Content uploaded by Lynn M. Zoch
Author content
All content in this area was uploaded by Lynn M. Zoch on Jul 27, 2017
Content may be subject to copyright.
Review
of the
Literature
Quarterly
Vol.
75,
No.
4
Winter 1998
762-775
WOMEN
MAKING
NEWS:
GENDER
AS A
VARIABLE
IN
SOURCE SELECTION
AND USE
By Lynn
M.
Zoch and Judy VanSlyke Turk
An important dimension of
agenda
setting
is the
way in which the
media
frame
the issues
and events the]/present
to
the public.
This
article foe uses
on one of the most important dimensions of framing:
choice
of informa-
tion source—the selections journalists make from among
the
many
possible
and
potential holders
of
information
of
those sources zvhose
information
and
viewpoints will actually be included
in
the news.
In
particular, this content analysis often years of three southern newspa-
pers focuses on the inclusion of female
sources
in
nezvspaper stories
and
analyzes whether
the
gender of the
reporter affects
that inclusion.
The media
of
mass communication
are
among
our
most powerful
social institutions, with
the
capacity
to set the
public agenda
by
attaching
salience to particular issues and events they cover. An important dimension
of agenda setting
is
the way
in
which
the
media frame
the
issues and events
they present to the public' This article focuses
on
one
of
the most important
dimensions
of
framing: choice
of
information source, that
is, the
selections
journalists make from among
the
many possible
and
potential holders
of
information of those sources whose information and viewpoints will actually
be included
in
the news. In particular, this article focuses on the inclusion of
female sources
in
newspaper stories.
A recent study
by the
Freedom Forum indicates that women
are
seldom used as sources for stories ofnational or international importance, but
rather are quoted as victims or because
of
their relationship with
a
male who
is central
to the
story.
The
study also found that female reporters were
no
more likely than their male colleagues
to
quote other women.- Thus, this
research focuses
on
gender
as a
primary variable
not
only
of
source
but of
reporter. Finally,
the
article explores
the
connection between
the
holder of
information
and
power: the absence
of
women as sources would reflect their
powerlessness, their symbolic annihilation
by the
media.'
Agenda
S^ffiMg.
The media clearly aremore than
a
mirror of or conduit
for the concerns and issues of others. There is significant research support
for
an agenda-setting model: that the public learns both facts and the salience of
those facts from
the
media.
The
literature supports
the
belief that public
perceptions
and
opinions toward issues/topics
and the
individuals
who
espouse those issues
are
shaped
by the
media, leading
the
public
to
view
certain issues/topics
as
more important than others."*
Lynn
M.
Zoch
is
an assistant
professor
and Judy VartSlyke Turk
is professor
and
dean
in
the
College
af journalism and Mass Communications,
the
University of South Carolina,
Columbia,
SC.
762
I0URNAUSM&
MASS,
CoMMtwKvii ION
The media, "in the process of transmitting others' concerns and issues
... reworks and translates them to focus attention and structure cognitions."^
From the array of available information, the media select for dissemination
those pieces of information reporters and editors think are important, the
media's
Agenda of salient information.^
While the media are not the only influence upon our sense of what's
important in our environment, the media do play a central role/ But the
world the public sees through the media's eyes
is
not an objective account
of
events, people, places, and issues. "While news may be viewed as a window
on the world through which Americans learn of their institutions and their
leaders, it is a window that reflects largely the media's own construction of
reality."*
If the media do indeed pass along their construction of reality, their
own agenda of what is important, the question then becomes, who sets the
media's agenda? Turk writes: "The sources of the raw material of informa-
tion upon which journalists rely and from which they choose what to use may
ultimately have as much to do with the media's agenda as the selection
processes of the journalists themselves. News is not necessarily what hap-
pens but what
a
news source says has happened because the news doesn't
'happen' until there is an exchange of information" between journalists and
their sources.^
Framing the Media Agenda. A frame is defined as a "schemata of
interpretation" which allows an individual to make sense of information or
an occurence,'" The media give their audiences frames to organize and
facilitate understanding. "(F)rames select and call attention to particular
aspects of the reality described, which logically means that frames simulta-
neously direct attention away from other aspects."''
In an article that focuses on political campaigns, Williams, Shapiro,
and Cutbirth point out both the importance and the dangers to news
consumersof this framing the
news.
They alsonote the importance of various
other message variables in addition to content—such as source attribution
and placement of stories—in the framing of a story.'~
Kosicki and Pan add to that list of framing variables when they suggest
that "media also have effects on the way issues are framed through the
choices of journalists, institutional traditions and workways, occupational
norms and values, and actions of policymakers who may be chosen as
sources."'-*
In the context of this article, the main variable of concern in the way the
media frame a story is the choice of sources. When a source is a woman, the
interaction of topic and status of source also are of concern in identifying the
role of the female source within the context of the story. It should also be
remembered that it is individual reporters who pick sources, decide what
questions
to
ask, which quotations to use and which to ignore, and what tone
the story should takeJ"' Thus how the gender of the writer affects the choice
of source and how much information is attributed to the source also are
considered important in this study.
