D: data analysis: manual and computer-supported Flashcards

1
Q

What to do with so much data?

A

Combine inductive and deductive logic with the data
First use induction to see patterns and give interpretations of the data
Next, use deductive approach by looking back at existing research and literature and making an analysis based on that data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Primary coding

A

Primary cycle of coding refers to inductive logic

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Secondary coding

A

Secondary cycle coding looks at deductive logic

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Primary and secondary coding diagram

A

Data → primary coding → theory/literature → secondary cycle coding
Specific instances → inductive → general principles → deductive

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Steps for iterative qualitative analysis

A
  1. Preparing raw data
  2. Primary cycle: immerse yourself in the data, create “open” codes
  3. Creating a code book
  4. Secondary cycle coding: create the hierarchical categories
  5. Revisit the scholarly literature
  6. Crete relationships between the categories
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Step 1: Preparing raw data

A

After you collect data
Organize your data, especially on a large scale, use the filed notes and transcripts
- assign pseudonyms for participants (match first letter of name)
- organize data in order you plan to analyze: chronological orders, by type (interview/observation), or through sources (employees/managers)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Step 2: primary cycle: immerse yourself in the data

A

Read through transcript, take it in, ask yourself “what is going on here, what do you see”
Identify data as belonging to or representing some type of phenomenon
Create codes that are words and short phrases that capture the salient attributes of your data, descriptive and from the data itself
Goal: to detail who, what, where, not why or how

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Step 2: primary cycle: create “open” codes

A

Find balance in how much detail have you in your open coding (not fruit, but orange)
Code manually by hand printing your transcripts and highlight/comment or use atlas.ti
Constant comparative method: Code on the level of the sentence of the paragraph, decide wether the line below it deserves a new code or not, create “in vivo” codes that stick closely to what the participants of your data said and what you observed
Goal: capture participants meanings

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What is a code

A

A label that you apply to a sentence or paragraph

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Constant comparative method

A

Constantly compare new uncoded data with your old codes
Try to apply the same codes to see if it fits with the data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Step 3: creating a codebook

A

Combine all your interviews in 1 list and generate 1 code book, clean up codes to avoid duplicates
Count amount of times each code appears, if there is only 1 of a code maybe remove the code or do more research
Assess your codes against your initial research, does it help you answer the RQ
Create definitions for all remaining codes, chose a quote that best represents the code

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is a codebook?

A

A map of your data, tells in an overview what the content of your data is
Can be created with excel or on atlas.ti

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Step 4: secondary cycle coding

A

A form of data reduction
The first step towards noticing broader patterns in your data
Help to answer your RQ: moving from describing to explaining and theorizing
You can begin to see where your analysis can “fill gaps” in existing research and show theoretical relevance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

By the end of the codebook you should have

A

Created hierarchical categories and have done research into literature to show linkage to your data and existing research

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How to you make hierarchical categories

A

Group codes that seem close to each other
Cut/paste one thats look like each other and put them together
Natural categories will form, name or color code them

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Computer-supported qualitative data analysis

A

Researchers should not fully rely on computers to analyze their data
Researchers should be able to interpret it by ourselves
Researcher knows the interviewee, computer doesn’t

17
Q

How can computers help with qualitative data analysis

A

Indexing: index words and address of a text message of line numbers or where to find something, code your data and then easily retrieve it through indexing, done through page numbers
Cross referencing: constructed through hyper links, click on them to go the other text or another part of your code