This paper was presented at the Association for Education in Journalism and
Mass Communication in San Antonio, Texas August 2005.
If you have questions about this paper, please contact the author
directly. If you have questions about the archives, email
rakyat [ at ] eparker.org. For an explanation of the subject line,
send email to
[log in to unmask] with just the four words, "get help info aejmc," in the
body (drop the "").
(Feb 2006)
Thank you.
Elliott Parker
====================================================================
Telestrator Terrorism: Fear Messages in the
Television News Coverage of Hurricane Ivan
A paper submitted to the RTVJ division
of the Association for Education in Journalism and Mass Communication
Nancy McKenzie Dupont
Associate Professor, Department of Communications
Loyola University New Orleans
Box 201, 6363 St. Charles Avenue
New Orleans, Louisiana 70118
(504) 865-3293
[log in to unmask]
Mary Blue
Associate Professor, Department of Communications
Loyola University New Orleans
Box 201, 6363 St. Charles Avenue
New Orleans, Louisiana 70118
(504) 865-3433
[log in to unmask]
Running Head: Ivan Fear
ABSTRACT
This paper examines the role fear-inducing messages in the coverage
of Hurricane Ivan in the New Orleans market. Content analysis and
personal interviews were used to gather data.
The research results find that fear-inducing message were common and
not limited to one kind of message-originator nor one television
station, which is surprising given that meteorologists admitted that
they knew the hurricane would not make a direct hit on New Orleans.
When dangerous Hurricane Ivan entered the Gulf of Mexico and began
moving northward as a category four storm, the four New Orleans
television news stations went into high gear. From Monday, September
13 through Wednesday, September 15, the stations went to
extraordinary lengths to inform the public about Hurricane Ivan's
predicted path. They interrupted programming, broadcast news
conferences live, and dispatched reporters throughout the viewing
area; some of the stations called on their sports anchors to report
the breaking story. All of this effort was undertaken even though
television meteorologists believed Hurricane Ivan would not hit the
metropolitan area (Arredondo, 2004, Katz, 2004, Thomas, 2004, and
Wicker 2004).
The number of people watching television during this coverage was
huge; at one point, Nielsen estimated that 50 percent of all homes in
New Orleans were watching television (Walker, September 20, 2004, p.
5). To the casual but reasonable observer of the coverage, the
number of fear-inducing messages would appear to be significant. For
example, there was repeated live coverage of public officials
pleading with citizens to leave the area coupled with predictions of
heavy rain and high winds that could cause widespread damage. In
addition, there were frequent reports of clogged traffic with cars
stuck for hours. At one point, WVUE chief meteorologist Bob Breck
suggested evacuees should wait to leave their homes until the traffic
situation eased; his suggestion prompted a litany of complaints again
him and the station from public officials (Walker, September 20, 2004, p. 1).
This study examines the following questions: First, did New Orleans
television stations use fear-inducing messages to capture and keep
the attention of the audience during three days of extraordinary
coverage? Secondly, did fear-inducing messages originate with public
officials more than from television station personnel
(meteorologists, anchors, reporters, etc.)? And finally, were there
any trends in the transmission of fear-inducing messages, such as a
difference among stations or in time of day that should be noted? In
this paper, the researchers are attempting to answer these questions
through content analysis.
The 2004 Atlantic Hurricane Season
The National Weather Service's Climate Prediction Center made its
final forecast for the 2004 Atlantic hurricane season on August 10,
and the news was not good. Using scientific probabilities and
meteorological jargon, forecasters outlined a 45 percent chance of an
above-average season for residents of the Gulf and South Atlantic
coastlines. The conclusions were based on data gathered from
sea-surface temperatures in the Atlantic (which were warmer than
normal), wind shear in the tropics (which was lower than normal) and
the likelihood of a developing El Nino weather phenomenon (National
Oceanic and Atmospheric Administration, August 10, 2004). In
non-scientific terms, that meant that the government was predicting
12-15 tropical storms with 6-8 becoming hurricanes and 2-3 becoming
major hurricanes, which are categories 3-5 on the Saffir-Simpson scale.
Residents of New Orleans are accustomed to such predictions, as they
are to massive amounts of media coverage of hurricane precautions and
dangers. Because of the major hurricane strikes near New Orleans in
1965 (Hurricane Betsy), 1969 (Hurricane Camille), 1992 (Hurricane
Andrew), and 1998 (Hurricane Georges), residents of southeast
Louisiana pick up free hurricane tracking maps, read special reports
on hurricane predictions in the New Orleans Times-Picayune and watch
special programs on hurricane preparedness on the four television
news channels every year. New Orleans television stations in
particular take severe weather seriously: There are eleven weather
reporters appearing regularly on television, all classified as
meteorologists by education and/or training.