Preference in Use of News Sources. Official sources, such as govern-
ment officials or police, are often preferred by journalists.'^ This may be
especially trueatsmall-or medium-circulation papers where the relationship
between reporters and their sources can be part of the power structure of the
community.""
Shoemaker and Reese write that "(s)ources have a tremendous effect
on mass media content, because journalists can't include in their news reports
WoMFN
MAKING NEWS:
GENUtkA^
A VAHIABLI: IN
SOURCL
SELECTION
AND USE 763
what they don't know... (sources) may also influence the news in subtle ways
by providing the context within which all other information is evaluated...
and by monopolizing the journalists' time so that they don't have an oppor-
tunity to seek out sources with alternative views."'"
Research shows that reporters repeatedly go to sources who are like
themselves. And journalists tend not to select or to quote sources who refute
their own ideas.'" Stempel and Culbertson suggest that a source's
assertiveness, credibility (as determined by the journalist), accessibility, and
quotability can affect both
a
source's prominence (frequency of mention) and
dominance (tendency to be quoted rather than paraphrased or just written
about) in news coverage.'"
Of course journalists "not only talk with those who are directly
involved (in news events), but they may also get information from sources
only indirectly associated with the event or reactions and opinions from
'people on the street.' But not all sources are equally likely to be contacted by
journalists—those with economic or political power are more likely to
influence news reports than those who lack power."^"
Cans argues that heavy reliance by media on officials representing
both government and private-sector entities presents
a
very specific (official)
picture of society and its institutions.-' Ht)w reporters gather news and the
types of sources they use are important because in framing that picture
journalists determine not only what informafion is presented to the public
but what images of society are presented.--
Women and the Poiver of News Sources. It is an unreal image of
society that is presented when women, who represent 52 percent of our
population, are relatively absent as news sources. And it is an unreal image
presented when women are rarely quoted on issues of national or interna-
tional importance, or are seldom quoted on the front page of
a
newspaper like
the
Nexv
York TimesP The issue of women as news sources, and the media-
presented image of power and legitimacy associated with those sources, is
another area of examination in this study.
A number of studies conducted over the last several decades have
found women badly underrepresented as sources in newspapers and on
televisions newscasts. The emphasis continues to be on mate authorities and
officials.^''
Reporters often give time pressure as a reason for lack of diversity in
news
sources.
They contend that the availability of
a
source is sometimes the
most important criterion to choice of that source.^'' Other research has shown
that newsgathering routines and peer group pressure within a news organi-
zation can affect a reporter's use of sources.^^
Critical theorists, on the other hand, blame gender inequality in the
media on power relations and the way they are embedded in the political and
social order in this country.-^ Bybee, for example, argues that power is
connected to knowledge in studying the social construction of women in
news.
He sees what he calls power/knowledge as "an integrated, mutually
reinforcing discourse of truth" that both keeps the media consumer from
seeing and hearing women's voices presented in the news, and women from
presenting the news.^"
One reason "elites" are used so often as sources, is that "(D)ue to their
centrality in power systems, they can supply a great deal of information
without unduly taxing their organizations or the resources of journalists.
Theyalsoare more likely
to
meet standard definitionsof reliability, trustwor-
thiness, authoritativeness and articulateness."-''
764
jauRNALKM
& MASS COMMUNICATION QUAHTFRIY
The reporter's use of sources creates a representation to the reader of
who has information important enough to cite in the news story. In this way
news
is a
representation of power and authority within the society.""'Feminist
researchers believe that the inequitable distribution of power befween men
and women creates a system where women "lose their voices" and become,
in effect, "invisible."^' Perhaps they are invisible, too, to reporters searching
for sources.
Within the context of the media, information or "news" indicates to
the reader who is in possession of knowledge that is important enough to
report. "At the same time that it informs about who are the authorized
knowers, it suggests, by relegation to a minor role and by omission, who is
excluded from having a say in important matters.""
From the review of the literature, several research questions emerged
that linked framing, source role in framing, and gender of bofh journalist and
source.The research questions examined in this study were:
(1) Do journalists still favor official sources of informa-
tion in their reporting of the news, as the literature indicates has
traditionally been the case?"
(2) Are the sources, and especially the official sources,
upon whom journalists rely still predominantly male, as has
been the finding in past studies?
(3) Does the gender of the reporter make
a
difference in
this tendency to rely upon official sources? Are male and
female journalists equally likely to rely on official sources?
(4) Does the gender of the reporter make
a
difference in
the extent to which women are used as sources? Are male and
female journalists equally likely to use women as sources?
(5) If journalists are, indeed, heavily dependent upon
official sources, how does this affect the media's construction of
realify, the "worldview" presented to media audiences and
consumers?
This study analyzed a sample of stories published between 1986 and
1996 on the front pages of the "A" (first section) and "B" (metro/local)
sections of three daily newspapers in the southeastern United States: the
Charlotte Observer
in Charlotte, North Carolina; the Stale in Columbia, South
Carolina; and the Augusta Clmmick' in Augusta, Georgia. These particular
newspapers were selected because they provided representation of different
circulation sizes—large, medium, and small, respectively^''—as well as dif-
ferent states.
