Legitimacy of need is evoked in studies of altruism by presenting study participants with a scenario depicting a need and the circumstances which have produced it. A photograph of the recipient of the need is shown with the "induction" scenario. By manipulating the legitimacy of need of the potential recipient of altruism the effect of legitimacy may be parsed from that of altruism. Manipulating the legitimacy of the need allows for the comparison, within varying categories of legitimacy, controlling for race of the potential recipient of altruism.
Bending the empathy-altruism hypothesis to the measurement of short term effects of media portrayals will require that the study participant be aware of the portrayal, then be challenged to assist another person (pictured in a post session questionnaire) described as being in need of assistance. if the subject is empathetic and finds the need of a person legitimate, then the subject will be generous in the amount of time s/he will be willing to commit to help the individual.
Dogmatism
The empathy-altruism hypothesis holds that altruism proceeds from a durable attribute of the helper (empathy), a sensitivity to the circumstances of the person requiring aid. However, not all subjects can be expected to feel empathy for others.
A measure of dogmatism assesses the reception, understanding, evaluation, and response to information with regard to the openness or closedness of a person's cognitive framework (Carlozzi, Bull, Eells, & Hurlburt, 1995). Highly dogmatic individuals are characterized by research as: closed-minded; possessing a general intolerance for outgroups; showing high respect for, and dependence on, authority figures; intolerant of ambiguity; more likely to distort incoming messages and meanings; and prone to make extreme judgments in an effort to reduce ambiguity (Carlozzi, Bull, Eells, & Hurlburt, 1995; Orysh, 1991; Rokeach, 1960). Those low in dogmatism are likely to be more open-minded, more tolerant of ambiguity, more accepting of outgroups, more likely to consider incoming information based on its own merits, and show less need to defend their thoughts or beliefs against those they see as different from their own (Carlozzi, Bull, Eells, & Hurlburt, 1995; Kerwin & Shaffer, 1991; Orysh, 1991; Rokeach, 1960). In other words, high dogmatics are guided by inner vision rather than by the circumstances of others, and their views of legitimacy of need are dictated by their dogmatic orientations. Thus, only moderate to low dogmatics will manifest empathy to a measurable degree and can be expected to manifest empathy and altruism after exposure to media portrayals of African Americans.
This Study
The model examined in this exploratory study requires that subjects be screened to be low in dogmatism (hence more likely to manifest empathy toward a person presented within the short time frame of an experiment). High dogmatics are unlikely to manifest a short term empathic response. Thus, they were removed from the sample of television viewers. The resulting sample of low dogmatics will then be exposed to TV news stories that differ in their portrayal of African Americans.
To identify news stories that differ in portrayal of African Americans, this research sought stimuli that differ in levels of symbolic racism (Sears, 1988). Symbolic racism provides a means for distinguishing between different contextual characteristics in portrayals of the same subject.
After presentation of the stimuli, subjects complete a post-session questionnaire to assess the perceived legitimacy of need for a person represented in scenarios designed to represent high or low legitimacy of need. The target of need will be shown in a photograph-depicting for some subjects a colleague who is African American, for others a colleague who is White. The willingness of the sample of television news viewers to assist the colleague in the photograph can then be influenced by the portrayal in the TV news story just presented.
This measurement model-if useful in future studies of short-term effects of television portrayals-will have to meet a series of standards:
1. The high versus low legitimacy of need scenarios must be perceived as evoking high or low perceived legitimacy of need.
2. The measure of willingness to help (altruism) must vary by news portrayal and by the racial depiction of the person in need.
Hypotheses
H1: Scenarios in the dependent measure perceived as high in legitimacy of need will produce higher altruism (willingness to help).
H2: Those subjects who view unfavorable television news portrayals of African Americans will rate the legitimacy of the recipient's need lower than those subjects who view favorable television news portrayals of African Americans.
H3: Favorable television news portrayals of African Americans will produce greater willingness to help African Americans in the post session questionnaire.
Independent Variable-News Portrayal of African Americans
Twenty-four news stories were selected based on a search of the Vanderbilt Television News Archives. A group of coders was recruited to determine which stories were high or low in symbolic racism. A total of five persons, three faculty members and two graduate students-including the experimenter, coded the 24 news stories. Coders were given a detailed definition for the construct of symbolic racism. They were then asked to view the selected news stories and rate them based on their congruence with that definition. Coders used a 7-point Likert scale which ranged from 1 (congruent with symbolic racism) to 7 (incongruent with symbolic racism). The literature has identified symbolic racism as negative, or unfavorable portrayals (Entman, 1990, 1992, 1994; Lule, 1995). Thus, those portrayals coded as congruent with symbolic racism were categorized as unfavorable whereas those portrayals coded as incongruent with symbolic racism were categorized as favorable.
