5
Mike Phillips
502 A01 Wiki Notes
Week 9: Qualitative Data
Overview
· Content Analysis
Exercise- News Videos
· Thematic Analysis
Exercise- Government Strategy for Single Mothers
· Grounded Theory
· Post-Structuralist Discourse Analysis
Last Week’s Wiki
· An outside reader would not be able to understand what that page meant; it is not a standalone document
· Its less about not enough detail, it’s about sentences that actually make sense- Ex. Hermeneutics is not a person, it’s a philosophy- “Hermeneutics do not conduct research”
· Avoid cryptic comments, typos (the more there are, the more difficult it makes it for the reader)
Aim of Unit 9 & outline
· Today, our class is mostly a workshop focusing on Content Analysis and Thematic Analysis. Its aim is to introduce four common approaches to qualitative research
Review of data collection
· Data collection (wk 8)
o Interviews
o Documents
o Observations
o Surveys
Content analysis
· The use of objective and systematic counting and classification procedures to produce a quantitative description of the symbolic content in a text
· Not used a lot in 598’s but it is useful because it gives you an understanding of one spectrum of qualitative research
· It is the most structured and it allows you to visualize the data.
o Content- words, meanings, pictures, symbols, ideas, themes, or any message that can be communicated
o Text- anything written, visual, or spoken that serves as a medium for communication
Source: W. Lawrence Neuman, Basics of Social Research, 2nd edition.
· Usually used within a positivistic or post-positivistic epistemological framework.
· Concerned with:
o Systematic and objective coding
o Creating a coding system that can be replicated by others
· Most commonly used with a large volume of data and it allows you to make generalizes to the whole population. Twitter’s trend analysis tool is a form of content analysis by showing the most common words.
Steps in Conducting Content Analysis
1) Choose your unit of analysis- newspaper article (is it the whole article, a line or a paragraph, what counts as your unit of analysis?)
2) Create a code book with:
o A coding frame: a list of items (variables you want to count) and how you want to count them (values)
Ex. Look at our printout, each time a data item appears, mark it in the box below: anchor (M or F), reporter (M or F), experts (M or F), also hard or soft news. You could do this week after week to create a data set.
Anchor / Reporter / ExpertM or F / M or F / M or F
Story # 1
Topic
Hard or Soft News
o -Values must be mutually exclusive and exhaustive- has to be either hard or soft, or M or F, not both. It must cover everything (no, medium news, transsexual anchor/reporter, etc.)
3) Coding notes: rules for coding Ex. how do you whether it is hard or soft news? What counts?
Exercise
National newscasts: code top stories based on the handout (as you watch the news, fill in the blanks of the exercise sheet above.)
o CBC The National http://www.cbc.ca/thenational/watch/
o CTV National News http://www.ctv.ca/ctvnews/
How did you classify the anchors, reporters, experts? (People in class differed in what they thought counted as experts; there were also slight differences in how people classified the news stories as hard or soft.)
Strengths and Weaknesses of Content Analysis
Strengths
· It can ensure that the analysis is:
o Rigorous (categories are precisely defined)- There are some differences, however, the amount of similarity was quite high and there was a lot of agreement between us
o Reliable (measures are applied consistently)- Another group would find their results to be similar ours. You would need to include your code book describing your framework and your definitions of the variables. In your 598 you would definitely attach this code book. You can do the same with interviews, articles, any sort of document.
o Replicable (the project can be replicated by another researcher).
· Works well with multiple coders – even in our exercise we used multiple codes with ease.
Weaknesses
· Measurement is tricky; how you count counts (seemed fairly straightforward but there were some differences in how we counted)
o You cannot count what is not there.
o Dealing with frequency, replication, intensity- as the number of codes used increases so too does the complexity
· Concerns about the validity of the texts- have you chosen your texts correctly? Is the news source the best source? Is this a valid measure? Etc.
· Not everything that counts can be counted- When talking about hard or soft news, are there variations? There are certain qualitative things that are missed when doing our observations. We can simplify things and possibly miss the bigger picture.
