as hospitals, managed care providers, and automobiles (Committee for the Study of Consumer Automotive Safety Information 1996, Sofaer et al 1992, and Wholey et al 1992).

We have embarked on an effort to evaluate how a number of different comparative product information programs have impacted certain markets. In this paper, we report on an initial study that examines some of the impacts of two private comparative product information programs. Speciically, we look at how the rankings of MBA programs done by Business Week and of national universities done by U.S News and World Report influence the behavior of prospective students, employers, and donors. We expected that information of this type would have a major impact because consumers (1) are highly motivated to acquire it, (2) find it inexpensive to acquire, and (3) find it helpful in assessing a complex offering that is high in credence qualities. Our results suggest that consumers have responded very sharply to the ratings, without showing evidence that the ratings have led to poor choices. Prospective students seem more likely to include highly-rated programs/colleges in their consideration sets and are willing to pay more in tuition to attend one of these programs/schools. Recruiters and donors also seem to be more generous when dealing with higher-ranked schools. The hypotheses, methods, results, and implications of our study are reported below.

HYPOTHESES

In this study, we were basically interested in measuring (1) how much consumers actually used comparative product ratings programs in making choices and (2) whether the ratings led them to make poor choices. We decided not to attempt to measure these variables by either observing consumers or asking them to recall their usage experiences. Such studies would be both expensive and subject to biased results caused by consumers behaving or answering in a socially desirable manner. Instead, we chose to examine consumer usage by looking at indicators of how consumers, as a group, actually behaved in the marketplace. Specifically, we looked for evidence that a school’s rank was related to the number of applications the school received and the amount of tuition charged by the university. We also looked for evidence that the ratings affected how much recruitersCanother category of consumers of the comparative product informationCpaid in starting salaries to hire MBA graduates of the ranked programs. Additionally, we looked at whether undergraduate institutions were able to spend more per studentCeither because of increased donations, higher tuition, or other resource-attraction effortsCbecause of attaining higher ratings. Our approach to studying behavior bears some similarity to what has been done in the class of studies known as "hedonic pricing studies" in that we are looking for a relationship between market information and price (see, for example, Asher 1992 and Stanley 1991).

Finally, we chose to study whether the ratings led prospective students to make poor choices by looking at the scores obtained on several "satisfaction" items of the Business Week surveys over the years. Our feeling was that scores on these items would probably decline in value over time if a significant number of students had experienced a bad match between their individual needs and the attributes of a schoolCalthough we clearly recognized that other factors could produce such a decline.

We hypothesized that we would find the following, controlling for the factors indicated:

H1: The better a school’s overall rating in one time period, the more applications it received in the following time period (controlling for the school’s tuition and, for MBA programs, what the starting salaries were for its graduates).

H2: The better a school’s overall rating, the more students paid in tuition to attend that school (controlling for whether it was a state institution or not).

H3: The better aschool’s overall rating for its MBA program, the more recruiters paid in starting salaries to hire its graduates (controlling for the GMAT scores, age, work experience, and pre-program salaries of the graduates).

H4: The better a school’s overall rating for its undergraduate program in one time period, the more expenditures per student it could afford in the next time period.

H5: The satisfaction scores of students with their MBA programs remained stable over time.

Support for all of these hypotheses would be consistent with the argument that the comparative product rating programs of these publications were clearly used by consumers in making their choices and decisions, without leading them to choose a program that did not match their educational needs.

METHODS

The data we used to test the hypotheses were drawn from several different sources. Our primary source was the series of articles and books published by Business Week that are intended to guide prospective students and employers about the best business schools (Byrne 1990-1995). From the Business Week publications, we obtained the following measures:

1. The overall rank awarded to approximately 20 different schools as a result of surveys done in 1988, 1990, 1992, and 1994. Every two years, the magazine surveys (a) graduating classes to determine their perceptions of various attributes of the programs they have just completed and (b) corporate recruiters to determine their perceptions of the attributes of each school’s graduates. An undisclosed weighting scheme is used to combine all the attribute ratings to form an overall rank (Byrne 1990-1995).

2. The average post-MBA starting salary of students who graduated from each of the schools during 1990, 1992, 1994, and 1996.

3. The average pre-MBA salary of students who graduated from each of the schools during 1990, 1992, 1994, and 1996 (i.e., what they were earning in 1988, 1990, 1992, and 1994).

4. The average entering GMAT score, GPA, age, and work experience of students who graduated from each of the schools during 1990, 1992, 1994, and 1996.

