Content-Type: text/html

Reality Programming on Television:
Realness and Meaning
 
 
W. James Potter
Department of Communication
University of California at Santa Barbara
Santa Barbara, CA  93106
<a href="/cgi-bin/wa?LOGON=A3%3Dind9602A%26L%3DAEJMC%26E%3D0%26P%3D5814103%26B%3D--%26T%3Dtext%252Fplain" >[log in to unmask]</a>
Work: (805) 893 - 7550
Home: (805) 568 - 0700
 
Jeremy Hagemeyer
Art Land
Misha Vaughan
Ron Warren
Kevin Howley
(Graduate students at Indiana University)
 
March 1995
 
Paper submitted to the Mass Communication and Society Division
of the Association for Education in Journalism and Mass Communication
Reality Programming -
Abstract
        This study poses two questions about the portrayals of aggressive behavior
 on non-fictional television programs.  First, how realistic is the pattern
 of portrayal?  This replicated reality is assessed by comparing the
 
      characteristics of televised portrayals to real world characteristics,
such
 as patterns of criminal acts, as well as the demographics of the
 
   perpetrators and victims.  Second, what is the meaning of the aggression?
 
          An answer to this question is constructed through an analysis of
contextual
 variables of reward, consequences, intention, motivation, and style of
 
         presentation.  The data base for the analysis contains 2300 acts of
 
     aggression found in a composite week of 68 hours of non-fiction television
 
          programming.
        The TV patterns of aggression were not found to be a good replicable
 
       reality in terms of the patterns of seriousness of the aggression or on
 
         demographic patterns of perpetrators and victims.  Also, contextual
 
     patterns (low rates of punishment, high rates of absence of negative
 
      conseuqneces, high prevalence of megative motives and intentions) would
 
         increase the likelihood that viewers would interpret an anitsocial
meaning
 
          in favor of aggression
Reality Programming on Television:
Realness and Meaning
 
        It is well established in the media literature that news is a construction
 (Altheide, 1976; Fishman, 1980).  Viewers assume that this construction is
 a relatively accurate reflection of reality.  But is this a safe
 
   assumption?  Does reality programming (news, public affairs programs, and
 
          documentaries) in the aggregate present a realistic picture of the
world?
 
          This is a large question, too large to answer by any one study, so in
this
 
          paper we narrow the question by examining one facet of news--the
portrayal
 
          of aggression.
      We know from a long series of content analyses of entertainment
 
       programs that violence is greatly over-represented on television (for
 
       example, see Gerbner, Gross, Morgan, & Signorielli, 1980; Lichter &
 
     Lichter, 1983; "NCTV says," 1983; Schramm, Lyle, & Parker, 1961; Smythe,
 
          1954).  Also, there is a fair amount of violence and aggression on the
news
 (Cumberbatch, Lee, Hardy, & Jones, 1987) and documentaries (Williams,
 
        Zabrack, & Joy, 1982).  Also, high amounts of violence were found in a
 
        recent analysis of network, local, and superstation news in New York
City
 
          (Johnson, 1993, 1990).  Recently, there has been a growing popularity
among
 a new form of "reality programming" that deals with law enforcement--a
 
         genre that includes shows such as Cops, Top Cops, America's Most
Wanted,
 
          and the like.  These shows were also found to have very high rates of
 
       aggression (Oliver, 1994).
Issue of Reality
        Does televised non-fictional programming present a realistic picture of
 
          the world?  At first, this might sound like a strange question,
because
 
         isn't all news real?  In one sense, news is always real, that is, it
 
      presents accounts of what actually happened.  But in another sense, it can
 
          be unreal, that is, the accounts deviate from the events being
covered, and
 sometimes this deviation can be significant.  The deviations can be
 
      assessed by comparing the patterns of real world aggression with the
 
      patterns of that aggression on television programs that purport to
 
    replicate that aggression so as to inform the viewership about what happens
 in the real world.  For example, Oliver (1994) found that "reality
 
     programs" have been found to be less realistic than viewers might expect.
 
          On these television programs, 87% of criminal suspects were associated
with
 violent crimes, however only 13% of all crimes in the real world are
 
       violent.  This television world of non-fiction crime was also found to be
 
          unrealistic in terms of very high arrest rates (78%) and that among
 
     television police officers, blacks were under-represented in the TV world
 
          (9.0%) compared to the real world (17.0%).
        Therefore, an important issue is the degree of match between real world
 
          aggression and the aggression presented on television.  If the
television
 
          world in its non-fictional narratives replicates the real world then
we
 
         should expect to find four characteristics.
      First, the smallest percentage of aggression should be at the most
 
          serious end of the spectrum, that is, physical acts that result in
great
 
          harm. For example in 1991, the United States Criminal Justice
statistics
 
          indicate that of the 34.7-million crimes, only 18.5% were violent.  Of
all
 
          crimes, murder was less than 0.1%; rape, 0.5%; robbery 3.3% (and only
1.1%
 
          were robberies with injuries); and assault, 14.7% (and only 4.7% were
 
       assaults with injuries).  Less major, but still serious aggression includ
 
          es: personal larceny with contact 1.4%; burglary, 14.8%; household
larceny,
 24.5%; larceny without contact; and theft, 36.1% (all figures from
 
     Maguire, Pastore, & Flanagan, 1993, p. 244).  Also, there is a per year
 
         average of 1,062 bombings, which result in 171 personal injuries and 25
 
         deaths (p. 415).  Like other forms of high level aggression, bombings
are
 
          very serious, but the are a very small percentage of overall
aggressive
 
         acts.
      Second, if the serious forms of aggression are realistic, the
 
     perpetrators should be predominantly male, and ethnicity should be evenly
 
          split between black and white.  Among those arrested for violent
crimes,
 
          88.4% were males and 48.9% were blacks, but for property crimes, 74.6%
were
 male and 25.0% were blacks.  As crimes become less serious, it is more
 
         likely for women and whites to be the perpetrators.  To illustrate,
among
 
          those arrested for murder 89.7% are male and are 57.3% black; rape,
98.7%
 
          male and 43.4% black; robbery, 91.4% male and 60.4% black; and
aggravated
 
          assault, 86.3% male and 41.2% black; larceny and theft, 68.0% male and
 
        23.8% black (Maguire, Pastore, & Flanagan, 1993, p. 432, 435).  And with
 
          less serious acts of aggression such as deceit and insults, we should
 
       expect a broader distribution across all demographic groups.
      Third, if the serious forms of aggression are realistic, the victims
 
          should be mostly male and an even split between black and white.  For
 
       example with murders, 78% of victims are male; 47% are white; 50% are
 
       black.  When examined proportionally, the murder rate per 100,000 people
is
 13.9 for males; 3.5 for females; 4.9 for whites; 34.0 for blacks.  Also,
 
          the murder rate is much higher for younger people: 12.4 for 100,000
people
 
          up to age 17; 22.3 for people between the ages of 18 and 24 (Maguire,
Pas
 
          tore, & Flanagan, 1993, p. 390-391).
      Fourth, most perpetrators of aggression should know their victims.
 
