Text Ch.11 Flashcards
“experiment”study =
studies that include at least these two criteria(random assignment and intervention)
-random assignment equals out groups
Temporal order and cross-sectional analysis
• Cross-sectional studies allow for correlational analysis, but rely on theory/logic to determine temporal order -Distinction between “correlational” designs and “experimental designs”
confounding variable
• the presence of some unobserved difference between groups that is correlated both with the outcome of interest and with the independent variable of interest, thereby making causal inference impossible.
random assignment vs. random sample
Random selection refers to how sample members (study participants) are selected from the population for inclusion in the study. Random assignment is an aspect of experimental design in which study participants are assigned to the treatment or control group using a random procedure.
Random assignment: is to get at that issue of confounding variables -used in experimental studies only -important for studying causality -internal validity -• Intervention (manipulation, stimulus) during data collection
Random Sampling:
- used to permit generalizations to larger population
- external validity
T OR F
SURVEYS use only random sampling and not random assignment
T
random sample assignment
May be used in experimental studies; if so, known as random sample assignment. In this approach, random sampling done first, random assignment done second.
Essentially
Random Sampling = _______Claims
Random Assignment = ______ Claims
Generalizability
Causality
Most experiments are ______while most observational studies are _______
random assignment, no random sampling
random sampling, no random assignment
Survey experiments
Survey Experiments: combination of survey and experiment
.like a normal survey but they build in the two requirements of a survey: random assignment and intervention
-treatment groups and control groups
Advantages: representative population
Disadvantage: not real world setting
T OR F
experimental research is declining in political science
F, its rising
WEIRD
WEIRD is the phenomenon that plagues a lot of psychology and other social science studies: Their participants are overwhelming Western, educated, and from industrialized, rich, and democratic countries.
Spillover
Spillover: possible information could spread from groups to others
Natural Experiments
Natural experiment: sometimes nature does the intervention for us
-gives researchers the opportunity to test ideas based on natural interventions
causal effect
the difference between the value of an outcome when a subject recieves a treatment and when a subject does not recieve a treatment
Fundamental problem of causal inference
a situation refering to the fact that causal effects cannot be observed in the real world and that causal inferences are therefore required
Internal validity in experiments
does A really change B in this setting or is there something else
-biased environment or measurement tool
Single Blind design
Double blind design
an experiment that blinds the subjects to the randomization process is a single-blind design
for experiments with an individual administering the treatment the gold standard is doubleblindin which both subjects and treatment administrators are unaware of the effect being tested
Textual Analysis
“Textual analysis is the systematic examination of the messages and meanings conveyed by texts.”
Why study texts? of all types
“Most of us are spectators to political life, and the information and impressions we form are based on texts. …. Systematic analysis of these texts can tell us what is being said about politics and what is assumed or understood to be ‘true’.”
2 structural features of texts
– Format of communication
– Ways in which content is conveyed
– E.g., type of story; length of story
What is said and what it means: How works, themes,
symbols, etc. convey meanings, norms, and assumptions
• Manifest content: surface meaning
• Latent content: underlying meaning
Latent content
actual underlying meaning of a text
manifest content
surface meaning of a text
Textual Analysis(2 types)
1.“Content analysis is a quantitative research
technique used to analyze the message
characteristics in any formof communication.”
2.“Discourse analysis, in contrast, is a qualitative
approach that focuses on the meanings reflected in,
and created by, discourses.”