Web clip 3 – Developing a conceptual model
Step 3 of research: Develop a theoretical framework
1. How to define variables?
What makes a good variable definition?
o Define a informative variable name (keep it short if possible)
o Define a variable definition without jargon
Based on a careful literature review
Unless they are very obvious (e.g. sales, profits)
Common pitfall: Examples do not substitute for a definition
o One or two supporting references per variable definition.
Variable definitions
What if many different definitions exist in the literature?
Acknowledge the major differences (accept or admit the existence or
truth of).
End with a definition that focuses on the shared meaning across
definitions ( so define your own definition). OR pick one definition and
justify why.
Always use EXACTLY the same variable names throughout your research.
2. How to build a conceptual model?
A conceptual model specifies how your variables are related to each other.
Variables are building blocks. There are four kinds of variables:
o Dependent variables
= the variable of primary interest. The phenomenon you are trying to explain.
Aka criterion variable or DV.
o Independent variables
Influences the dependent variable in a positive way, or in a negative way.
Aka predictor variable, IV.
If the independent variable changes, the dependent variable changes to.
o Mediating variables
= a variable that is situated between the dependent and independent variable.
The variable explains why (or how) the independent variable has a influence on
the dependent variable.
Aka mediator, intervening variable.
o Moderation variables
= a variable that effects the relationship between the independent variable and
the dependent variable. It changes the effect that the independent variable has
on the dependent variable. When or for whom the effect is stronger.
Aka moderator, interaction variable.
The relationships between the variables are arrows. There are three kind of relationships:
o Main effect
= Effect between the independent variable and the dependent variable.
o Direct / indirect effect
= the effect between a mediator and the dependent variable.
o Moderating effect
= the effect between the moderator and the dependent variable.
A conditional process model = a model that pieces together a mediator and a moderator.
1
Step 3 of research: Develop a theoretical framework
1. How to define variables?
What makes a good variable definition?
o Define a informative variable name (keep it short if possible)
o Define a variable definition without jargon
Based on a careful literature review
Unless they are very obvious (e.g. sales, profits)
Common pitfall: Examples do not substitute for a definition
o One or two supporting references per variable definition.
Variable definitions
What if many different definitions exist in the literature?
Acknowledge the major differences (accept or admit the existence or
truth of).
End with a definition that focuses on the shared meaning across
definitions ( so define your own definition). OR pick one definition and
justify why.
Always use EXACTLY the same variable names throughout your research.
2. How to build a conceptual model?
A conceptual model specifies how your variables are related to each other.
Variables are building blocks. There are four kinds of variables:
o Dependent variables
= the variable of primary interest. The phenomenon you are trying to explain.
Aka criterion variable or DV.
o Independent variables
Influences the dependent variable in a positive way, or in a negative way.
Aka predictor variable, IV.
If the independent variable changes, the dependent variable changes to.
o Mediating variables
= a variable that is situated between the dependent and independent variable.
The variable explains why (or how) the independent variable has a influence on
the dependent variable.
Aka mediator, intervening variable.
o Moderation variables
= a variable that effects the relationship between the independent variable and
the dependent variable. It changes the effect that the independent variable has
on the dependent variable. When or for whom the effect is stronger.
Aka moderator, interaction variable.
The relationships between the variables are arrows. There are three kind of relationships:
o Main effect
= Effect between the independent variable and the dependent variable.
o Direct / indirect effect
= the effect between a mediator and the dependent variable.
o Moderating effect
= the effect between the moderator and the dependent variable.
A conditional process model = a model that pieces together a mediator and a moderator.
1
