Upload
aldous-manning
View
215
Download
0
Tags:
Embed Size (px)
Citation preview
The Practical Aspects of Doing Research
An Giang UniversityJune, 2004
Dennis Berg, Ph.D.
Tool Box Analogy
When We Begin We Have A Small Number of Tools
Sampling
Measurements
Statistics
Research Designs
Content Analysis
SPSS - SAS
Tool Box II
As We Learn We Collect More Tools Through
Experience and Training
STUDYING AND PRACTICING RESEARCH AND ANAYLSIS
1. Learn the Logic of Science 2. Learn the Process of Developing
and Asking Research Questions3. Learn the Nature of Research
Designs4. Learn the Tools and Techniques
of Producing Evidence5. Learn the Tools and Techniques
of Analysis and Conclusions
Types of Research Based on Purpose
Two Type Classification Basic Applied
Within Applied Policy Analyst Evaluation Social Monitor Data Analyst
BASIC RESEARCH
Seeks to discover “new knowledge about SOCIAL PHENOMENA, hoping to establish GENERAL PRINCIPLES and THEORIES with which to explain them.”
The goal is “to produce new knowledge, including the discovery of the nature of RELATIONSHIPS between VARAIBLES.”
Applied Research
Seeks to “understand how basic research can help alleviate a demanding social problem and provide policymakers with well-grounded guides to remedial action.”
The goal is “to explore the value of basic knowledge in an applied setting….”
APPLIED RESEARCH
Concerned With
Focuses Upon
Policy Implications
Data Gathering and
Interpretation
Social Processes
PolicyAnalyst
EvaluationResearcher
Social Outcomes
SocialMonitor
DataAnalyst
Policy Analyst
Studies social processes and describes what
policy alternatives exist to solve and existing problem.
Evaluation Researcher
Studies social processes to
determine if a program or project is accomplishing what
it is intended to accomplish.
Social Monitor
Examines outcomes data to discover
patterns that require some organizational
or government action
Data Analyst
Uses and refines methodological tools to interpret outcome
data, often using advanced statistical
processes
TYPES OF RESEARCH BASED ON RESEARCH QUESTION
Descriptive StudiesCorrelational/Relational StudiesCausal Studies
ExplorationDescriptionExplanationEvaluation
Descriptive Research
When a study is designed primarily to describe what is going on or what exists. Public opinion polls that seek only to describe the proportion of people who hold various opinions are primarily descriptive in
nature.
Relational
When a study is designed to look at the
relationships between two or more variables. A
survey looking at the relationship between gender and make of motorbike would be
relational.
Causal
When a study is designed to determine whether one or more variables
(e.g., a program or treatment variable)
causes or affects one or more outcome variables.
RESEARCH DESIGNS
The detailed plan for the creation of
evidence.
Research Design Criteria
Reliability (repeatable) Replication (replicability) Validity (integrity of conclusion)
MeasurementInternalExternalEcological
Research Design Templates I
Experimental DesignsManipulationClassic Experimental
DesignsLaboratory ExperimentsQuasi-Experimental
DesignsNatural Experiments
Research Design Templates II
Cross Sectional Survey DesignsMore than one CaseSingle Point in Time
Longitudinal DesignsPanel StudyCohort Study
TIME IN DESIGN IA
Y EAR OF MEASUREMENT
Y EAR OF BIRTH
(COHORT)
1980 1990 2000
1960 A-20 B-30 C-40 1950 D-30 E-40 F-50 1940 G-40 H-50 I-60
TIME IN DESIGN IB
Rows Represent a cohort trough time
Columns represent a one point in time measurement
Numbers in cells is age of each population at time of measurement
TIME IN DESIGN IIA
Y EAR OF MEASUREMENT
Y EAR OF BIRTH
(COHORT)
1980 1990 2000
1960 A-20 B-30 C-40 1950 D-30 E-40 F-50 1940 G-40 H-50 I-60
TIME IN DESIGN IIB
cross sectional study highlighted
trend study: compares the cross sectional in 1990 with the cross sectional in 2000
TIME IN DESIGN IIIA
Y EAR OF MEASUREMENT
Y EAR OF BIRTH
(COHORT)
1980 1990 2000
1960 A-20 B-30 C-40 1950 D-30 E-40 F-50 1940 G-40 H-50 I-60
TIME IN DESIGN IIIA
cohort: a sample of the cohort would be taken at each point in time.
panel design: is a special case of longitudinal in that the very same people would be measured at the three points in time
Research Design Templates III
Case Study DesignsDetailed and Intensive Analysis of a Single Case
Comparative DesignsA Study Using Similar Methods To Study Two Or More Contrasting Cases.
Research Design II
II. Units of Analysis A. Individuals B. Groups C. Organizations D. Social Artifacts E. Considerations in Unit of Analysis
1. Ecological Fallacy 2. Reductionism
POINTS OF FOCUS
A. Characteristics
B. Orientations
C. Actions
Causality I
I. Deterministic vs. Probabilistic
II. Criteria for CausalityA. cause precedes the effect in timeB. two variables be empirically related
Causality I
C. correlation not explained by influence of a third
1. spurious relationships2. intervening variable3. specification variable
D. Necessary vs. Sufficient Causality
Steps in Designing a Research Project
A. Research QuestionB. Literature Review and ConceptualizationC. Choice of research methodD. OperationalizationE. Population and Sampling
Steps in Designing a Research Project
F. ObservationsG. Data processingH. AnalysisI. ApplicationJ. BibliographyK. Report WritingL. Journal Artical
Quantitative research (e.g., experiments and survey)
Qualitative research (e.g., face-to-face interviews; focus groups; site visits)
Case studies Participatory research
Some MoreResearch Methods
Involves information or data in the form of numbers
Allows us to measure or to quantify things
Respondents don’t necessarily give numbers as answers - answers are analyzed as numbers
Quantitative Research
Helps develop a deeper understanding of a topic
Often contrasted to quantitative research
Together (mixed methods)they give us the ‘bigger picture’
Examples include face-to-face interviews and focus groups
Qualitative Research
What do you want to know? How can you find out what you
want to know? Where can you get the
information? Who do you need to ask? When does your research need
to be done? Why? (Getting the answer)
Planning Your Research: Key Questions
What do I want to know? Keep in mind:
• Who your research is for;• What decisions your research will
inform;• What kind of information is needed to
inform those decisions. Literature Review is a Must Take another look at your
research question
Step 1: What?
How do I find out what I want to know?
Where can I get the information I need?
Who do I need to ask? Choose your methodology
quantitative or numbers information qualitative in-depth explanatory information case studies site visits or observation participatory research
Step 2: How? Where? Who?
When do all the different parts of the research need to be done?
List all your research work areas
Map them against a timeline Develop a work plan
Step 3: When?
Collect your data Keep returning to your research
question Organize your research results to
answer the question Keep in mind who you are doing
the research for Focus on what research results do
tell you Be creative, methodical and
meticulous
Step 4: Why?