komm.uni-hohenheim.de - Selective Exposure and Hostile ......Marko&Bachl& Selective Exposure and...

Marko Bachl

Selective Exposure and Hostile Media Perceptions in Election Campaigns

Presentation at the 67th Annual Conference of the World Association for Public Opinion Research, 4-6 September 2014, Nice, France

Marko Bachl

Theoretical Background: Selective Exposure & Hostile Media

3 Theoretical Background | Method | Results | Discussion

Selective Exposure & Hostile Media

Selec-ve exposure §  People select news which are consistent

with their poli-cal beliefs

§  Since Lazarsfeld et al. a dominant paradigm to explain exposure to and percep-on of news in elec-on campaigns, supported also by current experimental (GarreD & Stroud, 2014; Iyengar & Hahn, 2009; Iyengar et al., 2008; Knobloch-‐Westerwick, 2012; Knobloch-‐Westerwick & Meng, 2009, 2011) and observa-onal (Stroud, 2008; Wicks et al., 2013) research

à  If par-sans select news which agree with their beliefs, they should report the news coverage to be friendly to their beliefs

Hos-le media §  Par-sans perceive news coverage as

hos-le to their beliefs

§  Pioneered by Vallone et al. (1985) and demonstrated frequently thereaWer (Hansen & Kim, 2011)

§  Also prevalent in the context of elec-on campaigns (for example Beck, 1991; Dalton et al., 1998; Huge & Glynn, 2010; SchmiD-‐Beck, 1994), mostly rela%ve hos-le media percep-ons (Gunther et al., 2001)

à  Par-sans should report the news coverage to be hos-le to their beliefs.


Study Objectives

① ‘Real world’ es-mates of the prevalence and magnitude of selec-ve exposure and hos-le media percep-ons during two German general elec-on campaigns

② Addressing the contradic-ng predic-ons of the selec-ve exposure and hos-le media paradigms

§  Magnitude and direc-on of selec-ve exposure and bias percep-ons vary between media outlets

§  Varying effects are associated within media outlets §  Stronger selec-ve exposure à friendly media percep-ons §  Weaker selec-ve exposure à hos-le media percep-ons

Marko Bachl

Method, Measures, & Statistical Analysis


Method: Survey Data

German Bundestag elec-ons 2009 & 2013 Data Source:

§  Representa-ve survey data from rolling cross sec-on surveys in the run-‐up to the elec-ons, German Longitudinal Elec-on Study (GLES)

§  Geographically stra-fied random selec-on of telephone numbers and selec-on of respondents within households via last-‐birthday-‐method

2009 §  Field -me: 8 weeks §  Total interviews: n = 6008 §  AAPOR Response Rate: 19.6%

2013 §  Field -me: 11 weeks §  Total interviews: n = 7882 §  AAPOR Response Rate: 15.5%


Measures: IVs

IV of interest: LeW-‐right party alignment score §  Consists of party iden%fica%on, aktudes towards government of the

poli%cal camp, aktudes towards the poli%cal par%es and aktudes towards the top candidates

Important control or moderator variable: Poli-cal interest §  Mean score of two items (general poli-cal interest and interest in

current campaign), –2 to 2.

–5 Strongest alignment

with leW camp (SPD & Grüne)

0 +5 Strongest alignment with right camp (CDU/CSU & FDP)


Measures: DVs

Selec<on: §  Media outlets used for poli-cal informa-on during last 7 days

§  TV news of 4 most important broadcasters (ARD, ZDF, RTL, Sat.1) §  Na-onal tabloid BILD §  Up to 2 newspapers (na-onal newspapers taz, FR, SZ, FAZ, Welt;

generic: local or regional newspaper & online news (only 2009)) Bias Percep<on:

§  For all selected outlets, 3 poli-cal par-es favored by news coverage §  Bias differen-al according to leW-‐right party alignment

–1 Strongest bias

in favor of leW camp (SPD & Grüne)

0 +1 Strongest bias

in favor of right camp (CDU/CSU & FDP)

–.5 +.5


Statistical Analysis: Crossed Multi-Level Model

Cross-‐classifica-on of DV observa-ons in both respondents and media outlets

1.  Es-ma-on of effects of respondents’ characteris-cs which vary between media outlets

2.  Inves-ga-on of the rela-onship of the varying effects within media outlets

Bias Perc.(11)!

Respondent1! Respondent2! Respondenti!

Media Outlet1! Media Outlet2! Media Outletj!

Bias Perc. (12)! Bias Perc.(1j)! Bias Perc.(21)! Bias Perc.(2j)! Bias Perc.(i1)! Bias Perc.(i2)! Bias Perc.(ij)!

