Keeping the Gates' for Gatekeepers: The Effects of Wire News
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
By D. Charles Whitney and Lee B. Becker
'Keeping the Gates' for Gatekeepers:
The Effects of Wire News
Proportion of items in various vice, human interest, national, political,
categories found to have more international), McCombs and Shaw3 note
effect than any shared news values a Spearman's rho of .64between ranks of
seven news item content categories sup-
on choices by gatekeepers. plied by the wires and ranks of stories
selected by "Mr. Gates," and a Spearman r
An enduring concern in the study of of .80 for a replication study of the same
journalistic practice is the degree to which editor 17 years later.4
news is standardized. A number of com- Gold and Simmons, in a study of 24
mentators have noted that various con- Iowa daily newspapers relying solely on
straints reduce the variability of news one A P wire circuit for state, national and
available to audiences. Most of these con- international news found overall coeffi-
straints are tangible, concrete and rela- cient of concordance of .915 between ranks
tively well documented, such as time, of proportions of content supplied by the
'hews hole," or space, money, standard- wire service and ranks of proportion of
ized sources, organizational policy and content used by the newspapers in 13
craft norms.! Others are considerably less
apparent, and one such 'unseen" con- ' For time. scc Robert L. Jones. Vcrling C. Troldahl and J.K.
Hvistcndahl. "News Sclcction Patterns from a StateTTs-Wire."
straint is the subject of this paper. JOURhALlSM QUARTERLY. 38:303-12 (1961): and Guido H.
Two recent commentaries have reexam- Stempel 111. 'How Newspapers Use the Associated Press
Afternoon A-Wire." JOURNALISM QUARTERLY. 41:38&384
ined White's classic 1949 "Mr. Gates" (1964): for space. KC Gaye Tuchman. Mflking Nr
study of the news selection behavior of one r h r Con.\rrucrion of Rralir? (New York: Free P
David Manning White. 'The Gate-Keeper: A Case Study i n the
Midwestern wire news editor.2 Both have Selection Of News." JOURNALISM QUARTERLY. 27:383-390
argued that more remarkable than White's (1949): for money. see Edward J. Epstcin. New:v.from Nomhrre
(New York: Vintage. 1973): for standardiied sources. see
finding that the editor engaged in idiosyn- Warren Breed. "Newspaper 'Opinion Lcadcrs'and Processo o f
cratic, subjective selections and rejections Standardiiation." JOURhALISM QUARTERLY. 32277-284.
of news items was that the editor appar- 328 (1955); but for a contrary view. see Cuido H. Stempel 111.
'Uniformity of Wire Content in Six Michigan Dailies."
ently unconsciously mirrored selections JOURhA1.ISM QUARTERLY, 3745-48. 129 (1959): for policy,
already made for him by the wire services. see Brccd. "Social Control in the Newsroom: A bunctional
Analysis," Socirrl Forcr.\.33:32635 (1955). and John Dimmick.
Classifying the news available from the "The Gate-Keeper: A n Uncertainty Theory." Journnltsm
wire service into seven content categories M o n o g r a p h No. 37 (November 1974): for craft norms, sec
Tuchman. *Objectivity as Strategic Ritual." A m r r i m n Journal
(labor, accidents and disasters, crime and 4f ~lICl1JhR.V.77:wo679 (1972).
Paul M. Hirrch. "Occupational. O r ~ a n i i a t i o n a l and
W D. Charles Whitney is research assistant professor in Institutional Modclr in Mars Media Rocrrch: Towrrd an
the Institute of Communications Research at the Integrated Framework." in Hirrch. Peter V. Miller and k ,
(icrald Klinc. 4%. .SmreRir.$ ./or C'bmmunicarion Hr,wnrch
University of Illinois. Lee B. Bccker is associate (Beverly Hill,. CA.: Same. 1977). pp. 13-42. and Maxwell E.
professor of journalism at Ohio State University. The MFComL and Donald I.. Shaw. "Structuring the 'Ilnsccn
authors are grateful to the following graduate student Environment'." Journrrl , I / Cbmmuniwrron. 26: 18-21 (Spring
members o f the first author's spring 1979 xminar at 1976).