Unfavorable Portrayals. The two stories identified as the most unfavorable portrayals of African Americans dealt with topics characteristic of symbolic racism as discussed in the literature, related to racial quotas in education and special treatment by government. The first story was a report on a race-based scholarship program at the University of Maryland (M= 2.2, SD = 1.3). The second story detailed legislation that mandated higher entrance requirements for college athletes (M=2.0, SD = 1.2). This story profiled an African American athlete who did not meet those requirements and also pointed out that 92% of those athletes who failed to qualify were African American.
Favorable Portrayals. To be rated as favorable, a story had to portray African Americans exhibiting characteristics counter to those of symbolic racism; especially as it relates to the "traditional values,"or "work ethic" illuminated in the discussion of symbolic racism. The most incongruent story (M= 6.4, SD = .89) was a profile of Osceola McCarty. McCarty is a former domestic who saved her money and endowed a scholarship worth more than $150,000 to the University of Southern Mississippi. The second story chosen (M= 5.8, SD = 2.17) was a profile of a middle-class African American family in Washington, D.C.
Screening for Dogmatism
The potential sample of television news viewers completed the short-form dogmatism scale of Troldahl and Powell (1965). The short-form dogmatism scale employs a 6-point Likert scale with responses ranging from 1 (agree very much) to 6 (disagree very much). To maintain continuity with other measures, scales were adapted to a 7-point Likert form. Lower scores on the scale indicate more agreement with the statements and thus, a higher level of dogmatism.'
To categorize the sample by level of dogmatism, their responses to questions in the dogmatism scale were added to form a composite score. The lowest possible score was 20; the highest possible score, 140. Actual scores ranged from 51 to 117 with a median of 87. Subjects whose scores fell below the median were categorized as high dogmatics (more agreement with statements in the dogmatism scale), whereas subjects whose scores fell above the median were categorized as low dogmatics (showing less agreement with statements in the dogmatism scale). Overall, 49% of the subjects (N = 62) were categorized as high dogmatic and 51 % (N = 64) were categorized as low dogmatic. Dogmatism scores were not normally distributed. The short-form dogmatism scale yielded a Cronbach's alpha of .65. This procedure of categorizing dogmatics is consistent with previous studies employing the dogmatism scale (see Kerwin & Shaffer, 1992; Orysh, 1991). To assess whether a median split could have affected results, supplemental analyses were run in which study participants were categorized by quartile and tercile splits on dogmatism: the results did not yield any differences from the median split.
Dependent Variables
Legitimacy. To find scenarios evoking need which accurately represented high and low legitimacy, six potential scenarios were pretested on 109 students in a lower division general education course. Of these six scenarios, three were intended to be high in legitimacy whereas the other three were intended to be low in legitimacy. Legitimacy was manipulated by altering the cause for the perceived need in the post-session questionnaires. For example, one scenario had a classmate expressing a need for some tutoring. In the high legitimate scenario, the cause behind that need was a student missing class because of a death in the family. In the low legitimate scenario, the cause behind the need was a student missing class because of partying too much. Pretest participants were asked to rate the legitimacy of the recipient's excuse on a 7 point Likert scale with 1 being "very low," and 7 being "great." The most and least legitimate scenarios (M= 6.5, SD = .76 and M = 1.8, SD = 1.1, respectively) were selected for inclusion in the main experiment. Both scenarios contained an almost identical number of words per sentence and characters per word. For the scenario rated the least legitimate, the Flesch Reading Ease was 69.4 and the Flesch Grade Level was 8.0. For the scenario that was rated the most legitimate, the Flesch Reading Ease was 69.2 and the Flesch Grade Level was 8.0.
Analysis of pretests showed significant correlations between the legitimacy of need and subject's willingness to help. As subject's ratings of the legitimacy of the need went up, so did willingness to help (for the six scenarios, r= .21 to r= .63, all sig. at p < .02) and vice versa.2
In the main experiment, each scenario referred to "the student pictured above." To examine whether the race of the potential recipient made a difference in the study participants' willingness to help, pictures of either African Americans or White Americans were attached to the scenarios. Since each subject read two scenarios, each subject saw pictures of two different White Americans, or pictures of two different African Americans.