Alternate years were sampled, and the sample alternated between
even-numbered and odd-numbered months. The new.spaper published on
the first day of the selected month was chosen, and all stories on the front
pagesofthe"A"and"B"sectionsof those
issues,
including the jumps of those
WoMLN
MAKING
NEWS:
CENOEI^ AS A
VAHIAKII
IN SOURCE SILECTIONAND USE
L/eVi'lopint'nt
o
Methodol-
ogy
765
Findings
and
Discussion
stories continued onto a subsequent page, were analyzed. Selecting the first
day of the month provided a sample that represented all seven days of the
week.This
samp!
ing method yielded a total of 1,126 stories.The stories were
quite evenly divided among the three newspapers:
341
(30.3%) of the stories
were published in the [Charlotte]
Observer,
398 (35.3%) in the [Columbia]
State, and 387 (34.4%) in the [Augusta]
Chronicle.
The unit of analysis was the news story. Information coded for each
story included:
(1) date and year of publication;
(2) placement variables (page, column width of head-
line,
location on the page, length of story, and use of photo or
other illustration);
(3) topic covered and its scope (international, national,
state,
or local). Topics were selected based on a review of the
newspaper content analysis literature;^^
(4) variables on the source of the story (whether the story
was generated by staff or wire service, if it was bylined or not,
and gender of the byline if one was used);
(5) number and identification of sources to whom infor-
mation in the story was attributed. For each attribution, coders
recorded whether the source was identified by name, the gen-
der of the source, any organizational or official title used for the
source, any identification of the source's employer or the orga-
nization the source represented, and the length of the material
attributed to the
source.
Length was measured in picas because
that is currently the dominant measure used in the newspaper
industry.
The coding manual was pretested extensively and revised several
times before coding began to increase intercoder and intracoder reliability.
Intercoder reliability also was enhanced by limiting the coding to only four
individuals, the two researchers and two graduate students trained by the
researchers.^''
766
Data on the Stories. Almost three quarters of the stories (72.4%) were
news,
while
27.6%
were features.-*^ The
Chronicle
used proportionately more
news stories (78.3%) than the State (71.3%) or the
Observer
(66.9%).
The largest number of stories—465, or 41.4"/"—were local stories,
dealing with city, town, village, or county topics and issues. Given the
tradition in the United Statesof
local,
"hometown" newspapering, with most
newspapers clearly identified with and circulating within a particular com-
munity, that local emphasis was not surprising. That pattern held across all
three newspapers, although the
Chrojiicle,
the smallest paper, was more local
than the other two papers:
46.6%
of its stories were local, compared to 40.0%
in the
Obserzvr
and only 37.4"/(i in the State.
Neither was it surprising that the smallest number of stories—104, or
only 9.3%—dealt with international topics and
issues.
U.S.
newspapers carry
louRNAUSM
&
MASS
COMMUNICAIION QIIAKILRLY
much less news about events in other countries and on other continents than
donewspaperselsewhere,
a
phenomenon U.S-journalists sometimes explain
by noting that the
U.S.
is less dependent upon actions and events beyond its
national boundaries than most other countries in the world.
Almost half of the
stories—551,
or 49.6%—were short: fewer than 90
lines of type. "Shorter is better" seemed particularly true in the Augusta
paper:
42.2%
of the stories analyzed from the
Chronicle
wero shorter than 60
lines of type, and it ran the smallest number of long stories.
Data on Reporters. More stories came from the newspapers' own
staffs
(815,
or 73.2%) than from wire services or syndicates
(295,
or
26.5%).
The
Chronicle
used proportionately more wire service or syndicated stories than
the other two
papers:
34.7%
compared to only
23.1%
in the
State
and
21.3%
in
the Ofescri'iT.^**
Bylines were common: 84.8% of the stories carried the byline of the
reporter or reporters who wrote the story. Of the
954
stories that had bylines,
more than half—592, or 55.6%—carried the byline of
a
male, while only
26.3%
(280) had female bylines. The gender of the reporter in the remainder of
stories with bylines {192, or 18.0%) could not be determined from the
reporter's name or initials.
There were substantial differences among the three newspapers:
60.3%of the bylines in theSffl/c were thoseof women, compared to only
26.8%
in the Observer and 21.8% in the
Chronicle.'^''
It was interesting although
probably not significant that the largest number of bylines whose gender
could not be determined (28.1%) was in the
Chronicle.
Although the change was not statistically significant
(x^=11.19,
df=10,
ns),
the proportion of female bylines increased noticeably over the eleven-
year-period covered by the coded stories (from 21.3% in 1986 to 41.8% in
1996),
perhaps an indication that the proportion of women on the newspa-
pers'
staffs had increased.""'
Women's bylines appeared more frequently than men's over stories
whose topic was courts/court cases, crime, religion, health/science/medi-
cine,
and human interest profiles. Women were noticeably more likely than
men to write local stories
(54.3%
of all local stories carried female bylines) but
male bylines predominated over international, national, and statewide sto-
ries
(x^-15.01,
df-4, ;)=.005).
Data on Sources and Their Status. The 1,125 stories analyzed included
3,624
attributions of information to sources.*' The number of sources per
story ranged from zero
(96,
or 8.6%, included no source attributions) to
16
in
one story. The mode of sources per story was two.