Ohio State University for assistance in fieldwork and ' Op. ,"I., p. 21.
data coding: John Liston. Louis Sanford. Paul Rolfes. ' P r u l Snider. "'Mr.
Gate%'Revisited: A 1966 Version crl the
Jayne Rolfes, Joseph McKnight, Francis Raj and Joy 1949 Case Study." JOURhAI ISM Q t l A R l r R l Y. rr:419-427
Gray. (1967).
60
Downloaded from jmq.sagepub.com by FELICIA GREENLEE BROWN on April 12, 2012Wire News Effects on Gatekeepers 61
categ0ries.s Whitney, for example, recorded several
Stempel, in a 1959 content analysis of such non-substitutions in story play in
wire stories used by six small Michigan dai- news scripts in a large metropolitan radio
lies. found agreement to be relatively low, newsroom.9
with only eight of 764 stories used by all A second and perhaps more plausible
papers and with overall agreement at 31% explanation of the correspondence
of stories across papers.6 His 1964 study of between wire copy provided editors and
21 metropolitan papers offers findings their editorial choices is that wire service
more directly relevant to the research editors and news editors in media outlets
reported here.’ Average use of A P A-wire share highly similar news values, and thus
items by all papers was 22% ranging from a a finding that each select news items in
low of 11% by a New York paper to 34% similar proportions in news content
by the Rochester (N.Y.) Times-Union. In categories merely reflects similarity of
the study period, the afternoon A P A-wire judgment.
transmitted 97 Washington, D.C.- If Gold and Simmons and the “Mr.
datelined stories (1 7% of all stories), 298 Gates” studies are correct, the wire services
U.S.-datelined stories (54%) and 159 “set agendas” for the news play “mix” of
foreign-datelined stories (29%). While the these various sorts of stories by
numbers and proportions of stories bear- transmitting them in varying proportions:
ing these datelines varied substantially, if 5% of what is transmitted is labor news,
proportions selected by papers within cate- then 5% of what is selected should be as
gories did not: 22% of the D.C.-datelined well. Also implied is that the proportions
stories, 20% of the U.S.-datelined stories in categories are consistent across time.
and 24% of the international stories were The proportion of news within the content
selected by the papers. type becomes a n added piece of
These findings suggest that wire service information for editors to use in making
editors, in broad terms, “set the news story selections, information that would be
agenda” for newspaper news editors, by absent if equal proportions of news were
suggesting the proper “news mix” and transmitted in each category.
proportions within news categories such as An adequate test of an hypothesis that
accidents and disasters, crime and vice and wire service editors’ assignment of items in
human interest news. Two possible varying proportions to news categories
explanations present themselves. The first influences other editors’ selections of a
is that the structure of each day’s wire file, subset of those items in similar
independent of proportions of content, p r o p o r t i o n s . t h e n , would r e q u i r e
influences editors’ selections. For example, variations in the wire file proportions
“soft news” such as human interest stories assigned to various news categories. Where
may be transmitted early in a wire cycle, the file is “stacked,” or proportioned much
leading to a higher proportionate selection as news is routinely transmitted, editors’
for such early-moving copy, while “hard selections should mirror proportions
news” might move closer to deadline.8 transmitted. Where the file is ”balanced,”
While in an absolute sense editors might several outcomes, amenable to varying
value later-moving stories more highly as interpretations, a r e possible: a ) if
news, they might not ordinarily alter selections mirror the “balanced” nature of
previously-made news j u d g m e n t s . the “balanced” wire file, the outcome
strongly supports the notion that wire
‘ David Gold and Jerry L. Simmons. %cws Seknion
editors “set the agenda“ for news editors;
Patterns Among Iowa Dailies.” Public. Opinion Querrcrh
m425430 (IW5). b) if selections instead follow the
* Op.62 JOURNALISM QUARTERLY
news values; c) if, however, selections in a vice, human interest, national, political
“balanced” condition reflect neither the and internatibnal. Stories which could not
“usual” proportions nor the “balanced” be reliably coded into a single category by