          With murder, among those who police knew the relationship between the
 
       perpetrator and the victim, 33.4% were family members, 42.2% were
 
   acquaintances; and 24.4% were strangers (Maguire, Pastore, & Flanagan,
 
        1993, p. 386-387).  With less serious forms of aggression (such as lying
 
          and insults), we should expect the parties to know each other well.
        In sum, if non-fictional programs exhibit a function of informing the
 
        viewership about aggression in the real world, then in the aggregrate,
 
        those programs should present a pattern where 1) there is a wide
 
  distribution of types of aggressive acts with the most serious acts being
 
          the least prevalent; 2) the perpetrators of serious aggression should
be
 
          predominantly male with an even split between black and white; 3) the
 
       victims of serious aggression should be mostly male and evenly split
 
      between black and white; and 4) about three-quarters of aggression should
 
          be between people who know each other well.  If instead, the
non-fictional
 
          programs exhibit a function of stimulating an aroused state of fear in
the
 
          audience, then those programs will present a pattern of 1)
          over-representing the most serious forms of aggression, 2) show no
regard
 
          for a demographic balance in the portrayals of perpetrators and
victims,
 
          and 3) portray victims not knowing their aggressors.
Meaning of Aggression
        Most reviewers of the media effects literature conclude that the viewing
 
          of violent television increases subsequent viewer aggression.  This is
the
 
          finding of narrative reviews (Andison, 1977; Baker & Ball, 1969;
Comstock &
 Strasburger, 1990; Comstock, Chaffee, Katzman, McCombs, & Roberts, 1978;
 
          Friedrich-Cofer & Huston, 1986; Geen, 1994; Liebert & Schwartzberg,
1977;
 
          Roberts & Maccoby, 1985) as well as meta-analyses (Carlson,
Marcus-Newhall,
 & Miller, 1990; Comstock & Paik, 1991; Hearold, 1986; Wood, Wong, &
 
      Chachere, 1991).
        Most of these reviews (especially the more recent ones) acknowledge the
 
          importance of the context of the portrayals as an important factor in
the
 
          effect.  It is the context of the portrayal that provides cues for the
 
        viewer to interpret the meaning of the action.  Some cues such as
 
   punishment and harmful consequences serve to present aggression in a
 
      negative context and would therefore lead the viewer to interpret the
 
       aggression as an undesireable act.  In this study, we look at five
 
    contextual characteristics: reward, consequences, intention, motive, and
 
          presentation style.
        Reward:  Social learning theory predicts that people can learn about
 
       behaviors vicariously and that a behavior that is portrayed as rewarded
is
 
          more likely to be learned (Bandura, 1977).  This theoretical position
has
 
          been reinforced by the findings of several major reviews of the
empirical
 
          literature.  For example, Comstock et al. (1978) and Comstock and
 
   Strasburger (1990) concluded that the link between exposure to television
 
          violence and subsequent aggressive or antisocial behavior is enhanced
if
 
          the violence is rewarded, or at least not punished.  When violence is
 
       punished, the aggressiveness is likely to be inhibited.
        A few content analyses have looked at the degree of reward presented for
 
          violence and antisocial acts on television.  Potter and Ware (1987)
report
 
          that about 88% of antisocial acts were portrayed as being rewarded.
 
      Williams, Zabrack, and Joy (1982) found that aggression was portrayed as a
 
          successful way to solve conflicts.
        Consequences:  The consequences of aggression (i.e., emotional or physical
 pain or injury) is another important contextual element that viewers use
 
          to interpret the meaning of those portrayals.  For example, Comstock
et al
 
          (1978) and Comstock and Strasburger (1990) concluded that the
aggression
 
          will have a stronger influence on viewers if the violence is shown
with
 
         negative consequences, that is, pain and suffering.  When violence is
 
       regarded as having no consequences, the probability of aggression
 
   increases.  Liebert and Schwartzberg (1977) in their review of the media
 
          effects literature report that with children, the delay of
consequences
 
         (negative motives and undesirable outcomes) leads to a more positive
 
      feeling about the aggression.
        Gunter (1985) also concluded that one of the major factors that explains
 
          how viewers interpret the degree of seriousness of violence is the
 
    consequences.  He found that in his series of 12 experiments, TV viewers
 
          rated violence with observable harm as more serious than scenes that
showed
 no consequences.  However, scenes portraying a clear fatality were rated
 
          as less serious than scenes showing non-fatal injuries.  Gunter
concludes
 
          that the rating of consequences is based on the amount of pain and
sufferin
 
          g shown by the victim; if a victim quickly dies, the suffering is
rated as
 
          less than if the victim is portrayed in agony.
        A few content analyses have looked for the consequences of violence and
 
          found that consequences are rarely portrayed.  For example, Williams,
 
       Zabrack, and Joy (1982) report that in over 81% of violent acts on
 
    television, there was no depiction of an impairment to the victims.  In
 
         Finland the same finding holds where Mustonen and Pulkkinen (1993)
report
 
          that consequences were usually mild: among physical aggression, 64%
was
 
         mild and only 12% severe consequences; among psychologicalaggression,
37%
 
          was mild and only 9% severe.  Also, Cumberbatch, Lee, Hardy, and Jones
 
        (1987) found that injuries were rarely shown as a result of violence on
 
         British TV.  However, a different pattern was found in Japan.  Iwao, de
 
         Sola Pool, Hagiwara (1981) report that Japanese programs emphasize the
 
        suffering of victims, while this seldom occurs in Western TV.
        Intention:  Gunter (1985) noticed that the legal or moral context of
 
       behavior is an important mediator of public perceptions of violence.
 
       Defensive or altruistic aggression may be interpreted as milder than
 
      offensive, intentional, or sadistic aggression.  He argues that unusual
 
         forms of aggression in which a strong motive (sadistic or sexual) is
 
      apparent are perceived as more serious.  Thus the intention of the
 
    character can alter a viewer's meaning of a violent episode.  Also,
 
     Williams, Zabrack, and Joy (1982) examined the contextual factor of
 
     intentionality, and found that in the 81 hours they examined, over 97% of
 
          the aggressive acts were intentional.
        Motive:  Motive is closely linked to intention.  For example, Mees' (1990)
 social norms approach to defining aggression found three modes of
 
    intention that underlie conceptions of motivation for aggressive acts: (1)
 
          thoughtlessness (the aggressor should have taken possible dangers into
 
        consideration but did not); (2) inconsiderate (the aggressor knows that
the
 action will cause distress or harm but accepts this and places his/her own
 interests above those of others); and (3) malicious (wickedness is
 
     accepted and intended by the aggressor.
        Mustonen and Pulkkinen (1993) measured motivation by dividing aggression
 
          into defensive and offensive groups.  Within offensive aggression
there
 
         were five values: instrumental, masochistic, reactive-expressive,
sadistic,
 and altruistic.  They report finding that spontaneous acts (57%) were
 
        higher than planned aggression (27%) and that first strike acts (76%)
were
 
          much higher than retaliatory ones (12%).  Aggressors achieved their
desired
 ends fully or almost fully in about 25% of the acts.  Also, Potter and
 
         Ware (1987) analyzed the antisocial acts in 88 hours of primetime
 
   television for motivation, which they defined as the locus of control for
 
          the act--either internal or external.  They found that 39% of the
1,664
 
         antisocial acts were internally motivated.
        Presentation Style:  Both Hart (1986) and Gunter (1985) make the point
 
         that the salience or arousing nature of the violence makes a
difference.
 
