Surveying the Impact · vie, a genre infam-ous for weak plots and artistic license yet spec-tacular...

08

7482064645

11>

November 2004Volume 46 Number 9$5.00 U.S.$6.50 Canada

W H E R E S C I E N C E A N D P O L I C Y M E E T

®

ALSO . . .

SUVs:Perceptionversusperformance

Stormwateron the Web

Surveyingthe ImpactThe Day After Tomorrow

Surveyingthe ImpactThe Day After Tomorrow

Before and After TheA U.S.A U.S.

Before and After The

©T

WE

NT

IET

H C

EN

TU

RY

FO

X

On Memorial Day weekend 2004, Twentieth Century Fox releasedThe Day After Tomorrow, a disaster movie depicting an abruptand catastrophic climate change. In the movie, a global warming–

induced shutdown of the North Atlantic thermohaline circulation system1

triggers extreme weather events worldwide and subsequently a new iceage, with wrenching global consequences. Before it even hit the the-

aters, however, the movie generated an intense storm of media con-troversy as scientists, politicians, advocacy groups, and political

pundits debated the scientific accuracy and political implicationsof the movie and global climate change.

Numerous predictions were made as to how the movie wouldinfluence risk perceptions and attitudes of the U.S. public

toward global warming. Some commentators feared that thecatastrophic plotline of The Day After Tomorrow would be

so extreme that the public would subsequently dismiss theentire issue of global warming as fantasy. At the other

end of the spectrum, others spun a scenario in which,panicked by the movie, the U.S. public would

force Congress to pass climate change legisla-tion, President George W. Bush would subse-

quently veto the bill, and challenger JohnKerry would exploit public hysteria over

global warming to win the U.S. presi-dential election. Some predicted

the film would do more to raisepublic awareness of global

warming than any numberof scientific papers or

documentaries, whileothers opined that

the film would

by Anthony A. Leiserowitz

Day After Tomorrow. Study of Climate Change Risk Perception. Study of Climate Change Risk PerceptionDay After Tomorrow

have no impact at all.2 The Day AfterTomorrow went on to become one of themost commercially successful movies ofall time, grossing nearly half a billiondollars worldwide in a little more than amonth.3 But what impact did the moviehave on U.S. risk perceptions, behavioralintentions, and political preferences?

To answer this question, a nationalstudy was conducted to explore the pub-lic impact of The Day After Tomorrow.The study included two nationally repre-sentative surveys of the U.S. public. Thefirst survey was implemented a weekbefore the movie’s release and the sec-ond was done four weeks later, after themovie had played in theaters for threefull weekends. The second survey alsooversampled movie “watchers” to allowcomparative analysis with “nonwatch-ers.” Finally, a media content analysiswas conducted with two purposes inmind: to determine the quantity and

quality of news coverage about themovie and to compare The Day AfterTomorrow with two other recent contro-versial films and two real-world newsevents. The research tested the hypothe-sis that representations of risk in popularculture can have a powerful influence onpublic risk perceptions—in some casesmore powerful than official risk commu-nications from scientists, governmentofficials, or special interest groups.4

Anecdotal evidence suggests that thisinfluence can, at times, be profound. Forexample, the film Jaws (1975) is thoughtby many to have greatly amplified pub-lic risk perceptions of shark attacks. Thevivid imagery and theme music fromthis movie still reverberate in the publicmind, stoking individual fears, influ-encing behavior (such as vacation andswimming preferences), and generatingcountless secondary ripple effects,including re-emergent, media-driven

“shark panics” such as was seen in the United States in the summer of2001. Likewise, the dramatic portrayalof a nuclear accident in The China Syn-drome (1979), combined with the sub-sequent real-world accident at ThreeMile Island, arguably shaped the publicdebate about the safety of nuclearpower. This synergism of fiction andreality may have greatly amplified theperceived risk of nuclear power, withripple effects that still reverberate inpublic opinion and fundamentally con-strain the industry today. According toresearchers Roger and Jeanne Kasper-son, although these events resulted in“no fatalities, [they] shut down nuclearplants worldwide, cost billions of dol-lars, and eroded public confidence innuclear power and (perhaps) other hightechnologies, industry and regulatoryinstitutions.”5 Despite these intriguingexamples, however, and even though

24 ENVIRONMENT NOVEMBER 2004

©T

WE

NT

IET

H C

EN

TU

RY

FO

X

the massmedia are

widely rec-ognized as

having a largeimpact on oth-

er public atti-tudes and behav-

iors, almost noresearch has ex-

plored the role ofmotion pictures in

public risk perceptionand behavior.6

Public Risk Perceptionsof Global Climate Change

Global climate change has beendescribed as a prototypical exam-

ple of a class of “hidden hazards”—risks that, despite their serious conse-

quences for society, generally passunnoticed or unheeded until they reachdisaster proportions.7 Public opinionpolls and academic studies consistentlyshow that Americans regard climatechange as a relatively low national pri-ority, despite decades of scientific warn-ings.8 For example, in a 2000 Galluppoll, the environment ranked sixteenthon Americans’ list of the most importantproblems facing the country, while cli-mate change ranked twelfth out of thir-teen environmental issues, just belowurban sprawl.9 Further, climate changeis commonly understood as a geograph-ically and temporally distant concern.Climate change is often described as aglobal problem, with particularly severeconsequences for marginalized peopleand places (such as small island or poortropical countries)—not for the UnitedStates itself.10 The Day After Tomorrow,however, depicted disasters spawned byclimate change and impacting present-day New York City, Los Angeles, Wash-ington, DC, and other global centers ofeconomic, political, and cultural power.The film thus had the potential to signif-icantly alter American risk perceptionsof the likelihood and severity of climatechange in the United States and to shift

public conceptions of climate changefrom a linear warming trend to abrupt,nonlinear, and catastrophic change.

