doing qualitative research: study design, sampling, data collection elizabeth boyd, phd epi 240...
TRANSCRIPT
Doing qualitative research: study design, sampling, data
collection
Elizabeth Boyd, PhD
EPI 240
January 15, 2008
Recap: Using qualitative methods To interpret, illuminate, illustrate To understand why or how To describe previously unstudied
processes or situations To learn about subjects who are few or
hard to reach To brainstorm ideas
Research questions - what do you want to understand? Meaning Context Unanticipated
phenomena Process Explanations
Opinions Attitudes Understandings Actions
Formulating research questions Pub bias example Your examples Common problems:
Too broad Impossible to find answers or ‘how to operationalize’ Really asking a quantitative question
Data collection Based on what you want to know:
Where do you go to find out? Site selection
Who do you ask/observe? ‘Sample’
What do you ask/do? Data
Choosing your site Justification: why is this site the best for
answering your research questions? Naturalistic?
Ethnography; video Public vs. private settings
Neutral? Interviews; focus groups
Choosing your site Implications
Ethical Role as researcher-caregiver
Logistic How to gain access? How to gather data?
‘Sampling’ Who will you include in your study and
why? Everyone (ethnography) Sample -- need sampling strategy
Random Convenience Purposeful
Typical Hetereogenous Extreme cases/comparisons
Data collection Once you have identified your site and
participants, practical matters include: Timing -- when to go/how long to stay? Fitting in Establishing relationships -- Who? How? How much? Equipment:
Recording devices Hand notes Video/digital Impact on participants Cost/transportability/impact
Quality in qualitative research Two ‘phases’ -- data collection and write-up Overall: quality = credibility Credibility is achieved through depth,
clarity, nuance
Rich and sufficient data Enough background to understand and portray
full range of persons, processes, settings Detailed description of range of views & actions --
multiple perspectives Beyond superficial Analytic categories -- complexity Comparisons -- generative or general? Saturation -- stop seeing new cases/instances
Bottom Line Regardless of the type of data you are
working with, Ground ALL observations, analyses in the
particular details of your data
Managing your data … You’ve interviewed 10 (or 20, or 30, or 100)
people, now what? Transcription Coding Analysis
Transcription Written representation of the interview
Types of transcription: “Cleansed” transcript “Just the words” “Jeffersonian” transcript
The “cleansed” transcript Dr. E: I’m deputy editor of Annals of
Internal Medicine. I was associate editor from 1978 to 1999, and I was deputy editor from 1999 to 2003. My sub-specialty is pulmonary disease which I practice every day at the University of Pennsylvania. Most of the editors at Annals do practice, though not as extensively as I do. …
“Just the words” IR: So today is March seventh. I’m at Annals of
Internal Medicine and I’ll be interviewing Dr. P.E. And for the record can you state your name and position?
DrE: It’s P.E. I’m deputy editor of Annals of Internal Medicine.
IR: Okay. And how long have you been working at Annals?
DrE: Since 1978. It’s a long time. I was associate editor from 1978 to 1999 and I’ve been deputy editor from 1999 to 2003.
“Jeffersonian” transcript IR: So: today is March seventh, I’m at Annals of
Internal Medicine and I’ll be interviewing doctor Pete Ernest. (0.4) A::nd um for the record can you state your name and position?
IE: It’s Pete Ernest, I’m deputy editor of Annals of Internal Medicine.
IR: Okay. And how long have you been working at Annals?
(0.4) IE: Since nineteen seventy eight. It’s a lo::ng
time. I was uh:: associate editor from nineteen seventy eight t nineteen ninety ni:ne, …
Which transcription method to use? Speed versus detail and accuracy versus
cost What are you most interested in learning --
Content? Narrative? Interaction?/Context?
Assignment If necessary, refine your research questions Describe your sample and data collection
protocol Describe how you will prepare your data for
analysis Be sure to defend your choices!