          Thus the style of presentation can make a difference in terms of
providing
 
          cues to the viewer about how the story should be interpreted.  This
would
 
          seem to be an especially important element in the presentation of
 
   non-fictional programming where we should expect a low key recitation of
 
          the facts if the information was the primary focus of the story.  The
tone
 
          of this style would serve to inform viewers about the events of the
day,
 
          and the appeal would therefore be to the mind.  The contrasting
approach
 
          woould be to sensationalize the story so that it appeals to the
emotions of
 the viewer.  With sensationalized stories the intention is to arouse the
 
          emotions and upset the viewer.
        Given the effects studies reviewed above, a certain pattern of context
 
         would seem to increase the likelihood that viewers will interpret
 
   aggression as permissible and good.  Such a pattern would show aggression
 
          as being rewarded (or not punished), as being free from harmful
 
 consequences, as intentional with a clear motive, and with an arousing
 
        style such as presenting an appeal to the emotions.
Method
Sample
      One composite week of television programming was compiled for
 
     analysis.  Included in this week were all programs broadcast by local
 
       affiliates on the four commercial networks (ABC, CBS, Fox, and NBC)
 
     broadcast in the XXX market from 6 p.m. to midnight.  The 28 nights (4
 
        networks by 7 nights) of programming were recorded from March 28, 1994
 
        until June 19, 1994.  No recording was done during the May sweeps
period.
 
          When a night was selected for the sample, the entire set of six hours
was
 
          videotaped.
      The total sample was composed of 168 hours.  The coding was performed
 on all the programming in the sample with the exception of non-program
 
         material (ads, PSAs, and promos separated from the programs they were
 
       promoting).
Coders
        Six graduate students were the coders.  Their training consisted of being
 
          part of the development of all of the codes and procedures. Also, one
 
       formal and several informal pilot tests were conducted and the results
were
 discussed among the team members.
Coding
      Coding began in April.  Each coder analyzed from 12 to 40 hours of
 
          videotaped programming.  For each tape, they recorded program
identifier
 
          information for each aggressive act.  This identifying information
included
 network (ABC, CBS, Fox, and NBC), time of day the program started, length
 
          of program, coder and type of program.
      Each time a coder saw an act that fit the definition of aggression,
 
          he/she wrote a short descriptor for that act on a separate line on the
 
        coding data sheet.  Along with the act of aggression, contextual
 
  information was also coded.  The contextual factors analyzed in this study
 
          are as follows: profile of perpetrator (gender, race, and age),
profile of
 
          victim, relationship of the actors, intention of the perpetrator,
motive of
 perpetrator, motive, consequences, and presentation style.
      After the coding for a block of shows was completed, the coder
 
      entered his/her data into the mainframe computer and ran some initial
 
       frequencies programs to clean the data file by checking for the range of
 
          codes on each variable.  Each of these individual data files was then
 
       combined into a master file which contained over 6,000 acts of
aggression.
Measures
      Type of Act:  We began with a list of 42 acts of aggression, which
 
          were then used by coders (see Table 1 for the complete list).  In our
 
       initial analysis, we found that many of the act types accounted for very
 
          few occurrences, so the 42 types of aggressive acts were grouped into
eight
 categories as follows:
        1. Serious Assault is a group that includes all direct physical aggression
 resulting in death or great physical harm of self or other.  Serious
 
       assaults include: unsuccessful suicide attempts, killing, maiming,
 
    beatings, and severe attacks.
        2. Minor Assault refers to any direct physical aggression resulting in
 
         minor harm or no harm of self or other.  It includes the acts of
slapping,
 
          punching, kicking, failed assaults, detention resulting in loss of
freedom,
 and deprivation of basic needs: food, clothing, shelter.
        3. Harm to Property refers to any direct physical aggression aimed at
 
        another or one's own possessions.  Harm to Property includes: bombing,
 
        arson, vandalism, such as to a person's dwelling, or place of business,
 
         robbery, burglary, and larceny.  Likewise, harm to property includes
any
 
          indirect physical aggression, such as tampering with property in order
to
 
          cause harm to individuals (e.g. "fixing" another's car in order to
cause an
 accident), and theft from a store or an entity other than an individual.
        4. Intimidation refers to any symbolic aggression placing another under
 
          duress and intended to pressure, constrain, or persuade in a noxious
 
      manner.  It includes:  extortion, coercion, threats, abridgment of
privacy,
 and harassment.
        5. Deception refers to any symbolic aggression utilizing fabrication,
 
        untruths, and trickery intended to mislead and/or cheat another.  It
 
      includes:  fraud, cheating, lying, and assuming a false identity.
        6. Hostile Remarks refers to any symbolic aggression meant to diminish
 
         another's or one's own self-concept.  It includes: rejection; such as
 
       insults, criticism, or cursing another, resentment, suspicion, and
 
    malicious remarks; such as hate speech, put downs, and yelling, and
 
     screaming in another's presence, and self-deprecation.
        7. Societal Harm refers to any physical and/or symbolic aggression
 
     detrimental to society at large.  It includes: mayhem or other random acts
 
          of physical aggression (looting, rioting, trashing, wilding),
vandalism
 
         directed against public property, resentment, such as hate speech
directed
 
          at societal groups, suspicion (i.e. the projection of hostility onto
other
 
          groups), negativism, such as oppositional behavior directed at
authority
 
          and/or social institutions, and antisocial behaviors such as abuses of
 
        power, or "fixing" a parking ticket.
        8. Accidents refers to any physical aggression, such as a mishap or
 
      natural catastrophe or calamity resulting in death, great harm, minor
harm,
 or no harm.  It includes: fire, earthquake, auto, air, or other transit
 
          related mishaps, and any other natural or man-made disasters.
        This eight part scheme is an elaboration of the ideas of Greenberg et al's
 (1980) four component definition of antisocial behavior: (1) Physical
 
