research by design
TRANSCRIPT
Interdisciplinary Research by Design
MAIS 796
Paul M. Rogers PhD
George Mason University
You can’t fix through analysis
what is flawed by design.
You can’t fix through analysis
what is flawed by design.
Professional contextsKind of problems
Options in research
Contexts• What counts as
a professional context? What elements make up “the context”
Problems• What kinds of
problems warrant systematic research?
Questions• What do we
need to understand?
What is context?
Situated Global
Industries, Sectors, Clusters, Organizations
Four Major Industries
• Natural Resources – Farming, mining, fishing, logging etc …
• Manufacturing – Processing materials from natural resources
• In-person services – Restaurants, styling salon, janitorial, landscaping, construction, teachers?
• Knowledge work – Scientists, creatives, engineers, architects, writers
Clusters
Clusters are geographic concentrations of interconnected companies, specialized suppliers, service providers, and associated institutions in a particular field that are present in a nation or region. Clusters arise because they increase the productivity with which companies can compete.
Professional contextsKind of problems
Options in research
Contexts• What counts as
a professional context? What elements make up “the context”
Problems• What kinds of
problems warrant systematic research?
Questions• What do we
need to understand?
Concrete problems lead to clear questions
From contexts to problems to questions
Professional contextsKind of problems
A
c
a
d
e
m
i
c
s
P
r
a
c
t
i
t
i
o
n
e
r
s
From contexts to problems to questions
Professional contextsKind of problems
Options in research
Quantitative Research
otherwise known as
Statistics
random assignment random sample
population mean
median mode
sample external validity
internal validity variable
discrete variable continuous variable
measurement data independent variable
dependent variable frequency distribution
histogram boxplot
measures of central tendency degrees of freedom
standard deviation null hypothesis
hypothesis testing statistical significance
Three types of statistical research
• Descriptive
– Describes the way things are; not why.
• Relational
– Relationships between factors: correlation not causation
• Experimental
– Establish causality through experimentation.
Experimental Studies
• Random selection of subjects
• Use of control groups
• Random assignments to control and experimental groups
• Random assignments of groups to control and experimental conditions
– Greater internal validity, causal claims can be checked
– Less external validity, not very practical
(RAD)
The beauty of experimental studies
comes for the clear inferences you can
draw from the research
Quasi-experimental Studies
• Have control groups, but without random assignment and selection
– Greater external validity
– Easier to carry out
– Less control
– Harder to make causal claims
Qualitative – Quantitative
Multiple variable, synthetic-single variable analytic
Interpretive and impressionistic-numerical
Low control – high control of the research context
Hypothesis generating-hypothesis testing
3 Things to Remember
• Explore many design options before adopting a plan, especially a weak one
• Pay attention to detail
• Conduct yourself respectfully at all times
The importance of research questions
• Well crafted questions guide the systematic planning of research
– Who or what do you want to study?
– How should you collect and filter data?
– What time frame is appropriate for your research?
• influence the choice of research methods
• identify your target population from which you draw your sample
• determine your unit of analysis
• clarify terms
• identify predictors
• determine the type of study you will design
• help you account for different background characteristics
• raise challenges related to data collection
• Influence the number of people, texts, tools, or processes you will study
Research questions …
Influence and possibly determine every facet of your research design
P.S. make sure at the end of the day your project answers your questions
Research questions …
Exercises
• Goal: Articulate a set of research questions
1. Identify, before you begin, the industries, sectors, organizations, people, texts, processes, and/or tools you are most interested in.
2. What practical problems interest you? Why?
3. Make a list of questions that would need to be answered to help inform solutions to the problem? Rank your list in order of importance?