The Chronicle was most likely of the three newspapers to run a
story without source attribution
(14.5%
of the stories mentioned no sources)
while the Observer was least likely, using no sources in only 2.9% of its
stories.
A
large majority^-89.3%—of the sources were named,
a
finding which
may be of comfort to journalists and media critics concerned about the extent
to which media rely upon unnamed or anonymous
sources.
The Observer was
most likely to name its sources (91.8% of all sources were named), and the
State was least likely, naming sources 86.9% of the time.
Most of the sources (86.3%), whether they were named or not, were
identified by title. All but
14.5%
of those with titles had "official" titles. Half
(50.7%) had middle management titles: director, manager, senator, council-
man, detective, etc. Another 25.0% had top management titles: president,
chief executive officer, chairman/woman,
chief,
general, owner, etc. An-
WoMEN
MAKING
NEWS:
CFMUR
AS A
VARIABU:
IN SGURCK
SeLEajtw
AND USE
TABLE 1
Combined Demographics
Tvpe of Story
News
Feature
# of Stones
# oi Sources Used
Sexof Snnrrp
Female
Male
Undetermined
Scope of Storv
International
National
State
Local
Mixed
Topic of Storv
Governance
Courts/Cases
Crime
Military
Education
Business/Finance
Culture
Religion
Health/Medicine/Science
Political/Campaigns
"Common Man" Profile
Other
Male Bvline
592
2,104
18.9%
(398)
68.97o (1,449)
12.2'^
(257)
55
87
156
242
51
153
36
60
24
45
47
47
8
56
39
40
33
Female Bvline
280
894
26.5'!<, (237)
65.87.
(588)
7.7% (69)
17
28
60
152
23
64
20
38
10
22
15
21
7
37
11
19
16
Total
814
311
1,125
104
164
280
465
111
271
66
135
49
83
80
88
20
130
63
77
64
Note:
Since some of the stories did not carry
a
byline,
the "Total" column may be greater tban the sum
of male and female reporters.
other 6.5% had the title "spokesperson," generally indicating a public rela-
tions official within an organi/ation, and the remaining 33% were identified
simply as an "official" or "officials."
The
Charlottf Observer
was most likely of the three newspapers to run
a story quoting nonofficial sources:
47.4"/"
of the nonofficial sources used
across the three papers were in
Observer
stories, and almost
20%
of all sources
mentioned by the
O/jscnitT
were nonofficial. The Augusta
Chronicle
was most
likely to run stories that attributed information to officials.
The employer of almost
8
in 10 sources (79.5%) was identified. Sources
were most likely to work for a government agency or department (63.0"/i)) or
for
a
business
(19.6"''a).
Another
11.3'!^)
of the sources worked for not-for-profit
organizations, and the smallest number,
6.1"'ii,
worked for an association.
The researchers concluded from this data that the answer to their first
researchquestion, "Do journalists still favor official sources of information in
768 &
MASS
CUMMUNKATION QLMKTEKLY
their reporting of the news?" is yes. Journalists for these three newspapers
DO heavily favor {lfficial sources of information in their reporting of news.
And because male officials significantly outnumbered female officials
as
sources, it appears journalists heavily favor official MALE sources of infor-
mation. The gender of the reporter seems to make little or no difference in this
preference: male and female reporters gravitate equally to official sources in
their reporting (x-^=3.449, df-4,ns).
Data on Source and Reporter Gender. Almost 7 in 10
of
the sources
used were male: 2,466
or
68.0"''". Only 746,
or
20.671.,
were female.
The
remaining
11.4'%i
could not be identified as to gender, either because initials
were used instead of the first name or because the first name was one that
could be used by either a male or a female.
The Charlotte Observer used proportionately more female sources
(28.1%
of all its attributions were to female sources) than either the State in
Columbia (18.8"4) or the Aiigustii CUnmkk (13.67o).
Of all the attributions to female sources, 26.1"/i> (195) were in stories
that used between one and throe sources, compared
to
33.6%> (829)
of
all
attributions
to
male sources occurring
in
stories that used this same small
number of sources. In "grading" the importance
of
information, one might
argue that the fewer sources per story, the more "burden of
proof"
on each
of those few sources to lend veracity and credibility to the information in the
story. If that is the case, however, then the over.ill proportion
of
female to
male sources (746 female sources compared
to
2,466 male sources) would
indicate women more often than men were relied upon for this veracity and
credibility.
If was not surprising, given the preponderance of male sources used,
that male sources outnumbered female in stories dealing with every topic.
The representation was most balanced in coverage of education (56.7"'ijof the
education sources were male, 34.7''''o female) and culture
(56.Q'Vn
male, 32.9%
female).
Male sources also outnumbered female sources regardless
of the
scope of the story. The imbalance was most evident in international stories,
with
61.5%
male sources and only
14.0%
female, while in local stories, women
were used as sources
a
bit more frequently: 24.87o of the time. The smallest
proportion of attributions to both female (6.7'^) and male sources (8.9"/;,) was
in international stories. And the largest proportion for both female and male
sources (47.8% and
39.3%,
respectively) were in local stories, not surprising
given the preponderance of local stories.