Less than a month after National Oceanic and Atmospheric
Administration (NOAA)'s last season forecast, however, it became
clear to scientists and television meteorologists that the 2004
season was worse than predicted. Eight disturbances reached tropical
storm status in August alone, breaking the record of seven that were
observed during August 1933 and August 1995 (National Weather
Service, September 1, 2004). The most intense of the storms was
Hurricane Charley, which moved ashore on August 13 near Cayo Casta on
the Florida Gulf Coast with maximum winds of 145 miles per hour, and
it was strong enough to maintain hurricane intensity across as it
crossed the Florida peninsula west to east. It reemerged in the
Atlantic and made two landfalls in South Carolina before it merged
with a frontal zone. Charley caused an estimated 13-15 million
dollars in damage and was responsible for ten deaths (National
Weather Service, September 1, 2004).
In addition to its destruction, Hurricane Charley had another impact
on the southeastern coastline. Forecasters were expecting the storm
to hit the Tampa-St. Petersburg area, but shortly before landfall it
intensified from a category two to a category four storm and took a
hard right turn. Hurricane Charley hit 70 miles south of Tampa, and
people in its path were caught off guard. Meteorologists at the
National Hurricane Center blamed the media, which they said focused
too much on the evacuations of Tampa and St. Petersburg and too
little on the entire hurricane warning area, which extended south to
the Florida Keys (Los Angeles Times, August 15, 2004).
Ivan is Born
Hurricane Ivan had it origins in the classical tropical cyclone
model, forming off the coast of Africa near Cape Verde and gathering
strength quickly as it moved east to west across the Atlantic
Ocean. It formed on August 31, became a tropical depression two days
later, a tropical storm a day after that, a hurricane early on
September 5 and a major hurricane later that same day (National
Hurricane Center, October 1, 2004). It caused major damage in
Grenada, Jamaica, the Cayman Islands, and Cuba. It entered the Gulf
of Mexico through the Yucatan Channel as a category four hurricane;
it had been a category five three previous times (National Weather Service).
As Ivan moved northward in the Gulf of Mexico, the National
Hurricane Center predicted consistently that its point of impact
would be somewhere around the Alabama-Mississippi state line. But
the New Orleans metropolitan area was covered in the Center's
forecast "cone," and by the time hurricane warnings were issued at
4:00 pm on Tuesday, September 14, New Orleans Mayor Ray Nagin had
held two news conferences, had ordered a state of emergency, and was
predicting 8-12 inches of rainfall in the city (Schleifstein, 2004).
New Orleans television stations (CBS affiliate WWL-TV, NBC affiliate
WDSU-TV, Fox affiliate WVUE-TV, and ABC affiliate WGNO-TV) began
preempting programming early in the day on Tuesday to cover weather
predictions, evacuation orders, traffic conditions, and public
official news conferences. By 9:00 am Wednesday, September 15, all
four television news stations left network programming to cover
Hurricane Ivan exclusively until 7:00 pm when three stations began a
gradual return to regular programming. WDSU-TV, however, stayed with
hurricane coverage throughout prime time even after it was determined
that Ivan would not seriously affect the city of New Orleans (Walker,
September 20, 2004, p. 5).
The payoff for the stations was huge. On Monday evening, the second
half hour of WWL-TV's Nightwatch newscast got a 33.4 local
rating. WDSU-TV was number one in prime time ratings Wednesday night
when it stayed with live hurricane coverage rather than return to
network programming (Walker, September 20, 2004, p. 5). And WVUE-TV
reported that its aggressive coverage of New Orleans's evacuation
problems had received more positive feedback than any other story in
the station's history (Walker, September 20, 2004, p. 1).
The Meteorologists Remember
Two months after the coverage of Hurricane Ivan, on November 17,
2004, four of the city's eleven television meteorologists were
questioned on their recollections of the coverage. Those
participating in the debriefing were Carl Arredondo, chief
meteorologist at WWL-TV, Dan Thomas, meteorologist at WDSU-TV,
Crystal Wicker, morning meteorologist at WVUE-TV, and Bruce Katz,
chief meteorologist at WGNO-TV.
All said they believed all along that New Orleans would not have a
direct hit from Hurricane Ivan but they did not want to say that on
the air. "We had the prediction correct, but all along we were
supporting local officials saying it's time to get out of town
because science is inexact, " said Thomas. He said he told a friend
off camera to not evacuate, but on camera "…you err on the side of
caution" (Thomas, 2004). But both Arredondo and Katz said they told
family members to leave town. The intensity of the storm may have
been a factor; "I think with Ivan, too, the thing that got us
alerted by it was that it was a category five in the Gulf, and we all
know what hurricanes can do, so in a situation like that you have to
lean towards an air of caution," said Wicker (Wicker, 2004).
