Advances in Consumer Research Volume 26, 1999 Pages 527-532
TERROR MANAGEMENT AND MARKETING: HE WHO DIES WITH THE MOST TOYS WINS
Naomi Mandel, University of Pennsylvania
Steven J. Heine, University of Pennsylvania
In recent years, reports of deaths in American news programs have increased considerably. It is impossible to watch the local news without being bombarded with reports of the latest gory murders, fatal car crashes and deadly fires. How does this death-related content affect consumers' perceptions of products advertised within these programs? This paper demonstrates that high-status items are evaluated more favorably by individuals who are subtly reminded of their own impending mortality than by control subjects. In contrast, low-status and non-status items are rated slightly less favorably by mortality salient subjects than by their control counterparts.
Are there situations where advertising during a depressing, death-related television program can increase interest in a product? Previous research has shown the contrary-content of such programs can negatively influence viewers' attitudes toward the commercials. For example, subjects who saw a documentary film on Nazi war crimes had more negative attitudes toward the products advertised than those who saw a more neutral program (Axelrod 1963). However, this paper proposes that certain products, such as high-status items, may be evaluated more favorably when embedded in a negative, emotion-laden context.
One psychological theory that supports this idea is Terror Management Theory (TMT; Solomon, Greenberg & Pyszczynski 1991), which posits that one purpose of culture is to minimize the anxiety associated with the awareness of one's own mortality. Culture provides immortality via religious beliefs such as an afterlife or by allowing individuals to feel that they are key performers in a meaningful cultural drama. Individuals may avoid mortality anxiety by making an enduring mark on the world through group memberships, awards, books written, or, in the case of the present investigation, by acquiring tangible symbols of value.
How can terror management be applied to a marketing context? We hypothesize that subjects who are exposed to material which reminds them of their own impending deaths, such as a news story about a particularly gory murder, will have a heightened interest in purchasing an item which symbolizes their value within their culture. Such items might include status symbols such as a Lexus or designer clothing, or membership in a prestigious organization. For example, an individual who has just read a news story on the web about the latest tragedy may be more likely to click on a banner ad stating, "Click here to find out more about owning a Lexus." A similar effect is expected for advertisements on the obituary page of a newspaper or during a particularly morbid television program, such as "When Animals Attack." An examination of these effects is particularly timely in the United States, where reports of murders have increased considerably in recent years despite the substantial drop in homicide rates nationwide.
TERROR MANAGEMENT IN THE LITERATURE
Several studies in marketing have shown that the mood produced by the content of a movie or television program can impact attitudes toward consumer products that are advertised within the programs. Axelrod (1962) had subjects rate products both before and after they had seen a documentary on Nazi war crimes, and the ratings were significantly lower for most products after watching the movie. On the other hand, happy programs have been shown to induce greater perceived commercial effectiveness, more positive cognitive responses, and better recall for the products advertised (Goldberg and Gorn, 1987). While these studies have shown that programs that contain distressing material result in lower ratings of products and advertisements, these studies have collected ratings for non-status items, such as ketchup and pantyhose. The goal of this paper is to show that, as terror management theory predicts, the opposite effect will obtain for status items. When a consumer is reminded of her own mortality through exposure to death-related content, she may seek to reduce this anxiety through acquiring material possessions.
Terror Management Theory (Greenberg, Pyszczynski, & Solomon, 1986; Solomon, Greenberg, & Pyszczynski, 1991; Pyszczynski, Greenberg & Solomon 1997) is inspired by the works of Becker (1973; 1975) as well as Darwin, Freud, and others who have argued that one of humans' most basic instincts is the drive toward self-preservation. Becker argued that people counter their fear of death by creating a cultural worldview, which gives meaning and order to the world. This establishes an "anxiety buffer" between the individual and the uncertain and uncontrollable universe. TMT posits that terror is managed through two mechanisms: (1) the individual's cultural worldview, containing a set of standards through which the individual attains personal value and thus immortality; and (2) self-esteem, which is acquired by achieving the standards of value which are defined as part of the cultural worldview.
These two mechanisms result in two basic hypotheses. The first, the anxiety-buffer hypothesis, suggests that strengthening a person's self-esteem should relieve anxiety due to death-related threats. This hypothesis was supported through experiments which showed that increasing a subject's self-esteem reduced self-reported anxiety in response to viewing death-related stimuli (Greenberg et al., 1992; Greenberg, Pyszczynski, et al., 1993). Second, the mortality salience hypothesis states that reminding people of their impending mortality should increase the need for cultural structure, leading the individual to have unusually positive reactions to those who support the cultural worldview and unusually negative reactions to those who threaten it. Many experiments have shown that reminders of death lead to more favorable evaluations of people who personify cultural values and negative evaluations of people who defy those values (see Pyszcyzynski, Greenberg & Solomon, 1997, for a review). For example, Rosenblatt et al. (1989) asked subjects to recommend a bail amount for a woman accused of prostitution, and those subjects who were manipulated to think about their mortality were more punitive than those for which mortality was not salient. This and dozens of other experiments have controlled for the possible alternative explanation that responses could have been influenced by the affect generated through the manipulation.