        Aggression:  Any overt behavior intended to frighten, injure, or damage
 
         oneself, another individual, an animal or property; (2) Verbal
Aggression:
 sending noxious symbolic messages; (3) Theft:  the intentional and
 
     deliberate taking of another person's or institution's property without
 
         right or permission; and (4) Deceit:  intentional misleading of someone
for
 purposes that are  detrimental to an individual, group, or institution.
        Profile of Perpetrator and Victim:  For both perpetrator and victim, the
 
          coder identified the gender (female, male, unknown, or multiple); race
 
        (white, black, Hispanic, Asian, other, or multiple); and age (young
child,
 
          child, adolescent, young adult, mid- age, elderly, and multiple).
Also,
 
          the Relationship of Actors was coded using the following values:
strangers,
 acquaintances, close (friends or family), self, and multiple.
        Intent: This was coded from the perpetrator's point of view and had four
 
          values as follows: Maliciousness (consciously planned and executed);
 
      Inconsiderate (aggressor knows that the action will cause harm or distress
 
          but accepts this and places his/her own interests above those of
others);
 
          Thoughtlessness (the aggressor should have taken possible dangers into
 
        consideration but did not); and No chance for pre-thought.
      Motive: Again, this was coded from the point of view of the
 
   perpetrator and had seven values as follows: To hurt the other physically;
 
          To hurt the other symbolically (to cause emotional pain or
embarrassment);
 
          Desire for external gain (usually economic); To make one's self feel
better
 (ego gratification, etc.); Perpetrator cannot control him/herself
 
    (instinctual action); Prosocial motive; No motive shown.
        Reward:  This had five values: Aggression is approved AND rewarded
 
     physically; Aggression is approved AND rewarded symbolically; Aggression is
 neither approved nor punished (neutral); Aggression is punished
 
  symbolically; and Aggression is punished physically.
        Consequences:  This had four values:  No consequences shown (merely the
 
          aggressive act); Minor consequences shown (victim gets over easily and
 
        quickly); Major consequences shown (victim is portrayed in major hurt
for
 
          short time); and Major consequences shown (victim is portrayed in
major
 
         hurt for long time).
        Presentation Style:  This had two values: Rational/intellectual and
 
      Emotional (sensationalized).
Intercoder Reliability
        A seventh person served as an additional coder to run the reliability
 
        check.  Seventeen hours of randomly selected programming from the sample
 
          was coded independently.  This included 786 acts of aggression.  The
codes
 
          of this seventh coder were compared to the codes of each of the other
six
 
          coders to determine a percentage of agreement on each variable for
each
 
         coder.  On any given variable, the percentages of agreement among
coders
 
          varied to only a small degree (less than six percentage points).  The
perc
 
          entages of agreement were then entered into the Scott's pi formula to
 
       determine the intercoder reliability.  The variables and Scott's pi's are
 
          as follows: Act Type .873; Perpetrator Gender .913; Perpetrator Race
.918;
 
          Perpetrator Age, .909; Victim Gender .913; Victim Race .918; Victim
Age
 
         .909; Relationship of Perpetrator and Victim .911; Intention .824;
Motive
 
          .818; Reward .853; Consequences .901;  and Presentational Style .866.
Type of Program
        The full data base contained over 6,000 acts of aggression across both
 
         entertainment and informational types of programs.  This analysis is
 
      limited to the information set of programs, which originally contained
nine
 programming categories.  This set of nine was condensed into specific
 
        program types better suited for comparative analysis across genres.  The
 
          resulting five program types used for comparison are:  1) Local News -
as
 
          produced by the ABC, NBC, CBS and FOX affiliates of the sample market
and
 
          commonly appearing at 6 and 11 pm;  2) National News - as produced by
the
 
          ABC, NBC and CBS network news operations and usually appearing at 6:30
pm;
 3) News Magazine - network and syndicated programs concentrating on
 
      current events reporting in a magazine format, examples include A Current
 
          Affair, Inside Edition, 20/20, 48 Hours, 60 Minutes and Eye-to-Eye;
4)
 
         Talk/Interview - a construct of network, syndicated and local programs
 
        included in the categories coded as Celebrity News/Talk, News
          Talk/Interview, Social Issues Talk, and Sports Talk and specifically
 
      including such examples as Ricki Lake, Entertainment Tonight, Nightline
and
 Sports View;  5) Non-Fiction Narrative - a construct of the Crime News and
 Documentary categories, programs which place actual events (more of a
 
        recent than current nature) into a narrative context such as Cops,
Rescue
 
          911, Ancient Prophecies, and Unsolved Mysteries.
Results and Discussion
      Out of the total 168 hours coded in the entire sample, 68 (40.5%)
 
         were information type programs composed of several different formats
(see
 
          Table 1).  In this set of informational programs, there were 2300 acts
of
 
          aggression for an average rate of 33.8 per hour.  National news
exhibited
 
          the highest rate with local news showing the lowest rate.
The Issue of Reality
        First, real world figures indicate that there is a wide distribution of
 
          types of aggressive acts with the most serious acts being the least
 
     prevalent.  In this analysis, we found a wide distribution of aggressive
 
          acts (see Table 2).  The most prevalent single acts were: assaults
 
    resulting in death (4.58 acts per hour on average), harsh
          criticism/insults/put downs (4.01), and suspicion (3.04).  Of the 42
 
      different types of aggression for which we coded, these three types
 
     accounted for 34.7% of all the aggression on the information type programs.
 
        Up until this point in the analysis, it appears that the pattern indicates
 a similarity with the real world.  However, when we look at the proportion
 of aggression that is physical (52.4%) compared to symbolic (44.4%), it
 
          begins to look unrealistic.  While there are no reliable real world
figures
 for such proportions, it is reasonable to conclude that on a daily basis
 
          many more people are likely to engage in symbolic acts (such as
deceit,
 
         rejection, threats, resentment, suspicion, malicious remarks, coercion,
and
 harassment) than in physical acts (such as assaults, attacks on property,
 
          theft, detention, and mayhem).  Also, within the sub-category of
criminal
 
          violence, it appears that there are much higher rates on informational
 
        television than in the real world.  In the real world, only 18.5% of all
 
          crimes were violent, but in this study there were 832 acts of a
criminal
 
          nature (those in the assualt, attack on property, and theft
categories), a
 
          conservative estimate of violence (only those acts of assault
resulting in
 
          harm) is 75.7%.  In the real world, assualts account for only 14.7% of
all
 
          crimes, and only 4.7% were assaults with injuries.  Less major, but
still
 
          serious aggression includes larceny, burglary and theft which in the
real
 
          world account for 76.8% of all crime, but in the television world it
is
 
         only 2.9%.
        Second, real world figures indicate that the perpetrators of serious
 
       aggression should be predominantly male and split between black and
white.
 