Journalists or quasi-journalists reporting on the possibility that
something may happen when they don't believe it will happen could
lead to charges of providing extensive coverage for the sake of
getting ratings. Such a scenario could deteriorate into providing
fear for the sake of revenue. Three of the four meteorologists said
they are aware that weather forecasting can be a ratings booster and
they have, at times, felt the pressure to perform. "We all realize
that when you have a hurricane warning, when you are wall-to-wall,
people are looking at every station," Katz said. "In the back of
your mind you know viewership is higher" (Katz, 2004). "Every news
director, every manager is trying to hype the news department so if
you watch the promos, you're looking at weather promos, especially in
this market," said Wicker (Wicker, 2004). But all agreed that hyping
the weather means walking a fine line. "I think over time if you
exaggerate, if you sensationalize, it'll come back and bite
you. They'll say, 'He was wrong, and the storm before that he was
wrong' and you'll have the reputation of 'That guy doesn't know what
he's talking about'" (Katz, 2004).
The only real public evaluation of the weather coverage came days
later when Times-Picayune media critic Dave Walker called WWL-TV's
morning meteorologist David Bernard "the star of the storm," faulting
all others with "telestrator terrorism" (Walker, September 20, 2004,
p. 1). "Local weathercasters….used their on-screen graphics to
repeatedly scrawl arrows, lines, and squiggles pointed directly at
the city of New Orleans, even though that was never the National
Hurricane Center's track" (Walker, September 20, 2004, p. 1).
Literature Review
Barry Glassner's 1999 book The Culture of Fear: Why Americans Are
Afraid of the Wrong Things was a popular examination of a topic
scholars had been studying for years and asked the question, "Does
modern American society create and perpetuate fears that are not
real?" Glassner held public officials and television news culpable
for creating artificial fears, and he was particularly critical of
television news magazines which compete with entertainment for prime
time audiences:
Producers of TV Newsmagazines routinely let emotional accounts trump
objective information. In 1994 medical authorities attempted to cut
short the brouhaha over flesh-eating bacteria by publicizing the fact
that Americans are fifty-five times more likely to be struck by
lightening than die of the suddenly celebrated microbe. TV
journalists brushed this fact aside with remarks like 'whatever the
statistics, it's devastating to the victims'" (Glassner, 1999).
In other similar cases, Glassner demonstrates how alarm grows today
much as it did during and following the 1938 "War of the Worlds"
broadcast when radio listeners believed the earth was being attacked
by Martians. Then, as now, journalists expressed alarm, "experts"
commented, and anecdotal evidence was valued above scientific
evidence. The result was that people tuned in to their fears and
ignored the obvious explanation that the story was being told for
dramatic effect.
In Creating Fear: News and the Construction of Crisis, social
scientist David Altheide agrees with Glassner that America is
experiencing a postjournalism period in which audiences have come to
expect entertainment from their news. As a result, the art of
interviewing has changed dramatically. No longer do journalists seek
interviews or "sound bites" for the purpose of obtaining information;
such "sound bites" would run too long on the air and would take away
from the visual and dramatic elements of the story. "Sound bites"
have become quick answers to questions and sources of emotions in
journalistic story-telling, and people who want to present their
message on television have learned to adapt to the formula (Altheide, 2002).
Issues, says Altheide, become linked with emotion, which are
psychologically linked with fears. Following the theory of symbolic
interactionism, the topic becomes the signifier and fear becomes the
signified. In one quick step, the entire culture becomes involved in
the fear mechanism. He argues that private life and public issues
have melded into popular culture, making fear messages more common in
the mass media. His data support his view: fear messages have
increased in his sampling of the mass media from 1992 to 2002 (Altheide, 2002).
A television news audience is not privy to the judgments used to
select and present stories, but they may have expectations. In All
the News That's Fit to Sell: How the Market Transforms Information
Into News, James T. Hamilton asserts that audiences have clues about
what they will see and hear when they tune into a television news
broadcast. "These expectations are the brand images consumers have
about these products," he writes (Hamilton, 2004, 126). Those
expectations are built by the news departments, who select and write
news stories based on demographics, he writes. To do otherwise might
be economic suicide because of the large cost involved with doing a
story. News must fit consumer tastes (Hamilton, 2004, 137).
Put together Altheide's data and Hamilton's observations and you
have a new guiding theory for television news research: if audiences
want fear messages, that may be one explanation of their
presence. The question of why the public would find value in fear
messages could open a new line of inquiry.
A comprehensive study of television weather has not been published in
a communication journal since 1982. Bogart (1968) found that
television weather reports were preferred by more than half of the
sample, and concluded "It must be the personality of the
weathercasters who make this mundane subject come to life." Tan
(1976) determined that television was used more often (53 percent of
the respondents) than any other medium to obtain weather information.
