 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
If It Bleeds It Leads, If It's White It's Right: Local TV News in 12 Cities
Howard J. Ehrlich, Jason Weller & Allison Eden
The Prejudice Institute
2743 Maryland Avenue
Baltimore, MD 21218
www.prejudiceinstitute.org
This is the fifth in a series of studies documenting institutional discrimination in the treatment of news. Specifically, in this study which focuses on the late evening local newscast, we look at the news story and the manner in which it is shaped by race/ethnicity and gender. This study has four components. We begin by revisiting the basic findings of the research literature with respect to the appearance of persons by their demographic characteristics. We then extend that analysis to look at the structure and content of the news story and how this intersects with race/ethnicity and gender. Third, we look at the portrayal of “major characters” and “authority figures” in the news stories. And finally, we examine the implications of these portrayals for the “social capital” manifest by the broadcasts.
From our review of the past television news research (Ehrlich and Weller, 2005), we generated the following guiding hypothesis:
1. The major characters are predominantly White, Anglo, and male.
2. Whites will disproportionately appear as major characters in more positive roles.
3. Black males are more likely to be depicted as major characters in criminal roles, especially as suspects or perpetrators of violent crimes.
4. Depictions of major characters as authority figures will follow the pattern: white, male, female, black.
5. Major characters and positive authorities are predominantly White, Anglo, males.
6. Crime and violence are the major content areas of news reports.
7. The primary function of TV news is to erode the social capital of the communities in its broadcast area.Design and Methodology
The data to be presented are based on two samples. The primary sample, reported here, comprises 13 newscasts drawn from 12 cities around the country. The cities and stations represented are: Boston (WBZ), Bryan, TX (KRHD), Dayton (WHIO), Denver (KUSA), Fairbanks (KTVF), Grand Rapids, MI (WOTV), Harrisburg (WHP, WGAL), Los Angeles (NBC**), Louisville (WLKY), San Francisco (KGO), Tucson (KOLD), and Wilkes Barre , PA (WBRE). Network affiliations included six stations allied with CBS, five with NBC, and two with ABC. The Fox network stations which were caught up in our sampling were excluded because they were hour-long newscasts and thus not directly comparable.
In each city during a three-week period in early spring of 2005, we recorded a randomly selected 30-minute, late news (10 or 11 p.m.) program. Each program was viewed and coded by two coders, occasionally by three. The basic unit of analysis was the discrete story, but total times were obtained for sports, weather, and self-promotional advertising. Table 1 presents the basic categories for observation and coding.
On the Air
From the opening logo and introduction to the closing banter of the anchors and typically a promo for the station’s early morning news show, the average news presentation occupies 14 minutes and 23 seconds. The range was from 11:30 (Fairbanks) to 17:40 in Bryan, Texas (located near Texas A & M University, in east Central Texas). In that time, news directors squeezed in an average of 16 stories. The total number of stories ranged from 10 to 23. The typical story was slightly under one minute, although one station in Harrisburg spun stories at half-a-minute each, while San Francisco allocated an average of one and one-quarter minutes to their stories.
These data are not surprising, appearing only slightly higher than in earlier studies with respect to the number of stories and story length. The number of local stories reported ranges across stations from 2 to 16 with a median of 7 stories. (A local story was operationally defined as one that reported a news event within the metropolitan area served by the station. The metro area typically covers a central city and its surrounding counties.)