The Movie’s Plot

The Day After Tomorrow tells thestory of National Oceanic and Atmos-pheric Administration (NOAA) paleo-climatologist Jack Hall (played by Dennis Quaid), who survives the disin-tegration of a massive ice shelf breakingoff Antarctica and returns to warn theworld about the possibility of an abruptclimate change due to global warming.A few weeks after presenting his theoryto a world climate conference, scientistsmonitoring the North Atlantic thermo-haline circulation system discover thatthe system is rapidly shutting down.Using his paleoclimatic computermodel, Hall forecasts that “the world ison the verge of a major climate shift”and attempts to warn the U.S. vicepresident. His warnings, however, gounheeded. Meanwhile, extreme weatherevents begin to occur throughout theworld, with grapefruit-sized hail inTokyo, tornadoes destroying downtownLos Angeles, and ultimately, a stormsurge–driven tidal wave drowning Man-hattan. Based on his model projections,Hall determines that three massive,hurricane-like supercells will formacross the Northern Hemisphere andrapidly pull sub-zero air from the uppertroposphere down to the ground, quick-freezing everything in their path andleading to the onset of a new ice age.Dr. Hall is called to brief the U.S. presi-dent, draws an east-west line throughthe center of the United States, and rec-ommends that all people south of theline be evacuated to Mexico, which ulti-mately opens its borders after the U.S.president forgives all Latin U.S. debt.Meanwhile, Hall’s teenage son (playedby Jake Gyllenhaal), in Manhattan for ascholastic decathlon, survives the tidalwave and takes shelter with his friendsin the New York Public Library, wherethey resort to burning books to keepwarm as the ice age begins. The rest of

the movie fol-lows Dr. Hall,who must braveAntarctic condi-tions as he treksto New York torescue his son.

Movie Reviews

The Day AfterTomorrow sparkeda heated nationaldebate about the sci-entific accuracy andpolitical implicationsof the film and thebroader issue of glob-al warming. The sci-ence underlying thefilm was criticized bymany climatologistsand other scientists,who were dismayed bysome of the main ele-ments of the movie. In particular, some of the scientists complainedabout the physical impos-sibility of a “quick-freeze”or a storm surge–driventidal wave hundreds offeet tall and the fact that athermohaline circulationshutdown would neitherhappen so quickly nor havesuch far-reaching conse-quences.11 Other scientists,however, used the film andthe controversy surroundingit as a “teachable moment”—an opportunity to not only cri-tique the film but to more con-structively educate the publicabout climate change.12 Like-wise, the political implicationsof the movie were debated byvarious pundits, ranging fromArianna Huffington to RushLimbaugh. Finally, a number ofenvironmental and political advo-cacy groups organized to greetmoviegoers with leaflets and peti-

VOLUME 46 NUMBER 9 ENVIRONMENT 25


tions,w h i l e

o t h e r sworked the

media to al-ternately hail or

decry the messageand politics of the

film.13

Movie critics alsogreeted the film with

widely divergent reviews.Some critics held the movie

up against the standards of finetheater and subsequently blasted

it for a weak plot, hokey situations,and pervasive use of standard movie

clichés. Others approached the moviewith the standards of the scientific doc-umentary and found, despairingly, thatwhile it had some elements of truth, italso included numerous scientific distor-tions and outright fabrications. Other

critics,however,

approachedthe film ex-

pecting a Hol-lywood block-

buster disaster mo-vie, a genre infam-

ous for weak plots andartistic license yet spec-

tacular visual effects. Thesecritics often found themselves

pleasantly surprised at howmuch better the film was than

many past disaster movies. In theend, the only thing most critics could

agree on was the outstanding quality ofthe special effects.

The general public, however, over-whelmingly liked The Day After Tomor-row. By mid-July, the film had grossedmore than $183 million in the UnitedStates alone and an additional $335 mil-lion overseas, for a total of more thanhalf a billion dollars.14 In the UnitedStates, an estimated 30 million ticketswere sold. The study reported herefound that 70 percent of adult moviego-ers rated the movie as good or excellent,18 percent rated it as average, and only13 percent as poor or terrible.

In responding to the critics, themoviemakers, including director RolandEmmerich, screenwriter Jeffrey Nach-manoff, and producer Gordon Smith,repeatedly pointed out that their primarygoal was to create a “popcorn movie”—a summer thriller that would draw amass audience. However, the filmmak-ers also admitted to having the sec-ondary goals of raising public con-sciousness and concern about globalwarming. These divergent goals—ofmass entertainment, education, andpolitical pressure—coexist in an uneasy

tension within the film, with con-sequences (intended and unintend-

ed) that are hard to disentangle. Whatimpact, if any, did the film have on pub-lic risk perceptions and conceptual mod-els of climate change? Did the filmmake moviegoers more or less willing totake personal actions to reduce their owngreenhouse gas emissions? Did itchange their political priorities or vot-ing intentions?

Risk Perceptions: Moviegoersversus Nonwatchers

To answer these questions, a represen-tative survey (n = 529) of the U.S. adultpopulation was conducted after themovie had played in theaters for threeweekends.15 The survey questionnairemeasured public climate change riskperceptions, conceptual models, behav-ioral intentions, and political prefer-ences. As of mid-June, approximately21 million U.S. adults had seen The Day After Tomorrow. Demographically,moviegoers were more likely to be male(57 percent), 18 to 29 years old (38 per-cent), Hispanic (26 percent), and politi-cally liberal (31 percent) than nonwatch-ers, who were demographically identicalto the U.S. public as a whole.

Global warming risk perceptions weremeasured using broad questions aboutgeneral concern and worry and likeli-hood estimates of specific climatechange impacts. Moviegoers (“watch-ers”) were found to have significantlyhigher risk perceptions than “nonwatch-ers.” When asked, “How concerned areyou about global warming?” 83 percentof moviegoers said they were somewhator very concerned, compared to 72 per-cent of nonwatchers (see Table 1 on page27). Likewise, a higher proportion ofmoviegoers (40 percent) than nonwatch-ers (31 percent) said that they worryabout global warming “a fair amount” or“a great deal.” A separate question askedmoviegoers directly whether the moviehad made them more or less worriedabout global warming. Forty-nine per-cent said that it had made them some-

©T

WE

NT

IET

H C

EN

TU

RY

FO

X

what or much more worried, 42 percentsaid it had not changed their level ofworry, and only 1 percent said theybecame less worried. Again, some com-mentators had warned that The Day AfterTomorrow would so trivialize global cli-mate change that the public would sub-sequently dismiss the whole issue. Thisforecast was clearly incorrect.