We hypothesize that subjects who are manipulated to think about death will be more interested in purchasing high-status products than those who do not receive the manipulation. Belk and Pollay (1985) have shown that consumption is increasingly portrayed in the media as an end in itself, and that material possessions can create a feeling of well-being and happiness. By purchasing a high-status product, such as a luxury car or designer clothing, an individual may feel more valuable within the culture, thus relieving anxiety in response to threats to his mortality. Solomon, Greenberg & Pyszczynski (1991) support this idea by stating that the acquisition of material objects such as jewelry or automobiles may give a person evidence of value, at least in the United States. As Becker (1975) noted, "By continually taking and piling and computing interest and leaving to one's heirs, man contrives the illusion that he is in complete control of his destiny."
THE ROLES OF CONSCIOUSNESS, INDIVIDUAL DIFFERENCES, AND AFFECT
The manipulation of mortality salience must be handled very delicately. Studies have shown that subtle manipulations, such as "Please briefly jot down emotions that the thought of your own death arouses in you", are more effective in generating the expected worldview defense than more blatant manipulations, such as "consider your deepest emotions about your own death, imagining that you have just been diagnosed with cancer." (Greenberg, Pyszczynski, et. al, 1994, Study 1). In addition, mortality salience effects were only found when the subject was distracted from the problem of death immediately after the manipulation, while subjects who were forced to keep death-related thoughts in memory did not produce the expected effects (Greenberg, Pyszczynski et. al., 1994, Studies 2 & 3). Study 4 of the same paper showed that accessibility of thoughts relating to death were higher after delay and distraction than immediately after the manipulation. The authors suggest that mortality must be made salient and then removed from focal consciousness, in what Freud would call preconsciousness. This is because when people are aware that they are having death-related thoughts, they become defensive and try to control them.
Due to the sensitivity of mortality salience to differing experimental methods, the experiment used a manipulation that is identical to the one used in most successful TMT experiments, which asks subjects a series of death-related questions in a questionnaire format. They were then distracted by answering some affect questions, followed by ratings of product advertisements. This study focuses on establishing the predicted relationship between mortality salience and product preferences by using this traditional manipulations. However, future research will explore whether manipulations such as death-related news articles or websites produce the same effect.
Individual differences have been shown to moderate how a person reacts to mortality anxiety. People seem to have unique styles of coping with the threat of impending death. For example, authoritarians have been shown to be especially likely to react positively to similar others and negatively to dissimilar others when their mortality is threatened (Greenberg et al. 1990). Similarly, Greenberg et al. (1991) demonstrated that for conservatives, mortality salience increased favorable reactions to in-group members and unfavorable reactions to out-group members, while for liberals, mortality salience did not change reactions. Other factors which might cause differences in reactions to mortality salience include gender or age (where people who are closer to impending death might experience more anxiety). Because of individual differences, some consumers may deal with their anxiety by purchasing a new car, while others may turn to religion or contribute to charity. Thus, the experiments must control for possible covariates such as age, gender, authoritarianism and conservatism.
One possible side-effect of manipulating mortality salience is that subjects who have received the manipulation may experience more negative affect than control subjects. This negative affect might then cause products to receive lower ratings, as in Axelrod (1962), which would confound the expected results. Thus, the experiments in this study must carefully control for negative affect. One way to control for this result is to have target subjects answer questions about death, while control subjects answer questions about depression. In addition, measures of affect were taken after the manipulation and compared for the two groups.
In the first study, 74 students in an introductory social psychology course filled out a questionnaire at home in exchange for partial course credit. They were told that the purpose of the study was to explore their political attitudes, personality traits and preferences. All questions were presented on a 7-point Likert scale. The first 25 questions measured potential covariates such as authoritarianism (Adorno, 1950) and conservatism. Next, the subject answered 18 questions about either fear of death or depression, which were intentionally placed in the middle of the questionnaire to avoid arousing the subjects' suspicions. Treatment subjects completed the fear of death scale (Boyar, 1964), which contained questions such as, "The idea of never thinking again after I die frightens me," while control subjects completed the depression scale, containing such items as, "I have trouble sleeping through the night." (Zung 1965). Thus, both groups completed the same number of negatively framed questions. Immediately after the manipulation, subjects completed a 20-question affect scale (Watson, Clark & Tellegen, 1988). The purpose of the affect scale was twofold: to control for potential differences between experimental conditions in affect and to distract subjects so that mortality thoughts were accessible but not their central focus.