Table 1. The Basic Categories of Observation —
Story Characteristics
Story Characteristics
• Story Name
• Story length
• Reported by: station reporter, network, independent, weather person, sportscaster, text, anchor
• Race, ethnicity, gender of reporters
• Story designation: news, feature, press release, product plug, lottery, stock market, indeterminate, other
• Location: city, metro area, state, national, international, other
• Interactivity: n/a, URL, address, contact
• News category: accident/aftermath, arts, civil unrest, civil trial/lawsuit, consumer, demonstration/protest, disaster/clean up, economic/business, school/education, environment, fire, international relations, health, immigration, labor, media, military/war/terrorism, obituary, prison/prison issues, police, poll, religion, science, sports, technology, transportation, weather as news, other, incomprehensible
• Entertainment: holiday/special event, celebrity, features, other
• Politics–location: city, county, state, federal, out-of-state, international
• Politics–events: campaign, legislature/legislation, appoint-ments, symbolic, scandal, other
• Crime: violent, personal, white collar, misdemeanors, bias crimes, property offense, criminal justice system, criminal trial, other
• Crime appearance: suspect, victim, convicted criminal, witness, suspect’s relative/friend, victim’s relative/friend, neighbor, police, judge/juror, prosecuting attorney, defense attorney, expert witness, pundit, otherStay Tuned
In this section we describe the content categories of the observed stories and their geographic location. We then examine the stories by their race/ethnic and gender populations.
The 335 news stories were coded into the 36 categories presented in Table 1. The leading categories were crime (15.8%), military/war/terrorism (9.6%), environment (7.5%), accidents and their aftermath (7.2%), health (6.6%), and education (5.1%). All other categories accounted for less than 5% each. The third quartile of stories ranged from campaign politics (4.5%), sports as news (3.9%), economics/business (3.0%), and police (3.0%). .
The location of these stories was not entirely expected. One-third were local and another one-third were national. The low percentage of local stories likely reflected the skill set of reporters and the added costs of enterprise reporting. The national stories, in contrast, were pretaped from the networks. Stories which took place outside the local area or specifically dealt with state issues or state political events amounted to 20%. International stories—those not coded as war or military—accounted for 9%. There were approximately 4% whose locations were undetermined.
Violence. There were 56 crime stories. Almost half of these were about violent crimes. Another quarter were stories of criminal trials. Many of those dealt with violent crimes but were coded as a trial when the dominant focus was on the trial itself. Property offenses, personal crimes, and white collar crimes each appeared 5% of the time. Since violent crimes typically account for 2% of all crimes nationally, it is clear that this coverage is strongly skewed.
The second most frequent category — military/war/terrorism — dealt exclusively with the Iraqi insurgency and terrorist incidents mainly in the Middle East. Like the crime stories their focus was on violent acts.
Introducing the Major Characters
Not everyone who appears on camera is necessary for the story being told. In fact, most appearances are fleeting or accidental. The people caught on camera are basically being used by the reporter or camera person to fill out a story. Of course, there are characters who are the story or are otherwise vital to the story. We established three criteria for defining a major character: (1) S/he appears for seven or more seconds in the story; (2) s/he appears in more than two screen shots; or (3) s/he is the main focus or subject of the story. The basic counts appear in Table 2.
As this table clearly shows, Latinos and Asians are virtually invisible. White women are three times less likely than men to appear as a major character but twice as likely as black women. In sharp contrast, three-fourths of the major characters are White and seven out of ten are male. Our basic hypotheses are clearly confirmed.
Next we examine the stories in which the major characters are displayed (Table 3). As can be seen, almost half the stories (44%) deal with crime. While it is apparent from this table that blacks are more likely than whites to be a major character in a crime story, (37% versus 12%) the differences are more dramatic when you look at the focus of the story. Our observational protocol allowed for the dual coding of crime stories (see Table 1). These stories were coded by type of crime and by criminal appearance. The latter refers to the “reason” for the character’s appearance in the story. Violent crimes and criminal trials account for three-fourths of the crime stories. We observed that the color of the major characters strongly influences their appearance. Black characters are mainly shown as convicted criminals or as suspects in a crime. White characters tend to be victims, witnesses, legal workers such as lawyers, prosecutors, and the like.
Other differences between groups seem less apparent. However, one substantial difference in the category “military/war/terrorism,” while unexpected, may likely reflect the fact that the major characters selected for interviews or shown at press briefings were officers. The absence of black characters may have thus reflected the mutual level of discrimination, that is, mutual between the news reporter or editor and the military. (Reporters who were “embedded” were often limited in their coverage and whom they could speak with by their military supervisors.)