Overall, watchers and nonwatchersdemonstrated high levels of concernabout global warming yet lower levelsof worry.16 While many Americans areconcerned about global warming, fewerof them actively worry about it. Thishelps to explain the seeming paradoxbetween public opinion surveys thatshow Americans expressing high con-cerns about the issue yet giving it lowpriority in either national or environ-mental issue rankings.17

This study also included a series ofquestions measuring public likelihoodassessments of various global-warmingimpacts on the United States (see Figure1 on this page). Again, across the board,moviegoers perceived global warmingas a greater threat than the rest of the

general public. More than 80 percent ofwatchers responded that global warmingis somewhat or very likely to producemore intense storms, hurricanes, andtornadoes over the next 50 years, versus72 percent of nonwatchers. Likewise,higher proportions of moviegoersbelieved that the flooding of majorcities, food shortages, and a decrease inliving standards are likely to happen inthe United States over the next 50 years.

Most telling, however, moviegoers weremuch more likely than nonwatchers tobelieve that global warming could leadto a shutdown of the Gulf Stream oceancurrent or a new ice age—two underly-ing premises of The Day After Tomor-row. Across the board, the movieappears to have had a strong influenceon watchers’ risk perceptions of globalwarming. To test this conclusion, multi-ple regression analyses were conducted


0

20

40

60

80

100

Per

cen

t

More intensestorms, hurricanes,tornadoes

Flooding ofmajor cities

Food shortages

Standards ofliving decrease

Gulf Streamshutdown

New ice age

Watchers

Nonwatchers

“In the United States, how likely do you think it is that each of the following will occur during the next 50 years due to global warming?”

NOTE: Nonwatchers weighted (n = 390), watchers weighted (n = 139).SOURCE: A. Leiserowitz.

Figure 1. Percent of watchers and nonwatchers who found each item somewhat or very likely.

.


to control for the possible influences ofgender, age, education, income, race,political identification, and political ide-ology on each result.18 Even after con-trolling for these demographic and polit-ical factors, however, watchers were stillsignificantly more likely than nonwatch-ers to perceive global warming as agreater risk.

Overall, these results show that TheDay After Tomorrow had a considerableimpact on the global-warming risk per-ceptions of those who saw the movie.Further, a majority of Americans, watch-ers and nonwatchers alike, currentlybelieve that global warming will have arange of important impacts on the Unit-ed States over the next 50 years—in par-ticular, more intense storms, hurricanes,and tornadoes. From a scientific stand-

point, by far the least likely of theseimpacts—the onset of a new ice age—iscurrently and correctly perceived asunlikely by a clear majority of Ameri-cans. The movie, however, does appearto have led a substantial minority of mo-viegoers (41 percent) to believe such anevent is likely to happen.

Conceptual Models of theGlobal Climate System

How do Americans conceptualize theglobal climate system? Do they viewglobal climate as stable and stronglyresilient to human interference, or dothey view the climate system asextremely sensitive and vulnerable toabrupt and catastrophic shifts? The story

line of The Day After Tomorrow wasbased not on a gradual, linear warmingbut rather an abrupt and catastrophic cli-mate change, greatly compressed intojust a few weeks. Did the movie shiftpublic conceptual models of how globalwarming and the climate system work?To answer this question, the survey pre-sented respondents with five differentand highly simplified models of the cli-mate system. Respondents were askedto pick the one that best reflected theircurrent understanding. The models wereprovided in graphic and textual form andcan be seen in Figure 2 on this page. Sig-nificantly, moviegoers were much morelikely (39 percent) than the nonwatchers(28 percent) to choose model (A),which depicts a threshold model of theclimate system. This model describes a

0

10

20

30

40

Per

cen

t

A

A B C D E

B C D E

Climate is stable within certain limits.If the changes are small, climate will return to equilibrium.If they are large, there will be abruptand catastrophicimpacts.

Climate is randomand unpredictable.We do not knowwhat will happen.

Climate is slow tochange. Global warming will gradually lead to dangerous impacts.

Climate shows a delicate balance.Small changes willhave abruptand catastrophicimpacts.

Climate is verystable. Global warming will have little to no impact.

Watchers

Nonwatchers

“Which one of the five pictures below best represents your understanding of how the climate system works?”


Figure 2. Conceptual models of watchers and nonwatchers, percent of respondents

systemthat is

resilient todisturbance

within certainlimits; however,

forcings beyondthese thresholds lead

to abrupt and cata-strophic impacts. This

was the model implied byThe Day After Tomorrow,

in which global warminggradually increased until it

reached a critical tipping point,causing the thermohaline circula-

tion system to collapse and climatechaos to ensue.

Interestingly, moviegoers were nomore likely than those who did not seethe movie to choose the most extrememodel (D), which depicts the climatesystem as extremely sensitive to humandisturbance. Thus, the catastrophicimpacts depicted by The Day AfterTomorrow did not lead moviegoers tosuddenly adopt an extreme model of cli-mate sensitivity. Further, the movieappears to have influenced moviegoersto reject the other extreme models: (E),which depicts a very stable system and(B), a totally random and unpredictablesystem. Overall, however, the randomand unpredictable model (B) is still pre-ferred by a large proportion of watchers(29 percent) and nonwatchers (34 per-cent) alike. This most likely reflects thecommon (mis)interpretation of climatechange using widespread cultural mod-els and the personal experience of dailyweather, well known for its unpre-dictability beyond short time horizons.19

Behavioral Intentions

The Day After Tomorrow had signif-icant impacts on public risk percep-tions and conceptual models of climatechange, but did it influence respon-dents’ willingness to undertake indi-vidual actions to address global warm-ing? The survey asked respondents,“How likely are you to do the follow-ing because of your concerns aboutglobal warming: Purchase a more fuel-efficient car? Join, donate money to orvolunteer with an organization work-ing on issues related to global warm-ing? Make your views on global warm-ing clear to politicians (by writingletters, telephoning, sending e-mails,signing petitions, etc.)? Talk to friendsand family about how to reduce or pre-vent global warming?” The first itemreflects willingness to take action toreduce one’s own emissions, while thenext three indicate willingness to takeactivist, political, or social action. Thelast item, the willingness to talk tofamily and friends about global warm-ing, is particularly important as itreflects issue salience. The moreimportant an issue is perceived to be,the more people talk about it, which inturn leads to an increase in perceivedissue importance, and so on, in a posi-tive feedback loop. This process iscommonly referred to as “word ofmouth” or “buzz” and is a critical ele-ment in social change. Moviegoerswere found to be much more likely toengage in all four behaviors thannonwatchers (see Figure 3 onpage 30). With the exceptionof respondants’ likelihoodto express their view-points to politicians,the difference be-tween watchersand non-watchers

remained statistically significant evenafter controlling for the influence ofdemographic and political variables.