Expression of Altruism. After viewing the news stories, participants were asked to read two scenarios in which a potential recipient of altruism expressed a need. The scenarios presented the potential recipient of altruism as a classmate who needed assistance with course material for a class in which the subject is enrolled. Altruism was measured by responses to the question "How much time would you be willing to spend with this student to help him catch up on the material he missed?" Potential responses were: 0 = no time at all, 1 = 15 minutes, 2 = 30 minutes, 3 = 45 minutes, 4 = 60 minutes, 5 = 75 minutes, 6 = 90 minutes, and 7 =120 minutes.
Altruism appeared to provide an unobtrusive measure. Of the 126 subjects in the sample, none of the subjects-in answering open-ended questions in the postquestionnaire, was able to identify altruistic behavior as the true aim of the research.
Results
The Sample. Subjects were drawn from three introductory, undergraduate journalism courses at a large, public university in the Southeast. By gender, 72% of the subjects in the sample were female. All African American participants were removed from the sample. The subjects ranged in age from 18 - 32 with more than 95% of the sample falling within the 18 - 25 age range. Dichotomizing the dogmatism variable and removing the high dogmatics yielded 46 subjects who were likely to show empathy, and thus included in the study.
The first measurement requirement was that the high versus low legitimacy of need scenarios will have to be perceived as evoking high or low perceived legitimacy of need. A one sample t-test indicated that the two scenarios did evoke high (M = 6.57, SD = .76, p = .001) or low (M = 1.75, SD = 1.03, p 001) perceived legitimacy of need, as required.
A mixed-model ANOVA tested the hypotheses. This ANOVA provided the opportunity to test the within-subject manipulation of the legitimacy of the need in conjunction with the between-subjects manipulations of the valence of the news portrayal and the race of the potential recipient of help in the post-session questionnaire.
The ANOVA yielded no main effect for the race of the potential recipient or the valence of the news story. There was a significant main effect for legitimacy, F = (1, 90); p = .001, W2 = .89. The independent variables of legitimacy-the race of the potential recipient and the valence of the news story portrayal-were submitted to a multiple regression employing stepwise extraction. The dependent variable was the subject's overall willingness to help, measured in minutes.
The subject's rating of the legitimacy of the need in the low legitimacy scenario was the only predictor of willingness to help (R2 = .18, F = 4.30, p = .02). The results of the mixed-model ANOVA, in conjunction with the multiple regression, provide support for Hypothesis 1, that legitimacy (high vs. low) must be perceived.
Hypothesis 2 stated that those subjects who view unfavorable television news portrayals of African Americans will rate the legitimacy of the recipient's need lower than those subjects who view favorable television news portrayals of African Americans. To test this hypothesis, ANOVA included subject ratings of the legitimacy of need as the dependent variable and the race of the potential recipient, with the valence of the news portrayal as independent variables. The ANOVA yielded only one significant main effect, that of the valence of the portrayal, F = (1, 4.4); p = .04. Subjects who saw the unfavorable television news portrayals of African Americans rated the legitimacy of need significantly lower (M = 1.50) than those who saw the favorable television news portrayal of African Americans (M = 1.91). These results provide support for Hypothesis 2.
Hypothesis 3 stated that favorable television news portrayals of African Americans will indicate in the post-session questionnaire greater willingness to help African Americans. An ANCOVA examined the interaction between the valence of the news portrayal and the race of the potential recipient on a subject's willingness to help. Since legitimacy was a within-subject manipulation, it was included in the analysis as a covariate.
The valence of the portrayal, and the race of the potential recipient had no main effect on the subject's willingness to help. However, the ANCOVA did reveal a significant interaction between the valence of the portrayal and the race of the potential recipient, F = (1, 6249); p = .006, -eta 2 = .31. The means for this interaction are presented in Table 1.
The Newman Keuls' test for differences among multiple means was employed to examine the means in Table 1. There were two significant differences in means. First, subjects were significantly more likely to help an African American than a White American after seeing news stories which contained unfavorable portrayals of African Americans. Second, subjects were significantly less likely to help an African American than a White American after seeing news stories which contained favorable portrayals of African Americans. Thus, while the results in Table 1 provide support for the second measurement requirement, these results do not provide support for Hypothesis 3.
Discussion
This study set out to adapt the empathy-altruism hypothesis to measure the effects of favorable or unfavorable news portrayals of African Americans in a possible sequential relationship between media portrayals and later behavior.
In order for the model to be valid, two criteria had to be met. First, scenarios which were intended to be high or low in legitimacy had to produce perceptions of high and low legitimacy. Second, altruism had to vary by the race of the person depicted in the induction scenario and to be affected by news portrayal. Results showed that both of these criteria were met. It appears the measurement model is potentially useful as a tool for gauging the short-term effect of racial portrayals in television news.
