Content Analysis Flashcards
(26 cards)
what is quantitative content analysis?
the systematic and replicable examination of symbols of communication or media materials for patterns
what are characteristics of content analyses?
1) objective
2) systematic
3) focused on manifest content
what is manifest content?
content that is observable
T or F: when conducting content analyses, it is important to define measurable scoring units
True
what are measurable scoring units?
the basic or standard unit or measurement
what are different kind of measurable scoring units?
words, phrases, minutes, images, entire documents
T or F: researchers are interested in the appearance of a phenomena of interest in a single document
True: ex - does a tweet contain at least one swear word
T or F: content analyses do not require a sample from a population
False: difference is we’re sampling media artifacts/content rather than people
T or F: content analyses require researchers to establish very clear/explicit definition of the behaviors/symbols of investigation
True
What are the different kinds of content definitions?
constitutive and operational definitions
what are constitutive definitions?
definitions found in dictionaries
what are operational definitions?
states how one measures something and forces an explanation to how one understands or interprets a concept
what are the uses of operational definitions?
- attitudes towards a brand
- website usability
- purchase intention
- organizations trust and reputation
T or F: most content analyses require the use of two or more coders who review the sample/population of media materials
True: used to make judgements regarding the presence of a variable of interest
T or F: operational definitions are used to generate a “codebook”
True
what is a codebook?
the guide that coders use to code media content
- set of instructions
T or F: quantitative content analyses require only one coder
False: requires at least two coders
what is intercoder reliability?
when two or more independent coders agree on the coding of the content of interest with an application of the same coding scheme
how is intercoder reliability demonstrated?
coders will jointly code a random subset of the overall sample
what is the general rule of them for intercode reliability?
the subset equal 10% of the overall sample
T or F: we need the coders to agree on 50% or more of the coded cases to claim intercoder reliability
False: need coders to agree on 70% or more of the coded cases
how is intercoder reliability measured?
assessing the degree to which something is present or absent
- this yield nominal-level data
- ex: does this tweet contain name called?
0 = no, 1 = yes
- can use more than two categories
T or F: it is possible to ask for interpretative/subjective evaluations
True, BUT these rarely result in intercoder reliability
- example: - 0 = non-violent, 1 = somewhat violent, 2 = violent, 3 = very violent
what are the advantages of content analysis research?
1) unobtrusive
2) relatively inexpensive
3) deal with current events and topic of present-day interest
3) user material that is relatively easy to obtain and work with
4) yields data that can be quantified