National Policy Preferences and Politics

Did this increased willingness to takeindividual action also translate to thenational policy agenda? The surveyexamined whether The Day AfterTomorrow led moviegoers to elevate thepriority of global warming as a nationalor environmental issue or whether themovie influenced presidential votingpreferences. The survey asked respon-dents, “Here are some issues now beingdiscussed in Washington. Among these,which do you think should be the toppriority for Congress and the Presi-dent?” Among the general public, glob-al warming ranked tenth out of tennational issues and sixth out of nineenvironmental issues (see Table2 on page 30). Moviegoers,however, ranked globalwarming higher: eighthamong national is-sues and fifthamong envi-ronmentalissues. ©

TW

EN

TIE

TH

CE

NT

UR

Y F

OX


The differences between watchers and nonwatchers remained statisticallysignificant even after controlling fordemographic and political variables.Interestingly, moviegoers also rankeddamage to the ozone layer higher thannonwatchers, which may reflect thecommon conflation of global warmingwith ozone depletion among manymembers of the public.20

Finally, what impact did the moviehave on U.S. voter preferences? Manycommentators noted that the film castthe Bush administration in a relativelynegative light. For example, the actor(Kenneth Welsh) who played the U.S.vice president looked strikingly like cur-rent Vice President Dick Cheney andplayed the role of a global-warmingnaysayer. At one point in the movie,after his warnings are again brusquelyignored, the hero, Professor Hall, raiseshis voice in urgency: “Mr. Vice Presi-dent! If we don’t act now it’s going to betoo late!” At a later point, once the enor-mity of the climate shift has becomeapparent, the director of NOAA saysangrily to the vice president, “You didn’twant to hear about the science when itwould have made a difference.” But did


0

20

40

60

80

Per

cen

t

Purchase a morefuel-efficient car

Talk to friends or familyabout how to reduceor prevent global warming

Join, donate money,volunteer with global warming group

Express your viewson global warmingto politicians

Watchers

Nonwatchers

“How likely are you to do the following because of your concerns about global warming?”


Figure 3. Behavioral intentions of watchers and nonwatchers, percent somewhat or very likely.

these less-than-subtle characterizationshave an impact on public attitudestoward the Bush administration?

The survey measured this in twoways. First, respondents were askedhow much they trusted a number of dif-ferent groups—including the Bushadministration, the National Aeronau-tics and Space Administration (NASA),NOAA, the U.S. Environmental Protec-tion Agency (EPA), scientists, and envi-ronmental groups—to tell them thetruth about global warming (see Table 3on page 32). Moviegoers were morelikely to distrust the Bush administra-tion and more likely to trust scientistsand environmental groups than non-watchers were.

A second question asked respon-dents, “If the 2004 presidential elec-tion were held today, who would youvote for?” Moviegoers were less likelyto vote for George Bush and more like-ly to vote for John Kerry (see Figure 4on page 32). This difference betweenwatchers and nonwatchers remainedstatistically significant even after con-trolling for demographic and politicalvariables. Thus, it is likely that TheDay After Tomorrow did have animpact on voter preferences, if only onthose individuals who saw the movie.

As a whole, these results suggestthat popular movies can have a consid-erable influence on the risk percep-tions, conceptual models, behavioralintentions, policy preferences, andeven the voting intentions of themovie-going public. The Day AfterTomorrow, although hailed by somecritics and reviled by others, was wellreceived by the movie-going publicand became an enormous commercialsuccess. Individuals who saw The DayAfter Tomorrow were more likely toperceive global warming as a threat, tobe willing to act as consumers and cit-izens to mitigate this threat, and totranslate their heightened concern intopolitical action. Given these results,however, was there a measurable shiftin public opinion at the national level?

To answer this question, a represen-tative survey (n = 472) of the U.S. pub-

lic was conductedone week before therelease of the movie(14–23 May), measur-ing the same variablesdescribed above.21 Ac-ross the board, no differ-ences were found in U.S.risk perceptions, policypriorities, or behavioralor voting intentions. As awhole, Americans beforeand after The Day AfterTomorrow were no morelikely to be concerned or toworry about global warm-ing—or to believe that cli-mate change impacts weremore likely to occur. Theyalso were no more likely toprioritize global warming as anissue, take personal actions, orto vote differently.

How could this be? The an-swer is simply a matter of num-bers. Based on box office totalsand survey data, The Day AfterTomorrow was seen by approxi-mately 21 million adults aged 18and older—an enormous movie audi-ence. Yet this represents only 10 per-cent of the U.S. adult population, notenough to change public opinion as awhole. A movie, even one as commer-cially successful as The Day AfterTomorrow, is rarely viewed by a major-ity of all Americans.22

In addition, moviegoers did not uni-versally leave the theater transformedinto global warming alarmists ornaysayers. While the results reportedabove suggest that on balance, the filmtended to make people more concerned,not all people responded in the sameway or with the same intensity. This isa critical point almost completelymissed by many pundits, scientists, andcritics before the film was released.Many commentators treated the publicas a single, homogenous mass, like agreat herd about to be spooked into amass stampede, either toward climatechange alarmism or outright denial. Atruism of social science and public

©T

WE

NT

IET

H C

EN

TU

RY

FO

X



opinion research, however, is that the“U.S. public” is in fact many publics—a plurality of different groups and inter-pretive communities, each predisposedto attend to certain risks and issues andto discount or ignore others. Very fewevents have the power to move publicopinion en masse in the same direction;September 11 is an obvious exception.

Further, the mass media plays a crit-ical role in this process of risk amplifi-cation or attenuation. As mentionedabove, The Day After Tomorrowsparked a heated debate about the sci-ence and politics of global warming inthe U.S. press. This debate was con-ducted at all levels of the media, frombroadcast networks to local papers andInternet Web sites. But how many andwhat kind of news stories were generat-ed? A media content analysis wasundertaken to answer these questions.

