A QUESTIONS AND ANSWERS APPROACH TO A+ IN COM
312 EXAM.
Content analysis
The quantitative study of recorded human communication;
Involves placing observed items (units) into well-defined categories (representing
variables)
Content Analysis Myth #1
Content analysis is easy; Objectivity (intersubjectivity) ◦ Inter-coder reliability is hard to
achieve ◦ Manifest vs. latent content (e.g. issue vs. valence)
Content Analysis can be applied to?
Can be used to analyze written, spoken, or visual communication:
content analysis myth #2
The term "content analysis" applies to all examinations of message content
content analysis myth #3
Anyone can do content analysis; it doesn't take any special preparation
what kind of prep does content analysis require
Coding rules , Coder training , Pilot studies
content analysis myth#4
Content analysis is for academic use only
what are the goals of content analysis
1. Generality (theoretical relevance)
2. ◦ Description: What's the problem/phenomena
3. ◦ Explanation: What inferences can we make about people/sources that created the
material
what is the first step of CA
Develop a proposition to test
what is the second step of CA
Review the literature
what is the third step od CA
3. Develop hypotheses and/or research questions
,what is the fourth step of CA
Use previous measures or adjust/adapt/create coding instructions/classification system
what is the fifth step of CA
Define your population, sampling units
what is the 6th step of CA
. Code messages
what is intercoder reliability
The level of agreement among coders
Cohen's Kappa
nominal
Scott's Pi
two coders
Krippendorf's alpha
two or more coders, any type of measurement
what is the 7th step of CA
Analyze
what is the 8th step of CA
Interpret
threat to reliability 1
Poor coding scheme
threat to reliability 2
Inadequate coder training
threat to reliability 3
coder fatigue
threat to reliablility 4
Rogue coder
what is a strength of experiments
Causal mechanism determined
what is a weakness of experiments
Hard to make generalizable (artificial environment/non-representative sample)
what is a strength of surveys
, Generalizable, good for variables we cannot manipulate
what is a weakness of surveys
Causality determination not absolute, working with perceptions
what is a strength of CA
Not tampering with environment, looking at communication in context
what is a weakness of CA
Hard to determine source motivations, causal effects on human behavior
what is a population
Universe of events
what is a sampling
Selecting "events" (often people) from a population
important aspect #1 of sampling
Representative sample: A sample must reflect the same degree of variance in the
actual population
important aspect #2 of sampling
Generalizability
important aspect #3 of sampling
Sampling error: The degree to which a sample differs from the population; Random
error&Sampling bias
Random Sampling
probability sampling): Each event in the population has an equal chance of being
selected. First two require a sampling frame.
Stratified random sample
Sampling in a way that represents known portions within a population (e.g. by race,
gender, age, etc.)
Cluster sampling
Requires moving through the different stages (clusters) within a sample
Nonrandom Sampling
Simple convenience sampling; Volunteer sampling &Inclusion/exclusion criteria
Quota sampling
The nonrandom version of stratified sampling
312 EXAM.
Content analysis
The quantitative study of recorded human communication;
Involves placing observed items (units) into well-defined categories (representing
variables)
Content Analysis Myth #1
Content analysis is easy; Objectivity (intersubjectivity) ◦ Inter-coder reliability is hard to
achieve ◦ Manifest vs. latent content (e.g. issue vs. valence)
Content Analysis can be applied to?
Can be used to analyze written, spoken, or visual communication:
content analysis myth #2
The term "content analysis" applies to all examinations of message content
content analysis myth #3
Anyone can do content analysis; it doesn't take any special preparation
what kind of prep does content analysis require
Coding rules , Coder training , Pilot studies
content analysis myth#4
Content analysis is for academic use only
what are the goals of content analysis
1. Generality (theoretical relevance)
2. ◦ Description: What's the problem/phenomena
3. ◦ Explanation: What inferences can we make about people/sources that created the
material
what is the first step of CA
Develop a proposition to test
what is the second step of CA
Review the literature
what is the third step od CA
3. Develop hypotheses and/or research questions
,what is the fourth step of CA
Use previous measures or adjust/adapt/create coding instructions/classification system
what is the fifth step of CA
Define your population, sampling units
what is the 6th step of CA
. Code messages
what is intercoder reliability
The level of agreement among coders
Cohen's Kappa
nominal
Scott's Pi
two coders
Krippendorf's alpha
two or more coders, any type of measurement
what is the 7th step of CA
Analyze
what is the 8th step of CA
Interpret
threat to reliability 1
Poor coding scheme
threat to reliability 2
Inadequate coder training
threat to reliability 3
coder fatigue
threat to reliablility 4
Rogue coder
what is a strength of experiments
Causal mechanism determined
what is a weakness of experiments
Hard to make generalizable (artificial environment/non-representative sample)
what is a strength of surveys
, Generalizable, good for variables we cannot manipulate
what is a weakness of surveys
Causality determination not absolute, working with perceptions
what is a strength of CA
Not tampering with environment, looking at communication in context
what is a weakness of CA
Hard to determine source motivations, causal effects on human behavior
what is a population
Universe of events
what is a sampling
Selecting "events" (often people) from a population
important aspect #1 of sampling
Representative sample: A sample must reflect the same degree of variance in the
actual population
important aspect #2 of sampling
Generalizability
important aspect #3 of sampling
Sampling error: The degree to which a sample differs from the population; Random
error&Sampling bias
Random Sampling
probability sampling): Each event in the population has an equal chance of being
selected. First two require a sampling frame.
Stratified random sample
Sampling in a way that represents known portions within a population (e.g. by race,
gender, age, etc.)
Cluster sampling
Requires moving through the different stages (clusters) within a sample
Nonrandom Sampling
Simple convenience sampling; Volunteer sampling &Inclusion/exclusion criteria
Quota sampling
The nonrandom version of stratified sampling