However, even though respondents used television more often, only 41
percent considered it their preferred source for weather. Hyatt et.
al. (1978) tested recall of television weather reports, and concluded
that the "amount of weather information retained from a forecast
seems to be minimal indeed." Gantz (1982) attempted a study of
accuracy, as well as redundancy, in weathercasts in Indianapolis. The
data from this research suggested that forecasts frequently vary from
station to station. Since weather technology, budgets and personnel
have changed significantly since 1982, these studies are of little value.
While no previous research could be located that specifically
discussed the use of fear in television weather forecasting,
countless studies have been conducted on the use of fear appeals
since the early 1950s. Articles on the use of fear most relevant to
the current study can be placed in two categories: the use of fear
in persuasive messages such as public speeches and advertising, and
studies of the use of television news images that induce fear in viewers.
Studies of Fear Appeals in Persuasive Messages. Most of the research
on the study of the use of fear in persuasion grows out of an
experiment conducted by Janis and Fishbach in 1953. They used three
different intensities of fear appeals in a communication on dental
hygiene. Results indicated that though all three forms were equally
effective in teaching factual content, the greatest conformity was to
the position advocated in the "minimum" threat communication. The
"moderate" threat communication was somewhat less effective. The
"strong" fear appeal was the least effective for producing conformity
to the recommended actions. (Janis and Fishbach 1953)
As a result of this study, most subsequent research has assumed the
principle that the greater the threat, the less opinion change in the
direction advocated. However, extensive research in this area has
demonstrated that the relationship between fear arousal and
compliance is complex. Some studies have suggested variables that
have an effect on this basic relationship.
Higher levels of fear can be used when the message is presented by a
highly credible source. (Boster and Mongeau 1984). A higher level
of fear can also be used when the message provides reasonable
solutions to prevent the fearful consequences. According to research
in advertising, a receiver asks two questions when evaluating a fear
appeal: 1) How effective are the recommendations in preventing the
fearful outcome? And 2) Am I capable of enacting the
recommendations? So a persuader can use a higher level of fear
appeal if the audience is provided with a reasonable solution for the
prevention of the fearful consequences.
Rogers' (1983) Theory of Protection Motivation argues that a fearful
response is dependent on two factors: the perceived magnitude of the
noxiousness of the threat and the perceived probability of being
effected by the threat. When evaluations of these two dimensions are
high, a greater fearful reaction occurs which triggers a greater
motivation to protect oneself.
Another condition influencing the reception of a fear appeal is
whether or not the risk information is specific. The receiver is
likely to process and remember more specific information. So
concrete and explicit risks to the receiver are more likely to have
an impact on the receiver's recall and consequent attitudes, beliefs,
and behaviors than are vague, abstract, or uncertain
risks. Therefore, a higher level of fear may be used when specific
risk information is also included.
Finally, fear appeals are often short-lived. Researchers have found
that the effects of fear appeals begin to dissipate within
twenty-four hours, and one study reports that fear levels decreased
only ten minutes after exposure to a fearful message. For this
reason, it may be necessary to use a high level, rather than a
moderate level of fear, assuming that a residual level of fear may be
sustained to influence the receiver. One researcher suggested that
if one desires any enduring change in beliefs or behaviors, receivers
must be periodically reexposed to the fearful stimuli. (Smith 1982)
Advertising can make audiences fearful of a variety of situations –
from contracting AIDS to smelling bad. Many types of fears are
evoked in ads, including physical, social and psychological. Studies
have shown that audiences are more likely to remember ads that use
fear appeals than those using positive, upbeat appeals. (Hyman and
Tansey 1990) However, again, the research shows that high levels of
fear can cause receivers to feel helpless and without control over
the situation, thus decreasing attention and effectiveness of the message.
Studies of Fear-Inducing Television News Images. Recently, research
in the use of fear has moved into the area of television news
coverage. These studies are different from the earlier research in
at least one significant way. They do not examine the fear present in
a message as a "fear appeal," which assumes an intent to persuade on
the part of the originator of the message. Instead, the researchers
discuss messages as "fear-inducing," which places the responsibility
for the perception of fear on the audience. A television news story
may cause a viewer to be fearful even though the news reporter was
not specifically attempting to persuade the audience by using fear as
an appeal.
Young (2003) examined the implications of fear to the perceived
importance of news stories. Subjects who were put in the role of a
news editor were asked to evaluate news clips for an evening news
broadcast. Results indicated that the more noxious and the more
likely to have a personal impact, the more likely a story was
considered important.
Newhagen (1998) found that television news images that induced anger
were the most likely to be remembered by subjects, but news images
that induced fear were significantly likely to be remembered as well.
Most studies of fear-inducing messages in television news centered
on the effect of reporting crime stories on the level of fear of
crime on viewers. For example, Lowry, Nio, and Leitner (2003) found
that network television news variables account for almost four times
the amount of variance in people's perceptions of crime as the most
important problem facing the country than did the actual crime
rate. They also found the sheer amount of time devoted to crime
coverage was much more influential than the number of crime stories;
story rank was unimportant. Gross and Aday (2003) found that by
emphasizing certain issues such as crime, local television news has
the effect of leading audiences to think about those issues more than
others. And a study by Romer, Jamison, and Aday (2003) showed that
across a wide spectrum of the population and independent of local
crime rates, viewing local television news is related to increased
fear of and concern about crime.