Using Lexis-Nexis, media coverageof the film The Day After Tomorrowwas analyzed from 1 April through 30June 2004 (see Figure 5 on page 33).“The Day After Tomorrow” was usedas the search term, and analysis wasrestricted to substantive articles (movie

Figure 4. Presidential politics of watchers and nonwatchers, percent of respondents

listings were not included). Three lev-els of media were examined:

• national television and radio broad-casts, including national networksABC, CBS, NBC, FOX, PBS, and NPRand cable networks such as CNN andMSNBC;

• national newspapers such as TheNew York Times, The Washington Post,USA Today, and Los Angeles Times; and

• major metropolitan newspaperssuch as The Boston Globe, ChicagoSun-Times, The Denver Post, and TheSan Diego Union-Tribune.

News stories were coded into four categories:

• science stories that focused on theveracity of either the movie or globalwarming;

• political stories that focused on thepolitical implications of the movie;

• entertainment stories that inter-viewed the stars of the movie orfocused on the special effects; and

• movie reviews.These categories were not mutually

exclusive, as a number of news storiesdiscussed the scientific and politicaldimensions of the movie. Entertain-ment stories, however, were almostalways focused exclusively on theentertainment aspects of the film.

Overall, thesemedia sources gen-erated 151 substantivenews stories about themovie. Of these, 39 percentaddressed the science underlyingthe movie, 37 percent focused on thepolitics, 29 percent provided moviereviews, and only 23 percent were enter-tainment stories. The articles and broad-casts included numerous editorials,opinion pieces, interviews with leadingclimatologists, and debates between glob-al warming advocates and opponents.News stories also included coverage ofefforts by Al Gore, MoveOn.org, and envi-ronmental groups’ efforts to use the movieas a “teachable moment”; a leaked memofrom NASA administrators allegedly sti-fling comment on the movie from NASAscientists; and a Pentagon-commissionedreport on the geopolitical implications ofabrupt climate change.23 Many news sto-ries addressed the science underlying themovie and the broader political implica-tions in the midst of a presidential election.

Media controversy helped drive peo-ple to the theater. But how big was this

m e d i astorm, re-lative to otherrecent controver-sial films or real e-vents? To answer thisquestion, an analysis ofmedia coverage of two otherrecent movies was conducted forcomparison: Fahrenheit 9/11, MichaelMoore’s documentary on alleged tiesbetween the Bush family and SaudiArabia and the consequences of the IraqWar, and The Passion of the Christ, MelGibson’s controversial depiction of thelast days of Jesus Christ. News storiesgenerated about two real-world eventswere also analyzed to put media


0

500

1,000

1,500

2,000

2,500

3,000

TV and NPR

Newspapers

Total

Nu

mb

er o

f st

ori

es in

maj

or

U.S

.b

road

cast

med

ia a

nd

new

spap

ers

2001 IPCC Report(1/1–3/30/04)

The Day AfterTomorrow(3/28–6/28/04)

Fahrenheit 9/11(4/25–7/20/04)

The Passionof the Christ(12/25/03–3/25/04)

Abu Ghraib(4/28–7/20/04)

SOURCE: Lexis-Nexis.Films/Events

Figure 5. Media coverage

c o v -e r a g e

of TheDay After

Tomorrowin context:

the release ofthe 2001 Inter-

governmenta lPanel on Climate

Change (IPCC) syn-thesis report on cli-

mate change and theAbu Ghraib prison

scandal in Iraq.24

Some commentatorshad predicted that the film

would bring more publicattention to the issue of glob-

al warming than the publica-tion of most scientific articles,

reports, or congressional testi-monies, and this prediction appears

to have been correct. The Day AfterTomorrow generated more than 10

times the news coverage of the 2001

IPCC report—which summarizes thelatest international scientific consensuson the causes, consequences, and solu-tions to global climate change andserves as the scientific basis for inter-national negotiations (see Figure 5).However, while The Day After Tomor-row did generate media controversyand attract national attention, it paled incomparison to either Fahrenheit 9/11 orThe Passion of the Christ. MichaelMoore’s Fahrenheit 9/11 generatedthree times more news stories than TheDay After Tomorrow despite earningonly half as much at the box office inits first month of play. Likewise, ThePassion of the Christ generated nearlyfive times more news stories whileearning only about 60 percent more inits first month.

Dwarfing the coverage of all thesestories, however, was coverage of theAbu Ghraib prison scandal, which hadmore than 10 times the coverage of TheDay After Tomorrow. Note as well thatthis event, as important and shocking asit was, was still only a subtheme of themuch larger story about the Iraq war.

Conclusions

The Day After Tomorrow had a signif-icant impact on the climate change riskperceptions, conceptual models, behav-ioral intentions, policy priorities, andeven voting intentions of moviegoers.The film led moviegoers to have higherlevels of concern and worry about glob-al warming, to estimate various impactson the United States as more likely, andto shift their conceptual understandingof the climate system toward a thresholdmodel. Further, the movie encouragedwatchers to engage in personal, politi-cal, and social action to address climatechange and to elevate global warming asa national priority. Finally, the movieeven appears to have influenced voterpreferences. These results demonstratethat the representation of environmentalrisks in popular culture can influencepublic attitudes and behaviors.

Critically, however, this influence waslimited by the level of national expo-sure. Surveys conducted immediatelybefore The Day After Tomorrow wasreleased and three weekends afterward


©T

WE

NT

IET

H C

EN

TU

RY

FO

X

foundno shift

in broadpublic at-

titudes or inbehaviors .

More than 21million U.S.

adults went tosee The Day Af-

ter Tomorrow inthe theaters (and

millions more world-wide), making the

movie an enormouscommercial success.

Nonetheless, this repre-sents only 10 percent of

the U.S. adult population—not enoughto measurably shift public opinion as awhole. This percentage, however, willcertainly increase once the film isreleased on video and later broadcast onnational and international television.Will the movie thus influence a largerproportion of the U.S. public? It alsoremains to be seen whether the moviewill have the same influence overtime—in other words, was the influencetemporary? It is possible that theobserved shift in public perceptionsand behavioral intentions repre-sents a momentary blip—thatafter time, the experienceand imagery of the moviewill recede in publicmemory, along withheightened worryand concern a-bout globalw a r m -

ing. A follow-up studywill be conducted incoming months toaddress these crit-ical questions.

