The Role of Researchkapil Deb Subedi
2010
Research Methods
Kapil Deb SubediLecturer of Management, Saptagandaki Multiple Campus
Overview
Why Study Research Methods? What are the Stages in Research? Elements of Research Not Frequently Discussed Why writing and the language of research is so
important The Flow of Research Time, Units & Questions Causal Model Building Next Class
What is Research?
“The discovery of nature, of the ways of planets, and plants and animals, required first the conquest of common sense. Science would advance, not by authenticating everyday experience but by grasping paradox, adventuring into the unknown. Novel instruments, telescopes and microscopes among others, would offer disturbing new perspectives. …Here were the new incentives to ransack the world for undiscovered species and seek clues to the mystery of an ever-changing nature.”
Daniel Boorstin (1983) The Discoverers
Why Study Research Methods?
To Help:– conduct your own research– evaluate the soundness of theories– assess the validity of others’ claims– understand and apply research results– solve organizational problems.
Objectives of A Thesis or Dissertation
Demonstrate the ability to:1. Do independent research
2. Make a contribution to knowledge with the research
3. Document the research and make it available to the scholarly community (i.e., publish!)
Source: Davis, Gordon and Clyde A. Parker (1997) Writing the Doctoral Dissertation, p. 15
Stages in the Research Process
Problem
Model &Hypotheses
ResearchDesign
MeasurementData
Collection
DataAnalysis
Generalization
Real Stages in the Research Process
Problem
Model &Hypotheses
ResearchDesign
MeasurementDataCollection
DataAnalysis
Generalization
Research is NOT a linear process!!
Elements of Research Not Frequently Discussed
• Research is not a linear process• It is just written up like it is
• Note: articles are written backwards: once you know the results you know what literature is important
• Research is a social process• Not because research is social but because results must
enter into a social “learned” society
• Research value (impact) is more a question of importance than volume
• But volume makes for a wonderful, simple measure of productivity• But, one good published idea is worth more than a hundred articles
• How do you know value? CITES!!! [isi web of science]
Little Discussed Elements (II)
Research is for posterity– i.e., it has a different time scale than consulting
Refereed archival journals versus the Internet
Research builds upon the past– …by tearing it down (theory building), – Or by supporting it (replications studies)
• Research not published is virtually worthless• The importance is more to be read than to read!
Research demands a special form of writing and language
Writing
Not like anything you have ever done before– Larger audience– More risk– Structured completely differently– Rewrite-rewrite-rewrite
Social system expectations– Expects that conventions will be met– Expects that the authors will get to the point!
"There are two moments worthwhile in writing, the one when you start
and the other when you throw it in the waste-paper basket." (Samuel Beckett)
Drucker’s view
“The man of knowledge has always been expected to take responsibility for being understood. It is barbarian arrogance to assume that the layman can or should make the effort to understand him, and that it is enough if the man of knowledge talks to a handful of fellow experts who are his peers.”
P. 62
Examples of Types of Writing
Academy of Management Journal:– “We examine how firms search, or solve problems, to create new products.
According to organizational learning research, firms position themselves in a unidimensional search space that spans a spectrum from local to distant search. Our findings in the global robotics industry suggest that firms' search efforts actually vary across two distinct dimensions: search depth, or how frequently the firm reuses its existing knowledge, and search scope, or how widely the firm explores new knowledge.
Katila, Riita, Ahuja, Gautam (2002) “SOMETHING OLD, SOMETHING NEW: A LONGITUDINAL STUDY OF SEARCH BEHAVIOR AND NEW PRODUCT INTRODUCTION.” Academy of Management Journal, 45(6): 1183
Academy of Management Review– “We propose a model in which between-individual differences in performance
(heterogeneity) and within-individual differences in performance over time (variability) affect flow line performance. The impact of heterogeneity and variability is contingent upon the flow line context, particularly the rules governing the way work moves between employees (work flow policy). We show how subtle changes in this policy can have a motivational effect on heterogeneity and variability and how these, in turn, can impact the relationship between work flow policy and flow line performance. “
Doerr, K et. al, (2002) HETEROGENEITY AND VARIABILITY IN THE CONTEXT OF FLOW LINES. Academy of Management Review, (4): 594-607
Language of Research - I
Research Issues:– Theoretical v. Empirical
a.k.a., pure v. applied
– Generalizable (nomothetic) v. Local Truth
Language of Research - II
– Ecological Fallacy (individual like group) v. Exception Fallacy (group like individual; law of small numbers)
– Positivism (external “truth”) – v. Post-Positivism (never know “truth”, only
probabilities)
Levels of Research
Reporting– Telling what someone else did
Descriptive– Describing who, what, where, when, how
Explanatory– Answers “WHY?”