Marko Bachl

Results


Media Selection: Idealized Results

Selective exposure (left) No selective exposure Selective exposure (right)

0.00

0.25

0.50

0.75

1.00

−5 0 5 −5 0 5 −5 0 5Left−Right Party Alignment

Estim

ated

Pro

babi

lity

of S

elec

tion

(at l

east

onc

e we

ekly

)


Media Selection: Empirical Results (I)

taz FR SZ FAZ Welt

0.00

0.05

0.10

0.15

0.00

0.05

0.10

0.15

20092013

−5 0 5 −5 0 5 −5 0 5 −5 0 5 −5 0 5Left−Right Party Alignment

Estim

ated

Pro

babi

lity

of S

elec

tion

(at l

east

onc

e we

ekly

)

Political Interest less than average average above average

taz FR SZ FAZ Welt

0.00

0.05

0.10

0.15

0.00

0.05

0.10

0.15

20092013


Estim

ated

Pro

babi

lity

of S

elec

tion

(at l

east

onc

e we

ekly

)

Political Interest below average average above average


Media Selection: Empirical Results (II)

BILD Lokal ard zdf rtl

0.00

0.25

0.50

0.75

0.00

0.25

0.50

0.75

20092013


Estim

ated

Pro

babi

lity

of S

elec

tion

(at l

east

onc

e we

ekly

)

Political Interest below average average above averageNote Not shown: Sat.1, Online news


Bias Perception: Idealized Results

Friendly bias perceptions No bias perceptions Hostile bias perceptions

−1.0

−0.5

0.0

0.5

1.0

−5 0 5 −5 0 5 −5 0 5Left−Right Party Alignment

Estim

ated

Bia

s of

Med

ia O

utle

t (ne

gativ

e: le

ft; p

ositi

ve: r

ight

)


Bias Perception: Empirical Results (I)

taz FR SZ FAZ Welt

−0.50

−0.25

0.00

0.25

0.50

−0.50

−0.25

0.00

0.25

0.50

20092013


Estim

ated

Bia

s of

Med

ia O

utle

t (ne

gativ

e: le

ft; p

ositi

ve: r

ight

)

Political Interest less than average average above average

taz FR SZ FAZ Welt

−0.50

−0.25

0.00

0.25

0.50

−0.50

−0.25

0.00

0.25

0.50

20092013


Estim

ated

Bia

s of

Med

ia O

utle

t (ne

gativ

e: le

ft; p

ositi

ve: r

ight

)

Political Interest below average average above average


Bias Perception: Empirical Results (II)

BILD Lokal ard zdf rtl

−0.50

−0.25

0.00

0.25

0.50

−0.50

−0.25

0.00

0.25

0.50

20092013


Estim

ated

Bia

s of

Med

ia O

utle

t (ne

gativ

e: le

ft; p

ositi

ve: r

ight

)

Political Interest below average average above averageNote Not shown: Sat.1, Online news


Results: Media Selection and Bias Perception

Political Interest: less than average average above average

ardBILD

FAZ

FR

LokalOnlinertl

sat1

SZ

taz

Welt

zdf

ard BILD

FAZ

FRLokalrtlsat1SZ

taz

Welt

zdf

ardBILD

FAZ

FR

LokalOnlinertl

sat1

SZ

taz

Welt

zdf

ardBILD

FAZFR

Lokalrtl

sat1SZtaz Welt

zdf

ardBILD

FAZ

FR

LokalOnlinertl

sat1

SZ

taz

Welt

zdf

ard

BILD

FAZ

FR

Lokal

rtlsat1

SZ

taz

Weltzdf

−0.050

−0.025

0.000

0.025

0.050

−0.050

−0.025

0.000

0.025

0.050

20092013

−0.5 0 0.5 −0.5 0 0.5 −0.5 0 0.5Effects of Left−Rigth Party Alignment on News Selection

Effe

cts

of L

eft−

Rig

th P

arty

Alig

nmen

t on

Med

ia B

ias

Perc

eptio

n(n

egat

ive: h

ostil

e m

edia

; pos

itive

: frie

ndly

med

ia)

Marko Bachl

Discussion


Discussion: Summary

Selec-ve exposure: §  Evidence for selec-ve exposure to na-onal newspapers (quality press

and tabloid BILD) and RTL news §  Broad reach of local/regional newspapers and public TV broadcasters

across the poli-cal spectrum

Bias percep-ons: §  Some evidence for friendly media percep-ons (quality press in 2009) §  Some evidence for rela-ve hos-le media percep-ons among the highly

involved, but of limited magnitude

Selec-ve exposure & bias percep-ons: §  Some support for the proposed paDern of both phenomena within

media outlets (2009 and among the highly involved in 2013)


Discussion: Limitations & Strengths

Limita-ons: §  Cross-‐sec-onal survey design: no causal claims, “de facto

selec-vity” (Sears & Freedman, 1987, p. 198) §  Limita-ons of media outlet sample, especially (poli-cally aligned)

internet sources

Strengths: §  Credible ‘real world’ es-mates for prevalence and magnitude of

selec-ve exposure and hos-le media percep-ons during two German na-onal elec-on campaigns

§  Empirical test of a proposal to reconcile the contradic-ng conclusions of two prominent paradigms