Addition-ally, a keycom-


More accessible than typical journals and much more timely than textbooks,Environment delivers solid analysis of environmental science and policy issues. Thispeer-reviewed magazine offers first-rate scholarship on complex topics in a formatthat is clear and easy to understand. Students, faculty, and policymakers rely onauthoritative and insightful articles from experts in the field and international perspectives. The magazine alsooffers in-depth reviews of governmental and institutional reports, pointers to the best environmental Web sites,book recommendations, and commentaries that put controversies in perspective.

Subscribe today! Monthly (Combined Jan./Feb., Jul./Aug.); ISSN 0013-9157

Individual Rate $48 • Institutional Rate $98Add $18 for postage outside the U.S.

Heldref Publications1319 Eighteenth Street, NW, Washington, DC 20036-1802 • Phone: (800) 365-9753 • Fax: (202) 293-6130

www.heldref.org

Libraries may order through subscription agents.

HELDREF PUBLICATIONS

Subscribe Todayand get 10 issues of Environment for only $48


ponent of the risk

ampl i f i ca -tion process is

media attention.Issue salience and

priority is, in no smallpart, driven by the sheer

number and repetition ofnews stories. This research sug-

gests that, relative to other newsstories, global warming is a rarely

reported issue. These results help contex-tualize and explain why global climatechange remains a relatively low nationaland even a low environmental priorityfor most Americans. Unfortunately,

without strong and concertedleadership from the local tointernational levels, it maytake a series of real-worldextreme events linked toclimate change to perma-nently raise the salienceand priority of globalwarming among the massmedia and the broaderU.S. public.

Anthony A. Leiserowitz is aresearch scientist at DecisionResearch, Inc. and an adjunctprofessor of Environmental Stud-ies at the University of Oregonin Eugene. His research focuseson environmental risk perception,decisionmaking, and behavior.Leiserowitz may be reached at(541) 485-2400 or by e-mail [email protected] authorthanks his colleagues at DecisionResearch, particularly Paul Slovicand research assistants PhilipSolomon Hart and C. K. Mertz, as

well as independent scholar Robert W.Kates, for their help, encouragement,

and constructive criticism. This paper isbased on research supported by a grant

from the National Science Foundation(SES 0435622).

NOTES

1. Thermohaline circulation refers to a system ofocean currents that distribute heat from the tropics

northward to the North Atlantic. These warm oceancurrents maintain a relatively warm, temperate climatein northern Europe. Recent paleoclimatology researchhas demonstrated that this system has sporadicallyflipped on and off, resulting in abrupt climate shifts, atleast at the regional scale. While a future thermohalineshutdown is currently considered a low-probabilityevent, growing scientific concern about the potentialfor abrupt climate change from this and other possibletriggers has led recently to a report by the U.S. Nation-al Research Council, a Pentagon-commissioned studyon the geopolitical implications, and widespread mediaattention. See National Research Council, Committeeon Abrupt Climate Change, Abrupt Climate Change:Inevitable Surprises (Washington, DC: National Acad-emy Press, 2002); P. Schwartz and D. Randall, AnAbrupt Climate Change Scenario and Its Implicationsfor United States National Security (Washington, DC:Institute for National Strategic Studies, 2003),http://www.ndu.edu/library/docs/pentagon%5Fclimate%5Fchange.pdf; and W. Steffen et al., “AbruptChanges: The Achilles’ Heels of the Earth System,”Environment, April 2004, 8–20.

2. S. Connor, “Warming Up: The Debate over aMovie That Claims to Be a Vision of the Future,” TheIndependent, 8 May 2004; G. Easterbrook, “Blast-Frozen Nonsense,” The New Republic Online, 10 May2004; A. Freedman, “Disaster Movie’s Focus on RapidChange Expected to Set Off Renewed Debate,” Green-wire, 5 April 2004; G. Lean, “How Rupert MurdochSaved the Planet (and Other Tall Stories),” The Inde-pendent, 16 May 2004; and P. J. Michaels, “ApocalypseSoon? No, but This Movie (and Democrats) HopeYou’ll Think So,” The Washington Post, 16 May 2004.

3. BoxOfficeMojo.com, The Day after Tomorrow(24 July 2004), available at http://www.boxofficemojo.com/movies/?id=dayaftertomorrow.htm.

4. This research also contributes to recent develop-ments in risk perception theory, including work on therole of affect and emotion in risk perception and TheSocial Amplification of Risk Framework, which “aimsto examine broadly, and in social and historical con-text, how risk and risk events interact with psycholog-ical, social, institutional, and cultural processes inways that amplify or attenuate risk perceptions andconcerns, and thereby shape risk behavior, influenceinstitutional processes, and affect risk consequences.”J. X. Kasperson et al., “The Social Amplification ofRisk: Assessing Fifteen Years of Research and Theory,”in N. Pidgeon, R. E. Kasperson, and P. Slovic, eds.,The Social Amplification of Risk (Cambridge, UK:University of Cambridge Press, 2003), 13–46; P.Slovic, The Perception of Risk (London: Earthscan,2000); and P. Slovic et al., “Risk as Analysis and Riskas Feelings: Some Thoughts About Affect, Reason,Risk, and Rationality,” Risk Analysis 24, no. 2 (2004):311–22.

5. R. E. Kasperson and J. X. Kasperson, “HiddenHazards,” in Deborah G. Mayo and Rachelle D. Hol-lander, eds., Acceptable Evidence: Science and Valuesin Risk Management (New York: Oxford UniversityPress, 1991), 9–28.

6. Very few studies have examined the impact ofpopular movies on risk perceptions. But see W. C.Adams et al., “Before and After ‘The Day After’: ANationwide Survey of a Movie’s Political Impact,”paper presented at the Annual Meeting of the Interna-tional Communication Association, San Francisco,CA, 27 May 1984; C. A. Anderson et al., “The Influ-ence of Media Violence on Youth,” Psychological Sci-ence in the Public Interest 4, no. 3 (2003): 81–110; andC. M. Bahk and K. Neuwirth, “Impact of Movie Depic-tions of Volcanic Disaster on Risk Perception andJudgments,” International Journal of Mass Emergen-cies and Disasters 18, no. 1 (2000): 65–84.