5. The tuition charged by each of the schools during 1990, 1992, 1994, and 1996.

6. The average scores graduates of each of 23 schools gave on three questions that reflected their overall satisfaction levels with the programs they had just completed. These questions were: (a) To what extent did your MBA experience fulfill or fail to meet your expectations of what a good business school should be?(1=Failed expectations; 10=Vastly exceeded) (b) Do you believe your MBA was worth its total cost in time, tuition, living expenses, and lost earnings?(1=Return was 0%; 10=100% return), and (c) How would you judge the school’s performance in providing you with numerous ways of thinking or approaching problems that will serve you well over the long haul? (1=Poor; 10=Outstanding).(Note: This is the wording for these items that was used in 1988. In subsequent years, slight changes were made.)

In addition, information on the number of applications received by each of the 20 schools during 1990, 1992, and 1994 was obtained from Barron’s Guide to MBA Programs (Miller 190, 1992, 1994). Since the 1996 Barron’s Guide was not readily available to the authors, information for 1996 was obtained from REA’s Authoritative Guide to the Top 100 Business Schools and from Princeton Review’s Student Access Guide to the Best Business Schools.

Our major source for data about the undergraduate programs of "national universities" was the series of articles and books published by U.S. News & World Report (1990 through 1994). From this source we obtained data on rank, tuition, and expenditures per student for each of 25 universities for the years 1990 through 1994. We supplemented these data with information on the number of applications received by each of the 25 universities for the years 1991 through 1994 obtained from The College Handbook (College Entrance Exam Board 1993-1996).

We ran a series of ordinary least squares regressions to test the hypotheses. To test H1, we examined the following model:

Applicationst = f ( Rankt-1, Tuitiont , Post-MBA Starting Salaryt-1)

Thus, we looked at how applications received by MBA programs in 1996 were related to rank as published in 1994, controlling for relative tuition in 1996 and relative post-MBA starting salary in 1994 (and so forth for other years). We also looked at how applications in 1994 to undergraduate national universities were related to rank in 1993, controlling for relative tuition in 1994 (and so forth for other years). Since we pooled data for several years, we needed to adjust for inflation for the variables that reflected dollar figures. We chose to handle this problem by using relative tuition and relative starting salary for a given school for a given year instead of dollar figures. In other words, we calculated a mean tuition and mean starting salary for each year and calculated what each school’s tuition and starting salary was in relation to the mean figures (i.e., (tuition for each school in t)/(mean tuition in t)). The regression done for the sample of national universities did not include relative starting salary because this information was not available. In addition, since many students from ranked schools go on to graduate school instead of the work force, we believed that starting salary information for undergraduate programs would not be a good predictor of applications.

TABLE 1

REGRESSIONS USING MBA PROGRAM DATA

To test H2, we examined the following model:

Tuitiont = f ( Rankt-1, State)

We looked at whether relative tuition in 1996 was related to rank in 1994 for the MBA programs, and whether relative tuition in 1994 was related to rank in 1993 for undergraduate national universities, controlling for whether the school was a state institution or not (and so forth for other years).

To test H3, we examined the following model:

Starting Salaryt = f (Rankt-1, GMATt-1, Aget-1, Workt-1, Pre-Salaryt-1)

We looked at whether the average relative starting salary offered by MBA recruiters to the class graduating in 1996 was related to rank in 1994, controlling for the average GMAT, age, years of work experience, and relative previous salary the class had when it entered in 1994 (and so forth for other years).

To test H4, we examined the following model:

Expenditurest = f (Rankt-1, tuitiont)

We looked at whether average relative expenditures per student in 1994 for the national universities was related to rank in 1993, controlling for relative tuition in 1994 (and so forth for other years).

Finally, to test H5, we compared scores on the three satisfaction questions from the Business Week survey over the 1988-1996 time period. Not all schools were rated in all time periods. Indeed, the books published by Business Week only report the scores for the 23 top-rated schools in 1988Cand thereafter they only report, from among a total sample of 40 schools, the twelve highest scoring schools on each question and the twelve lowest scoring schools on each question. Nevertheless, there are enough data to obtain a reasonably good picture of whether students became more or less satisfied with their choices over time.

RESULTS

The regression results for the data on MBA programs are found in Table 1 and those for the data on undergraduate national universities are found in Table 2. For a given dependent variable, results are typically shown for both complete models and for more parsimonious models that included only those independent variables that achieved at least a .05 level of significance in a stepwise entering procedure. In addition, Table 3a and 3b present the results for the MBA program satisfaction scores.