The time allocated major characters also reflects the implicit priorities of the reporter. Overall, major characters who were black appeared for 280 seconds, summing across all stations. In sharp contrast, whites captured 2,252 seconds of air time.
Searching for Authority
Depending upon deadline, the size of the newshole, and the easy availability of a person who is knowledgeable about the event being covered, reporters may craft their story by turning to an authority figure. This occurred approximately one-fourth of the time. That is, one out every four major characters were interviewed because they had some specific authority on the matter being reported. We coded four dimensions of authority: giving reassurance, providing an analysis, presenting information, or asserting an oppositional analysis. In addition, we coded authority figure utterances which we agreed were vacuous and without meaning.
Most authority figures were highlighted to present information, 54%, while 15% gave reassurance to the viewers. Paradoxically, given the expert status of the figures, only 14% were called on to provide an analysis of the event being covered. Half of those, 7%, asserted an analysis that could be labeled as critical or in opposition to the dominant interpretation being screened. Finally, another 17% fulfilled the stereotype of the TV pundit, namely by making a set of essentially vacuous comments.
The race and gender of these authorities was as expected, predominantly white and predominantly male. Their distribution is presented in Table 4. Eighty-six percent of the authority figures are white, and seven out of ten are male. The selection of persons for expert analysis is a more voluntaristic decision than interviewing, for example, victims, witnesses, or others directly involved in the event being reported. And that being the case, women, black and other ethnic authorities are either too obscure or scarce in the population, or the selection of white men as experts reflects an unwitting discrimination on the part of the reporter and producer.
Social Capital: Where It All Counts
Following the coding of a local news story, judges were required to rate the story on a five-point scale identifying its value along a continuum of social capital. The instructions for a “very negative” story read: This story portrays a social problem which may be construed as a malignant property of the city (for example, violent crime, lead poisoning, air pollution, political corruption, police abuse, bankrupt schools, disruptive city council meetings). At the other pole, the story displayed a very positive or celebratory aspect of the city: (for example, coverage of the arts festival or city book fair, opening of the farmers’ market, Mayor cuts the ribbon at a new wing of the aquarium). Stories were assigned to the intermediate point if their affective tone was basically neutral or indifferent.
There is a core meaning to the concept of “social capital.” Social capital is viewed as a resource of communities and is based on trust in institutions and individuals, social participation, cooperation and mutual aid. Social capital derives from the social norms and institutional processes that enable people to trust one another and to work together cooperatively, providing mutual aid as well as in the formation of new groups and social arrangements.
The central theorem of this developing theory is that the greater the social capital of a community the lower the social pathologies to be found in that community. There is a growing body of evidence to support this generalization (See the preceding chapter, but also Brehm and Rahn, 1997; Portes, 1998; Paxton, 1998; Lederman, Loayza, and Menéndez, 2002; Messner and Baumer, 2004.).
There were 113 local stories in our sample, amounting to one-third of the stories covered. Categorizing the stories by their polarity: 55% were negative, 32% were neutral and 13% were positive. Table 5 displays the distribution of stories by their assessed social capital. Notable in Table 5 is the dramatic difference between the frequency of the very negative and very positive stories, approximately 20% versus 2%.
There was considerable variation by station with three stations having no positive stories and an additional nine stations with only one positive social capital story. Negative stories predominated. At one station more than half its stories were negative. Over all, the reporting of stories impacting on a locality’s social capital appeared at a ratio of three negative stories to one positive story.
Gender and race also play a role in the social capital of the newscast. As Table 6 indicates, white women are featured in the positive social capital stories. Their mean rating is 4.4 on a five-point scale. The other differences by race and gender are quite small, show only minor variation, and tend towards the negative.
