research methods - the political economy of war

Download Report

Transcript research methods - the political economy of war

RESEARCH METHODS
Research Methods:
Procedures designed to establish
facts about the social world.
Please Note:
This section is not from the book.
Take Notes.
SOURCES OF KNOWLEDGE
Tradition: Received Wisdom, part of cultural heritage;
Common sense, “gut feeling”
Authority: Speaking as a result of status as “expert”
Guard against, the “expert” speaking outside his or her
Area of expertise. The “expert” is not infallible.
Personal Experience: It must be true because “I
experienced” such and so.
VERSUS
SCIENCE
1. Systematic+
2. Logical framework+
3. Empirical Evidence+
4. Falsification
A system of checks and balances to extract truth.
Fallacies of Thinking
1. Inaccurate Observation
Casual versus systematic Observation
Using instruments that guard against inaccuracy
2. Over-generalization
One or a few cases is representative of the whole
Choose large sample, randomly; replicate study
3.Selective Observation
Observation that fits in with your biases
4.Ecological Fallacy
(Ecological refers to things bigger than the individual,
like groups and systems)
Concluding what is characteristic of a group is
characteristic of every single individual
5.Reductionism
Explaining a variety of complex events or behaviors by
a single or narrow set of causes
E.g.: Psychiatric Reductionism
Scientific Generalizing VS Stereotype
Scientific Generalizing is:
1.
2.
3.
4.
Systematic
Logical
Empirical
Falsifiable
Stereotyping is:
1.
2.
Casual
Doesn’t guard against fallacies of thinking.
Types of Research
1. Cross Sectional
Covers the entire spectrum at one point in time.
E.g. US Census Data
2. Longitudinal
Studies changes in a population over time
a) Trend Studies: Changes in the population over time
b) Cohort: Changes in a cohort (sub population over
time.
c) Panel Studies: Changes in the same sample of the
population (the panel) over time.
Quantitative VS Qualitative
I. Quantitative: measuring with numbers
Used in the Positivistic school in sociology: A school that seeks
scientific measurement to extract laws about the social worldnumbers are then used with statistical techniques
II. Qualitative: your interpretation/description of what you
observe
Max Weber: Versthein: understanding
Understanding why a behavior occurred from the
perspective of the actor.
PURPOSE OF RESEARCH
Exploratory: To find out more about something
Descriptive: Detailed description of a social
phenomena (who, what, where, when, how)
Explanatory: To establish causation, one thing
causing another- answers the question WHY.
Types of EXPLANATION
Idiographic
(from Idio meaning unique e.g. idiosyncrasy)
Seeks to explain a social event by an exhaustive set of
unique causes- seeks total explanation, all causes
Nomothetic
Seeks to explain a social event by narrowing down
causes to a few.
Social Research is Probabilistic
Social Empirical Research does not establish 100%
causation, rather it is PROBABILISTIC i.e X does not
cause Y 100%; Rather we say X will probably cause Y,
other things being equal (Ceteris Paribus)
Causation is probabilistic because it depends on many
factors and their combination
Quantum Physics VS Newtonian Physics..
Objectivity VS. Subjectivity
Max Weber: Value-Free social research.
Objective: Uniform application of standards to
phenomena. Inter-subjective- shared understanding
Subjective: Arbitrary application of standards to
phenomena.
Subjectivity reveals Ulterior Motive.
Subjectivity and Propaganda.
CONCEPTS
A concept is a mental construct that represents
some part of the world in simplified form
E.g. Gender
VARIABLES
Variables are concepts whose values can vary.
1= Constant
X= 1,2,3……. I.E. Variable
Variables are either:
1. Independent: the one that causes the change
2. Dependant: the one in which change is caused.
Measurement
Procedure designed to ascertain the value of a
variable in a specific case
Variables have attributes, e.g. Gender can be male
(attribute) or female (attribute)
Some variables cannot be measured directly. They are
measured through indicators e.g. Religiosity
When you specify the indicator you are going to use to
measure the variable, it is called
OPERATIONALIZATION.
Reliability & Validity
Measurements should be
1. Reliable- consistency in measurementobjective
2. Valid- you are measuring what you are
supposed to measure.
Attributes of Variables
Variables are composed of Attributes
E.g. Religiosity can have the attributes
a. Very Religious
b. Religious
c. Not Religious etc;
Hypothesis
A carefully formulated statement that can be
verified or discarded after examination of the
data.
Theory
A set of statements that explains or predicts
social life
Mills: Research without a connecting theory are
mere bits of information
Theory without Research is mere speculation.
TWO TYPES OF THEORY
DEDUCTIVE THEORY
From the General to the Specific
Theory, Research, Confirmation or Rejection
INDUCTIVE THEORYFrom the Specific to the General
Also called Grounded Theory
Theory formation is an On going process.
DISCRIPTIVE STATISTICS
Tools of summarizing your reporting.
Sociologists deal with thousands and even
millions of cases, to make sense of our
description we need to present some average
of the observations, e.g. age, income etc.
1. Mean: arithmetic average
2. Median: Divides the distribution in half
3. Mode: The single score that occurs the most.
Correlation & Causation
Correlation: A relationship in which two variables move
together e.g. X changes and Y changes (together)
Causation: A relationship in which the change in one
variable causes a change in the other X changes which
causes Y to change
Correlation is not Causation.
When Correlation is mistaken as Causation it is called a
SPURIOUS relationship. Usually means that a third
(intervening) variable is causing the movement in both
variables.
Conditions for Establishing Causation
1.Two variables must be correlated (means
when one changes the other should change as
well)
2.The cause (independent variable) must
precede the effect (dependant variable) in timealso known as direction of determination.
3.There is no evidence that the relationship is
spurious
Things from this section to learn for the first exam
Definition of Science
Difference Between Generalizing and Stereotyping
Cross Sectional and Longitudinal Study
Objectivity Versus Subjectivity
Deductive Versus Inductive Theory
Correlation Versus Causation
Conditions for Establishing Causation.
Mills’s comment on Theory & Research.