Lecture 4 - Quantative Study Design

Question 1

What is study design?

Answer 1

The set of decisions regarding how a research question will be answered in terms of the structure in which data will be collected
» Quantitative study designs can be divided into Descriptive designs, Correlation designs, and Causal designs.

Question 2

What is Correlation?

Answer 2

The quantitative relationship between two variables is referred to as covariation or correlation

Question 3

When is a correlation present?

Answer 3

A correlation is present when the value of one variable provides information about the value of another variable

Question 4

When is no correlation present?

Answer 4

There is no correlation present when the value of one variable gives us no information about what to expect on the other variable

Question 5

What kind of relationships can correlations have?

Answer 5

Positive (as one variables values go up, so do values on the other variable)
OR
Negative (as one variables values go up, the values on the other variable go down)

Question 6

What possible relationships can there be between correlated variables (A and B)?

Answer 6

A causes B (direct causation)
B causes A (reverse causation)
A and B are both caused by C (common causation)
A causes B and B causes A (bidirectional or cyclic causation)
There is no connection between A and B; the correlation is a coincidence.

Question 7

Can you tell what relationship a correlation has based on the discovered correlation?

Answer 7

No - it is impossible

Question 8

What is a cause?

Answer 8

The explanation for some characteristics, attitudes, or behaviours of groups, individuals, other entities (e.g. families, cities) or for events.

Question 9

When does a causal effect occur?

Answer 9

Variation in the independent variable leads to variation in the dependent variable (the outcome) - when all other things are equal !!!!

Question 10

What are the criteria for causal inference?

Answer 10

Temporal precedence + non spuriousness (required) and association + mechanism (not required)

Spuriousness = coincidental

Question 11

Why do experiments allow causal inference?

Answer 11

True experiments include the following
To address non spuriousness:
> Comparison groups
> Random assignment

To address association and temporal precedence:
> Manipulation of the IV
> Manipulation of the DV