Most of the attributions to sources were quite
brief.
Four in
10
(40.4%)
were
6
or fewer lines of type in length, and
69.97..
were
11
or fewer lines long.
Women sources were quoted most briefly, 6 lines or
less:
42.27u of all female
sources used were
in
this category, compared to 36.77..
for
men. Female
sources were also mentioned most frequently in the shortest stories, those 6
lines or less in length:
43.5"'o
of ail female source use was in these shorf stories,
compared to 39.9% for men.
The difference between
the
number
of
male and female sources
quoted at length was more striking, with
84.3%
of those quoted for
30
or more
lines being men. The same was true
for
story length with 76.4%
of the
attributions to men in stories 24 to 29 lines
of
type long, compared to only
17.17u of the female attributions.
Since length is one cue journalists give to importance in
a
story (longer
is more important),
it
appears, then, that male sources were quoted more
frequently in the longer, more important stories, and were more trusted than
WOMEN
MAUNC
Niwi:
CEHDER
Ai
A
V.^RIABU
IN SOURCE
SoirniiN
AND USE
Total #
Mode per Storv
Identified bv Name
Yes
No
Sex of Source
Female
Male
Undetermined
Title Description
Spokesperson
Top Manager
Middle Manager
Nonofficial
Official
lob Type
For Profit
Business
Not For Profit
Organization
Association
Gov't Agency,
Department,
Commission,
or Committee
Charlotte
1,322
2
91.8%
8.2%
28.17o
62.5'!i.
9,5';i.
4,3%
24.8%
48.4%
19.7%
2.8%
22.2%
10.67o
8.27o
TABLE
2
Source Data
Columbia
1,443
4
86.9%
13,1%
18.87a
67.87.,
13.37o
7,1%
26,6%
52,8%
11.67o
1.87a
18,2%
13.8%
3.9%
64.1%
Augusta
991
2
89,5%
10,5%
13.67o
75.5%
10.97o
8.3%
23,3%
50.9%
11.7%
5.87o
20.77«
6,3%
7.27o
65.8''/l.
Combined
3,756
2
89.3%
(3,222)
10.77a (387)
20.6%
(746)
68.0%
(2,466)
11,4%
(412)
6,5%
(204)
25.07a (789)
50.7%
(1,598)
14.57,.
(456)
3.3%
(103)
20.2%
(586)
10.67o (306)
6.37a
(181)
62.9%
(1,821)
770
women to give the longer, more in-depth quotes, Chi square statistics
showed a high level of significance (;7<.000) when sex of the source was cross-
tabulated by length of attribution.
Just as length is an indicator of the importance journalists attach to
information, so
is
placement of
a
story in the newspaper. The front page of the
"A"
section is THE most important placement, and generally the farther back
in the newspaper a story appears, the less important it's been judged by the
editor. In this study, almost 7 in 10 of the sources (69,2%) to whom informa-
tion was attributed in stories on the "A" front page were men, compared to
only 17.37o women (x'=23,19, df=l,
p<.OOQ).
Additionally, 57.77a of all male
source mentions were on the "A" front page, compared to only 47.57i> of the
attributions to women.
Based on analysis of this data, the researchers' second research ques-
tion, "Are the sources, and especially the official sources, upon whom
journalists rely still predominantly male?" can also be answered in the
affirmative.
Women reporters were more likely than men to attribute information
to sources with middle management official titles while stories with male
bylines tended to favor sources with top management, spokesperson, or
jounNAUiM
& Alisi
CvMMUNiCAriot^
other official titles. Mate and female reporters were equally likely to quote
non-official sources: only 14.5"/i> of stories with female bylines and 14.6% of
maie-hylined stories attributed information
to
individuals without official
title or position.
Thus,
the researchers' third research question, "Does the gender of the
reporter make a difference in the tendency to rely upon official sources?" is
answered no; male and female reporters share a preference for and reliance
on official sources.
There was little difference between male and female reporters as to the
number of sources mentioned in each story. But women reporters were more
inclined than their male colleagues to include female sources in their stories.
More than one quarter (26,5%)
of
the sources used
in
the 280 stories that
carried female bylines were women, compared
to
only 18.9'/o
in
the 592
stories with malehylines (x'=16.22, df=l,
p<,000).
Male reporters were more
likely than females to attribute information to sources whose gender could
not be determined:
12.2%
of the sources in stories with male bylines were of
indeterminate gender, compared to only 7,7% in female-bylined stories.
The fourth research question
("Does
the gender of the reporter make
a difference in the extent to which women are used as sources?") thus can be
answered in the affirmative. Women reporters are more likely than their male
colleagues to attribute information to female sources.
The fifth research question addressed in this study asked, "If journal-
ists
are,
indeed, heavily dependent upon official sources, how does this affect
the media's construction
of
reality, the "worldview" presented
to
media
audiences and consumers?"