In the second part of the experiment, subjects rated product advertisements. We hypothesized that mortality salient subjects would be more interested than control subjects in purchasing high-status products because these products make them feel more valuable within the culture. Each subject viewed four product advertisements. There were two high-status products (a Lexus automobile and a Rolex watch), and two low- or non-status products (a Chevrolet Metro automobile, whose ad emphasized low price, and an ad for Pringles potato chips). The order of these advertisements was counterbalanced. After each ad, the individual answered three questions on a 7-point Likert scale. For example, the questions for the Lexus advertisement were:
1. How effective did you find this advertisement? (Effectiveness)
2. After reading this advertisement, how interested are you in buying a luxury car? (Interest)
3. If you were thinking about buying a luxury car, how likely would you be to purchase a Lexus? (Purchase Intentions)
These questions were modified for each ad, so for example, Metro question #2 read, "After reading this advertisement, how interested are you in buying an economy car?" Since the subjects were students, they were told to assume for all of the questions that they had graduated from college and were earning a comfortable salary. Thus, they could afford any of the items, but it would probably involve trading off other purchases.
After completing the product ratings, mortality salient subjects completed the depression scale, while control subjects completed the fear of death scale. Since they had already completed all of the dependent measures, this additional scale could not influence their answers on these critical tasks. However, we felt that it was important for all of the subjects to complete all of the questions, so we could use information such as the subject's fear of death score as covariates in the analysis.
PRODUCT RATINGS FOR MORTALITY SALIENT AND CONTROL SUBJECTS
In the product rating task, subjects were asked three questions about each advertisement (Effectiveness, Interest, and Purchase Intentions). A Cronbach test of reliability showed that these three questions were measuring the same underlying construct. The items formed a cohesive package for Pringles (a=.82), Rolex (a=.76), Lexus (a=.82), and Metro (a=.86). Thus, it seems reasonable to use an average of the three questions as a general measure of the subject's interest in the product.
Test of Hypotheses
For each of the four products, a t-test was used to examine the difference in product ratings between the mortality salient subjects and the control subjects. Table 1 summarizes the results of these analyses.
Rolex. Since we had predicted that mortality salient subjects would give higher ratings to status products, we used a one-tailed t-test. As predicted, subjects in the mortality salient group rated the Rolex higher than did subjects in the control group on each of the three questions. Mortality salient subjects gave Rolex an average rating of 3.54, while control subjects gave it an average rating of 2.83, and this difference was significant (t=2.07, p=.02). This provides support for our hypothesis, that subjects who were reminded of their impending mortality would have an increased interest in material possessions.
Lexus. Mortality salient subjects also rated the Lexus higher than did the control group on each of the three questions. Treatment subjects gave Lexus an average rating of 4.68, while control subjects gave an average rating of 4.12. When the three questions were averaged together, this effect was significant (t=1.65, p=.05). Since mortality subjects again rated a high-status product higher than did their control counterparts, these results provide additional support for our hypothesis.
Metro. We did not have an a priori hypothesis for low-status products, and assumed that mortality salient and control subjects would rate these products similarly. Thus, we used a two-tailed t-test to compare the ratings in the two groups. Mortality salient subjects rated the Chevrolet Metro lower than did control subjects in two of the three questions, but this difference was not significant. Overall, mortality salient subjects rated the Metro a 3.28 on average, while control subjects rated it a 3.50. The fact that individuals who were thinking about their mortality rated low-status products marginally lower than did control subjects is an interesting result that needs to be explored further in future research.
Pringles. An analysis of the Pringles ratings produced similar results to those for the Metro. Mortality subjects rated Pringles nominally (but not significantly) lower than did control subjects on two of the three questions, and when averaged together, the mortality salient ratings were 4.08, while the control ratings were 4.36.
A repeated measures analysis of variance was performed using one between-subject factor (mortality salient vs. control manipulation) and two within-subject factors (high-status vs. low-status product; automobile vs. non-automobile category). This analysis revealed a significant STATUS*MORTALITY interaction (F[1,73]=6.54, p=.01). As illustrated in Exhibit 1, control subjects gave a higher average rating to the low-status products (3.93) than they gave to the high-status products (3.48), while mortality salient subjects gave a higher average rating to the high-status products (4.09) than they gave to the low-status products (3.63).
