Quasi-experiments

Question 1

Quasi-experiments

Answer 1

A quasi experiment is where the independent variable is not deliberately manipulated.

Question 2

Two types of quasi-experiments

Answer 2

Those with an independent variable - natural experiment.
Those with no independent variable - difference studies.

Question 3

Natural experiments

Answer 3

We use natural experiments when it isn’t possible (either ethically or practically) to manipulate the independent variable deliberately.
The independent variable varies naturally - they have naturally occurring change.

Question 4

Example of a natural experiment

Answer 4

Comparing the long term development of children who have been adopted, fostered or returned to their mothers with a control group of children who spent all their lives with their biological families.
Independent variable - separation from their biological mother.
Dependent variable - long term development.

Question 5

Rutter et al (1999)

Answer 5

Romanian orphans and how soon they were adopted and their development progress.

Question 6

Fearn (1999)

Answer 6

Studied women in Fiji.
No western TV before 1995 - no eating disorders.
By 1998 - 74% of females stated they were too fat and eating disorders became more prevalent in their society.

Question 7

Difference studies

Answer 7

In a difference study, the apparent independent variable also occurs naturally.
However, the independent variable has not been made to vary by anyone.
It is simply a difference that already exists between people, so it isn’t actually a variable.
They have a naturally occurring difference.

Question 8

Baron - Cohen (1985)

Answer 8

Studied neurotypical and ASD children.

Question 9

Example of a difference study

Answer 9

Investigating differences in aggression levels between teenagers and the elderly.
Independent variable - age (a difference that hasn’t been manipulated.)
Dependent variable - aggression.

Question 10

Evaluation of quasi-experiments

Answer 10

Because the independent variable is not directly manipulated by the researcher, we can’t be sure that effects on the dependent variable were not due to confounding or extraneous variables.
Random allocation to conditions is not possible.
This means there may be biases in each condition.
The unique characteristics of the sample cannot be generalised to other groups.
A naturally occurring event may only happen very rarely, reducing the opportunities for research.
This also may limit the scope for generalising findings to other similar situations.