Switching Replications design

Description

A two-group experiment is done as normal, with a treatment group and a control, with parallel pre- and post-tests. The roles are then switched and the experiment reproduced with the control becoming treatment group and the initial treatment group becoming the control.

R	O	X	O		O
R	O		O	X	O

Example

An experiment to test a method of increasing suggestibility shows unexpected success. As there is time available (the participants agreed to full day of activity), the experiment is continued with a switch of treatment, applying it to the control group. The results show that the control group are similarly affected. The re-test of the original treatment group also shows that the suggestibility is sustained, at least for the hour between the first and second post-tests.

Discussion

Sometimes there may be a suspicion that the control group and treatment group have some fundamental difference that is distorting results. This effect is more likely when the sample groups are small. It can also appear when there has been a problem in the assignment of people to groups, such as when a couple people from a minority end up in a single group and cause bias in the results.

This method effectively switches their roles and after the initial treatment/no treatment stages, the control group becomes the treatment group and vice versa for a second round.

This is not a completely independent affair as the repeated pre-test for the control group could cause priming effects, in which case the Solomon Four-Group design could be used an additional check.

Quick links

Switching Replications design

Description

Example

Discussion

See also

Site Menu

You can buy books here

Quick links

Disciplines

Techniques

Principles

Explanations

Theories

And

Site Menu

Quick links

Disciplines

Techniques

Principles

Explanations

Theories

And