          The television perpetrators of each type of aggressive act are
          predominantly male (see Table 3).  Of all aggressive acts, males alone
are
 
          the perpetrator 48.4% of the time and are partially responsible
(multiple
 
          gender) 7.9%, while females are the perpetrator only 20.4%.  These
 
    percentages underestimate the participation of the two genders because
 
        23.3% of the aggressive acts did not have an identifiable perpetrator.
For
 example, news programs will report a crime where the perpetrator is
 
      unkown.  When we base the gender figures on only those acts where there is
 
          a clear depiction of the perpetrator's gender, we see that males are
 
      responsible for 62.3%, females for 27.5% and multiple for the remaining
 
         10.2%.
        Comparing the female and male lines on Table 3, it is clear that males are
 much more likely to be the perpetrator of the more serious aggression,
 
         including acts of Serious Assaults, Minor Assaults, and Harm to
Property.
 
          Looking at only these three types of aggression for males and females,
 
        males were the perpetrator over 85.7% of the time, which is very close
to
 
          the the real world statistic that 88.4% of those arrested for violent
 
       crimes are male.         There is a sharp contrast, however, between television
 
          and real world statistics when the perpetrator variable is race.
Using the
 same construct of comparison as above (Serious Assaults, Minor Assaults,
 
          and Harm to Property compared to real world satistics for violent
crime
 
         arrests), blacks were the perpetrator in about 26% of the television
cases
 
          -- in contrast to the real world statistic of 48.9% of those arrested
for
 
          violent crimes being black.  In each act type, whites were the
predominant
 
          perpetrators, representing 67.6% of all cases where the perpetrator is
 
        identifiable.  Across all act types, 20.2% were perpetrated by blacks
(when
 comparing only those cases identifying either a white or black
 
 perpetrator), thus weakly confirming the expectation that as the
 
  seriousness of the act decreases, the breadth ascribed to perpetrator race
 
          increases.
        As for age, most of the perpetrators who age was codable were in the 19 to
 60 age level.  Overall the 31 to 60 age level showed the highest
 
   prevalence of aggressiion compared to the other age levels; however, the
 
          highest rates were with the relatively less serious forms of
aggression.
 
          The number of acts perpetrated by 19 to 30 year-olds also come close
(but
 
          do not supersede 31 to 60 year-olds) in the act types of Minor Assault
(99
 
          compared to 114) and Accidents (15 compared to 23).
        Third, if the serious forms of aggression are realistic, the victims
 
       should be mostly male and an even split between black and white.  With
 
        gender, the pattern indicates that males are more likely than females to
be
 a victim.  With race, whites are primarily the victims.  In looking at the
 age of victims one can see a similar tendency as with perpetrators.  The
 
          numbers tend to centralize around the middling range of 19 to 60.
        Fourth, if aggression is portrayed as realistic, most perpetrators of
 
        aggression should know their victims.  In real life, over 70% of the
 
      perpetrators of serious assaults are family members or acquaintances of
the
 victim.  However, our data show that in the television world, on only
 
        17.2% of serious assaults did the victim know the perpetrator (close
 
      friend/family, self, or acquaintance).  This figure is higher (26.6%) with
 
          minor assaults, but it is still far from the real world figure.
Contextual Cues for Interpreting Meaning
        The most harmful contextual pattern for aggression would have high rates
 
          of reward (or low rates of punishment), low rates of harmful
consequences,
 
          high intentionality with a clear motive, and with an arousing style
such as
 presenting an appeal to the emotions.  This is close to the pattern that
 
          was found in this analysis (see Table 4).
        A high proportion of aggressive acts are neither rewarded or punished
 
        (between 40.5% and 74.9% depending on type of aggression).  Some acts
(such
 as serious assault, deception and accidents) show a pattern of higher
 
        rates of punishment than rewards.  But this finding is secondary to the
 
         main finding that the majority of aggressive acts are portrayed as
neutral.
  According to social learning theory, a non-punished act is often
 
    interpreted the same as one that is rewarded, that is, the viewer does not
 
          learn that the act should not be performed.
        In real life most of the serious acts of aggression go unpunished.  For
 
          example, in 1991 only 21.2% of all reported crimes were cleared by an
 
       arrest; in the others no arrest was made.  Arrest rates are higher with
 
         violent crimes (44.7%) than with property crimes (17.8%), but still
most
 
          crimes remain unsolved, much less punished (Maguire, Pastore, &
Flanagan,
 
          1993, p. 452).  These real world figures contrast with the public's
 
     attitudes and perceptions about crime.  For example, 59% of American rate
 
          the police's record of solving crimes as excellent or very good
(Maguire,
 
          Pastore, & Flanagan, 1993, p. 169).  Given these results, it would
seem
 
         that people are not aware of the low rate of clearing crimes.  Where do
 
         people get this misinformation?  Is it possible that the context of
 
     portrayals on television non-fiction lead people to believe that most
 
       crimes are punished?
        There was a high percentage of aggressive acts portrayed with no
 
   consequences -- from 27.8% to 81.0%.  The highest rates of major long term
 
          consequences was with accidents and serious assault, which is a
positive
 
          indication that viewers are getting cues that aggression has negative
 
       ramifications.  However, the number of aggressive acts shown without any
 
          consequences is very high.
        When we examine the patterns of intent and motive, we see that the
 
     aggression is portrayed in a very conscious, premeditated and antisocial
 
          manner.  With intent, maliciousness is the dominant value on all act
types
 
          except accident.  The dominant motive for the more serious aggression
of
 
          assaults and property harm is to hurt the other physically.  It is
 
    interesting to note that the dominant motive for property harm (which
 
       includes burglary, robbery, and larceny) is to hurt the other physically;
 
          external gain is far behind as a motive.
        The presentational style is predominately intellectual, but there is also
 
          a high prevalence of emotional style, especially with hostile remarks
and
 
          minor assaults.
        The above analysis by act type present some puzzling patterns.  This is
 
          perhaps because of the differing nature of the types of programs
analyzed.
 For example, we should not expect the Evening News to present aggression
 
          in the same way that Cops does.  When we look at these same contextual
 
        variables in terms of patterns across shows, we see some revealling
 
     differences (see Table 5).  With presentation style, the news shows (both
 
          local and national) emphasize an intellectual approach by generally
letting
 the facts speak for themselves and giving the viewers information.  In
 
         contrast, the news magazine, talk/interview, and non-fiction narrative
 
        shows appear to rely much more strongly on emotional appeals where the
 
        facts of the aggression are less important that the arousal of the
audience
 with suspense, fear, and re-enactments.  The news programs are also more
 
          likely to show long term major consequences of the aggression and be
more
 
          likely to attach punishment, especially symbolic punishment.
Differences
 
          across program types are more difficult to see with the variables of
motive
 and intent.  It appears that news programs are more likely to attribute a
 
          motive of hurting another physically, while the non-news shows are
more
 
         likely to use the motives of hut someone symbolically or to make the
 
      perpetrator feel better.
Conclusion
        The findings of this analysis are very troubling.  It appears that
 
     non-fictional television presents an extremely high rate of aggression, and
 that the most serious forms of that aggression (physical violence and
 
        crime) are presented at rates far above the rates in the real world.  As
a
 
          set of programming, this is misleading the viewing public about how
much
 
          violent crime there is, underestimates the percentage of black law
 
    enforcement officers, under-estimates the young as perpetrators and victims
 of serious criminal violence, and fosters a false belief that perpetrators
 are unknown to their victims.
        Furthermore, the contextual cues in the portrayals are such that they
 
        serve to encourage the learning of aggression.  That is, there is a low
 
         rate of punishment (less than one-quarter), a fairly high rate of lack
of
 
          consequences (about half), clear antisocial intentions (especially
 
    maliciousness) and motives (to hurt the victim physically and for external
 
          gain), and a fairly high rate of an appeal to the emotions (about
40%).
 
