Web clip 6 – Building Hypotheses
Is part of step three of the deductive research process: Develop a theoretical framework.
A theoretical framework consists of: variable definitions, a conceptual model and hypotheses.
1. What is a hypothesis?
A tentative statement
o = preliminary translation (in Dutch: voorlopige vertaling)
o You are going to test, using data, whether this statement is true or false.
about the coherence
o The relationship between variables.
between two or more variables.
o Main hypothesis is about two variables
o Mediating or moderating hypothesis is about the relationship between three
variables.
2. What makes a good hypothesis?
Testable (= in terms of measurable variables)
Derived from theory
Unambiguously phrased
3. How to phrase testable hypotheses?
Directional versus unidirectional hypotheses
o Directional hypotheses: One in which the specific direction indicates (positive
or negative)
VB: A merger decreases employee morale. (negative effect)
VB: The effect of employee morale on employee productivity is weaker
for older employees. (negative effect)
o Unidirectional hypotheses: Indicates that there is an effect, but do not
indicate the direction of the effect.
VB: A merger affects employee morale.
VB: The effect of employee morale on employee productivity depends
on employee age.
Main effect hypotheses:
o H: X positively affects Y.
R&D investments positively affects firm profitability.
(Continuous X = a variable whose value is obtained by measuring)
o H: The effect of X on Y is positive.
The effect of R&D investments on firm profitability is positive.
(Continuous X)
o H: If X increases, Y decreases.
o…
Public companies invest more in R&D than private companies.
(Discrete X = a variable whose value is obtained by counting)
Developed countries have higher private label shares than developing
countries.
So the X is: Whether a country is a developed country or a developing country.
1
Is part of step three of the deductive research process: Develop a theoretical framework.
A theoretical framework consists of: variable definitions, a conceptual model and hypotheses.
1. What is a hypothesis?
A tentative statement
o = preliminary translation (in Dutch: voorlopige vertaling)
o You are going to test, using data, whether this statement is true or false.
about the coherence
o The relationship between variables.
between two or more variables.
o Main hypothesis is about two variables
o Mediating or moderating hypothesis is about the relationship between three
variables.
2. What makes a good hypothesis?
Testable (= in terms of measurable variables)
Derived from theory
Unambiguously phrased
3. How to phrase testable hypotheses?
Directional versus unidirectional hypotheses
o Directional hypotheses: One in which the specific direction indicates (positive
or negative)
VB: A merger decreases employee morale. (negative effect)
VB: The effect of employee morale on employee productivity is weaker
for older employees. (negative effect)
o Unidirectional hypotheses: Indicates that there is an effect, but do not
indicate the direction of the effect.
VB: A merger affects employee morale.
VB: The effect of employee morale on employee productivity depends
on employee age.
Main effect hypotheses:
o H: X positively affects Y.
R&D investments positively affects firm profitability.
(Continuous X = a variable whose value is obtained by measuring)
o H: The effect of X on Y is positive.
The effect of R&D investments on firm profitability is positive.
(Continuous X)
o H: If X increases, Y decreases.
o…
Public companies invest more in R&D than private companies.
(Discrete X = a variable whose value is obtained by counting)
Developed countries have higher private label shares than developing
countries.
So the X is: Whether a country is a developed country or a developing country.
1