Thematic Analysis
Thematic Analysis is a little less structured than content analysis. An example by Ritchie and Spencer in their article can be found here: Ritchie, J. & L. Spencer (2002). Qualitative data analysis for applied policy research. In A.M.
o Fits with positivist, post-positivist and hermeneutic epistemology
Steps in Thematic Analysis
o Data is reviewed- Read each of the transcripts and highlight key points and themes as well as make notes on the range of questions being answered
o A thematic framework is identified- Using your themes and your research topic/question, create an index that you can use to code the transcripts.
o An index is created and applied to the data- Code each transcript by assigning the codes to sentences in the transcripts
o Summaries of data are entered into a chart-
o These summaries are charted and brought together- You bring together responses to all the questions to answer the overall research question using your common themes.
Strengths and Weaknesses of Thematic Analysis
Strengths
o Data is clearly defined by your index, and easy to use
o Process is quite:
§ reliable (measures are applied consistently)
§ replicable (the project can be replicated by another researcher)
Weaknesses
o Analysis may be reductive-you are being encouraged to break down the data and might lose the context.
o Themes are considered in isolation from broader social context- you are not considering issues of what people might be willing to reveal in an interview.
o Issues of power not explicitly considered- They might not be explicit in disagreeing with government because of power concerns.
Grounded Theory
There is an exercise in the handout on how to do grounded theory. If you are interested in doing grounded theory, you can do this exercise on your own time after class.
Features
· Developed by Glaser & Strauss: The Discovery of Grounded Theory. 1967
· Used a lot by applied disciplines: (nursing, teaching, physical education, public administration).
· Purpose: theory grounded in people’s experiences. Ex. Research a social group and understand their world view. Trying to develop a process where you understand their world view.
· Epistemology – hermeneutic / post positivist. A lot more structured than other hermeneutic research data collection techniques.
Coding Steps
Quite similar to the thematic analysis steps, although there are less steps and it is quite broad compared to thematic analysis.
o Open coding
o Open up, break down and fragment the data similar to indexing data
o Create categories to code the data
o Creating memos
o Axial coding- more emphasis on trying to establish and create relationships between the two
o Selective coding-
o It is similar to the maps in the final step in thematic analysis
o Establish your ‘story line’ – another way of stating an answer to your research question. It involves creating memo’s and summaries.
Strengths and Weaknesses of Grounded Theory
Strengths
o May be less reductive than thematic analysis- you are not focusing so much on breaking down and categorizing data in grounded theory as you do in other qualitative research techniques
o Data analysis steps are flexible yet reasonably clear-
Weaknesses
o It is supposed to have no theoretical lens You come in without a theory and try to develop a theory through your understanding and observations of people– not clear if this is possible.
o Themes are considered in isolation from broader social context (and texts)
o Issues of power not explicitly considered- Trying to understand the world view of the group and how the relationships between people are based on power.
Post-Structuralist Discourse Analysis
What is Discourse?
· “System of statements which constructs an object” (Parker) language creates the object. Ex. Single mothers and all the associations we have with that word
· Language and statements that allow us to make sense of and see things.
· They are systematic and should be repetitive
· Discourses involve power and knowledge.
Post-Structuralism
· Rejection of simplistic causality- we can’t simply find cause and effect in the world,
· Do not believe in inevitable and linear historic progress
· Concerned with relations between power, knowledge and language – Ex. how bullying is defined in the text and how it is handled in the workplace based on how it is defined.
Strengths and Weaknesses of Post-Structuralist Discourse Analysis
Strengths
o Themes are not considered in isolation from broader social context (and texts)
o Issues of power are not explicitly considered
o Least reductive approach
Weaknesses
o Data analysis steps not necessarily clear
o Communication between client can become an issue
o Not suitable for positivistic research questions and designs
Summary
· A range of approaches to qualitative data analysis.
o Content Analysis
o Thematic Analysis
o Grounded Theory
o Post-Structuralist Discourse Analysis
· Whichever method you use will depend on your research question and your epistemology.