proportions, this invites an “added infor- two judges were excluded. When the story
mation” interpretation-that where the files were completed, items were
proportioning approximates “normal” tentatively assigned to two decks. In the
wire service: distributions in the categories, first, or “balanced“ deck, 14 items were
it is followed, but when it does not, idio- assigned to each of the seven categories; in
syncratic selection will apply. the second, or “unbalanced” deck,
p r o p o r t i o n s of items were used
Methods approximating those reported by Snider
Editorial managers of the morning and for wire items read by “M r. Gates”: labor 5
evening newspapers in Columbus and items (5.1%); accidents and disasters, 7
Dayton, Ohio, and of the three Columbus items (7.1%); crime and vice, I 1 items
and two Dayton commercial television ( I 1.2%);human interest, 14 items (14.3%);
stations with regularly-scheduled news national, 16 items (16.3%); political, 21
broadcasts of 30 minutes per day or more items (21.4%); and international, 24 items
were asked for lists of news employees (24.5%). All items were presented to a
whose duties included selecting wire panel of five Ohio State University
service news one or more days per month. journalism faculty and staff members with
Fifty-two such editors were identified, and news editing experience for ranking on a
46 (88.5%) agreed to participate in the Likert-type 1-5 “newsworthiness” scale.
study. Fieldwork dates were May 1-14, Their mean item rankings were then used
1979, and each editor completed a news to balance the overall “newsworthiness”
selection task and answered a dozen within each category across the two decks.
personal interview items. Administration A repeated measures counterbalanced
was completed at subjects’ offices at the design called for the experimental subjects
beginning or end of their working days, (the news editors) to select 21 stories, or
and about 30 minutes was required of each. about the number of wire service items that
Fieldworkers ‘were journalism graduate the largest newspaper in the two cities ran
students. in an average day, from each file.
The selection task was as follows: Two Fieldworkers decided by a coin toss
dummy files of what the editors were told whether the first subject to whom they
were lead paragraphs of wire service news administered the selection task would
stories were printed on cards. Each file select stories from the “unbalanced” or
included 98 news items, roughly the “balanced” deck, or file, first; in
number of items that a content analysis of subsequent administrations, each field-
a week of the Ohio A P newspaper wire worker systematically varied the order of
indicated would be transmitted during a administration. Twenty-four editors
typical day’s morning o r evening selected from the “unbalanced” day first,
newspaper cycle. About 275 items were and 22 from the “balanced.”
selected from current and old newspaper
files, from four-year-old wire service items Results
and from fictionalized accounts similar to Of the 46 editors who participated, 38
the newspaper and wire items. Omitted (82.6%) were male and eight ( 17.4%)were
were sports, state and local items, weather (;old and Simmcrn%.qi. ( I I . . found such catcgorio to he
and stock market quotationsltland items of inbariahl!. applied hy ncwrpapcr\. which u x d a fixed p r o p o r ~ i o n
of w c h copy rcpardle.i\ofuhat uarccrmingin. Whltnc).qr. d . ,
especially important current interest. pp X-V. ha, rclcrrcd I n w c h s t o r m a* ”policy” \toric\, vnce the
References to a day of the week were rule* for their rclccticin or rcjeation arc fixed a n d
ni)nciintempor;tnel,u,: A ucather forcca%t for a particul;lr
altered to “yesterday” or “today.” Stories rcpiiin or rlalc h m ii .I)or a 1.0 chance crl k i n g \clectcd lor a
were coded into the seven categories particular neu*p;tpcr. repardlr\\ c r i 11, content on a g n c n d;i
and thc I ) ~ ~ u - . l o nmcra&c\
c~ ~ 1 1 1 . o r u i l l not. he u w d h?
mentioned in the “Mr. Gates” studies: particular cbcnlng n c u w \ t . rcgardlc\r of uhether Ihcy arc up.
labor. accidents and disasters, crime and d i i u n or unch;ingcd
Downloaded from jmq.sagepub.com by FELICIA GREENLEE BROWN on April 12, 2012Wire News Effects on Gatekeepers 63
FIGURE I
Percentages of stories in wire files, and percentages selected by editors, in balanced and unbalanced
- files in seven content categories.
