|
Table
3.2 Data Collection Techniques
|
|---|
| Data
Type
| Instrument
Design
| Administration
| Analysis
| Efficiency
| Limitations
|
|---|
|
1. Interviews
| | | | | |
|---|
| 1a.)
Semi structured
| - Questions
which encourage short answers. No suggested answers
| - Usually
self administered
| - Can
be largely precoded even machine entered
| - More
time spent in question development, fastest to analyze, best for large
numbers, good for detail work
| - Not
good for pilot work or new ideas
|
|---|
| 1b.)
Open ended
| - Questions
which encourage long responses
| - Can
be self administered, data usually better with an interviewer
| - Need
qualitative analysis techniques
| - Good
for exploring new areas or populations
| - If
self-administered answers can ramble and may not address the information
you seek
|
|---|
| 1c.)
Cognitive Testing
| - Exploration
of questions’ meaning e.g. "In your own words, what do you think
question 6 means?"
| | - Fast
to "eye-ball", can use qualitative analysis techniques
| - Assures
validity of survey or education content
| - Requires
interviewer only good with well developed materials
|
|---|
|
2.) Focus
Groups
| - Strict
protocol, group process, strict eligibility and carefully worded questions
| - Requires
organization, trained facilitator, and assistant, quiet room, usually
food, transportation and stipends
| - Need
qualitative analysis techniques
| - If
protocol followed and used well-formed questions yields excellent data
| - Takes
training and organization
|
|---|
|
3.) Narratives
| - Similar
to open-ended interviews, but usually only a few questions
| - Always
self administered, may be written or typed
| - Need
qualitative analysis techniques
| - Data
are in patients’ or clinicians’ voices, tell stories and are usually
very rich.
- Good
for exploration
| - Most
difficult data to collect.
- Variation
in value by respondents’ skills.
- Very
time consuming to analyze
|
|---|