The results in Table 1 tend to support H1, H2, and H3. Rank is significantly related to the number of MBA program applications a school receives in the next time period, the amount of tuition it can require in the next time period, and the starting salaries of its graduates in the next time period. The significance of rank holds up even when we controlled for other explanatory variables. In fact, rank explains how many applications a school receives better than post-MBA starting salary and tuitionCand it explains post-MBA starting salary better than age, GMAT score, GPA, or years of work experience, but not as well as pre-MBA salary.

The results in Table 2 tend to support H2 and H4. Rank is significantly related to the amount of tuition a school can require in the next time period, and the expenditures per student it makes in the next time period. The significance of rank holds up even when, to the extent possible, we control for other variables. However, rank is not significantly related to the number of applications an undergraduate national university receives in the next time period. This lack of association may be the result of a limited number of observations in our dataset. However, it may also be indicative of the importance of other influences when choosing an undergraduate program, such as geographic location, number of programs available for study, proximity to the student’s hometown, etc.

TABLE 2

REGRESSIONS USING UNDERGRADUATE NATIONAL UNIVERSITIES DATA

The results in Table 3a tend to suport H5 in that the 23 schools originally rated by Business Week seemed to be able to maintain relatively high overall satisfaction scores over time. If students felt misled or biased by whatever guidance they received from the ratingsCeither because schools genuinely were not as good as their ratings or because the ratings tended to overinflate student expectations to a point where they were hard for any school to matchCthen their satisfaction scores would probably decline over the years. Serious declines in scores on the three items did not occur and, for many schools, ratings went up. However, there seems to be a tendency for the 17 schools that were not part of the original top 23 of Business Week to have a decline in satisfaction ratings over time (see Table 3b). These schools only tend to have information revealed about them by Business Week through being listed as the bottom twelve scorers on questionnaire items. It is not clear whether satisfaction with these schools has declined because the students have become progressively poorer at finding matches for their needsCperhaps because of the lack of information about the programsCor because the students have become progressively more disappointed about not being highly ranked (especially if their schools raised expectations about being ranked).

Dissonance reduction may also be influencing the satisfaction scores. Essentially, when we encounter information that is discrepant with an existing attitude, dissonance is created. Past research suggests that there are two ways to reduce dissonance. The first is to change or update the existing attitude to reflect the new information (Festinger 1957). Another way, particularly when central attitudes are challenged, would be to bolster initial attitudes (Sherman and Gorkin 1980). Looking at the satisfaction results, we can hypothesize that students who went to highly ranked schools may feel a stronger tendency to bolster their initial attitudes that they went to a premier institution. Therefore, if they encountered information that would suggest otherwise, they would be more likely to counterargue the discrepant information and bolster their existing positive attitude toward the institution. This would explain the maintenance/increase in satisfaction rankings. On the other hand, graduates of unranked programs may not hold their program choice as a central attitude, since it would suggest that they were not as intelligent as their counterparts at highly ranked institutions. Perhaps, these individuals did not make their choice based on the overall rankings, but were more driven by other factors such as cost or location. Since their program selection is not as an important of a part of their self-image, they would respond to dissonant information by changing their initial attitudes rather than bolstering them. These attitude changes would be reflected in lower satisfaction scores over time.

Finally, students still could be making poor initial choices for themselves but end up satisfied because a shift in their utility functions might occur while they are in school. They may enter programs valuing attributes such as good teaching and opportunities for team experiences and find a school does not deliver to their satisfaction on these attributes. Still, they may come away basically satisfied in an overall sense because they may become socialized while in school to value getting a lucrative job offer much more than anything else. And, the higher the ranking of the school they attended, the more lucrative the offer they would have received. The possibility that this scenario takes place was assessed by examining, over time, several items that rated more specific attributes of schools such as teaching quality and curriculum. No decline in these scores occurred either.

DISCUSSION

The results reported here suggest that consumers use and respond sharply to at least certain types of comparative product inormation. We found evidence that is consistent with an explanation that the information (1) influenced consumers’ consideration sets (i.e., where students applied) and (2) influenced how much they were willing to pay (i.e., what tuition students paid, what salaries recruiters offered, what expenditures schools gained the ability to make). The information seemed to have these effects without misleading consumers substantially, since the users of the information seemed to feel highly satisfied with their choices and did not become significantly less satisfied over the time the information program was in place.

