chapter 8 Flashcards
What is a reactive study? How is it problematic?
Most social research, experiments and surveys. A situation where participants are aware that they are being studied. Their reactions will be influenced by their awareness of the study. It is problematic because people may change their behaviours because of awareness.
What is a non-reactive study? What are the 4 techniques?
This is a quantitative or qualitative study where participants are not aware that they are being studied. Researchers collect passive data (data that they created) The 4 research techniques are physical evidence, content analysis, existing stats analysis and secondary source analysis.
What are the advantages and disadvantages of non-reactive studies?
the advantages are that you get truer results because people cannot act differently because you are studying data about them, and it is cheaper/faster/easier to get passive data.
the disadvantage is that you have little control over how the data is collected, must rely on others’ organization+ integrity. You need to be creative in finding surrogates, there is no direct way to get data that you want. You have to infer yourself the data’s meaning and significance.
what are the limitations of physical evidence?
- they are indirect indicators, meaning that you must make educated guesses about the evidence. You must incorporate more evidence to be able to make inferences.
- You have to rule out other explanations. There can be many explanations for certain physical evidence
- privacy violation, you actively work to protect confidentiality and anonymity of participants, only using the data they created.
- you need creativity to think about what the evidence might indicate.
When is content analysis useful?
- When we are faced with a large quantity of data
- when you study topics from a distance (like from a different period or in another country)
- When recognizing the content is hard with casual observation
What is a coding system and why is it useful?
A coding system is about operationalizing variables in content analysis by measuring changes in concepts. Its a set of rules that helps to categorize and classify information from text. You need to create exhaustive and exclusive categories. It is useful because you need to change large amounts of data into quantitative data and it helps when someone want to copy your work.
what is a unit of analysis?
It varies depending on your content analysis. It can be an article, a character, a phrase. You have to tailor your coding system to it.
What do you measure with Content analysis?
- Direction (bad/good related to an issue)
- Intensity (strength of a variable)
- Frequency(How many time it comes up)
- Prominence (Where is it placed in the text)
- Space (amount of time)
What are the two types of coding? What are their ideal usage?
- Manifest coding: when you count number of time a word, symbol.. appears in the text.
- Latent (subjective): when you use context and overall text to see if it contains specific themes.
- Manifest is reliable but it doesn’t take meaning of words into account. Latent is less reliable but takes meaning. Manifest is usually more accurate.
- Ideally, you use them both even tho it takes longer. If they agree, that validates your point but if they disagree you should review your coding system.
What is intercoder reliability?
It is when you have different people using the same coding system to analyze different content. The assistants must understand variables and follow system. All your coders must code the same, so you must check all their consistency, which is the intercoder reliability, measure degree of consistency using a statistical coeff (0-1). You must report the coefficient with your results.
Explain how content analysis works with visual text and cultural symbols
Visual text: photos, paintings, buildings that communicate messages or emotional content through symbols and metaphors. Its mixed messages with different levels of meanings. It takes skill to interpret those meanings, it isn’t mechanical and depends on cultural context (diff cultures associate different meanings)
Culture symbols: Cultural symbols have many meanings, and visual stuff is multilayer with meanings. Symbols can be a source of conflict because of competing socio-politic groups who want to attach their own meanings to symbols.
Steps of Content Analysis research
1) formulate research question
2) Select the communication medium to analyze
3) Decide a unit of analysis
4) Draw a sample (randomly find article)
5) Create a coding system (manifest vs subjective)
6) Construct+refine coding strategies (distinctions in variables, levels)
7) Code data on recording sheet
8) Analyze the data
What are the limitations of content analysis?
With content analysis, we can’t:
- Determine the truthfulness of a message
- Can’t evaluate aesthetic quality of the visual text
- Significance or importance of the text
- The intentions of the author
- Influence on the receivers.
How is using existing statistics different from other techniques?
Statistics differ because you start off research by locating+ learning data before ever developing a research question.
what are the variants of verifying data quality in existing statistics?
1) Missing data: Major limitation, usually data that is lost or has never been collected. This often happens because of budget cuts and political reasonings who do not want certain information to be researched. Must deal with data from less refined categories.
2) Reliability: the organization has a different data collection definion or way to calculate things.
3) Validity: different conceptual definition that organization (what an injury is), use other incomplete statistics as proxy for your research, you must depend on others to get data(people make errors).
4) Topic Knowledge: researching about a topic that you know little about can cause many errors and bad assumptions.
5) Fallacies: Theres misplaced concreteness (when you quote stats in too much detail to seem more precise) and ecological (using wider statistics for a smaller subject)
