Introduction To Quantitative Methods
Introduction To Quantitative Methods
to Quantitative Methods
Business and Management MBA Programme
BAM0MBA—116
Week 5
Dr. Sang‐Bum Park
February 2021
Contents
• What is quantitative research?
• The core of quantitative research
– Variables
– Relationships between variables
– Validity
• Research modeling
– Mediation model
– Moderation model
• Regression analysis
– SPSS
– Stata
Types of research
Quantitative
Empirical
Research Qualitative
Conceptual
Quantitative research
Phenomenon
Interpretation Theory
Testing Hypothesis
Quantitative research
• The core of qualitative research
– Quantification
– Relationships between variables
Variables
• Variables
– “Variables are central to research. Most research
is concerned with relationships between
variables” (Schwab, 2005: 12).
– “[C]haracteristics of objects or events that can
take on two or more values” (Schwab, 2005: 12)
• Ex. Relationship between education and financial
success
Variables
• Conceptual variables
– At the level of abstraction and theoretical ideas
– Also called constructs
– Ex. education
• Operational variables
– At the level of operation and empirical studies
– Represented by measures
– Ex. The number of years a person has attended
school
Operationalization
The number
of years
Education
attending
school
Conceptual variable; Operational variable;
Construct Measure
Variables
• Dependent variables (Y)
– Outcomes, effects, consequences
– “Variables that researchers seek to understand,
explain, and/or predict” (Schwab, 2005: 12)
• Independent variables (X)
– Predictors, causes, antecedents
– “Those thought to influence or at least predict
dependent variables” (Schwab, 2005: 12)
Relationship
X Y
Independent variable Dependent variable
Exploratory variable Response variable
Relationship
Y
Dependent variable
“Dependent variables
typically are influenced
X by more than one
Independent variable independent variables”
Exploratory variable (Schwab, 2005: 13)
Empirical research model
Independent Dependent
Causal conceptual relationship
Conceptual X’ Y’
The measures X and Y
should validly represent
their respective
constructs X’ and Y’
(Schwab, 2005)
Operational X Y
Empirical relationship
Validity
• Validity
– “In research, when a conclusion or inference is
true” (Schwab, 2005: 20)
• Construct validity
– “Present when there is a high correspondence
between cases’ scores on a measure and the
mental definition of the construct the measure is
designed to present” (ibid.)
Validity
• Internal validity
– “Present when variation in scores on a measure of an
independent variable is responsible for variation in scores on a
measure of a dependent variable” (Schwab, 2005: 20)
– A research is internally valid when “the estimated coefficients
are unbiased and consistent, and if standard errors are
consistent” (Stock & Watson, 2012: 381).
• Internal statistical validity
– “Present when an empirical relationship is not due to chance”
(ibid.)
• Threats to the internal validity of a regression study
– Omitted variables, functional form misspecification, errors‐in‐
variables (measurement errors), sample selection, simultaneous
causality (see Stock & Watson, 2012: 378‐380).
Validity
• External validity
– “Present when findings obtained in a research
study, other than statistical generalization, are
correctly generalized” (ibid.)
– A research is externally valid when “its findings
can be generalized beyond the population and
setting studied” (Stock & Watson, 2012: 381).
Research modeling
• Research modeling using boxes and arrows
Sign Meaning
Relationship;
Direction of influence or cause
Variable
Regression model
• Simple regression model
– Explaining y in terms of x
– 0 1
– Ex. score 0 1attend
Classes attended Score on final exam
X Y
Independent variable Dependent variable
Regression model
𝑦 𝛽0 𝛽1𝑥 𝜇
X
Regression model
• A simple wage equation (Wooldridge, 2013)
– wage 0 1educ
• If wage is measured in dollars per hour and educ is
years of education, then 𝛽1 measures the change in
hourly wage given another year of education, holding
other factors (e.g., innate ability) fixed.
Education Wage
X Y
Independent variable Dependent variable
Stata
. reg wage education age married
The result is statistically significant if p‐value is lower than 5% (p < 0.05).
SPSS
Comparison
wage Coef. Std. Err. t P>|t| [95% Conf. Interval]
Weight of computer User satisfaction
X Y
Independent variable Dependent variable
Moderation
Type of computer
(Laptop: 1, Desktop: 0)
Weight of computer User satisfaction
Y = X * Mo
Moderation
Mo
X Y
1. X → Y
2. X + Mo + X * Mo → Y
The condition of Step 1 can be relaxed.
Coefficient of X * Mo in Step 2 should be statistically significant.
Mediation
Price
Weight of computer User satisfaction
Mediation
Me
X Y
1. X → Y
2. X → Me
3. X + Me → Y
The condition of Step 1 can be relaxed.
Coefficient of X in Step 3 should be nonsignificant (full mediation) or smaller than
coefficient of X in Step 1 (partial mediation).
Causal relationship
• Causal relationship needs more sophisticated
estimations.
Relationship
Causal
relationship
Quasi‐experiments
• Differences‐in‐differences (DiD)
• Instrumental variable estimators
• Regression discontinuity estimators
References
Schwab, D. P. 2005. Research methods for
organizational studies (2nd ed.). Mahwah, NJ:
Lawrence Erlbaum.
Stock, J. H., & Watson, M. W. 2012. Introduction
to econometrics (3rd ed.). Harlow, U.K.: Pearson
Education.
Wooldridge, J. M. 2013. Introductory
econometrics: A modern approach (5th ed.).
Cengage Learning, Canada: South‐Western.
Q&As