Summary Quantitative Research Methods

Lecture 1: Background
Experimental data are often collected in laboratory environments in the natural sciences, but they
are much more difficult to obtain in the social sciences. In the social sciences, we mainly focus on
analyzing nonexperimental (observational) data. Meaning: Not accumulated through controlled
experiments on individuals, firms, or segments. It is often impossible, prohibitively expensive, or
simply morally repugnant to conduct the kinds of controlled experiments that would be needed to
address economic issues.
 Quantitative research methods focus on the use of statistical methods for estimating
economic relationships, testing economic theories, and evaluating and implementing
government and business policy.

When do we use quantitative research in social sciences?
 Descriptive (what?):
If we are interested in a quantitative answer : How many students are enrolled?
If we are interested in a numerical change : Are the number of X rising compared to Y?
 Inferential / relational (why?):
If we want to test relationships : What is the relationship between self-esteem and average grade?
If we want to explain something : What factors predict changes in performance over time?

Be careful however, because numbers and statistics are seldom objective and disinterested. Data
have always been manipulated and sometimes with the wrong intentions.
 Simpson’s paradox refers to a phenomena whereby the association between a pair of
variables (X, Y) reverses sign upon conditioning of a third variable, Z, regardless of the value
taken by Z. Look at the aggregate ánd disaggregate data to expose the relationships and
therewith avoid selection bias.

,Step 1: Research Question

Exploratory/Descriptive?  What factors predict
team productivity?

Explanatory?  What is the impact of team
heterogeneity on team productivity?

Step 2: Literature Review

Gives you an overview of the research previously
conducted in a field and helps establish what
knowledge is lacking.



A theory identifies what variables are important and for what reasons, specifies how they are
interrelated and why, and identifies conditions under which they should be related and not related.
A variable  What we end up with when we apply our measure to something in the world.
That is, variables are the actual ‘’data’’ that we end up with in our data sets.

In any relationship, you will have an independent variable / predictor (education) and a dependent
variable / outcome (wage) to which there is a positive or negative correlation. In this example
relationship, a control variable could be age or gender.

A hypothesis defines the expected relationships between variables and therewith tests the existence
of a theory. An example of a hypothesis is that variable X has a positive relationship with variable Y.

Life is much more complex than X -> Y. Therefore, we must also account for moderator and mediator
variables.
A moderator variable influences the relationship between the IV and DV by strengthening or
weakening the relationship. This creates the interaction effect.
A mediator variable is located between the IV and DV and explains the relationship between
them. It mediates between the variables and can be used to see if the influence of the
mediator is stronger than the direct influence of the independent variable. The focus is on
mechanisms that make causal chain possible.

Moderator
When formulating a moderation
model, you need to:

1. First, test the relationship
between the independent
and dependent variable.
2. Second, you test the
manner in which the
moderator variable
strengthens or weakens the
relationship.

, Mediator




When formulating a mediation model, you need to specify several relationships my means of
hypotheses. An example is provided:




When formulating a mediation model, you need to test several Hypotheses:
1. First, the relationship between the independent and the dependent variable.
 High-quality relationships with supervisors are positively related to creative work
involvement.
2. Second, the relationship between the independent and the mediator.
 High-quality relationships with supervisors are positively related to employee’s
feelings of energy
3. Third, the relationship between the mediator and the dependent variable.
 Feelings of energy are positively related to creative work involvement
4. Fourth, the fact that the mediator mediates the relationship between the independent and
dependent variables.
 Individual’s feelings of energy mediate the relationship between relationships with
supervisors and creative work involvement