The studies discussed here can provide guidance for the current
research. In addition, they demonstrate the need for research on the
fear-inducing content of television weather reports, particularly
during catastrophic weather events such as hurricanes.
Hypotheses and Method
The researchers examined the role of fear and fear-inducing messages
in the coverage of Hurricane Ivan in the New Orleans
market. Specifically, the goal was to determine if fear-inducing
messages were more common among a certain type of message originator
(meteorologists, anchors, reporters, public officials, and the
general public), if fear-inducing messages were more common on one
station over another, and if fear-inducing message increased or
decreased as the coverage went forward.
Four 30-minute periods were recorded on Wednesday, September
15: 12:00-12:30 pm, 1:00-1:30 pm, 5:00-5:30 pm, and 10:00-10:30
pm. All four stations were recorded during the periods. The
researchers selected these time periods because they provided a
day-long sample, and all but one were at regular news times (Noon,
5:00, 10:00 pm). The 1:00-1:30 pm time was selected because the
broadcasters had heavily promoted that new hurricane coordinates
would be forthcoming from the National Hurricane Center at the top of
the hour.
Each researcher coded the results according to the following
variables: Station, Time period, Fear-inducing level, and Message
origin. A message was defined as a complete story from the anchor
lead-in of an element until the completion of that element. In
general, the unit of measure began again every time the coverage
returned to an anchor on set or changed anchors on set. For example,
one unit equaled anchor to live report and back to anchor, anchor to
package and back to anchor, anchor toss to meteorologist and back to
anchor, and an anchor conducting an interview until the end of the
interview. The exception of this was the few instances in which the
control of the information flow left the journalist (anchor or
reporter) and was turned over to someone else, as in the case of a
lengthy live interview or live coverage of a press conference. In
those cases, the live remarks were coded as separate message units.
Each researcher coded 143 cases independently. Fourteen of those
cases were selected using a random number generator to test for
intercoder reliability. The test produced an agreement rate of 85
percent (i.e. disagreement in two of 14 cases). Neither of the cases
produced a disagreement of more than one category. The researchers
then entered the agreed-upon data and negotiated an agreement in
those few cases of disagreement.
Three of the variables were self-explanatory, but the Fear-inducing
message variable required definitions:
Factual (no fear level): A message was defined as factual if it
contained mostly objective information. Examples are coordinates
from the National Hurricane Center, evacuation orders issued by
governments, school and work closings, preparation tips, shelter
listings, traffic conditions, evacuation routes, and curfew
orders. Facts that are in and of themselves fear-inducing
(e.g.: flash flood warnings), were coded as having some level of fear.
Low fear level: A message was defined as low fear level if it
contained low fear-inducing information and speculation.
Medium fear level: A message was defined as medium fear level if it
contained some fear-inducing information and speculation that did not
directly apply to the television coverage area (e.g.: high
winds/waves on the Alabama or Mississippi coastlines).
High fear level: A message was defined as high fear level if it
contained little or no objective information and was made up mostly
of speculation/danger warnings pertaining to the immediate coverage
area. Examples are telestrator maps showing impact at New Orleans,
or showing the hurricane turning towards New Orleans, pleas for
immediate evacuation, visuals of deteriorating weather conditions,
speculation (as opposed to scientific prediction) of flooding and
high winds in the city, and speculation on what will happen if the
storm veers off of its predicted path.
Since the coverage of Hurricane Ivan was so extensive (on Wednesday,
September 15 alone, the coverage extended from early morning until at
least 7:00 pm), and since the meteorologists interviewed agreed that
they believed the hurricane would not come ashore at New Orleans, the
researchers expected the following: Fear-inducing messages would be
present in a significant number, lower fear levels originating with
meteorologists, and lower fear levels later in the day as the
hurricane approached its predicted target (and New Orleans moved into
a low probability for landfall). Casual sampling of the coverage had
produced anecdotal information about the high number of dire
predictions of hurricane damage from public officials; the
researchers wanted to test significance in that area. In addition,
the researchers were curious about whether there were different
levels of fear-inducing messages among the four news
stations. Five hypotheses are offered:
H1: The frequency of fear-inducing messages will be larger than the
frequency of factual messages.
H2: The frequency of fear-inducing messages will be larger among
public officials than among broadcast professionals (meteorologists,
anchors, and reporters).
H3: Fear-inducing messages will be lower among meteorologists than
among all other message originators.
H4: The number of fear-inducing messages from members of the public
who are interviewed by reporters will be significant.
H5: The frequency of fear-inducing messages will be larger on the
station that brands itself "Your Weather Authority" (WVUE).