When we compare the local and nonlocal news stories by length, we observe only an insignificant five-second difference. However, when we compare their social capital ratings, we observe a robust difference. Negative stories are presented, on the average, four times longer (4:06 minutes) than positive stories. (See Table 8.)
There is little variation in the categories of stories that are local and which potentially impact on the community’s social capital. Of the 36 story categories (see Table 1), crime stories dominate. They accounted for 28 per cent of all social capital-relevant stories. These mainly negative stories were followed in frequency by entertainment stories, mainly features of no strong polarity (5.6 per cent), and by political events (4.7 per cent).
We turn now to the function of authority in the presentation of social capital. Table 7 presents the six authority roles we scored by the directionality of the social capital of the story. It is clear, yet surprising, that authority is used to support the negative stories. It is as if the story required additional support, especially corroborative information, to bolster its negativity. The mode in the positive stories, “gives reassurance,” is also surprising since one might have presumed that the audience was more likely to need reassurance in the negative events. On the other hand, it may be that it is the giving of reassurance that helps define the story as positive. Nevertheless, what does stand out in Table 8 is the overwhelming appearance of authority figures in the stories which have negative value.
Discussion and Conclusion
A randomly selected late evening half-hour newscast was recorded in 12 cities during the early spring of 2005. At a descriptive level, the study was designed to observe the content of the reported news and the race/ethnic and gender composition of the major characters in the newscast. To this end, two observers timed and coded approximately 40 variables in 210 stories with a total of 259 major characters.
Seven hypotheses extrapolated from past studies guided the research. All were confirmed.
• The major characters, operationally defined in the text, were predominately non-Hispanic white males. The sample included cities with substantial Latino and Asian-Pacific populations, and their absence from the newscast could not be attributed to a demographic bias.
• Whites appeared disproportionately as major characters and in more positive roles.
• Stories which dealt with crime, and violent crimes in particular, were more likely to have Black males as their major characters.
• Those major characters who were also portrayed as positive authority figures followed the predicted pattern: white, male, female, black, with whites appearing in more than eight out of ten instances.
• The most frequent story content was crime and violence. A substantial proportion of the local stories were negative. Generally, there were four minutes of negative stories to each minute of positive stories.
Two serendipitous findings, consistent with the predicted results, involved, first, the unique absence of black characters from stories about the ongoing war in Iraq. (This is highlighted by Pentagon estimates that 16-20 per cent of recruits are black.) The second finding, more precisely, set of findings, regards our unique entry into media research—the role of authority in news stories. With regard to the roles of an authority figure, we observed that they were used mainly to provide information or to give reassurance. Further, in only ten percent of their appearances did authority figures provide an opposing perspective, that is, another side to the story being reported.
The signal findings of this study are those which catalog the polarities of social capital as they are presented in the news stories. We argued that a story contributes to the positive social capital to the degree to which it is relevant to the well-being of the community and is egalitarian in its depiction of people by their social characteristics and event-specific statuses. It subtracts from the social capital to the degree to which the news stories presented are not relevant to the well-being of the community, tend to downplay its social problems, and are discriminatory in their selection of who appears on camera and the manner in which they are depicted. Of course, this study does not measure the direct impact of these stories on social capital. Our primary assumption, however, is that these stories represent a potential deposit or withdrawal from the community’s balance sheet. The discriminatory treatment of characters by their race/ethnicity and gender, the focus on black crime and violence, the emphasis on white males as major characters and as positive authority figures, and the outstanding focus on stories which have negative implications for the community and its citizens—all of these particulars lead to the conclusion that the social capital of the community is being eroded by the routine operation of the news media.
Bibliography
Brehm, John, & Rahn, Wendy (1997). Individual-level evidence for the causes and consequences of social capital. American Journal of Political Science, 41, 1999.