Predictive– Answers: “What will happen if...”
Flow of Research I:Top to Bottom Approach
Source: http://trochim.human.cornell.edu/kb/strucres.htm
Flow of Research II: Inductive v. Deductive
Deductive (“Traditional”)
Inductive (Theory Building; Qualitative )
Source: http://trochim.human.cornell.edu/kb/dedind.htm
Time & Units of Analysis
time in research:– cross-sectional versus longitudinal
units of analysis:– individual, – group, department– organizational, – inter-organizational, – cross-cultural
Types of Questions
Descriptive– Primarily used for qualitative or
demographic information Relational
– What is related to what (no temporal or causal inference)
Causal– X causes Y
Building a Model
Constructs– General ideas created for theory-building– Used to build conceptual/causal model
Variables– Need operational definitions– Types
Independent Dependent Moderating Intervening Extraneous
– That have some type of relationship
A Causal Model is made up of -
Independent(Predictor)
(X)Variables
Dependent(Outcome)(Criterion)
(Y)Variables
Hypotheses
Smoking……………..Causes……………….Cancer
Extraneous Variable: hair color, accent
And -
Independent(Predictor)Variables
Dependent(Outcome)Variables
Moderating
(Conditional)(Interaction)
(X*Y)
Variables
Smoking (by people with special gene) ……causes…………….Cancer
And -
Independent(Predictor)Variables
Mediating
(Intervening)
Variables
Dependent(Outcome)Variables
ModeratingVariables
Smoking ……………causes(by people with a special gene) cells to mutate that causes…….Cancer
And -
Independent(Predictor)Variables
MediatingVariables
Dependent(Outcome)Variables
ModeratingVariables
ControlVariables
Smoking ………………causes(by people with a special gene) cells to mutate that causes…….Cancer (when controlling for age)
Types of Relationships Among Variables
positive, negative, none:
Independent(Predictor)Variables
MediatingVariables
Dependent(Outcome)Variables
ModeratingVariables
ControlVariables
+ -
+
-
Why do You Need a Model?
In order to GENERALIZE!Source: http://trochim.human.cornell.edu/kb/introval.htm
What is written:
What you do:
Language of Research, cont’d.
Theory Model
Propositions/Hypotheses
Hypotheses
Statistical:– H0 (null: no difference)– HA (alternate: the relationship you want
to “prove”) Academic practice:
– H# (the relationship you want to “prove”)
Language ...
Theory Model Propositions/Hypotheses
Constructs/Variables
Constructs (i.e., the ideas behind the variables)
Independent(Predictor)Variables
MediatingVariables
Dependent(Outcome)Variables
ModeratingVariables
ControlVariables
ExternalVariables
Variable Measurement Types
Categorical– Red, blue, green
Sorting slots
Ordinal– Tall, medium, short
Non-equal spacing between items
Interval– 1, 2, 3, 4, 5
with equal spacing between items
Ratio– “true” zero; 0 to 1
Zero means something (absolute zero for example)
Levels in Quantitative Research
Theory Model Propositions Hypotheses Constructs
Operationalizations
“Operationalizing” Models -
we want to ensure that the measures represent the underlying (latent) variables (constructs) well.
Measures should be -
Reliable Valid
– Lots more on this later
Stages in the “Quantitative” Research Process, revisited
Problem
Model &Hypotheses
ResearchDesign
MeasurementData
Collection
DataAnalysis
Generalization