Formulating the research design

Report
Formulating the research design
• Research design - the general plan of how you
will go about answering your research
questions
- Objectives
- Sources
- Constraints
philosophy = question = design
The purpose of your research
• Exploratory studies: good means of finding out
- what is happening;
- to seek new insights;
- to ask questions;
- to assess phenomena in a new light
• Useful, if you want to clarify your understanding of a
problem
• Ways to conduct:
- a search of the literature;
- interviewing “experts” in the subject;
- conducting focus group interviews
Descriptive studies
• …to give an accurate profile of persons, events
or situations
• „…so what?“
• Mostly should be considered as a means to an
end rather than an end in itself
Explanatory studies
• …establish causal relationships between
variables
Different research strategies
• …does the particular strategy enables you to
answer your specific research question and meet
your objectives
• …choosinge your strategy according to:
- your question
- your existing knowledge
- the amount of available time/ resources
- your philosophical views
Strategies are not mutually exclusive
Experiment
• study causal links – whether a change in one
independent variable produces a change in
another dependent variable
• “how” and “why” questions
• Experimental vs control group
• Participants assigned at random to groups
• Internal vs external validity; constraints of
experiments
In sum, experiment will involve
typically…
• Definition of a theoretical hypothesis
• Selection of samples of individuals from known
populations
• Random allocation of samples to different
experimental conditions, the experimental group and
the control group
• Planned intervention or manipulation to one or more
of the variables
• Measurement of a small number of dependent
variables
• Control of all other variables
Survey
• “who”, “what”, “where”, “how much” and “how many”
questions
• Benefits:
- large data
- large population
- higly economical
- standardized tools: comparing samples
- perceived as trustable
- easy to explain and understand
- quantitative analysis
- possible to build models of relationships
- control over the process
- representativeness
Case study
• strategy for doing research which involves an
empirical investigation of a particular
contemporary phenomenon within its real
context using multiple sources of evidence
• .. and once again, context
• questions “why”, “what” and “how“
• techniques: various (interviews, observation,
documentary analysis, questionnaires)
• Triangulation - the use of different data collection
techniques within one study -> to ensure that the
data are telling you what you think they are
telling you
• Single case v. multiple case
• Holistic case v embedded case
Looks „unscientific“? :P
Action research
1. research in action rather the research about
action
2. collaborative democratic partnership
between practitioners and researchers
3. cycle of diagnosing, planning, taking action
and evaluating
4. the results should be applicable/ informative
in other contexts
Grounded theory
• … to predict and explain behavior
• … emphasis upon developing theory
• data collection starts without the formation of
the initial theoretical framework
• data generated by a series of observations ->
the generation of predictions -> predictions
tested in further observations -> confirm/ not
the predictions
Ethnography
• … to describe and explain the social world the
research subjects inhabit in the way they
would describe and explain it
• … researching the phenomenon within the
context in which it occurs
• … not using data collection techniques that
oversimplify the complexity of everyday life
Archival research
• … administrative records and documents as
the principal source of data
• questions about the changes
Multiple methods choices – combining quantitative
and qualitative techniques and procedures
Multi-method:
- multi-method quantitative study
- multi-method qualitative study
Mixed methods:
- mixed method research
- mixed model research
Time horizons
• Cross-sectional studies
- “snapshot”
- often survey strategy is used
• Longitudinal studies
- questions related to change
- control over measured variables
- possibility to use data collected earlier
The credibility of research findings
• Reliability
• Will the measures yield the same results on
other occasions?
• Will similar observations be reached by other
observers?
• Is the transparency in how sense was made
from the raw data?
Threats to reliability
•
•
•
•
Participant error
Participant bias
Observer error
Observer bias
Validity
• … is the relationship between two variables a
causal relationship
• The threats of validity:
- History
- Testing
- Mortality
- Maturation
Ambiguity about causal direction
Generalizability (external validity)
The ethics of research design
The main ethical issues are related to:
• Privacy of participants
• Informed consent; voluntarily nature of
participation and the right to withdraw from the
study
• Confidentiality of data provided by participants
and their anonymity
• Data collection/using/analysing/reporting
methods that may cause negative impacts to the
participants, including embarrassment, stress,
discomfort, pain, harm
Homework
• Please select the research strategy that would
be most appropriate for your own research
question; explain, why you choosed that one.
Write a short passage (approx 0,5 A4) and
email it to me on Friday, 29th March.
(If some of you feels that you do not want to
continue with his/her research question you
described in last homework, it is ok if you
generate new one )

similar documents