Analysis of Covariates
Did mortality salient subjects experience more negative affect than their control counterparts, thus impacting their product ratings? A Cronbach test of reliability showed that the 20 affect questions were highly intercorrelated (a=.85), so the mean affect was used in the analysis (where a higher number indicates more positive affect). Mortality salient subjects rated their affect levels only slightly lower than control subjects (5.10 vs. 5.16, respectively), and this difference was not significant (F[1,73]=0.09, p>.50). Since the level of affect was virtually identical in both groups, it could not have caused the differences in product ratings.
An analysis of covariance examined whether the following variables moderated the main effect of mortality salience on product ratings: authoritarianism, conservatism, score on the fear of death scale, and score on the depression scale. After successfully performing a Cronbach test of reliability for each construct, the subject's responses to individual questions were averaged to form overall measures of authoritarianism, conservatism, fear of death, and depression, which were then added to the original model. However, none of these variables was found to moderate the main effect. Moreover, analyses revealed that the order of the advertisements had no impact on our effects.
This experiment successfully showed that subjects who are thinking about their own deaths are more likely than others to be interested in products such as luxury cars or prestigious watches. Meanwhile, there was no significant difference between these subjects' ratings of neutral or low-status items. If anything, mortality salience seems to decrease the desirability of these non- or low-status goods. The practical application of these results is that marketers of luxury products should advertise during program content such as "When Animals Attack," while those who market low- or non-status items should steer clear of these programs.
These results are not inconsistent with those of Axelrod (1962), who found that watching Nazi war crimes decreased subjects' interested in items such as ketchup and pantyhose. However, subjects in the Axelrod study (unlike subjects in the current study) experienced increased negative affect as a result of the manipulation, which could explain Axelrod's stronger results.
ETHICS AND PUBLIC POLICY
The application of terror management theory to marketing is destined to bring up a host of ethical concerns. Many folks will argue that it is unethical to take advantage of people's misfortunes and anxieties in order to sell products. But while some may feel that using death to sell designer clothes, cars or jewelry is distasteful and even morally wrong, the methods suggested in this paper might be used for more community-oriented goals. For example, another way that people may respond to mortality salience is to try to behave more like an exemplary citizen of their culture, thereby upholding their cultural values. We might find that subtle reminders of one's inevitable mortality may increase memberships to volunteer organizations such as Big Brothers and Big Sisters of America. We are currently exploring this possibility. Other possible applications for marketing include: churches advertising during a particularly gory television program may find that attendance has significantly increased the following Sunday; social marketers may find that presenting public service announcements denouncing such things as drugs, drunk driving or smoking are more effective when embedded within death-related news stories. It is important for the individual marketer to determine whether these methods will be used in a positive manner.
CONCLUSIONS AND FUTURE RESEARCH
"People are constantly reminded of their vulnerability and mortality; one need only pick up a newspaper or turn on a television news program to find examples of such reminders of the fragile nature of human existence." (Greenberg et al., 1990)
The current study successfully showed that subjects who answer death-related questions have an increased interested in luxury items. However, in reality, it is unlikely that many marketers will include questions such as, "What do you think will happen when you die?" on their web pages or in their print advertisements. A more likely scenario is that consumers will happen to read obituaries or news articles and then be exposed to product advertising at the same site. We are currently exploring whether certain types of products are more appropriate to advertise after such death-related news items.
In addition, future research will further examine the antecedents of mortality salience. Little is known about what leads individuals to experience death anxiety, since TMT researchers consistently use the same manipulations. Does anxiety arise only when there is a personal threat of death, or also when one hears about the death of another? Does the manipulation have to involve a particularly untimely or gory death? One possibility is that mortality salience manipulations work because they force subjects to come to terms with their overall insignificance in the universe, and one instrument to counteract these feelings is the acquisition of material goods. We are investigating whether simply reminding an individual of his insignificance (through a scientific article about the size or age of the universe, for example) will produce the same effects. If successful, such an experiment would contribute to the Terror Management literature by demonstrating that the behaviors resulting from mortality salience are not the result of death-related thoughts, but of feelings of insignificance.
How does age moderate the effects of mortality salience? This study examined the responses of university students, but older individuals might react differently. For example, the Becker quote suggested that accumulating wealth for one's heirs might relieve death anxiety. Thus, interest in high-status products might diminish for older people, would might prefer to invest the money. Alternatively, individuals who are closer to death might be more susceptible to the mortality manipulation, thus showing even more interest in status goods.
