Refugees and Agenda Setting Evidence From Passive Media Measurement

Transcription

1 Refugees and Agenda Setting Evidence From Passive Media Measurement Daniel Prokop Lea Michalová MEDIAN, Czech Republic

2 admeter Methodology What we measure TV radio Internet (phone and PC) location How we measure audiomatching (TV, radio) user-centric net measurement Sample size 1000 respondents Representativity Sociodemographic Media consumption segments Outputs second-by-second single source data time-shift - 2 -

3 admeter: Audiomatching Scheme Feed-back Analysis Mobile app Audio-matching and data management Compressed record (1Mb per day) Measurement outputs - 3 -

4 Types of Analyses multiscreening campaign effectiveness out-of-home analysis timeshift ratings incremental reach e t c - 4 -

5 Out-of-home Ratings Ice Hockey Match Home Out-of-home Total - 5 -

6 Multiscreening - 6 -

7 Campaign Crossmedia Reach TV 3+ (91 %) Internet 3+ (56 %) Radio 3+ (47 %) = 99% 19 % 28 % TV+Web+Radio 3+, None of the media types alone reaches 3+ 1 % 22 % 22 % 1 % 2 % 4 % - 7 -

8 Probability of the respondent remembering the campaign (spontaneous) Campaign Effect 1 Print TV Online 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0, OTS (opportunities to see) - 8 -

9 Researching Agenda Setting 2 time series Media content Public attitudes Medium Declared media consumption Topic perception Inconclusive Inconclusive Our design Media content Viewers exposure to particular content (TV reportages) What they think - 9 -

10 2014 & 2015 Asylum Seekers, Attitudes and Numbers Number of asylum applications Amount Refugees are a major threat % % of refugeerelated media content 1,07x 2,06x 4,1x

11 Defining and Measuring Refugee Content Data July and August 2015 N = 249 respondents NEWTON MediaSearch 790 pcs of TV content either at least 2 occurrences of any of the words from the list (even the same twice), or at least one occurrence if used in conjunction with any of the four locations of Czech refugee centers

12 Agenda Setting First level Which topics What people think about Second level Their attributes and characteristics In what way Perceived salience of the topic of refugees Attitude to refugees (accept reject)

13 First-level Agenda Setting Exposure in last 2 months Explaining variable Sociodemographics Gender, age, education, region, size of community Control variables Salience Explained variable Media usage Favorite TV stations TV consumption Control variables

14 Importance of Topics Enforceability of laws and their quality Social security, pensions policy National economy and public finances 72, ,7 Unemployment Environment and environmental protection Immigration and granting asylum to refugees Corruption and political ties to business Schooling, science and research 61,6 61,2 59, ,8 War in Ukraine and relations with Russia 33 Support of sport and culture 28,6-14 -

15 Importance of Immigration and Refugee Topic Immigration and granting asylum to refugees 59,4 low and medium exposure (maximum 40 hits) 56 high exposure (41 and more hits) 67,7-15 -

16 Effect on Perceived Salience - Logistic Model Variable value Exp(B) Size of community Most frequently watched TV station Up to inhabitants * to inhabitants 0,362** inhabitants 0, inhabitants 1,162 Over inhabitants 0,676 Czech TV Group (Public) Major commercial stations (Nova, Prima, Barrandov) 1,286 Minor commercial sts. (Nova Cinema, Prima Cool) 2,239 Rate of exposure High exposure (top tercile, 41+) 1,898** 62,5 % Predicted correctly (59,4 % in zero model), Nagelkerke R-Sq 0,077 Modelling: mandatory media variables + backward stepwise ** p<0,05, * p<0,1-16 -

17 Second-level Agenda Setting Exposure in last 2 months Explaining variable Sociodemographics Gender, age, education, region, size of community Attitude Explained variable Media usage Favorite TV stations TV consumption Control variables Control variables

18 Czech Republic Definitely Should Not Accept Refugees Total 57,1 Czech TV Group (Public) 47,8 Major commercial stations (Nova, Prima, Barrandov) Minor commercial stations (Nova Cinema, Prima Cool) 38,9 64 low and medium exposure (maximum 40 hits) high exposure (41 hits and more) 44,6 62,3-18 -

19 Effect on Refugee Rejection - Logistic Model 1 Variable value Exp(B) Gender Female 2,023** Primary ** Education Lower secondary 0,418 Upper secondary 0,302* University 0,148** Most frequently watched TV station Czech TV Group (Public) ** Major commercial stations (Nova, Prima, Barrandov) 1,673 Minor commercial sts. (Nova Cinema, Prima Cool) 0,357* Rate of exposure High exposure (top tercile, 41+) 0,434** 68,3% Predicted correctly (57 % in zero model), Nagelkerke R-Sq 0,16 Modelling: mandatory meida variable + backward stepwise ** p<0,05, * p<0,1-19 -

20 Czech Republic Definitely Should Not Accept Refugees Total 57,1 High exposure on Czech TV (15 and more hits) 32,3 High exposure on Nova (15 and more hits) 52 High exposure on Prima (15 and more hits) 78,9-20 -

21 Effect on Refugee Rejection - Logistic Model 2 Variable value Exp(B) Gender Female 1,735* Primary Education Lower secondary 0,498 Upper secondary 0,369 University 0,256* Non-watcher ** Czech TV - type Watcher low exposure on Czech TV 1,082 Watcher high exposure on Czech TV 0,245** Non-watcher ** Prima TV - type Watcher low exposure on Prima 2,009* Watcher high exposure on Prima 2,936** 66,5% Predicted correctly (57 % in zero model), Nagelkerke R-Sq 0,19 Modelling: media enter + backward stepwise ** p<0,05, * p<0,1-21 -

22 Summary Electronic Media Measurement Actual exposure to content Higher exposure to content topics perceived to be salient Causal effects? Controlling for sociodemographic and TV-usage patterns eliminates other causal hypotheses Exposure on Czech TV -> lower refugee rejection, Exposure on Prima -> higher refugee rejection Limitations Small sample No pretest posttest Not controlling for news-broadcastconsumption

23 Thank you for