The apparent success of the comparative product information programs we studied can be attributed to several features that may not be prevalent with other programs. First, the information has been relatively low-cost for consumers to acquire and use. Magazines like Business Week and U.S. News & World Report are not expensive, nor are the paperback books that these magazines have published about their rankings. In addition, the rankings of these magazines have received enormous publicity and dissemination through other forms of media and through the communications of the schools that have been rated, making the cost of acquiring this information very low.

Second, these programs have been targeted toward a highly-motivated and skilled set of consumers: prospective students and employers. Other comparative product information programs may not be designed to serve such a goal-oriented and talented set of information-seekers.

TABLE 3A

OVERALL SATISFACTION FOR RATED SCHOOLS

Third, these programs have been designed to help guide decisions about offerings that are high in experience qualities and even credence qualities. Consumers can consider search attributes (e.g., size of campus, number of majors) when making a decision, but these are not really indicative of the school’s quality or its fit with a prospective student’s needs. Ratings are likely to be more valued in this kind of situationCespecially when shopping around (i.e., visiting campuses) to actually examine the search attributes of the options may be costly, and may not actually provide very helpful information.

Future research could examine whether it is possible for comparative product information programs to be effective when they operate in situations other than where they are providing very low-cost information about the experience and credence qualities of offerings that are highly-sought by talented and motivated consumers. With comparative information programs becoming a popular "remedy," there will be more opportunity to measure the impact of such programs and uncover the cases where the benefits of such programs outweigh the costs of implementing and managing them.

TABLE 3B

OVERALL SATISFACTION SCORES FOR UNRATED SCHOOLS

REFERENCES

Andrews, J. Craig (1995), "The Effectiveness of Alcohol Warning Labels: A Review and Extension," American Behavioral Scientist, 38(4).

Asher, Cheryl Carleton (1992), "Hedonic Analysis of Reliability and Safety of New Automobiles," Journal of Consumer Affairs, 26(2), pp. 377-396.

Barone, Michael J., Randall L. Rose, Kenneth C. Manning, and Paul W. Miniard (1996), "Another Look at the Impact of Reference Information on Consumer Impressions of Nutrition Information," Journal of Public Policy and Marketing, Vol. 15(1), pp. 55-62.

Beales, H., M. B. Mazis, S. C. Salop, and R. Staelin (1981), "Consumer Search and Public Policy," Journal of Consumer Research, 8(June), 11-22.

"Best National Universities," U. S. News and World Report, Vol. 109, No. 15 (October 15, 1990), pp. 118-119.

"Best National Universities," U. S. News nd World Report, Vol. 111, No. 14 (September 30, 1991), pp. 90-91.

"Best National Universities," U. S. News and World Report, Vol. 113, No. 13 (September 28, 1992), pp. 114-115.

"Best National Universities," U. S. News and World Report, Vol. 115, No. 13 (October 4, 1993), pp. 110-111.

"Best National Universities," U. S. News and World Report, Vol. 117, No. 12 (September 26, 1994), pp. 100-101.

Bettman, James R., John W. Payne, and Richard Staelin (1986), "Cognitive Considerations in Designing Effective Labels for Presenting Risk Information," Journal of Public Policy & Marketing, 5, 1-28.

Byrne, John A. (1990), Business Week Guide to The Best Business Schools, New York: McGraw-Hill, Inc.

Byrne, John A. (1991), Business Week Guide to The Best Business Schools, New York: McGraw-Hill, Inc.

Byrne, John A. (1993), Business Week Guide to The Best Business Schools, New York: McGraw-Hill, Inc.

Byrne, John A. (1995), Buenos Week Guide to The Best Business Schools, New York: McGraw-Hill, Inc.

College Entrance Exam Board (1993), The College Handbook, New York: College Entrance Exam Board.

College Entrance Exam Board (1994), The College Handbook, New York: College Entrance Exam Board.

College Entrance Exam Board (1995), The College Handbook, New York: College Entrance Exam Board.

College Entrance Exam Board (1996), The College Handbook, New York: College Entrance Exam Board.

Committee for the Study of Consumer Automotive Safety Information (1996), Shopping for Safety: Providing Consumer Automotive Safety Information, Special Report 248, Transportation Research Board, National Research Council, Washington, D. C.: National Academy Press.