§  Instant replica-on for two elec-ons, but further replica-ons needed

Universität Hohenheim Ins<tut für Kommunika<onswissenschaF Marko Bachl www.komm.uni-‐hohenheim.de/bachl marko.bachl@uni-‐hohenheim.de


Literature

Beck, P. A. (1991). Voters' intermedia-on environments in the 1988 presiden-al contest. Public Opinion Quarterly, 55(3), 371-‐394. doi: 10.1086/269269 Dalton, R. J., Beck, P. A., & Huckfeldt, R. (1998). Par-san cues and the media: Informa-on flows in the 1992 presiden-al elec-on. American Poli%cal Science Review, 91(1),

111-‐126. GarreD, R. K., & Stroud, N. J. (2014). Par-san paths to exposure diversity: Differences in pro-‐ and counteraktudinal news consump-on. Journal of Communica%on, online

first. doi: 10.1111/jcom.12105 Goldman, S. K., & Mutz, D. C. (2011). The friendly media phenomenon: A cross-‐na-onal analysis of cross-‐cukng exposure. Poli%cal Communica%on, 28(1), 42-‐66. doi:

10.1080/10584609.2010.544280 Gunther, A. C., Christen, C. T., Liebhart, J. L., & Chih-‐Yun Chia, S. (2001). Congenial public, contrary press, and biased es-mates of the climate of opinion. Public Opinion

Quarterly, 65(3), 295-‐320. doi: 10.1086/322846 Hansen, G. J., & Kim, H. (2011). Is the media biased against me? A meta-‐analysis of the hos-le media effect research. Communica%on Research Reports, 28(2), 169-‐179. doi:

10.1080/08824096.2011.565280 Huge, M., & Glynn, C. J. (2010). Hos-le media and the campaign trail: Perceived media bias in the race for governor. Journal of Communica%on, 60(1), 165-‐181. doi:

10.1111/j.1460-‐2466.2009.01473.x Iyengar, S., & Hahn, K. S. (2009). Red media, blue media: Evidence of ideological selec-vity in media use. Journal of Communica%on, 59(1), 19-‐39. doi: 10.1111/j.

1460-‐2466.2008.01402.x Iyengar, S., Hahn, K. S., Krosnick, J. A., & Walker, J. (2008). Selec-ve exposure to campaign communica-on: The role of an-cipated agreement and issue public membership.

The Journal of Poli%cs, 70(01), 186-‐200. doi: 10.1017/S0022381607080139 Knobloch-‐Westerwick, S. (2012). Selec-ve exposure and reinforcement of aktudes and par-sanship before a presiden-al elec-on. Journal of Communica%on, 62(4),

628-‐642. doi: 10.1111/j.1460-‐2466.2012.01651.x Knobloch-‐Westerwick, S., & Meng, J. (2009). Looking the other way: Selec-ve exposure to aktude-‐consistent and counteraktudinal poli-cal informa-on. Communica%on

Research, 36(3), 426-‐448. doi: 10.1177/0093650209333030 Knobloch-‐Westerwick, S., & Meng, J. (2011). Reinforcement of the poli-cal self through selec-ve exposure to poli-cal messages. Journal of Communica%on, 61(2), 349-‐368.

doi: 10.1111/j.1460-‐2466.2011.01543.x Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1968). The people's choice: How the voter makes up his mind in a presiden%al campaign (3 ed.). New York: Columbia University

Press. Sears, D. O., & Freedman, J. L. (1967). Selec-ve exposure to informa-on: A cri-cal review. Public Opinion Quarterly, 31(2), 194-‐213. doi: 10.1086/267513 SchmiD-‐Beck, R. (1994). Intermedia-on environments of West German and East German voters: Interpersonal communica-on and mass communica-on during the first all-‐

German elec-on campaign. European Journal of Communica%on, 9(4), 381-‐419. doi: 10.1177/0267323194009004002 Stroud, N. J. (2008). Media use and poli-cal predisposi-ons: Revisi-ng the concept of selec-ve exposure. Poli%cal Behavior, 30(3), 341-‐366. doi: 10.1007/s11109-‐007-‐9050-‐9 Vallone, R. P., Ross, L., & Lepper, M. R. (1985). The hos-le media phenomenon: biased percep-on and percep-ons of media bias in coverage of the Beirut massacre. Journal

of Personality and Social Psychology, 49(3), 577. doi: 10.1037/0022-‐3514.49.3.577 Wicks, R. H., Wicks, J. L., & Morimoto, S. A. (2013). Par-san media selec-ve exposure during the 2012 presiden-al elec-on. American Behavioral Scien%st, 58(9), 1131-‐1143.

doi: 10.1177/0002764213506208

Marko Bachl

Appendix


Measures: IVs


Measures: DVs

komm.uni-hohenheim.de - Selective Exposure and Hostile ......Marko&Bachl& Selective Exposure and...

Documents

Transcript of komm.uni-hohenheim.de - Selective Exposure and Hostile ......Marko&Bachl& Selective Exposure and...