                                 References
Altheide, D. (1976). Creating reality:  How TV news distorts events.
 
      Beverly       Hills, CA: Sage.
Andison, F. S. (1977). TV violence and viewer aggression: A cumulation of
 
          study results. Public Opinion Quarterly, 41, 314-331.
Baker, R. K., & Ball, S. J. (1969). Violence and the media. Washington, DC:
 U. S. Government Printing Office.
Carlson, M., Marcus-Newhall, A., Miller, N. (1990). Effects of situational
 
          aggression cues:  A quantitative review. Journal of Personality and
Social
 
               Psychology, 58, 622-633.
Comstock, G. A. (1985). Television and film violence. In S. Apter & A.
 
        Goldstein (Eds.) Youth violence: Programs and prospects. New York:
 
         Pergamon.
Comstock, G. A., Chaffee, S., Katzman, N., McCombs, M., & Roberts, D.
 
       (1978). Television and social behavior. New York: Columbia University
 
            Press.
Comstock, G., & Paik, H. (1991). Television and the American child. New
 
         York: Academic Press.
Comstock, G., & Strasburger, V. C. (1990). Deceptive appearances:
 
   Television violence and aggressive behavior. Journal of Adolescent Health
 
               Care 11(1), 31-44.
Cumberbatch, G., Lee, M., Hardy, G., & Jones, I. (1987). The portrayal of
 
          violence on British television. London: British Broadcasting
Corporation.
Fishman, M. (1980). Manufacturing the news. Austin, TX: University of Texas
   Press.
Friedrich-Cofer, L., & Huston, A. C. (1986). Television violence and
 
      aggression: The debate continues. Psychological Bulletin, 100, 364-371.
Geen, R. G. (1994). Television and aggression: Recent developments in
 
       research and theory.  In D. Zillmann, J. Bryant, & A. C. Huston (Eds.),
 
              Media, children, and the family (pp. 151-162). Hillsdale, NJ:
Lawrence
 
             Erlbaum.
Gerbner, G., Gross, L., Morgan, M. & Signorielli, N. (1980). The
 
  "mainstreaming" of America: Violence profile no. 11. Journal of
 
      Communication, 30(3), 10-29.
Greenberg, B. S., Edison, N., Korzenny, F., Fernandez-Collado, C., & Atkin,
 C. K. (1980).  In B. S. Greenberg (Ed.), Life on television: Content
 
       analysis of U. S. TV drama (pp. 99-128). Norwood, NJ: Ablex.
Gunter, B. (1983). Personality and perceptions of harmful and harmless TV
 
          violence.  Personality and Individual Differences, 4, 665-670.
Gunter, B. (1985). Dimensions of television violence. Aldershot, England:
 
          Gower Publishing Company Limited.
Gunter, B., & Wober, M. (1988). Television viewing and public trust.
 
       British Journal of Social Psychology, 11, 280-288.
Hart, A. (1986). Children and television: Audience and influence. Media in
 
          Education and Development. 19(2), 80-85.
Hearold, S. (1986). A synthesis of 1043 effects of television on social
 
         behavior. In G. Comstock (Ed.) Public communication and behavior (Vol.
1,
 
               pp. 65-133). San Diego, CA: Academic Press.
Iwao, S., de Sola Pool, I., Hagiwara, S. (1981). Japanese and U. S. media:
 
          Some cross-cultural insights into TV violence. Journal of
Communication
 
              Disorders, 31(2), 28-36.
Johnson, R. N. (1993). The portrayal of violence, conflict, and suffering
 
          on       television news broadcasts. Aggressive Behavior, 19, 42-43.
Johnson, R. N. (1990). The reporting of violence and conflict on television
       news. Aggressive Behavior, 16, 107-108.
Lichter, L. S., & Lichter, S. R. (1983). Prime Time Crime. Washington, D.
 
          C.:   The Media Institute.
Liebert, R. M., & Schwartzberg, N. S. (1977) Effects of mass media. Annual
 
          Review of Psychology, 28, 141-173.
Maguire, K, Pastore, A. N., & Flanagan, T. J. (Eds.) (1993). Bureau of
 
        Justice Statistics Sourcebook of Criminal Justice Statistics -- 1992.
 
             Washington, DC: U. S. Department of Justice.
Mustonen A., & Pulkkinen, L. (1993). Aggression in television programs in
 
          Finland. Aggressive Behavior, 19, 175-183.
NCTV says violence on TV up 16%. (1983, March 22). Broadcasting Magazine,
 
          p. 63.
Oliver, M. B. (1994). Portrayals of crime, race, and aggression in "reality
 based" police shows: A content analysis. Journal of Broadcasting &
 
     Electronic Media, 38, 179-192.
Paik, H., & Comstock, G. (1994). The effects of television violence on
 
        antisocial behavior: A meta-analysis. Communication Research, 21,
516-546.
Potter, W. J., & Ware, W. (1987).  An analysis of the contexts of
 
   antisocial acts on prime-time television. Communication Research, 14,
 
            664-686.
Roberts, D. F., & Maccoby, N. (1985). Effects of mass communication. In G.
 
          Lindzey & E. Aronson (Eds.) Handbook of social psychology: Vol 2.
Special
 
               fields and applications (3rd ed., pp. 539-598). New York: Random
House.
Schramm, W., Lyle, J., & Parker, E. B. (1961). Television in the lives of
 
          Our  Children. Stanford, CA: Stanford University Press.
Smythe, D. W. (1954). Reality as presented on television. Public Opinion
 
          Quarterly, 18, 143-156.
Williams, T. M., Zabrack, M. L., & Joy, L. A. (1982). The portrayal of
 
        aggression on North American television. Journal of Applied Social
 
          Psychology, 12, 360-380.
Wood, W., Wong, F. Y., & Chachere, J. G. (1991). Effects of media violence
 
          on viewers' aggression in unconstrained social interaction.
Psychological
 
               Bulletin, 109, 371-383.
 
