Writing a Scientific Research
Article
Language
•
Academic language
•
A poorly written article with
grammatical errors may not be
accepted
•
The entire manuscript should flow
naturally between sentences,
paragraphs, and sections.
Abstract
• 100 – 200 words (look up the guidelines)
– 4 – 5 sentences
• Key words
• 1st sentence - Background, general question
• 2nd sentence – Research Question/Hypothesis
• 3rd sentence – Model and data
• 4th sentence – Results
– Find the general story of the results
• 5th sentence –
Conclusion/Implications/Recommendations
Introduction/Background
• First 2 pages
– Global Statement: There is an important issue
addressed
– Specific Statement: Frame the issue • Significance (Contribution)
– Why is the topic important and interesting?
• What do you how to achieve
• Short summary of existing literature
• Some mentioning of methods and maybe
Literature review
•
Find a way to synthesize prior
knowledge
–
Turn it into a story to support/inform
your own research
–
Comprehensive yet targeted to your
study
•
Contribution of your study
Objectives
•
Objectives
–
What do you want to accomplish with
this study?
–
Be specific and refined to narrow the
emphasis
•
Hypotheses to test
Theoretical Model/Considerations
• Theoretical framework (descriptive theory)
– Examples:
• Principal-Agent theory
• Asymmetric information
• Utility/Risk Return
• Risk Aversion
• Formulation of the model
– Constrained maximization – Minimum problem
• Developments
– First Order Conditions
Empirical Model
•
Operationalize the theoretical model by
choosing specification, functional form,
variable measures, etc.
•
Specify the empirical model
•
Define variables for your study
– Justify their appropriateness
• Theoretically or
• Based on previous literature
– Provide expected relationship for each
Data
•
Description of data
–
Type and source of data
• Secondary • Primary
–
Description of the sample
–
Description of the variables
Method Types
•
Econometric
•
Simulation
Methods
• Identify the statistical model(s) used and explain it sufficiently
– Linkage with the statistical model
– Linkage with previous uses in the literature
• Indicate tests that need to be performed to get to the final statistical model
• Specify all the steps taken in the modeling process
Results
• Restate the problem and model
• Report results from testing of assumptions (if
any)
• Provide main story
– Significance and key relationship from hypothesis test – Significance and relationship of other variables
(demographics…)
• Compare and contrast
– With previous literature, Previous expectations,
Theory
Results (cont)
•
Explain in words what was discovered in
the research.
•
Clear, thorough, and in a logical sequence.
•
Tell everything you know about the
results; don’t just report the coefficient
estimates and their significance.
Conclusion/Policy Implications
• Same structure as the abstract but with more
emphasis on the results (longer than the abstract)
• Provide specific recommendations based on the
finding for the study
– For whom is it useful?
• Provide Implications for policy, businesses, consumers,
etc. that are linked to the results.
– How do they vary from previous research?
• Do the findings support or contradict previous
research? Theories?