H6: The frequency of fear-inducing messages will decrease throughout
the day as the threat to New Orleans lessens.
All of the data collected and examined was nominal. The results
were analyzed with SPSS using a combination of frequency tables and
chi-square computations.
Results
H1: The frequency of fear-inducing messages will be larger than the
frequency of factual messages.
H1 was tested using a frequency table with no-fear messages, compared
to the three categories for fear-inducing messages (low fear level,
medium fear level and high fear level).
Fear
Frequency
Percent
Valid Percent
Cumulative Percent
Valid
No fear
52
36.4
36.4
36.4
Fear
91
63.6
63.6
100.0
Total
143
100.0
100.0
No-fear messages appeared at a frequency of 36.4 percent, compared
with fear-inducing messages at a frequency of 63.6 percent, therefore
H1 is supported.
H2: The frequency of fear-inducing messages will be larger among
public officials than among broadcast professionals (meteorologists,
anchors, and reporters).
H2 was tested using a chi-square computation of fear-inducing
messages by message origin: broadcast professionals (meteorologists,
anchors, and reporters) and public officials.
Fear * Origin Crosstabulation
Origin
Total
B'cast pros
Public officials
Fear
No fear
Count
46
6
52
Expected Count
43.9
8.1
52.0
Low fear
Count
30
11
41
Expected Count
34.6
6.4
41.0
Medium fear
Count
33
3
36
Expected Count
30.4
5.6
36.0
High fear
Count
11
2
13
Expected Count
11.0
2.0
13.0
Total
Count
120
22
142
Expected Count
120.0
22.0
142.0
Chi-square=6.055, df=3, significance=.109
Broadcast professionals were more likely than public officials to
produce fear-inducing messages. Therefore, H2 is not supported.
H3: Fear-inducing messages will be lower among meteorologists than
among all other message-originators. H3 was tested with a chi-square
computation of fear level compared to message origin, with
meteorologists compared against all others (all other categories were
collapsed creating meteorologists and all others).
Fear * Origin Crosstabulation
Origin
Total
Meteorologists
All others
Fear
No fear
Count
9
43
52
Expected Count
11.6
40.4
52.0
Low fear
Count
10
31
41
Expected Count
9.2
31.8
41.0
Medium fear
Count
10
27
37
Expected Count
8.3
28.7
37.0
High fear
Count
3
10
13
Expected Count
2.9
10.1
13.0
Total
Count
32
111
143
Expected Count
32.0
111.0
143.0
Chi-square=1.329, df=3, significance=.722
The cell distributions in this crosstabulation are similar to those
that would be expected by chance alone. H3 is not supported.
H4: The number of fear-inducing messages from members of the public
who are interviewed by reporters will be significant.
This test measured messages from members of the public compared with
all other message originators.
Fear * Origin Crosstabulation
Origin
Total
All others
Members of public
Fear
No fear
Count
52
0
52
Expected Count
51.6
.4
52.0
Low fear
Count
41
0
41
Expected Count
40.7
.3
41.0
Medium fear
Count
36
1
37
Expected Count
36.7
.3
37.0
High fear
Count
13
0
13
Expected Count
12.9
.1
13.0
Total
Count
142
1
143
Expected Count
142.0
1.0
143.0
Chi-square=2.885, df=3, significance=.410
The comparison of message from members of the public and messages
from all other message originators shows no significant
difference. Therefore H4 is not supported.
H5: The frequency of fear-inducing messages will be larger on the
station that brands itself "Your Weather Authority" (WVUE). H4 was
tested using a chi-square computation comparing the stations to two
fear-level categories, factual and low fear-inducing and high
fear-inducing combined.
Fear * Station Crosstabulation
Station
Total
WVUE
All others
Fear
No fear
Count
14
38
52
Expected Count
12.0
40.0
52.0
Low fear
Count
8
33
41
Expected Count
9.5
31.5
41.0
Medium fear
Count
5
32
37
Expected Count
8.5
28.5
37.0
High fear
Count
6
7
13
Expected Count
3.0
10.0
13.0
Total
Count
33
110
143
Expected Count
33.0
110.0
143.0
Chi-square=6.523, df=3, significance=.08
There is a low significance in this computation, therefore H5 is not
supported.
H6: The frequency of fear-inducing messages will decrease throughout
the day as the threat to New Orleans lessens.
H6 tested time periods compared to fear-inducing messages.
Fear * Time Crosstabulation
Time
Total
Noon
1pm
5pm
10pm
Fear
No fear
Count
12
11
13
16
52
Expected Count
12.7
11.3
14.2
13.8
52.0
Low fear
Count
10
8
12
11
41
Expected Count
10.0
8.9
11.2
10.9
41.0
Medium fear
Count
10
8
9
10
37
Expected Count
9.1
8.0
10.1
9.8
37.0
High fear
Count
3
4
5
1
13
Expected Count
3.2
2.8
3.5
3.5
13.0
Total
Count
35
31
39
38
143
Expected Count
35.0
31.0
39.0
38.0
143.0
Chi-square=3.707, df=9, significance=.930
There is no significant difference among the stations as to the
number of fear-inducing message presented, therefore H6 is not supported.