Based
on
analysis
of
this data, the researchers would suggest
the
following impacts:
First, the "schemata of interpretation" the media, at least these three
newspapers, present to media audiences and consumers would lead them to
believe that news is made and information controlled almost exclusively by
men acting
in
some official capacity, with official status. This frame calls
attention to what men do and say and as Entman suggests, directs attention
away from women. Women were infrequently cited as sources,
a
signal to the
reader that they are relatively unimportant in both public and private sector
activities and events. A media consumer might infer that this lack of impor-
tance is the result of women not holding positions of authority and/or their
lack
of
credible, valuable information. And
of
course we know that's not
necessarily reality. Or the truth.
But do media audiences recognize that the reality with which they're
presented by the media isn't necessarily either real or truthful? Do they care?
We know from the literature that journalists are surprisingly unaware of the
frames they construct, but we know considerably less about how readers
deconstruct those frames. Further study—perhaps using
a
quasi-experimen-
tal design that measured reader interpretation
of
two identical stories but
with one attributing information to male sources and the other to females—
would be instructive.
Second,
to
the extent that information
is
power, and positions
of
power create opportunities for access to information, the media worldview
would have audiences and consumers believe that women are virtually
without power and thus have no access to information that would be of use
to the public. While there certainly are women who belie this particular
Conclusions and
Implications
for the Media's
Framing
of Reality
WoME.\
MAKINC
News;
ir.R
AS
A
V.'^IUABLL
yf.'
SOURCF.
Sl:-l.cno^ .vju
USE 771
construction of reality, the researchers contend that women are not repre-
sented in the higher echelons of organizations in proportion to their total
number in the population, and therefore are less likely to be identified as
viable
sources.
As mucii as one might like
to
blame the media tor NOT telling
it like it is, in tbis instance the media's "schemata of interpretation" may
actually be mirroring reality, as unfortunate and inequitable as that reality
might be. The fnime in this instance facilitates an understanding that is not
necessarily a distortion.
Third, the researchers expected that female journalists migbt attach
greater credibility to female sources tban do male reporters and thus use
proportionately more of them in their stories. After all, since they have had
theopportunityin their own professional lives
to
see that
a
woman journalist
can be just as competent as a male counterpart, they might be expected to
carry tbat attitude into their selection of sources and seek out the women—
spokespersons, officials, middle managers, and top managers—rather than
the men when both could be expected to have equally valuable information.
And in these three newspapers, at least, that did prove to be the case to a
statistically significant level
{p<.000).
The study found, however, that female reporters are more likely to
quote middle management sources—of either sex—while their male counter-
parts quote those considered top managers. Whik' the researchers did not
conduct interviews with reporters and thus can only speculate on the reasons
for this, two possible explanations come to mind. It is possible that many
female reporters are denied access to the highest levels in organizations or
government, and cannot reach those managers or officials for comment. It is
equally likely, however, that a female reporter with experience in the casual
slights of bureaucrats would consciously make the choice to speak to mid-
level sources, those most readily accessible, rather than take the chance of
being rebuffed by those in top-management positions. Tbis is an area in
which further research, particularly interviews with reporters of both sexes,
needs to be conducted.
A final point: the "maleness" frame that media employed in tbis study
extended not only to attaching greater importance to male sources but also to
diminishing the importance of women even when they were used as sources.
Across all three newspapers, female sources were quoted less frequently in
stories on the front page of the "A" section than on the front of the metro or
local section
{/K.OOO),
no matter the sex of the reporter. In addition, both the
length of attributions by female sources and tbe length of the stories in which
female sources appeared were shorter than those of male sources. Since
placement and length are cues journalists give to importance in a story
(longer is more important), it appears, then, that female sources were consid-
ered less important than males in the longer, more important stories, and
were trusted less to give the longer, more in-depth quotes.'*^
From these findings the researchers conclude that women journalists,
by and large, seem slightly more sensitive to the "symbolic annihilation of
women" in tbeir choices of news sources than their male counterparts. And
that may be positive news for journalism educators and editors, botb of
whom frequently explain gender diversity efforts by suggesting that adding
morewomen to newspaper staffs will enlarge and make more representative
the frame that's used to report the news. At least at these three newspapers,
their efforts appear to be working.
What does not appear to be working is any effort to increase the
visibility and thus the power of women journalists. While the numbers of
772
/OUR\AU.SM 6-
MA^S
QlMMLI.MtViniW
QLIAKILKLt
female bylines increased over the ten-year period of this study, the research-
ers found female reporters are still more likely to cover local stories, while
their male counterparts cover
state,
national, and international news (p<.005).
In addition, women's bylines appeared more frequently than men's over
stories whose topic was courts/court cases, crime, religion, health /science/
medicine, and human interest profiles—all of which are considered to be
traditionally local, family-oriented, or "women's" stories.
The interrelationship of power, information, and gender that this
study's research questions addressed is surely too complex to be fully
explored, much less definitively so, in a study of three newspapers in the
Southeast, The authors believe more comparisons must be made particularly
focusing on different geographic areas and by extending study to other
media, perhaps television. While there is no certainty that differences will
exist in various areas of the country or between different media formats,
possible differences should be explored.
In addition, further research could triangulate the findings of content
analysis with interviews with both reporters and media consumers to de-
velop a more complete understanding of why certain frames are used by
journalists and how they are perceived by readers and viewers.
NOTES
1.