Carroll, Joseph (2004). Public: Detroit, Washington, Los Angeles most “unsafe” cities: at least 7 in 10 say Minneapolis, Seattle, Dallas are safe. Retrieved November 11, 2004, from the Gallup Organization Web site: http://www.gallup.com
The Gallup Organization. (2004). Confidence in Institutions. Retrieved June 4, 2004, from The Gallup Organization Web site: http://www.gallup.com
Lederman, Daniel, Loayza, Norman, & Menéndez, Ana Maria (2002). Violent crimes: does social capital matter? Economic Development and Cultural Change, 50, 509-539.
McPherson, Miller, Smith-Lovin, Lynn, & Brashears, Matthew E. (2006). Social isolation in America: changes in core discussion networks over two decades. American Sociological Review, 71, 353-375.
Messner, Steven F., Baumer, Eric P., & Rosenfield, Richard (2004). Dimensions of social capital and rates of criminal homicide. American Sociological Review, 69, 882-903.
Paxton, Pamela (1999). Is social capital declining in the United States? A multiple indicator assessment. American Journal of Sociology, 105, 88-127.Portes, Alejandro (1998). Social capital: its origins and applications in modern sociology. Annual Reviews, 24, 1-24.
Table 2. Major Characters by Race/Ethnicity, and Gender
Male Female White n 144 48 192 % 55.6 18.5 74.1 Black n 36 23 59 % 13.9 8.9 22.8 Other n 4 4 8 % 2.1 1.0 3.1 Total n 184 75 259 % 71.0 29.0 100.0 Note: “Other” includes 1 Latino, 4 Asian, and 3 unidentified.
Table 3. Race of Major Character and the Content of News Stories
Story Category Race of Major Character White Black n % n % Crime 23 11.9 22 37.3 Military/war/terrorism 21 10.9 0 -- Environment 31 16.1 2 3.4 Accidents and their aftermath 17 8.8 11 18.6 Health 13 6.7 0 -- Education 3 1.6 6 10.2 All other stories 85 44.0 18 30.5
Table 4. Authority Figures by Race and Gender
Race Gender Total Male Female n % White n 22 9 31 86.1 Black n 4 1 5 13.9 Total n 26 10 36 100.0
Table 5. Distribution of Social Capital Ratings
Social Capital Ratings n % Very negative 21 19.6 Somewhat negative 38 35.5 Neutral 34 31.8 Somewhat positive 12 11.2 Very positive 2 1.9 Total 107 100.0
Table 6. Mean Rating of Social Capital by Gender and Race of Major Characters
Race and Gender n Mean 39 2.3 10 4.4 22 2.0 20 2.1
Table 7. Role of Authority in Social Capital-Relevant Stories
Authority Provides: Positive Stories Neutral Negative Stories 7 0 2 0 0 4 1 4 13 0 1 3 0 0 2 0 0 1 n 8 5 25 % 21.0 13.0 66.0
Table 8. The Late Evening News --- Number and Polarity of Stories
City Total News Stories News Time Local Stories Social Capital Positive Neutral Negative n Time n Time n Time 13 14:47 5 1 2:24 1 0:22 3 5:19 17 17:40 4 1 0:52 0 0 3 2:27 19 15:28 16 2 1:06 3 3:32 11 10:08 12 14:26 4 1 0:25 1 1:00 2 4:33 10 11:30 4 1 2:09 2 1:25 1 1:44 16 15:01 7 1 0:28 2 0:42 4 6:05 18 15:10 2 1 0:54 0 0 1 0:35 23 12:06 11 5 3:39 2 2:17 4 1:14 15 13:12 9 1 0:18 6 5:03 2 1:08 22 13:33 9 1 0:27 0 0 8 6:30 13 16:48 7 0 0 2 0:56 5 7:09 12 13:26 5 0 0 3 4:30 2 1:07 19 13:54 3 0 0 0 0 3 5:14 Means 16 14:23 7 1 0:58 2 1:31 4 4:06
Who We Are . Projects . Newsletter . Publications . Guest Book . Speakers . Internships . Factsheets . Links . Current Issues . HOME