Table 1
Rates of Aggression on Different Types of Information Programs
___________________________________________________________________
                     Number    Number of   Total  Number   Rate
Type of Program      of Titles Broadcasts  Hours  of Acts  Per Hour
 
Local News       9      49      25.5    719     28.2
National News    3      16       9      404     44.9
News Magazine   11      21       16     504     31.5
Talk/Interview   5      16       8      321     40.1
Non-Fiction Narrative    6      12       9.5    352     37.1
 
Totals  34      114     68      2300
 
___________________________________________________________________
 
    __________
 
Table 2
 
Types of Aggressive Acts
____________________________________________________________
 
I. Direct-Physical        Number  Percentage  Rate Per Hour
 
Assault
  Failed suicide attempt          1      0.0    0.01
  Resulting in death    318     13.8    4.58
  Resulting in great    170      7.4    2.50
     physical harm
  Resulting in minor    141      6.1    2.07
     physical harm
  Resulting in no harm   99      4.3    1.46
 
Attack on property
  Bombing        44      1.9    0.55
  Arson   8      0.3    0.12
  Vandalism      24      1.0    0.35
 
Accidents
  Death to other or self         80      3.5    1.18
  Great physical harm/  119      5.2    1.75
                damage
  Minor physical harm/   35      1.5    0.51
                damage
  No harm         4      0.2    0.06
 
Theft
  Extortion       3      0.1    0.04
  Robbery         8      0.3    0.12
  Burglary        1      0.0    0.01
  Larceny        15      0.7    0.22
 
Detention
  Loss of freedom        96      4.2    1.41
  Depriving other of     11      0.5    0.16
       basic needs
 
Mayhem   32      1.4    0.47
 
 
II. Symbolic
 
Coercion         38      1.7    0.56
 
Threats 118      5.1    1.73
 
Rejection        95      4.1    1.40
 
Abridgement of privacy   14      0.6    0.21
 
Harassment       78      3.4    1.15
 
 
Deceit
  Fraud  14      0.6    0.21
  Cheating       12      0.5    0.18
  Lying  28      1.2    0.41
  False identity         10      0.4    0.15
 
Resentment       20      0.9    0.29
 
Suspicion       207      9.0    3.04
 
Malicious remarks
  Hate speech    15      0.7    0.22
  Harsh criticism/insults/      273     11.9    4.01
            put-downs
  General yelling and    77      3.3    1.13
            screaming
 
Self Deprecation
  Self demeaning -       10      0.4    0.15
   unintentional
  Self demeaning -       14      0.6    0.21
     intentional
 
 
III. Indirect-Physical
 
Endangering another w/out         2      0.1    0.03
          their knowledge
 
Undirected negative
   physical acts
   Vandalism      1      0.0    0.01
 
 
IV. Harm to Society
 
Resentment        6      0.3    0.09
 
Suspicion        29      1.3    0.43
 
Negativism       11      0.5    0.16
 
Other "antisocial" behavior  19  0.8    0.28
____________________________________________________________
 
TOTAL                              2300 100.0            33.82
 
 
 
Table 3
 
Demographics of Perpetrators and Victims Profiled by Type of Act (In Percent)
_____________________________________________________________________________
 
                                                TYPE OF ACT
                 Sers    Minr    Prop                    Hstl    Socl
PERPETRATOR(S)   Aslt    Aslt    Harm    Intm    Dcpt    Rmrk    Harm    Acdt
Gender
        Female   8.0     9.2     8.8    12.4    20.3    36.7    19.4     2.9
        Male    45.8    65.4    40.2    64.5    51.6    51.2    42.9    19.3
        Multiple         6.7     8.4     4.9     7.6    14.1     6.5    10.2     1.3
        Unknown 39.5    17.0    46.1    15.5    14.1     5.6    27.6    76.4
 
                        (Chi Square = 763.2, df = 28, p < .0001)
 
Race
        White   24.3    45.5    34.3    53.8    57.8    70.2    48.0    10.9
        Black   10.0    14.7     1.0    14.3     9.4    14.5     7.1     5.5
        Hispanic         1.0     1.4     0.0     1.2     0.0     1.1     1.0     0.4
        Asian    1.4     1.4     2.0     0.8     1.6     1.1     2.0     0.0
        Other    2.9     4.9     1.0     2.8     0.0     0.8     4.1     2.1
        Multiple         9.2     9.5     5.9     6.8     6.3     3.4     6.1     1.3
        Unknown 51.1    22.5    55.9    20.3    25.0     8.9    31.6    79.8
 
                        (Chi Square = 689.6, df = 42, p < .0001)
 
Age
        Birth to 5       0.0     0.0     0.0     0.0     0.0     0.6     0.0     0.0
        6 to 12  0.2     0.9     0.0     0.0     0.0     0.4     0.0     0.4
        13 to 18         2.2     1.4     4.9     1.2     3.1     3.0     6.1     2.5
        19 to 30        15.7    28.5     8.8    15.5    17.2    25.9     8.2     6.3
        31 to 60        14.1    31.7    19.6    49.4    50.0    51.2    41.8     9.7
        61 and over      0.4     0.6     1.0     1.2     3.1     4.8     3.1     0.4
        Multiple        13.3    13.3     7.8     9.6     3.1     5.2    11.2     0.4
        Unknown 54.0    23.6    57.8    23.1    23.4     9.0    29.6    80.3
 
                        (Chi Square = 766.5, df = 49, p < .0001)
 
 
VICTIM(S)
Gender
        Female  17.4    14.4     6.9    33.1    26.6    27.3     3.1     9.7
        Male    40.1    63.4     9.8    35.5    31.3    49.8     3.1    26.1
        Multiple        21.3     7.8    21.6    13.9     9.4    13.1    40.8    34.5
        Unknown 21.3    14.4    61.7    17.5    32.8     9.9    53.1    29.8
 
                        (Chi Square = 531.2, df = 28, p < .0001)
 
Race
        White   28.2    53.6    14.7    51.8    48.4    60.3     8.2    23.9
        Black   16.4    16.7     2.0     7.6     3.1    12.4     4.1     5.5
        Hispanic         2.5     2.3     0.0     0.8     0.0     1.3     0.0     0.0
        Asian    2.7     0.9     1.0     1.6     0.0     0.7     1.0     0.0
        Other    4.3     2.9     2.0     3.2     0.0     0.1     1.0     1.3
        Multiple        11.0     4.9    16.7     6.4    10.9    10.0    33.7    30.7
        Unknown 35.0    18.7    63.7    28.7    37.5    15.2    52.0    38.7
 
                        (Chi Square = 525.9, df = 42, p < .0001)
 
 
Age
        Birth to 5       3.9     2.3     0.0     0.0     0.0     0.6     0.0     2.1
        6 to 12  3.7     2.0     0.0     2.4     0.0     0.1     0.0     5.5
        13 to 18         4.1     7.2     0.0     6.4    10.9     1.4     0.0     8.0
        19 to 30        14.5    36.9     6.9    22.3    12.5    24.8     1.0     9.2
        31 to 60        14.5    24.2     4.9    27.1    28.1    40.8     0.0     7.6
        61 and over      3.9     0.9     0.0     2.0     3.1     3.0     0.0     0.8
        Multiple        19.8     9.2    20.6    12.0     9.4    14.8    42.9    31.1
        Unknown 35.6    17.3    67.6    27.9    35.9    14.6    56.1    35.7
 