=denotes
I
percentage i n wire f i l e .
,\]denotes percentage s e l e c t e d by e d i t o r s
1-A: Balanced Condition
26- -26
24- -24
22- -22
20- -20
18- -18
16- -16
Labor Accidents Crime Human National Political Inter-
6Disaster 6Vice Interest nat ional
1-B: Unbalanced Condition
26- -26
2 4- -24
22-
20-
18-
16-
14-
12-
10-
Labor Accidents Crime Buman National Political Inter-
6Disaster 6Vice Interest national
female; 29 (63%) were newspaper editors the editors, and only seven stories (7.1%)
with the remaining 17 (37%) working for were selected by as many as half the
television stations. The editors spend a editors. In the “unbalanced” deck, only
mean of 4.04 days per week editing wire two stories (2%) were not selected, and 24
news. They had been wire editors for a (24.5%) were selected by as many as half
mean of 7.5 years with the most senior the editors. The most favored stories in the
editor having been one for 28 years, and “balanced“ and “unbalanced” set were
they had been professional journalists for a selected by 38 and 36 editors, respectively.
mean of 13 years. A test for an order-of-administration effect
Editors showed considerable variability was performed by computing 1-tests on the
in their selections of 21 stories from both mean number of items selected within each
sets of news leads: in the “balanced“ deck, c o n t e n t c a t e g o r y by o r d e r o f
only six stories (6.1%)were selected by no a d mi n i st r a t i o n (“ b a I a nce d ” o r
editors, 19 (19.4%) were selected by25%of “unbalanced“ first). In none of the 14
Downloaded from jmq.sagepub.com by FELICIA GREENLEE BROWN on April 12, 201264 JOURNALISM QUARTERLY
TABLE 1
Proportions in Wire File and Proportions Selected by Editors in
“Balanced” and “Unbalanced” Conditions
Balanced Condition Unbalanced Condition
% in wire % selected % in wire % selected
file by editors file by editors
Labor 14.3 11.0 5.1 5.3
Accidents & Disasters 14.3 20.7 7. I 7.3
Crime & Vice 14.3 16.0 11.2 14.5
Human Interest 14.3 11.7 14.3 17.9
National 14.3 22.2 16.3 19.8
Political 14.3 9.7 21.4 21.7
International 14.3 8.7 24.5 -13.4
100.1% 100.0% 99.9% w.wo
comparisons was the t-value significant at suggest, this is not the case; the Pearson r is
the .05 level (pooled variance estimate). -.41 (n.s.) Spearman r is -.33 (ns.).
Thus a n order of administration effect was Examination of the rankings of selected
considered unlikely. stories in the “balanced” condition reveals
Proportions of stories in the content no particular pattern of selection,
categories in the “unbalanced” conditions although, as previously noted, it shows
by and large serve as excellent predictors of that generally editors are least likely to
editor selections within the categories; only select international news, the category of
international news varies substantially news that in both White’s 1949 and
from the proportion of incoming news, Snider’s 1966 “ M r . Gates” studies
and international news was the lecist accounted for the highest proportion of
favored category in t h e balanced incoming wire news. A final internal check
condition. (Figure I). compared editors’ selections in the
A Pearson correlation coefficient unbalanced condition with selections in
between number of items incoming and the balanced condition; they were virtually
number selected in the categories in the unrelated (Pearson r -.046, n s . ;
“unbalanced” set is r = .71 (pz.037, n=7); Spearman r = .07, ns.), indicating that
the Spearman rank-order r is .62 (p = .025, selection influence was not closely related
n =7). For newspaper editors the Pearson r to editors’ news values.
is .71; for broadcast editors, Pearson r is Further Anrrlvsis. Newspaper editors
.66. Since there is no variation in the were more likely in the balanced condition
number of stories presented to the editors to select accident and diasaster, crime,
i n the “balanced” set. correlation human interest and international stories
coefficients between number of incoming than were their television counterparts,
and selected items in the “balanced” deck and the TV editors were more likely to
cannot be computed. select political, labor and national stories.