Another question of interest is how mortality salience might influence other shopping behaviors. For example, a recent study has shown that online shoppers display more loyalty than those shopping in traditional supermarkets (Degeratu, Rangaswamy & Wu, 1998). Would individuals who had recently read an online article about a murder seek comfort through familiar brands, or seek increased variety? Both of these hypotheses are feasible. On one hand, mortality salient subjects may feel overstimulated and wish to reduce their anxiety by consuming familiar brands. Conversely, when faced with the prospect of the end of their lives, people may want to remember lives that were filled with variety.
Finally, alternative explanations for the current results need to be considered. One possibility is that mortality salient subjects simply prefer higher-priced items, as opposed to prestige items. However, our study provides evidence against this alternative hypothesis. A Chevrolet Metro costs more than a Rolex watch in absolute dollars, but mortality salient subjects showed an increased interest in the Rolex and not the Metro.
In general, this study furthers our knowledge in the marketing discipline by borrowing an important theory from social psychology and applying it to a consumer context. To our knowledge, this is the first such application of Terror Management Theory to a consumer setting.
Adorno, T., Frenkel-Brunswick, E., Levinson, D., & Sanford, R.N. (1950). The authoritarian personality. New York: Harper.
Axelrod, J.N. (1962). Induced Moods and Attitudes toward Products. Journal of Advertising Research, 19-24.
Becker, E. (1973). The denial of death. New York: Free Press.
Becker, E. (1975). Escape from evil. New York: Free Press.
Belk, Russell W. and Richard W. Pollay (1985). Images of Ourselves: The Good Life in Twentieth Century Advertising. Journal of Consumer Research 11(March), 887-897.
Boyar, J.I. (1964). The construction and partial validation of a scale for the measurement of fear of death. Dissertation Abstracts, 25, 20-21.
Degeratu, A., Rangaswamy, A. & Wu, J. (1998). Consumer choice behavior in online and regular stores: the effects of brand name, price, and other search attributes. Working Paper, Penn State University.
Greenberg, J., Pyszczynski, T., & Solomon, S. (1986). The causes and consequences of the need for self-esteem: A terror management theory. In R.F. Baumeister (Ed.), Public self and private self (pp. 189-212), New York: Springer-Verlag.
Greenberg, J., Pyszcyzynski, T., Solomon, S., Rosenblatt, A., Veeder, M., Kirkland, S., and Lyon, D. (1990). Evidence for Terror Management Theory II: The Effects of Mortality Salience on Reactions to Those Who Threaten or Bolster The Cultural Worldview. Journal of Personality and Social Psychology 58(2):308-318.
Greenberg, J., Solomon, S., Pyszcyzynski, T., Rosenblatt, A., Burling, J., Lyon, D., Simon, L., and Pinel, E. (1992). Why Do People Need Self-Esteem? Converging Evidence That Self-Esteem Serves an Anxiety-Buffering Function. Journal of Personality and Social Psychology 63(6), 913-922.
Greenberg, J., Pyszcyzynski, T., Solomon, S., Pinel, E., Simon, L., and Jordan, K. (1993). Effects of Self-Esteem on Vulnerability-Denying Defensive Distortions: Further Evidence of an Anxiety-Buffering Function of Self-Esteem. Journal of Experimental Social Psychology 29, 229-251.
Greenberg, J., Pyszcyzynski, T., Solomon, S., Simon, L. and Breus, M. (1994). Role of Consciousness and Accessibility of Death-Related Thoughts in Mortality Salience Effects. Journal of Personality and Social Psychology 67(4), 627-637.
Goldberg, M. and Gorn, G. (1987). Happy and Sad TV Programs: How They Affect Reactions to Commercials. Journal of Consumer Research 14, 387-403.
Pyszcyzynski, T., Greenberg, J. and Solomon, S. (1997). Why do we need what we need? A terror management perspective on the roots of human social motivation. Psychological Inquiry 8(1):1-20.
Rosenblatt, A., Greenberg, J., Solomon, S., Pyszczynski, T., and Lyon, D. (1989). Evidence for terror management theory I: The effects of mortality salience or reactions to those who violate or uphold cultural values. Journal of Personality and Social Psychology 57, 681-690.
Singh, Surendra and Churchill, Gilbert A., Jr. (1986). Arousal and Advertising Effectiveness. Journal of Advertising 16(1), 4-10.
Solomon, S., Greenberg, J., & Pyszcyzynski, T. (1991). A terror management theory of social behavior: The psychological functions of self-esteem and cultural worldviews. In M. Zanna (Ed.), Advances in experimental social psychology (Vol. 24, pp. 93-159).
Watson, D., Clark, L.A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology 53, 1063-1070.
Zung, W.W.K. (1965). A self-rating depression scale. Archives of General Psychiatry 12:63-70.