Contrary to expectation, those subjects who viewed unfavorable portrayals of African Americans were more willing to help, whereas those subjects who saw favorable portrayals were less willing to help.
What factors in the study could explain this unexpected result? Participants in the study who demonstrated empathy toward the photo of an African American student in the post-session questionnaire and who judged the need of the individual depicted as legitimate, did demonstrate greater altruism or willingness to help. This willingness to help was greatest after the study participant had been exposed to a news story manifesting symbolic racism. This could have happened due to (1) a flawed understanding in the literature as to what constitutes a negative portrayal because of implied symbolic racism or (2) because viewers feel such unfair portrayals warrant greater altruism for all African Americans. If our understanding of the nature of symbolic racism in television news is inaccurate or incomplete, future research should systematically examine more levels of symbolic racism and levels marked by clear differences in the portrayals (i.e SES of person portrayed, statements made by person portrayed, etc. ). In time, a lexicon of features indexing portrayals with symbolic racism should result.
The second possibility-that contemporary viewers are aware of symbolic racism and respond with more generous levels of altruism toward African Americans is an intriguing possibility and could be the consequence of educational and social efforts to produce tolerance. To assess the likelihood of this explanation will require the news lexicon of symbolic racial portrayals called for above. As a consequence, it must be seen as a long-term research goal.
In addition, some research is required on specific measurement issues. How do viewers connect the race of those portrayed in the news stories to the race of those in the scenarios? Hypothesis 2 stated that those subjects who view unfavorable television news portrayals of African Americans will rate the legitimacy of the recipient's need lower than those subjects who view favorable television news portrayals of African Americans. This was the case. When subjects rated legitimacy of need lower, they did so for all potential recipients-both White and African American-there was no relationship between the rating of the legitimacy of the potential recipient's need and the race of the potential recipient. Yet differences in race appeared in altruism. Clearly, subjects were able to make some connection between the portrayal and the potential recipient when they were asked to help. But, the specific nature of this connection warrants further study.
While the overt, traditional stereotypes of African Americans that were found in the early days of the medium are not present today, research has found that African Americans are still misrepresented in television portrayals. However, the nature of this misrepresentation has not yet been clearly defined. Many researchers have found portrayals that are unfavorable toward African Americans, yet there is no framework to consistently identify what would make a portrayal positive or negative. Modern /symbolic racism (Sears, 1988) is ". . almost wholly abstract, ideological and symbolic in nature" (pp. 55-56). It appears the abstract nature of the construct may be a factor in studying portrayals manifesting symbolic racism.
In addition, when examining the specific content of the news portrayals, the overall context also should warrant examination. Much of the research into symbolic racism in television news has focused on news stories portraying crime and criminals. On the other hand, crime stories may offer a communication too overtly racist, thus alerting viewers of the intent of the research. But that proposition was not tested here and requires future study.
While this research suggests that different portrayals have different effects, a better system for classifying portrayals is needed. The combination of dogmatism and altruism appears to be unobtrusive and hold promise for future research into the effects of mediated messages. Much more study is required, but the approach is promising.
[Reference]
Batson, C. D. (1987). Prosocial motivation: Is it ever truly altruistic? In L. Berkowitz (Ed.) Advances in experimental social psychology (pp. 65-122). San Diego: Academic Press.
Batson, C. D., Batson, J. G., Slingsby, J. K., Harrell, K. L., Peekna, H. M., & Todd, R. T. (1991) Empathic joy and the empathy-altruism hypothesis. journal of Personality and Social Psychology, 61 (3), 413-426.
Batson, C. D., Dyck, J. L., Brandt, J. R., Batson, J. G., Powell, A. L., McMaster, R. M., & Griffitt, C. (1988). Five studies testing two new egoistic alternatives to the empathy-altruism hypothesis. Journal of Personality and Social Psychology 55(1), 52-78.
Batson, C. D., Fultz, J., & Schoenrade, P. A. (1987). Adults' emotional reactions to the distress of others. In N. E. Eisenberg & J. Strayer (Eds.), Empathy and its development (pp. 163-184). Cambridge: Cambridge University Press.
Batson, C. D., & Oleson, K. C. (1991). Current status of the empathy-altruism hypothesis. In M.S. Clark (Ed.) Prosocial behavior, Vol 12 (pp. 62-85). Newbury Park, CA: Sage Publications. Batson, C. D., & Shaw, L. L. (1991). Evidence for altruism: Toward a pluralism of prosocial motives. Psychological Inquiry 2 (2), 107-122.
