Objectives:
After Learning this lecture, you will be able to:
1. Describe the process by which Groseclose and Milyo measured media bias.
2. Identify three important findings in Groseclose and Milyos study: A Measure of
Media Bias
3. Describe the Hostile Media Phenomenon.
4. Identify three factors that trigger the Hostile Media Phenomenon
5. Identify three important findings in Gussin and Baum study: In the Eye of the
Beholder
Lecture 13
1. Bias in the media:
A. Evidence of partisan bias: idea that media favors one party over another (ex: Liberal or
Conservative bias)
1) Perceptions of partisan bias based on those involved in the news production
process
a) 1992 Journalists
(1) 88% of Washington journalists voted for Clinton
b) Washington-based journalists (1998)
1
(1) Social policy: Strong evidence that on certain social issues, there is some
type of bias in news media
(a) Not clearly liberal
(2) Economic policy: Ecnomically to the right because probably better off
financially
(a) Higher income shows that views respond to their economic group
b) Most media outlets owned by large corporations
(1) Social policy? Dont know
(2) Economic policy: tend to be more conservative
(a) Owners of media outlets have a conservative economic philosohpy,
will not hire people who argue against their own interests in their own
media outlets. Will hire people who can better articulate their interests
i. Gussin: when people rely on demographics, it is because of a lack of
evidence. Make assumptions based on numbers.
2) Perceptions of bias based on content of news coverage
a) Standard Method of Inquiry: Content analysis (come up with a set of rules for
determining bias and then look at news coverage)
2
(1) Non-academic analyses: not credible or scientific because they make their
conclusions before starting their study.
(a) http://www.mediaresearch.org/: conservative thinktank
i. skips conservative bias and only focsuses on evidence of liberal bias
(b) http://mediamatters.org/about_us/
i. vice versa for evidence of conservative bias
(2) Academic analyses: meta analysis
(a) Inconclusive findings for evidence
(3) Other forms of systematic bias
(a) Frontrunners: more negative coverage regarding front runners in the
election
(b) Increased negativity (Zaller)
i. Candidate-initiated vs. media-initiated: increase in both
Candidate-initiated: Doesnt make sense to accuse media of
negativity when the candidate iniates the negativity.
Candidates are being negative and media reports it
Media-initiated: Example is the flag factory
Less than we might think but both are on the rise
(c) Increased emphasis on game schema: idea that we treat elections like a
sporting event
i. Horserace: who is ahead?
Ex: Michigan Primary: emphasis on who is ahead/behind over
policies of candidates
ii. Strategy over policy
If someone goes over to NM for a speech, will not worry about
what was in the speech but the reasons why they would be in NM
(d) Fewer and shorter sound bites: hear less of what the candidates even
said, not anyhting to do with policy
i. Mitt Romneys speech ignored because he said couple of Cadillacs
3
B. Measure of Media Bias (Groseclose and Milyo): scientific anlysis of media bias
1) ADA scores: Americans for Democracic action
a) Counts how often cadidates vote correclty like the pulic
2) Perceptions
2007
ADA
0
5
10
15
20
30
55
60
70
75
80
85
90
90
95
100
MC
PID
CANTOR, ERIC
BOEHNER, JOHN
MCCONNELL, MITCH
PAUL, RON
GRAHAM, LINDSEY
GRASSLEY, CHUCK
COLLINS, SUSAN
SNOWE, OLYMPIA
LIEBERMAN, JOE
OBAMA, BARACK
BOXER, BARBARA
REID, HARRY
FEINSTEIN, DIANNE
SCHUMER, CHUCK
FRANK, BARNEY
STABENOW, DEBBIE
R
R
R
R
R
R
R
R
D
D
D
D
D
D
D
D
a) News FOX = 35_______
b) NYT = 65
WS____________
3) G&M measure media bias in novel way
a) Step One: ADA scores for Member of Congress (MC): looked up everytime a
candidate opnions; keeps track of when candidates mentioned thinktanks in a
favorable way
b) Step Two: Transfer ADA score from MC to Think Tank (TT)
(1) Rationale: Congress not rely on Conservatives to [i, Lib rely on Lib TT
c) Step Three: Transfer ADA score from TT to media outlet
4
(1) Rationale: Con media outlets rely on Con TT, Lib media outlets rely on
Lib TT
4) Pretty much generate ADA numbers and list thinktanks from liberal to conservatie
based thinktanks
5) Looked at news coverage from many media outlets (reports that menitoned the
thinktank in a favorable way; give that media outlet the thinktanks score. Total up
scores and take average
6) No opinion is interjected, only when determining when a member of Congress was
saying something favorable about a topic
7) Findings
a) Most media outlets are more liberal than the average American voter.
b) Every media outlet is:
(1) More liberal than: the average Republican in Congress
(2) More conservative than the average Democrat in Congress
(a) One exception: Wall Street Journal
8) Perception vs. Reality: study suggests that the media is not that extreme at all
a) FOX News = ___________
b) NYT = ________________
2. Perceptions of bias
A. Hostile Media Phenomenon (1985): people tend to view BALANCED coverage as
hostile to their own point of view.
1) Effect greatest among individuals with strong views: parties have become
ideologically extreme, more extreme your views; more likely to perceieve media as
hostile
a) Experiment
(1) Subjects: Pro-Israeli and Pro-Palestinian students.
(a) Viewed balanced coverage of the Beirut Massacre
(b) When asked, Pro-Israeli thought it favored Palestinians and vice versa
b) Findings:
5
(1) Different evaluations of fair coverage
(a) Palestinians: wanted them to ask Israelis about Palestinian occupation
(b) Israelis: how to justify terrorist attacks
(c) Both sides saids coverage wasnt fair, wanted tougher coverage
(2) Negative references more accessible than positive references: MBIP
(a) you remember all the negative stuff
(b) negative stuff was at the top of their head when asked
(3) Different perceptions of persuasiveness
(a) Thought it someone who was neutral watched the report, they would
be Pro-Israeli (by Palestinian responders) and vice versa
c) Source of bias: themselves, what they bring to the table because the news
report is balanced
2) Eye of the Beholder (Baum and Gussin)
a) Experiment:
(1) Subjects completed extensive pre-test survey about news media outlets
(a) FOX, CNN. NBC about liberal/conservative bias.
(b) Only interestd in FOX and CNN ratings but asked a lot because didnt
want them to know
(2) Randomly assigned to treatment (i.e., transcript of 2004 presidential
election)
(a) CNN, FOX, or KNWZ
(b) Made fake but balanced report but switched out CNN or FOX news
reporters, etc.
(c) Same coverage just where it came from
b) Findings
(1) If subject thought media outlet was liberal (pre-test): thought it was biased
in favor in John Kerry
(2) If subject thought media outlet was conservative (pre-test): thought it was
Bush biased
(3) Subjects who saw KNWZ: thought it was balanced because there was no
prior conceptions for the news outlet (just a little bit left of center)
6
c) Prior beliefs shaped perceptions of bias; had nothing to do with what was in the
news
(1) Citizens may create bias, even when it isnt present.
(2) High information people with strong views are most succeptible to having
prior beliefs shaped perceptions
7