7. Kasperson and Kasperson, note 5 above.8. R. J. Bord, A. Fisher, and R. E. O’Connor, “Public

Perceptions of Global Warming: United States and Inter-national Perspectives,” Climate Research 11 (1998):75–84; R. E. Dunlap and R. Scarce, “The Polls-PollTrends: Environment Problems and Protection,” PublicOpinion Quarterly 55 (1991): 651–72; J. J. Houghton,G. J. Jenkins, and J. J. Ephraums, Climate Change: TheIPCC Scientific Assessment (Cambridge, UK, and NewYork: Cambridge University Press, 1990); Intergovern-mental Panel on Climate Change, Working Group 1, Cli-mate Change 1995: The Science of Climate Change.Summary for Policymakers and Technical Summary ofthe Working Group I Report (Cambridge, UK, and NewYork: Cambridge University Press, 1996); Intergovern-mental Panel on Climate Change. Working Group I, Cli-mate Change 2001: The Scientific Basis. Summary forPolicymakers (Cambridge, UK, and New York: Cam-bridge University Press, 2001); National Academy ofSciences, Energy and Climate: Studies in Geophysics(Washington, DC: National Academy of Sciences,1977); and “Editorial: Costs and Benefits of CarbonDioxide,” Nature, 3 May 1979, 1.

9. R. E. Dunlap and L. Saad, Only One in FourAmericans Are Anxious About the Environment,http://www.gallup.com/poll/releases/pr010416.asp.

10. A. Leiserowitz, “Global Warming in the Ameri-can Mind: The Roles of Affect, Imagery, and World-views in Risk Perception, Policy Preferences andBehavior,” PhD dissertation, University of Oregon,2003.

11. See for example, K. Davidson, “Film’s Tale ofIcy Disaster Leaves the Experts Cold,” The San Fran-cisco Chronicle, 1 June 2004; S. Palmer, “GlobalWarming: The Warm, Hard Facts,” The Register-Guard(Eugene, OR), 23 May 2004; D. Vergano and S.Bowles, “Killer Weather, or Not?” USA Today, 26 May

©T

WE

NT

IET

H C

EN

TU

RY

FO

X

2004; and A. J. Weaver and C. Hillaire-Marcel, “Glob-al Warming and the Next Ice Age,” Science, 16 April2004, 400–02.

12. Likewise, a number of educational Web siteswere created to separate movie fact from movie fiction.See for example, The Woods Hole OceanographicInstitution Web site, http://www.whoi.edu/institutes/occi/currenttopics/abruptclimate_dayafter.html; andthe Union of Concerned Scientists Web site,http://www.ucsusa.org/global_environment/global_warming/page.cfm?pageID=1405.

13. Groups advocating a viewpoint on the film includ-ed MoveOn.org, the Rainforest Action Network, theNatural Resources Defense Council, the Union of Con-cerned Scientists, Greenpeace, Future Forests, TheCato Institute, and the Competitive Enterprise Institute.

14. See BoxOfficeMojo.com, note 3 above.

15. The survey was implemented 15–27 June byInternet survey firm Knowledge Networks (KN), usingtheir online research panel, which is representative ofthe entire U.S. population (see http://www.knowledgenetworks.com/ganp/index.html for moreinformation about the KN methodology). The surveyalso included an oversample of 98 randomly selectedadults who had seen the movie, for a total of 139 movie“watchers.” The within-panel response rate was 74 per-cent. The combined sample was weighted to corre-spond with U.S. Census Bureau parameters and thedemographic profile of moviegoers. For more details,please contact the author.

16. This is an important distinction to bear in mindwhen interpreting public opinion data. While “con-cern” and “worry” are often used synonymously, theycan produce different results. One may have a generalconcern for an issue without actively worrying about it.Worry is a more active emotional state and as such is

arguably a stronger predictor of action and behavior.

17. For a synthesis of public opinion data on globalwarming, see Program on International Policy Atti-tudes (PIPA), Global Warming (PIPA, 2003), http://www.americans-world.org/digest/global_issues/global_warming/gw_summary.cfm.

18. Multiple regression (partial correlation) is a sta-tistical technique used to determine whether anobserved relationship between an explanatory variableand a dependent variable persists, even after theeffects of one or more additional explanatory variablesare removed. For example, there is an observed nega-tive relationship between height and hair length—thatis, short people have longer hair than tall people. Thismay seem odd at first, but if the explanatory variablegender is added to the regression equation, theobserved relationship disappears (because womentend to be shorter and have longer hair than men). Ifthe relationship between height and hair length per-sists even after the effect of gender is removed, thenone can be more confident that the observed correla-tion is not spurious (StatSoft, Inc., “Multiple Regres-sion,” http://www.statsoft.com/textbook/stmulreg.html#cunique).

19. See W. Kempton, “How the Public Views ClimateChange,” Environment, November 1997, 12–21.

20. See Kempton, ibid.; and W. Kempton, J. S. Boster,and J. A. Hartley, Environmental Values in AmericanCulture (Cambridge, MA: MIT Press, 1995).

21. This survey was implemented by KnowledgeNetworks from 14–23 May (see note 15 above).The questionnaire used in both national surveyswas identical, with several questions added tothe second, post-movie survey. The within-panel response rate for the first survey was71 percent.

22. The Day After Tomorrow, however, will likelyinfluence a larger audience once it is released on video(scheduled for 12 October 2004).

23. G. Mahone, “No NASA Role in Movie,” TheNew York Times, 1 May 2004; A. Revkin, “Glob-al Freezing? Do Tell, NASA Says,” The NewYork Times, 4 May 2004; and A. Revkin,“NASA Curbs Comments on Ice Age Dis-aster Movie,” The New York Times, 25April 2004. See also Schwartz andRandall, note 1 above.

24. News stories about eachfilm and event were collectedusing a three-month sam-pling frame that includedstories from two monthsprior and one monthafter each movierelease and thethree monthsafter eachreal-worldevent.


rates second in Japan and Germany after view-ers have watched the film, it performed poorly in the United States and—even stranger from a non-U.S. view—is not affected by the film: About 7 percent of watchers and nonwatchers chose it.