[Reference]
Batson, C. D., & Weeks, J. L. (1996). Mood effects of unsuccessful helping: Another test of the empath-altruism hypothesis. Personality and Social Psychology Bulletin, 22 (2), 148-158.
Carlozzi, A. F., Bull, K. S., Eells, G. T., & Hurlburt, J. D. (1995). Empathy as related to creativity, dogmatism, and expressiveness. Journal of Psychology, 129 (4), 365-373.
Corea, A. (1993). Racism and the American way of media. In A. Alexander and J. Hanson (Eds.) Taking sides: Clashing views on controversial issues in mass media and society (pp. 24 31). Guilford, CT: Dushkin Publising Group.
Dates, J. (1990). A war of images. In J. Dates and W. Barlow (Eds.), Split images: African Americans in the mass media (pp. 1-25). Washington, D.C.: Howard University Press. Eisenberg, N. E. & Fabes, R. A. (1991). Prosocial behavior and empathy: A multimethod
developmental perspective. In M.S. Clark (Ed.) Prosocial behavior: Vol 12 (pp. 34-61). Newbury Park, CA: Sage Publications.
Eisenberg, N. E. & Miller, P. A. (1987a). Empathy, sympathy, and altruism: Empirical and conceptual links. In N. Eisenberg & J. Strayer (Eds.), Empathy and its development(pp. 292-316). Cambridge: Cambridge University Press.
Eisenberg, N. E. & Miller, P. A. (1 987b). The relation of empathy to prosocial and related behaviors. Psychological Bulletin, 10 (1), 91-119.
Entman, R. M. (1990). Modern racism and the images of blacks in local television news.
[Reference]
Critical Studies in Mass Communication, 7, 332-345.
Entman, R. M. (1992). Blacks in the news: Television modern racism and cultural change. Journalism Quarterly, 69 (2), 361-361.
Entman, R. M. (1994). Representation and reality in the portrayal of blacks on network television news. Journalism Quarterly, 71 (3), 509-520.
Lule, J. (1995). The rape of Mike Tyson: Race, the press and symbolic types. Critical Studies in Mass Communication, 12 (2), 176 - 196.
Orysh, L. K. (1991). Exploring a model to understand white students racial attitudes. Unpublished doctoral dissertation, University of North Carolina, Chapel Hill.
Penner, P. S. (1984). Altruistic behavior: An inquiry into motivation. Amsterdam: Rodopi. Piliavin, J. A., & Charng, H. (1990). Altruism: a review of recent theory and research. Annual Review of Sociology, 16,27-65.
Rokeach, M. (1960). The open and closed mind. New York: Basic Books, Inc.
Schroeder, D. A., Dovidio, ). F., Sibicky, M. E., Matthews, L. L., & Allen, J. L. (1988). Empathic concern and helping behavior: Egoism or altruism? Journal of Experimental Social Psychology, 24, 333-353.
Schwartz, S. H. & Fleishman, J.A. (1978). Personal norm and the mediation of legitimacy effects on helping. Social Psychology, 41, (4), 306-315.
Sears, D. 0. (1988). Symbolic racism. In P A. Katz and D. A. Taylor (Eds.) Eliminating racism: Profiles in controversy (pp. 53-85). New York: Plenum Press.
Shaffer, D. R., & Case, T. (1982). On the decision to testify in one's own behalf: Effects of withheld evidence, defendant's sexual preferences, and juror dogmatism on juridic decisions. journal of Personality and Social Psychology, 42, (2), 335-346.
Thornton, L. (1990). Smilin faces tell lies: The news industry. In J. Dates and W. Barlow (Eds.), Split images: African Americans in the mass media (pp. 388 - 420). Washington, D.C.: Howard University Press.
Troldahl, V. & Powell, A. (1965). A short-form dogmatism scale for use in field studies. Social Forces, 44, 211-215.
United States Commission for the Study of Civil Rights (1977). Window dressing on the set: Women and minorities in television. Washington, D.C.: U.S. Government.
United States Riot Commission Report (1968). Report of the National Advisory Commission on Civil Disorders. New York: Bantam.
Wilson, M. & Dovidio, J. F. (1985). Effects of perceived attractiveness and feminist orientation on helping behavior. Journal of Social Psychology, 125, (4), 415-420.
Zagona, S. V., & Zurcher, L. A. (1965). Notes on the reliability and validity of the dogmatism scale. Psychological Reports, 16, 1234-1236.