Research Workshop: Content Analysis
Week 1 - Introduction, RQs, and Sampling
What is content analysis?
Systematic and replicable examination of symbols of communication, which have
been assigned numeric values according to valid measurement rules, and the
analysis of relationships involving those values using statistical methods, to
describe the communication, draw inferences about its meaning, or infer from
communication to its context, both of production and consumption
Systematic Generalizable empirical evidence, identification of key
concepts, generation of testable hypotheses. E.g. the
scientific method.
Replicable Definitions are reported carefully so that readers can
fully understand the procedure, evaluate it and its
findings, and repeat it if desired. There must be
consistency among each coder.
Symbols of communication Different content is interpreted differently based on
medium, who is coding, function, etc. This can be
avoided with careful and detailed operationalization.
Numeric values or categories Numeric values are assigned to represent measured
differences. These are not assigned based on counting,
but on some sort of score or rating.
Describing and inferring Description is essential to the early phase of a research
program (e.g. in cases where events are very far from
reality, like hurricane Katrina).
Simple descriptive data invites inference testing
(conclusions about what is/isn’t observed). To draw
inferences, the researcher must be guided by theory
Research method that can be used to analyse characteristics of registered content of
communication
● Quantitative analysis of media content
○ Start with more manifest constructs to more latent
● Incredibly broad, encompasses a lot
● Deals with representation in the media
● Used by corporations to examine public view
,Why content analysis?
● Can connect media content data with survey data
○ Superior to asking people (more objective)
○ Independent variable in many studies about media impact
○ Explains media impact rather than only speculating
● Can be used for stimulus material in experiments
● Explores how certain groups are portrayed in the media
○ Offers insight into mediated reality
○ Media performance: How well are media representing reality?
○ Power relations: How does this content originate?
● Accessible
○ Costs very little money and can gather a lot of data
● Non-obtrusive
● Non-reactive (control over data collection process)
● Longitudinal in no time
● Can be used in basically every field
Content analysis is useful when
1. Data accessibility is a problem and the investigator is limited to using documentary
evidence
2. The communicator’s “own language” use and structure is critical
3. The volume of material exceeds the investigator’s individual capability to examine it
Centrality model of communication content:
● Cause of / prediction for media effects
● Outcome of media production and environmental factors
● Can be a “consequence” of dominant social ideologies at the time/place
● Fundamental to theory building in communication science
, Criticisms by Holsti
● Puts too much emphasis on comparative frequency on different symbol’s appearance
○ Sometimes specific symbols are crucial and it’s not only about the numbers
● Sometimes problems are chosen only because they are quantifiable
● Hard to distinguish between manifest and latent constructs
○ Content analysis makes no claim beyond manifest constructs
Research questions
Can be more descriptive or more explanatory, both on a spectrum
Descriptive Explanatory Comparative
● Describe the presence of ● Explain variation in (a) ● Makes descriptive
characteristics of media the presence of research questions more
content in a given period characteristics in media explanatory
content within a given
● Dominant in scientific time period or (b) ● Can compare
studies / practice characteristics of users of ○ Media categories
oriented content media content ○ Content
analyses characteristics
● Research about (causal) ○ Time periods
● Not ideal for latent relationships ○ Issue contexts
constructs ○ Environments
● Media content can be
● Raise “why?” questions explained but can also be
but don’t answer them an explanation
● Stimulate curiosity, can ● E.g. Is the frequency of
lead to looking for news about victims of a
potential causes and war positively related to
consequences the actual number of war
victims?
● E.g. How was Facebook
evaluated as a
corporation in economic
news stories of quality
Week 1 - Introduction, RQs, and Sampling
What is content analysis?
Systematic and replicable examination of symbols of communication, which have
been assigned numeric values according to valid measurement rules, and the
analysis of relationships involving those values using statistical methods, to
describe the communication, draw inferences about its meaning, or infer from
communication to its context, both of production and consumption
Systematic Generalizable empirical evidence, identification of key
concepts, generation of testable hypotheses. E.g. the
scientific method.
Replicable Definitions are reported carefully so that readers can
fully understand the procedure, evaluate it and its
findings, and repeat it if desired. There must be
consistency among each coder.
Symbols of communication Different content is interpreted differently based on
medium, who is coding, function, etc. This can be
avoided with careful and detailed operationalization.
Numeric values or categories Numeric values are assigned to represent measured
differences. These are not assigned based on counting,
but on some sort of score or rating.
Describing and inferring Description is essential to the early phase of a research
program (e.g. in cases where events are very far from
reality, like hurricane Katrina).
Simple descriptive data invites inference testing
(conclusions about what is/isn’t observed). To draw
inferences, the researcher must be guided by theory
Research method that can be used to analyse characteristics of registered content of
communication
● Quantitative analysis of media content
○ Start with more manifest constructs to more latent
● Incredibly broad, encompasses a lot
● Deals with representation in the media
● Used by corporations to examine public view
,Why content analysis?
● Can connect media content data with survey data
○ Superior to asking people (more objective)
○ Independent variable in many studies about media impact
○ Explains media impact rather than only speculating
● Can be used for stimulus material in experiments
● Explores how certain groups are portrayed in the media
○ Offers insight into mediated reality
○ Media performance: How well are media representing reality?
○ Power relations: How does this content originate?
● Accessible
○ Costs very little money and can gather a lot of data
● Non-obtrusive
● Non-reactive (control over data collection process)
● Longitudinal in no time
● Can be used in basically every field
Content analysis is useful when
1. Data accessibility is a problem and the investigator is limited to using documentary
evidence
2. The communicator’s “own language” use and structure is critical
3. The volume of material exceeds the investigator’s individual capability to examine it
Centrality model of communication content:
● Cause of / prediction for media effects
● Outcome of media production and environmental factors
● Can be a “consequence” of dominant social ideologies at the time/place
● Fundamental to theory building in communication science
, Criticisms by Holsti
● Puts too much emphasis on comparative frequency on different symbol’s appearance
○ Sometimes specific symbols are crucial and it’s not only about the numbers
● Sometimes problems are chosen only because they are quantifiable
● Hard to distinguish between manifest and latent constructs
○ Content analysis makes no claim beyond manifest constructs
Research questions
Can be more descriptive or more explanatory, both on a spectrum
Descriptive Explanatory Comparative
● Describe the presence of ● Explain variation in (a) ● Makes descriptive
characteristics of media the presence of research questions more
content in a given period characteristics in media explanatory
content within a given
● Dominant in scientific time period or (b) ● Can compare
studies / practice characteristics of users of ○ Media categories
oriented content media content ○ Content
analyses characteristics
● Research about (causal) ○ Time periods
● Not ideal for latent relationships ○ Issue contexts
constructs ○ Environments
● Media content can be
● Raise “why?” questions explained but can also be
but don’t answer them an explanation
● Stimulate curiosity, can ● E.g. Is the frequency of
lead to looking for news about victims of a
potential causes and war positively related to
consequences the actual number of war
victims?
● E.g. How was Facebook
evaluated as a
corporation in economic
news stories of quality