                        (Chi Square = 656.3, df = 49, p < .0001)
 
Relationship
        Stangers        40.5    45.2    29.4    41.4    43.8    48.9    36.7    20.6
        Acquaintances    5.7    17.0     1.0    19.5    17.2    21.1     1.0     6.7
        Close    8.8     6.1     2.9    13.1    17.2    15.3     3.1     1.3
        Self     2.7     3.5     0.0     0.0     0.0     4.6     0.0     5.5
        Multiple         2.0     1.4     0.0     0.0     1.6     0.8     5.1     0.4
        Unknown 40.3    26.8    66.7    25.9    20.3     9.1    54.1    65.5
 
                        (Chi Square  = 522.8, df = 35, p < .0001)
 
            n =  489     347     102     251      64     711      98     238
 
                                                                     n = 2300
 
 
 
Table 4
 
Profiles of Contextual Variables by Type of Act (In Percent)
_____________________________________________________________________________
 
                                    Type of Act
                 Sers    Minr    Prop                    Hstl    Socl
                 Aslt    Aslt    Harm    Intm    Dcpt    Rmrk    Harm    Acdt
 
Reward
 Physical reward        11.6    17.8    23.8     4.4    11.5     0.7     1.2     0.5
 Symbolic reward         3.8    17.5     3.6    17.1    19.7    23.3    25.0     1.0
 Neutral        48.5    48.1    40.5    54.4    24.6    65.3    47.6    74.9
 Symbolic punish        25.5    11.6    27.4    22.4    39.3    10.6    20.2    21.3
 Physical punish        10.5     5.0     4.8     1.8     4.9     0.0     6.0     2.4
             n =         447     320      84     228      61     678      84     207
 
     (Chi square = 461.1 , df = 28, p < .0001, reward unknown cases = 191)
 
Consequences
 None   39.1    48.1    55.3    60.9    55.4    68.0    81.0    27.8
 Minor   0.7    32.9    15.3    19.8    12.5    25.8     7.1     5.1
 Major-short term12.1   10.9    12.9    10.1    12.5     3.7     6.0    16.2
 Major-long term        48.2     8.1    16.5     9.2    19.6     2.5     6.0    50.9
             n =         448     322      85     207      56     678      84     234
 
 (Chi square = 738.1 , df = 21, p < .0001, consequences unknown cases = 186)
 
Intent
 Maliciousness  83.7    68.8    87.7    65.5    79.7    49.7    71.4     1.8
 Inconsiderate   8.6    29.7     8.6    34.1    20.3    47.3    28.6     2.9
 Thoughtlessness         5.8     1.6     2.5     0.5     0.0     2.7     0.0    32.4
 Reaction        1.9     0.0     1.2     0.0     0.0     0.3     0.0    62.9
                    n =  417     317      81     220      59     678      84     170
      (Chi square = 1681.7, df = 21, p < .0001, intent unknown cases = 274)
 
Motive
 Hurt physically        72.7    40.9    38.0     8.7     1.7     0.3    20.0     0.6
 Hurt symbolclly         0.2     4.8     2.5    20.4     0.0    46.2    16.5     0.0
 External gain   9.3     8.3    22.8    15.5    62.1    15.2    18.8     5.3
 To feel better  2.5     9.6    13.9    17.5    15.5    19.8    17.6     1.8
 Instinctual     1.5     1.0     1.3     1.0     0.0     0.5     1.2     4.7
 Prosocial motive 5.2   31.6     6.3    32.5    15.5    10.7    18.8     1.2
 No motive shown         8.6     3.8    15.2     4.4     5.2     7.4     7.1    86.5
            n =  407     313      79     206      58     666      85     170
 
     (Chi square = 2189.6, df = 42, p < .0001, motive unknown cases = 316)
 
Presentational Style
 Intellectual   65.8    41.8    59.8    52.6    68.8    35.2    60.2    64.7
 Emotional      34.2    58.2    40.2    47.4    31.3    64.8    39.8    35.3
             n =         489     347     102     251      64     711      98     238
 
     (Chi square = 158.9 , df = 7, p < .0001, style unknown cases = 0)
 
 
 
Table 5
 
Profiles of Contextual Variables by Type of Information Program (In Percent)
_____________________________________________________________________________
 
                                Type of Information Program
                     Local   National    News     Talk/      Non-Fiction
                     News      News      Mags   Interview     Narrative
Presentational Style
 Intellectual       76.2     67.3      32.5       28.7         25.9
 Emotional           23.8     32.7      67.5      71.3         74.1
             n =        719      404     504     321     352
 
     (Chi square = 447.8 , df = 4, p < .0001, style unknown cases = 0)
 
Consequences
 None   49.8    58.9    60.2    43.4    53.8
 Minor   8.6     5.7    16.7    48.4    18.6
 Major-short term       11.2     7.1    10.5     3.6    11.5
 Major-long term        30.5    28.3    12.5     4.6    16.0
             n =         650     367     455     304     338
 
 (Chi square = 357.7 , df = 12, p < .0001, consequences unknown cases = 186)
 
Reward
 Physical reward         5.4     8.5     4.9     5.3    14.2
 Symbolic reward        10.6    13.9    17.2    28.6     7.0
 Neutral        55.1    56.3    55.5    54.3    59.9
 Symbolic punish        21.3    19.3    18.7    11.2    16.3
 Physical punish         7.6     2.0     3.7     0.7     2.6
             n =         644     352     465     304     344
 
     (Chi square = 144.8 , df = 16, p < .0001, reward unknown cases = 191)
 
Motive
 Hurt physically        29.2    41.9    21.6     7.3    19.7
 Hurt symbolclly        16.2    16.7    20.7    36.2     9.7
 External gain  14.6     9.1    17.3    14.0    13.1
 To feel better  6.2     5.3    14.8    32.6     9.1
 Instinctual     2.6     0.6     1.1     0.0     0.6
 Prosocial motive        8.9    15.8    15.7     3.7    32.5
 No motive shown        22.3    10.6     8.9     6.3    15.3
             n =         582     341     440     301     320
 
     (Chi square = 473.9, df = 24, p < .0001, motive unknown cases = 316)
 
 
Intent
 Maliciousness  61.2    68.7    46.0    60.3    71.6
 Inconsiderate  18.9    25.9    48.9    34.1    15.0
 Thoughtlessness         8.8     4.3     4.0     4.0     2.1
 Reaction       11.1     1.1     1.1     1.7    11.3
             n =         603     348     446     302     327
 
           (Chi square = 244.6, df = 12, p < .0001, intent unknown cases = 274)