Untitled

In this study, we acknowledge that the plurality/diversity of media-generated news frames is crucial for covering the viewpoint of a multitude of societal actors, including the silent ones, and/or for presenting alternative viewpoints to these actors so that they can choose the most sense-making frame, compatible with their world-views. However, we argue that the existence of a framing contest alone cannot be considered as a sign of a consolidated democracy, democratic culture or a democratic public opinion. The contesting frames, despite their diversity, might commonly reveal an anti-democratic tonality. Here, the term tonality refers to the construction of a news frame. It implies the underlying political and cultural fundamentals that constitute the expression of that frame in terms of presenting the reality (Sayin and Toros, 2016). Accordingly, a research on the news frames must consider the tonality of news frames alongside their variety. This study will focus on the (1) distribution of news frames and their tonality, and (2) the relationship between these frames and tonality regarding the headline news of six Turkish newspapers during the Gezi Park protests in Turkey in 2013.

Data prep

rm(list=ls(all=TRUE)) 
setwd("D:/emre/SkyDrive/makale/consolidation book/PRR")
library(foreign)
library(nnet)
library(effects)
library(car)

## 
## Attaching package: 'car'

## The following object is masked from 'package:effects':
## 
##     Prestige

library(reshape)

mydata <- read.dta("r.dta")
mydata$Tone <- relevel(mydata$tone, ref = "Neutral")
mydata <- rename(mydata, c(RESPONSIBL="Responsibility"))
mydata <- rename(mydata, c(HUMAN="Human"))
mydata <- rename(mydata, c(CONFLICT="Conflict"))
mydata <- rename(mydata, c(MORALITY="Morality"))
mydata <- rename(mydata, c(ECONOMIC="Economic"))

Model

model1 <- multinom(Tone~ Responsibility  + Human  + Conflict + Morality + Economic , data=mydata)

## # weights:  21 (12 variable)
## initial  value 907.453750 
## iter  10 value 791.380623
## iter  20 value 759.194636
## iter  20 value 759.194635
## iter  20 value 759.194635
## final  value 759.194635 
## converged

summary(model1)

## Call:
## multinom(formula = Tone ~ Responsibility + Human + Conflict + 
##     Morality + Economic, data = mydata)
## 
## Coefficients:
##                 (Intercept) Responsibility      Human   Conflict  Morality
## Anti-democratic -1.28225450      0.2849648 -0.1794296  0.0336153 1.5963198
## Democratic      -0.08745288     -0.1399092  0.3415845 -0.1188080 0.9158738
##                   Economic
## Anti-democratic 0.22047112
## Democratic      0.09377321
## 
## Std. Errors:
##                 (Intercept) Responsibility      Human   Conflict  Morality
## Anti-democratic   0.2460515     0.05510812 0.05683265 0.03513069 0.2377163
## Democratic        0.2268129     0.06240497 0.06076433 0.04263227 0.2545402
##                  Economic
## Anti-democratic 0.1012143
## Democratic      0.1111371
## 
## Residual Deviance: 1518.389 
## AIC: 1542.389

Interpretation (check “1” among the exponentiated confident intervals, if there is “1” it is insignificant)

exp(coef(model1)); exp(confint(model1))

##                 (Intercept) Responsibility     Human  Conflict Morality
## Anti-democratic   0.2774112      1.3297152 0.8357468 1.0341867 4.934838
## Democratic        0.9162620      0.8694371 1.4071756 0.8879783 2.498958
##                 Economic
## Anti-democratic 1.246664
## Democratic      1.098311

## , , Anti-democratic
## 
##                    2.5 %    97.5 %
## (Intercept)    0.1712713 0.4493278
## Responsibility 1.1935772 1.4813809
## Human          0.7476506 0.9342234
## Conflict       0.9653741 1.1079043
## Morality       3.0969089 7.8635262
## Economic       1.0223407 1.5202085
## 
## , , Democratic
## 
##                    2.5 %    97.5 %
## (Intercept)    0.5874301 1.4291677
## Responsibility 0.7693412 0.9825562
## Human          1.2491816 1.5851522
## Conflict       0.8167962 0.9653637
## Morality       1.5173785 4.1155129
## Economic       0.8833344 1.3656055

effects plots

Responsibility <- effect("Responsibility", model1)
plot(Responsibility, 
     rug = FALSE, band.transparency = 0.1,
     xlab = "Frequency of News Using Responsibility Frame",
     main = "The Effect of Responsibility Frame on the Tone of the News" ) 

Human <- effect("Human", model1)
plot(Human, 
     rug = FALSE, band.transparency = 0.1, 
     xlab = "Frequency of News Using Human Frame",
     main = "The Effect of Human Frame on the Tone of the News" )

Conflict <- effect("Conflict", model1)
plot(Conflict, 
     rug = FALSE, band.transparency = 0.1, 
     xlab = "Frequency of News Using Conflict Frame",
     main = "The Effect of Conflict Frame on the Tone of the News" )

Morality <- effect("Morality", model1)
plot(Morality, 
     rug = FALSE, band.transparency = 0.1, 
     xlab = "Frequency of News Using Morality Frame",
     main = "The Effect of Morality Frame on the Tone of the News" )

Economy <- effect("Economic", model1)
plot(Economy, 
     rug = FALSE, band.transparency = 0.1, 
     xlab = "Frequency of News Using Economic Frame",
     main = "The Effect of Economic Frame on the Tone of the News" )