For a discussion of the four phases of agenda setting, see Maxwell
McCombs, "Explorers and Surveyors: Expanding Strategies for Agenda-
Setting Research,"
joiirnalism
Qiiartcrh/
69
(winter
1992):
H13-24;
and Maxwell
McComhsand Tamara Bell, "The Agenda-Setting Role of Mass Communica-
tion" in An Inle^imtcii Approach to Coniiminiaihoii Tlwori/ ami Research, ed.
Michael B. Salwen and Don W. Stacks (Mahvvah, N|: Lawrence Frlbaum
Associates, 1996).
2.
Debra Gersh Hernandez, "Are Women Being Annihilated by the
Media,"
Editor
&
Publisher.
1
July 1995.
3.
Gaye Tuchman, Making News: A Stiuiif in
the
Construction
of
Reality
(New York: Free Press, 1978).
4.
For an overview of the agenda-setting process and various relation-
ships that affect agenda setting, see James W. Dea ring and Everett M, Rogers,
Agenda
Setting (Thousand Oaks: Sage Publications, 1996), For
a
review of the
literature, sec Maxwell McCombs and Sheldon Gilbert, "News Influence on
Our Pictures of the World," in
Perspectives
on Media Effects, ed. Jennings
BryantandDolfZillman(Hillsdale,NJ: LawrenceErihaum Associates, 1986).
For a discussion of general media agenda-setting effects on the public, see
Jack M. McLeod, Lee B. Becker, and James E. Byrnes, "Another Look at the
Agenda-Setting Function of the Press," Communication
Research
1 (April
1974):
131-65.
5.
Donald
L.
Shaw and Maxwell
E.
McCombs,
The
Emergence of American
Political Issues: The Agenda-Setting Futictioi: of the Press (St. Paul: West Publish-
ing, 1977), 151,
6. Bernard C. Cohen in
The
Press
and
Foreign
PoZ/Vi/(Princeton: Princeton
University Press, 1963),
13,
stated that "The press is significantly more than
a purveyorof information and opinion- It may not be successful much of the
time in telling people what to think, but it is stunningly successful in telling
its readers what to think about."
7.
Walter Lippman,
Ptibtic
Opinion (New York: M^icmillan, 1961), 203.
WOMEN MAKOIG
IVEIVS.-
GENDER AS
A
VARIABLE IN SOUHCH
SEucnoN
ANP USE
773
8. Judy VanSlyke Turk, "Subsidizing
the
News: Public Information
Officers and their Impact on Media Coverage
of
State Government" {Ph.D.
diss.,
Syracuse University, 1985), 34.
9. Turk, "Subsidizing the News,"48.
10.
Erving Goffman,
Frame
Analysis (NY: Harper & Row, 1974), 21.
11.
Robert M. Entman, "Framing: Toward Clarification
of a
Fractured
Paradigm,
"
Journal
of
Communication
43 (autumn 1993): 54.
12.
Wenmouth Williams Jr., Mitchell Shapiro, and Craig Cutbirth, "The
Impact
of
Campaign Agendas on Perceptions
of
Issues," in
Aj^enda
Scttin;^:
Readings
in
Media,
Public
Opinion,
and
PoUcymaking,
ed. David
L.
Protess and
Maxwell McCombs (Hillsdale, NJ: Lawrence Erlbaum, 1991), 253.
13.
Gerald
M.
Kosicki
and
Zhongdang
Pan,
"Framing Analysis:
An
Approach
to
Media Effects," (paper presented
at the
annual meeting
of
International Communication Association, Chicago, IL, 1996).
14.
Adapted from
W.
Breed, "Social Control
in the
News Room:
A
Functional Analysis,"
Social Forces 33
(1955) in
Media
Management:
A
Casebook
Approach.
Stephen Lacy, Ardyth
B.
Sohn, and Jan LeBlanc Wicks (Hillsdale,
NJ: Lawrence Erlbaum, 1993).
15.
For more in-depth discussion
of
these issues, see D. L. Paletz and
R.
M. Entman,
Media,
Power,
Politics
(New York: Free Press, 1981); Charles
S.
Steinberg,
The Information
Fstablishincnt:
Our
Government and the Media
(New
York, Hastings House, 1980); Oscar H. Gandy,
Beyond Agenda
Setting: Infor-
mationSubsidiesand
Public
Poiicif(NaTwood,N]: AhlexPubhshmf;,,'\982); and
Mark Fishman, Manufacturing the News (Austin, TX: University
of
Texas
Press,
1980), 50.
16.
John Soloski, "Sources
and
Channels
of
Local News," journalism
Quarterly
66 (winter 1989): 864-70.
17.
Pamela
J.
Shoemaker and Stephen D. Reese, Mediating the
Message:
Theories
of
Influences on Mass Media Content
(New York; Longman, 1991), 150.
18.
Angela Powers and Frederick Fico, "Influences on Use of Sources
at
Large U.S. Newspapers,"
Nezospaper Research
journal 15 (fall 1994): 87-97.
19.
Guido Stempel
HI and
Hugh Culbertson,
"The
Prominence
and
Dominance
of
News Sources
in
Newspaper Medical Coverage," journalism
Quarterly
61
(autumn 1984): 671-76.