There is much more to comment and com-pare about the studies mentioned, and the par-ticipants of the Potsdam workshop agreed to unite forces to create such a comparison. For now it is worth noting that the impact studies of The Day After Tomorrow have entered a new, reflexive area of climate change research: the area of the impacts of impacts. Twentieth Century Fox Germany has established an initia-tive to facilitate emissions trading rights and reducing CO2 emissions of services, events, and traffic (see http://www.climatepartner.de). One might take it as image work, but it is also an indication that The Day After Tomorrow might not be the last of the global warming movies. Thus, it will be helpful for climate scientists to continue researching media and film representations of climate change and the public’s response to them. It is doubtful that the creators of the United Nations Framework Convention on Climate Change had Hollywood on their minds when they drafted Article 6, which asks for improved communication and education on the issue of climate change. But the entertainment industry seems to have done quite a lot for the public awareness of climate change, and Anthony Leiserowitz gave us a very useful look at this new domain of climate impact research.

Fritz ReusswigPotsdam Institute

for Climate Impact ResearchGermany

1. See http://www.boxofficemojo.com/movies/?page=main &id=dayaftertomorrow.htm. 2. A. Leiserowitz, “Before and After The Day After Tomor-row: A U.S. Study of Climate Risk Perception,” Environment, November 2004, 22–37. 3. F. Reusswig, J. Schwarzkopf, and P. Pohlenz, Double Impact. The Climate Blockbuster The Day After Tomorrow and its Impact on the German Cinema Public, Potsdam Institute for Climate Impact Research (PIK) Report No. 93 (Potsdam, Germany: PIK, 2004), http://www.pik-potsdam.de/publications/pik_reports. 4. M. Aoyagi-Usui,“The Day After Tomorrow: A Study on the Impact of A Global Warming Movie on the Japanese Public,” National Institute for Environmental Studies (NIES) Working Paper (unpublished), October 2004; T. Lowe et al., “Does Tomor-row Ever Come? Disaster Narrative and Public Perceptions of Climate Change,” Draft Tyndall Working Paper (unpublished), October 2004; and A. Balmford et al., “Hollywood, Climate Change, and the Public,” Science, 17 September 2004, 1713.

A s the article “Before and After The Day After Tomorrow” was going to press, I was very pleased to learn that some-

what similar studies had been conducted in the United Kingdom, Germany, and Japan. Thanks to the generous hospitality of Fritz Reusswig and the Potsdam Institute for Climate Impact Research, the primary investigators of all these studies gathered for a workshop in October 2004 to share our respective findings. This meeting was quite stimulating and led to the formation of an international research team to conduct cross-cultural experimental research.

I thank Reusswig for his comments on the paper and would like to take this opportunity to address his primary concern. We conducted three national surveys of the American pub-lic—before, during, and several months after the movie played in theaters. The article reported results from the first two waves, in particular the second, which compared a ran-domly selected group of movie watchers and nonwatchers from a national sample in June 2004—several weeks after the movie debuted. The first two surveys were not based on a within-subject (panel) design, so this study was unable to directly measure whether watch-ing the film changed an individual’s attitudes toward climate change. Thus Reusswig raises a legitimate question: Are the significant dif-ferences observed in the U.S. study between movie watchers and nonwatchers really due to the impact of the film, or did movie watchers already have “more pro-climate or pro- environment attitudes before entering the cine-ma”? In other words, perhaps moviegoers went to the film because they were already more concerned about global warming.

Three streams of convergent evidence sug-gest this hypothesis is incorrect. First, our own and other previous national surveys have found that climate change is not a highly salient con-cern of the American public, yet by the time of our second survey, 21 million American adults had seen the movie in the theater. Our respon-dents were randomly selected to represent this group. On its face it seems unlikely that 21 million Americans went to the film because they were already highly concerned about glob-al warming. It is more likely that most people went to see the film because it was a summer-time, blockbuster disaster movie.

Nonetheless, we explicitly tested this hypoth-esis in our third and final survey, completed in

Are the significant differences observed in the U.S. study between movie watchers and nonwatchers really due to the impact of the film, or did movie watchers already have “more pro-climate or pro-environment attitudes before entering the cinema”?


November 2004. In this survey (not reported in our article because it had not been conducted yet) we re-interviewed the same respondents as in wave two, including movie watchers. We asked them, “Why did you watch this movie?” Of all movie watchers, only 17 percent said they went because they were “interested in global warming.” By contrast, 83 percent of moviegoers went because they “liked the trailer” (29 percent), “like disaster movies” (21 percent), “like to see all big films” (21 percent), or “another reason” (12 percent). In contrast, Reusswig’s team found that among German moviegoers, 36 percent said a prior interest in climate change led them to watch the film. As he writes, “The German panel study demonstrates a rather strong self- recruitment of . . . more engaged visitors of the film.” Again, by contrast, only 17 percent of American moviegoers said they went because of a prior interest in global warming. Thus, the results on which he bases his conclusion that “there is a significant self-selection effect” are probably more indicative of very interesting cross-cultural differences between German and American climate change risk perceptions.

Second, as reported in the article, we deter-mined that movie watchers were demographi-cally different from the general public—they tended to be slightly younger, male, Hispanic, and politically liberal. We therefore used mul-tiple regression to control for sociodemograph-ic and political variables, including sex, age, education, income, race, political party, and political liberalism. In almost all cases and as

reported in the article, we found that even after controlling for these variables, there remained significant differences between the attitudes of watchers and nonwatchers.

Third, as reported in the Environment arti-cle, we directly asked movie watchers whether the movie made them more worried about global warming. Forty-nine percent of movie-goers said the film made them somewhat (36 percent) or much more worried (13 percent), 42 percent said it did not change their level of worry, and finally, only 1 percent said it made them less worried. These three streams of con-vergent evidence all suggest that indeed, the reported differences in perceived risk between watchers and nonwatchers were due to the impact of the film.

During the meeting in Potsdam, the principle investigators of all five studies identified a number of other intriguing cross-cultural dif-ferences in American, British, German, and Japanese responses to the movie, which we intend to investigate further with a multination-al experimental study, using exactly the same research design and instruments in these and other cultural contexts. We have only scratched the surface, however, in the effort to under-stand the role of popular representations of risk (such as movies, books, television, fiction, and nonfiction) or of cross-national differences in public risk perception and behavior.

Anthony A. LeiserowitzDecision Research

Eugene, Oregon

44 ENVIRONMENT APRIL 2005

We have only scratched the surface

in the effort to understand the role of popular

representations of risk.

Surveying the Impact · vie, a genre infam-ous for weak plots and artistic license yet spec-tacular...

Documents

Transcript of Surveying the Impact · vie, a genre infam-ous for weak plots and artistic license yet spec-tacular...