The notion that wire editors and news In the unbalanced condition, newspaper
editors share similar conceptions of how editors and television editors are virtually
many stories should be selected within identical in their selection patterns, except
each of the seven categories can be tested that newspaper editors were much more
by comparing the number of incoming likely to pick human interest stories (T =
stories in the unbalanced set with the 2.34. 44 d.f., p = .02), and TV editors were
number o f stories selected by editors from more likely to select national news items.
the balanced set. As both examination of Several content variations seem
Figure I and reference to correlations counterintuitive and in some cases
Downloaded from jmq.sagepub.com by FELICIA GREENLEE BROWN on April 12, 2012Wire News Effects on Gatekeepers 65
contradictory to previous research. Becker newspaper and television editors than was
has noted that broadcast newsmen are anticipated.
more likely t o report they cover Gold and Simmons, in a reexamination
controversy and conflict than are print of content analysis data of small Iowa daily
journalists," and Buckalew has noted that newspapers, finding a similar pattern,
for television editors, a visuality news noted that "This similarity in patterns of
determinant must be added to the list of news emphasis may represent similarity of
traditional news elements judged by print news judgments. An alternative
journalists.12 As such, then, the content interpretation is that the pattern of
variations may represent structural emphasis of the wire service, represented
variations not controlled in the ex- by the frequency with which various types
periment. of stories are set out, is more o r less
uncritically accepted by these ...daily
Summary newspapers for their own patterns of
Forty-six Columbus and Dayton, Ohio, emphasis." This study undercuts the
editors cooperated in a counterbalanced- notion that the wires and the editors
design field experiment to test whether routinely share news values. It supports the
proportions of news items in seven content idea that news as routinely transmitted in
categories transmitted by wire services stock categories is indeed "uncritically
served to cue editors as to proportions accepted" in newspaper and television
which should be selected from these newsrooms.
categories. Support for such an hypothesis Sprinkled through the recent research
was found, but little support for an literature on the role of the mass media in
alternative explanation that wire editors setting the political agenda are references
and newspaper and television editors share to the nagging question of how the media
an ongoing set of news values was found. formulate the agenda they present to their
In addition, it was suggested that audiences.13 This study suggests quite
international news was less valued by the strongly that the local media. at least, are
influenced greatly by the decisions of a
relatively few editors operating at the
1 ' Lee Becker. "Organimtional Variables and the Study o f
Ncwsroom Behavior." Paper presented to the International regional, national and international
Association for Mass Communication Research, Warsaw. bureaus of the wire services. In other
Poland. August. 1978. p. 10.
words, the agenda being presented by the
' 2 James K . Huckalcw. *A @Analysis of Television News
Editors' Decisions." JOURNAL.ISM QI1ARTEKI.Y. 46: 135-137 media audiences is influenced by the
(1969). newsgathering procedures of the media
1 ' Maxwell E. M K o m b s and Donald I..Shaw."ThcAgcnda- and the relationships among the media.
Setting Function or Mass Media." Piihlic Opinion Qiiwtrrlr.
37:176-187 (1972): Shaw and McComhs. The Eiiiertrnce of The local media are hardly actingalone in
Ameriutn Poliricul I.uues (St. Paul. Minn.: W e t . 1977). shaping the political agenda.
RESPONSE TO PHOTOGRAPHS
(Continuedfrom page 20)
the way in which the reproduction process content of what they report and on the
changes the connotative meaning of the interpretation of that content by message
photograph. receivers. The variants of photographic
Further, this research on the effects of style and structure are many. Relatkely
still another "stylistic variation" on the few of these variants have been 5ub.iected
perception of a photographic message to systematic study, and the author\ hope
adds to our knowledge of the potential that readers o f this report will he chal-
influence which journalists have on the lenged to isolate other variants and t o
investigate them.
Downloaded from jmq.sagepub.com by FELICIA GREENLEE BROWN on April 12, 2012You can also read