20.
Shoemaker and Reese, Mediating
the
Message,
151.
21.
Herbert Gans,
Deciding
What's
News
(New York: Vintage Books, 1979),
145.
22.
Tuchman, Making Neivs.
23.
Hernandez, "Are Women Being Annihilated."
24.
See Jane Delano Brown, Carl
R.
Bybee,
Stanley T.Wearden, and Dulcie
Murdock Straughan, "Invisible Power: Newspaper News Sources and
the
Limits
of
Diversity,"
journalism Quarterly
64 (spring 1987): 50;lY(nrfoi(i
Dress-
ing
on the
Set:
Women and Minorities
in
Television.
A
Report
of the U.S. Commis-
sion on Civil Rights, August 1977, updated January 1979. U.S. Government
Printing Office; Carol
M.
Licbler
and
Susan
J.
Smith, "Tracking Gender
Differences: A Comparative Analysis of Network Correspondents and Their
Sources,"
Journal
of
Broadcasting and Flectronic Media 41
(winter 1997): 58-68;
and Kathleen
A.
Hansen, Jean Ward, Joan L. Connors,
and
Mark Neuzil,
"Local Breaking
News:
Sources, Technology and News Routines,"
journalism
Quarterly
71
(autumn 1994): 564.
25.
Marc Cooper and Lawrence
C.
Soley, "All the Right Sources,"
Mother
Jones,
February-March 1990, 20-27.
26.
Powers and Fico, "Influences on Use of Sources."
774 jouKNMJSM &
MASS
COMMUMCATJON
QLMKrERiy
27.
See
I.
Diamond and
L.
Quinby,
Foucault
and
Feminism
(Boston: North-
eastern University Press, 1988); M. Foucault,
Power/Knowledge
(New York:
Pantheon, 1980); and D. Scholle, "Critical Studies: From the Theory of
Ideology to Power/Knowledge," Critical Studies in Mass
Communication
5
(March 1988):
16-41,
28.
Carl R. Bybee, "Constructing Women as Authorities: Local Journal-
ism and the Microphysics of Power,"
Critical
Studies in Mass
Communication
7 (September 1990): 199.
29.
Brown et al., "Invisible Power," 46.
30.
Richard V. Ericson, Patricia M. Baranek, and Janet B.L. Chan,
Negoti-
ating
Control:
A Study of News
Sources
(Toronto: University of Toronto Press,
1989).
31.
LindaA.M.Perry, "Differences, Dominance, and Dialectics: A Call for
Change,"
in
Differences that Make
Difference:
Examining
the
Assumptions
in
Gender
Research,
ed. Lynn H. Turner and Helen M. Sterk (Westport, CT:
Bergin & Garvey, 1994), 187-94.
32.
Ericson, Baranek, and Chan,
Negotiating
Control,
4.
33.
For this research, official sources were operationalized as those
sources holding a position of authority and responsibility in, or representing
as a spokesperson, an organization or governmental agency.
34.
According
to the
1997
Editor
ami
Publisher International
Yearbook,
as
oi
30 September 1996 the daily circulation of the
Charlotte Observer
was 236,050
and Sunday circulation was 301,412; the State |Cokimbia, SC| was 122,053
and Sunday circulation was
160,381;
and the Augusta
Chronicle
was 90,581
and Sunday circulation was 100,606.
35.
For instance, see McCombs and Gilbert, "News Influence"; Shaw and
McCombs,
The
Emergence
of American
Political
hsues; Turk, "Subsidizing the
News";
Williams, Shapiro, and Cutbirth, "The Impact of Campaign Agen-
das";
Dan Berkowitz and Douglas W. Beach, "News Sources and News
Context: The Effect of Routine News, Conflict and Proximity,"
Journalism
Quarterly 70 (spring 1993): 4-12; and Soloski, "Sources and Channels."
36.
Intercoder reliability across the four coders was
83%,
and intracoder
reliability averaged 92'^.
37.
Stories were categorized as news if they had a timely news peg,
emphasized the traditional "5 Ws and H" news values, and were written in
traditional inverted pyramid news
style.
Stories were categorized as features
if they lacked particular timeliness and focused more on human interest than
on the traditional news values.
38.
The
Chronicle,
which as of October
1997
employs fewer reporters and
editors (63) than either the State (133) or the Observer (218), thus has less
capacity to generate its own stories.
39.
In October of 1997 the
Chronicle
employed 20 women and 43 men in
the newsroom, the State employed 44 women and 89 men, and the
Observer
employed 79 women and 139 men. The newsroom staffs are composed of
reporters and editors, and human resources personnel at the papers did not
provide a breakdown of solely male and female reporters.
40.
Because data regarding staff size and composition was not collected
for each year analyzed, that possible correlation cannot be verified.
41.
Each different source was counted in each story. If information was
attributed to the same source multiple times in the same story, that source
was counted only once.
42.
Chi square statistics showed
a
high level of significance
{p<.000)
when
sex of the source was cross-tabulated by length of attribution.
WOMEN
MAKING
NEWS: GENPER AS
A
VAUABLE W SOURCE SELECTION AND
USE
775