Discussion/Conclusions
Early on September 16, Hurricane Ivan came ashore near Gulf Shores,
Alabama, as a category 3 storm. Southeastern Louisiana was spared
the brunt of the storm, and suffered only 7.9 million dollars in
damage compared to the 8 billion dollars in damage to Alabama and
Florida (Yerton, 2004). Louisiana residents declared themselves lucky.
Most would agree that fear of catastrophic hurricanes is the
appropriate emotion for those living in coastal areas. The fact that
New Orleans residents are told that they are statistically overdue
for a major hurricane strike compounds that level of fear. In this
study, the researchers are not suggesting that an absence of fear
would be expected during hurricane coverage; however, the foremost
goal of any news broadcast should be to inform the public and not
frighten the audience unnecessarily. But in today's marketplace, the
quest for high ratings may overtake the goal of providing objective
information and inducing fear may be an easy way to capture and keep
an audience.
The research results indicated that the highest number of
fear-inducing messages during the television coverage came from
reporters. This could be explained by the fact that reporters are
traditionally doing live reports from the heart of the activity,
providing scenes that are inherently frightening. If they find
themselves in areas of low storm activity, they move either of their
own volition or as directed by the assignment desk. It is logical
that scenes would appear frightening to viewers when they are
provided by reporters who have every reason to be frightened themselves.
With reporters in harm's way, the people they interview should
produce a high level of fear on their own, yet this research did not
bear that out. This unsupported hypothesis might suggest even
further a hyping of the information by broadcast professionals as
opposed to the general public. However, the number of messages
originated from the public at large was small in comparison to other
message-originators. A different time period, particularly earlier
time periods before Hurricane Ivan got close to land, could have
produced an entirely different result in coverage of the same storm.
Anchors were more likely to present factual information than would
be expected by chance alone. In their duties of voicing over
graphics and edited video, they rely on facts gathered in the
newsroom and written for them to read over the air. They may have
less of an opportunity to ad lib than do other broadcasters, and that
may tend to make them less likely to speculate in a fear-inducing
manner. They are also the traditional question-askers putting them
in a position of searching for facts rather than succumbing to emotion.
Surprisingly, meteorologists were no more likely to present a no
fear message than any other message originator, including other
broadcasters, public officials, and members of the public. It should
be remembered that in post-Hurricane Ivan interviews, four
meteorologists who covered the storm said they always thought the New
Orleans area was not in any danger. If that is true, what was the
motivation to present fear-inducing messages? In addition, this
research shows that there was no major difference between the number
of fear-inducing messages produced early in the day as opposed to
later in the day when it was obvious the storm would hit
Alabama. Again, what was the motivation to present fear-inducing
messages when the viewing area was clearly out of danger? The answer
to both of those questions could be the quest for high
ratings. Another possible explanation could be that reporters,
photographers, meteorologists, and anchors, who have given their all
in covering the approach of the storm, may continue looking for any
slight change in the projected path and may cover the story longer
than necessary. Hurricane Charley was an example of a storm that
missed the target the media had set for it. On the other hand,
WDSU-TV was rewarded for its efforts to continue looking for a story
by some of the highest ratings of the two-day period.
This study indicates public officials were the source of more
fear-inducing messages than broadcast professionals. The question
about public officials was based on the researchers' observation the
day before that the television stations were broadcasting news
conferences live and that many of the public officials' statement
were frightening. Some of their live statements included information
about the availability of enough body bags and advice on taking an
axe into one's attic to escape rising water through the roof of one's
house. Although in the time periods studied, public officials
presented more factual information than fear-inducing messages, a
recording on the previous day might have produced vastly different results.
Finally, the number of fear-inducing messages originating from
members of the public during reporter interviews was
significant. This observation can be directly linked to the number
of fear-inducing messages from reporters. Their job is to be sent
into harm's way, and there is a high probability that they will
interview people on the scene who are either frightened or have
reason to communicate warnings to others. The likelihood is low that
someone living or working in a non-threatened area would be
interviewed during day-long hurricane coverage.
Recommendations for future study
The researchers hope to use this study as a guide for future
examinations of fear-inducing messages in television weather
forecasting. Suggestions for possible additions include the following:
1. "Fear-inducing" is a construct that needs to be further
defined. Should it just be a frightening image that is coded
fear-inducing, or should the threat to personal safety be a factor?
2. The coding method of messages could be restructured. For
example, if any part of the weather forecast had a fear-inducing
element, the entire message had to be coded as fear-inducing. During
this coverage, weather segments were lengthy with many messages
imbedded. Revising the definition of a message could provide more
information.