Discenza, Richard and Jeffrey M. Ferguson (1992), "The Instrumental Role of Product Information: A Study of Warning Labels for Non-Prescription Drugs," Health Marketing Quarterly, 10(1,2), pp. 155-168.

Festinger, L. (1957), A Theory of Cognitive Dissonance, Evanston, IL: Row, Peterson.

Gilbert, Nedda (1996), Student Access Guide to The Best Business Schools, New York: Princeton Review Publishing Company.

Hjorth-Anderson, Chr. (1984), "The Concept of Quality and Efficiency in Markets for Consumer Products," Journal of Consumer Research, 11(September), 708-718.

Hoffer, George, E., Stephen W. Pruitt, and Robert J. Reilly (1992), "Market Responses to Publicity-Provided Information: The Case for Automotive Safety," Applied Economics, Vol. 24(7), 661-667.

Levy, Alan J. and Brenda Derby (1995), "Consumer Use of Food Labels: Impact of the New Food Labeling Regulations," working paper, U.S. Food and Drug Administration.

Levy, Alan J., Sara B. Fein, and Raymond E. Schucker (1996), "Performance Characteristics of Seven Nutrition Label Formats," Journal of Public Policy and Marketing, Vol. 15(1), pp. 1-15.

Magat, Wesley A. and W. Kip Viscusi (1992), Informational Approaches to Regulation, Cambridge, MA: MIT Press.

Mazis, Michael B., Richard Staelin, Howard Beales, and Steven Salop (1981), "A Framework for Evaluating Consumer Information Regulation," Journal of Marketing, 45(Winter), 11-21.

Miller, Eugene (1990), Barron’s Guide to Graduate Business Schools, 7th edition, Woodbury, NY: Barron’s Educational Series.

Miller, Eugene (1992), Barron’s Guide to Graduate Business Schools, 8th edition, Woodbury, NY: Barron’s Educational Series.

Miller, Eugene (1994), Barron’s Guide to Graduate Business Schools, 9th edition, Woodbury, NY: Barron’s Educational Series.

Moorman, Christine (1996), "A Quasi Experiment to Assess the Consumer and Informtional Determinants of Nutrition Information Processing Activities: The Case of the Nutrition Labeling and Education Act," Journal of Public Policy and Marketing, Vol. 15 (1), pp. 28-44.

Moorman, Christine (1997), "Market-Level Impacts of Information: Competitive Responses and Consumer Dynamics," working paper, University of Wisconsin.

Research and Education Association (1996), REA’s Authoritative Guide to the Top 100 Business Schools, Piscataway, NJ: Research and Education Association.

Rudd, Joel and Karen Glanz (1991), "Providing Consumers with Quality of Health-Care Information: A Critical Review and Media Content Analysis," in R. Mayer, ed., Enhancing Consumer Choice, Columbia, MO: American Council on Consumer Interests.

Sherman, Steven J. and Larry Gorkin (1980), "Attitude Bolstering When Behavior is Inconsistent with Central Attitudes," Journal of Experimental Social Psychology, 16,388-403.

Sofaer, Shoshanna, Erin Kenney, and Bruce Davidson (1992), "The Effect of Illness Episode Approach on Medicare Beneficiaries’ Health Insurance Decisions," Health Services Research, 27(5), pp. 671-693.

Stanley, Linda R. (1991), "A Market Test of Consumer Response to Information Disclosure," Journal of Public Policy and Marketing, 10(2), 202-218.

Stewart, David W. and Ingrid M. Martin (1994), "Intended and Unintended Consequences of Warning Messages: A Review and Synthesis of Empirical Research," Journal of Public Policy and Marketing, 13(1), 1-19.

Tellis, Gerard, J. and Birger Wernerfelt (1987), "Competitive Price and Quality Under Asymmetric Information," Marketing Science, 6(Summer), 240-254.

Venkatesan, M., Wade Lancaster, and Kenneth W. Kendall (1986), "An Empirical Study of Alternate Formats for Nutritional Information Disclosures in Advertising," Journal of Public Policy and Marketing, 5, 29-43.

Wholey, D. R., J. Christianson, S. Sanchez, R. Feldman, and M. Peterson (1992), "The Voluntary Dissemination of Performance Information by Health Care Organizations," Advances in Health Economics and Health Services Research, 13, 1-26.

Wright, Peter (1979), "Concrete Action Plans in TV Messages to Increase Reading of Drug Warnings," Journal of Consumer Research, 6(December), 256-269.

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