2. Political Communication Methods Flashcards
dominant methodology - media systems
Comparative case studies
dominant methodology - media and democracy
Case studies (democratization)
Experiments (democratic virtues)
Content analysis
Survey research
dominant methodology - impact of media on politics
Content analysis (case studies)
Time series (policy change)
Experiments (political knowledge)
dominant methodology - electoral campaigns
Survey research
Content analysis
Experiments
universe (content)
content of what
population (content)
complete set of media you are interested in
cases (content)
what specific media you are interested in
units of analysis (content)
what specific elements of specific media do you want to analyse
what is content
universe
population
cases
units of analysis
big data consequences
can be messy and perhaps incomplete or poorly encoded
from causation to correlation
saliency
what is the weight of content
levels of saliency
- word count
- word combination
- counting the argumentation
trade-off saliency
The more interesting the measure, the more complex it is to code, and the less easy it is to code it quantitatively (hand coding vs. computer coding)
valance
what is the direction of the content (the degree of attraction of aversion that one feels towards a specific object or event)
levels of valance
- mere direction (pro / against)
- intensity of positivity / negativity (degree)
- object of positivity or negativity
wat is de direction / hoe erg / tegen wie?
language modelling
big data and new types of patterns.
pre-training on an example corpus (copyright issues), like Wikipedia / parliamentary debates
This makes windows of 6-12 worlds before and after and does preprocessing (removes stop words)
windows of 6-12 worlds before and after specific word
experiments
a research design in which the researcher both controls and randomly assigns values of the independent variable to the participants
- controlling / randomly assigning is important!
research design / controls and randomly assign values of variables
experiments - controlling
control group + treatment group
experiments - randomly assigning
randomly assigned to treatment or placebo
framing experiments
- sample
- pretest
- treatment
- post-test
sample (experiment)
randomly assign students to one of three groups
pretest (experiment)
before reading test the students are surveyed
treatment (experiments)
two groups get a different text to read, third one is control group
post-test (experiments)
survey again to see effects of the treatment
Problems of surveys
selection effects vs. media effects
false recalls
endogeneity problem
false recalls (problem survey)
exposure is a necessary condition for the direct effect to happen / inaccurate reporting has significant effects on results of studies
inaccurate reporting –> effect on results study
samples checklist
- check sample size
- check the type of sample
- what is the bias
- is the bias a problem
- modest in claims
endogeneity problem
occurs when the explanans (X) may be influenced by the explanandum (Y) or both may be jointly influenced by an unmeasured third
X influenced by Y or a third variable