3. Providing viewers with live "wall-to-wall" coverage is not only
expensive in terms of salaries and overtime for personnel, few
commercials are run – causing a revenue shortfall, and national
spots are not aired since network programming is preempted. The
factors station managers use to determine when to "go live" and an
approximate cost of that decision would provide information important
to the research.
4. Even in a metered market, the accuracy of ratings during
catastrophic weather events is questionable. Many families with
meters many be ordered to evacuate, others may decide to leave on
their own volition. In addition, electrical power outages and cable
television outages occur almost immediately. On September 16, 2004,
Nielsen did not report ratings for New Orleans because less than 350
meters were in operation. How does this factor effect a station's
decision to stay on the air with live coverage?
In all potentially life-threatening weather events, accurate
objective information is essential to the public interest.
Fear-inducing messages may capture and keep the attention of the
audience during these events, but again, the foremost goal of any
news broadcast should be to inform the public and not frighten the
audience unnecessarily.
Works Cited
Aftermath of a Hurricane; Forecasts Still Works in Progress, Los
Angeles Times, August 15, 2004, p. A18.
Altheide, D. (2002) Creating Fear: News and the Construction of
Crisis. New York: Aldine de Gruyter.
Boster, F. J., and Mongeau, P. Fear-arousing persuasive messages. In
Communication yearbook 8. R. N. Bostrom and B. H. Westley, eds.
330-375. Beveral Hills, CA: Sage Publications, 1984.
Gantz, W. (1982). Redundancy and accuracy of television station
weather reports. Journalism Quarterly, 59(3), 440-446.
Hyatt, D., Riley, K. & Sederstrom, N. (1978). Recall of television
weather reports. Journalism Quarterly, 55(2), 306-310.
Hyman, M. and Tansey, R. The Ethics of Psychoactive Ads, Journal of
Business Ethics, 1990, 9 (2): 108
Glassner, B. (1999) The Culture of Fear: Why Americans Are Afraid
of the Wrong Things. New York: Basic Books.
Gross, K. and Aday S. The Scary World in Your Living Room and
Neighborhood: Using Local Broadcast News, Neighborhood Crime Rates,
and Personal Experience to Test Agenda Setting and Cultivation
Journal of Communication, 1 September 2003, vol. 53, no. 3, pp. 411-426
Hamilton, J.T. (2004) All the News That's Fit to Sell: How the
Market Transforms Information Into News. Princeton, N.J.: Princeton
University Press.
Janis, I. L. and Feshbach, S., Effects of Fear-Arousing
Communications. Journal of Abnormal Psychology, 1953, 48, 78-92
Lowry D.T.; Nio T.C.J.; and Leitner, D.W. Setting the Public Fear
Agenda: A Longitudinal Analysis of Network TV Crime Reporting, Public
Perceptions of Crime, and FBI Crime Statistics Journal of
Communication, 1 March 2003, vol. 53, no. 1, pp. 61-73
National Oceanic and Atmospheric Administration, August 2004 Update
to Atlantic Hurricane Season Outlook, August 10, 2004 (US Government document).
National Weather Service, Tropical Weather Summary, September 1, 2004
(US Government document).
National Weather Service, Tropical Weather Summary, October 1, 2004
(US Government document).
Newhagen, J. E., TV News Images That Induce Anger, Fear, and Disgust:
Effects on Approach-Avoidance and Memory. Journal of Broadcasting &
Electronic Media, Spring 1998, Vol. 42, Issue 2
Rogers, R. W. (1983).Cognitive and Physiological Processes in Fear
Appeals and Attitude Change: A revised theory of protection
motivation. In J. T . Cacioppo and R. E. Petty (Eds.), Social
Psychophysiology: A Sourcebook. (pp. 153-176) New York: Guilford.
Romer, D.; Jamieson, K.H.; and Aday, S. Television News and the
Cultivation of Fear of Crime. Journal of Communication, 1 March
2003, vol. 53, no. 1, pp. 88-104.
Schleifstein, M. Throngs Flee Ivan, Times-Picayune (New Orleans),
September 15, 2004, p. 1.
Smith, M. J., (1982). Persuasion and human action. New York: Wadsworth.
Walker, D. Anchors away, clichés galore, Times-Picayune (New
Orleans), September 20, 2004, p. 5.
Walker, D. Star of the Storm, Times-Picayune (New Orleans),
September 20, 2004, p. 1.
Yerton, S. Insurers Thankful for Louisiana's luck; State Escaped
Worst Hurricane Effects, Times-Picayune (New Orleans), November 18,
2004, p. 1.
Young J.R. The Role of Fear in Agenda Setting by Television News.
American Behavioral Scientist, August 2003, vol. 46, no. 12, pp. 1673-1695.
|
