Repeated Measures Design

Repeated Measures Design

Each subject participates in all the conditions of an experiment. Subjects serve as their own controls because they participate in both the experimental and control conditions.

Why Repeated Measures Design

1. conduct an experiment when few participants are available
2. conduct the experiment more efficiently
3. increase the sensitivity of the experiment
4. study changes in participants’ behavior over time.

Sensitivity of an experiment

\
Refers to the ability to detect the effect of the independent variable even if the effect is a small one. An experiment is more sensitive when there is less variability in participants’ responses within a condition of an experiment, that is, less error variation.

Error variation

Can be due to variations in the procedure each time the experiment is conducted or to individual differences among the participants

**Practice effects**

\
Should be balanced across the conditions in repeated measures designs so that practice effects “average out” across conditions. Key to conducting interpretable experiments using the repeated measures designs is to use appropriate techniques

Counterbalancing

General term used to refer to balancing techniques used in the two repeated measures designs

Complete design (balancing technique)

Practice effects are balanced for each participant by administering the conditions to each participant several times, using different orders each time. Each participant can thus be considered a “complete” experiment.

Incomplete design (balancing technique)

Each condition is administered to each participant only once.

Complete design

* Participants are given each treatment enough times to balance practice effects for each participant.


* When the task is simple enough and not too time consuming, it is possible to give one participant several experiences with each treatment.
* Researchers have two choices in deciding how to arrange the order in which the treatments in a complete design are administered: block randomization and ABBA counterbalancing.

Block Randomization (balancing effects in complete design)

Technique for assigning participants to conditions in the random groups design. Block randomization can also be used to order the conditions for each participant in a complete design.

ABBA Counterbalancing (balancing effects in complete design)

* Can be used to balance practice effects in the complete design with as few as two administrations of each condition.


* Involves presenting the conditions in a random order followed by the opposite of that order.
* Even number of repetitions is required.
* ABCCBA

Incomplete design

* Each participant is given each treatment only once. 


* Any differences in the participant’s performance between the experimental and control conditions could be due to the effect of the independent variable or to the practice effects resulting from the order.
* To break this confounding of the order of conditions and the independent variable, we can administer different orders of the conditions to different participants, e.g B→A. 

Confounding effect

\
Effects resulting from the A→B order.

All Possible Orders (balancing effects in incomplete design)

* The preferred technique for balancing practice effects in the incomplete design is to use all possible orders of the conditions.


* N! possible orders with N conditions. 

Selected Orders (balancing effects in incomplete design)

Practice effects can be balanced by using just some of all the possible orders. 

The number of selected orders will always be equal to some multiple of the number of conditions in the experiment.

Latin square (selected orders)

* Each condition appears at each ordinal position once


* Each condition precedes and follows each other condition exactly once.

Random starting orders with rotation (selected orders)

Begin with a random order of the conditions and to rotate this sequence systematically with each condition moving one position to the left each time.

Data Analysis of Random groups designs

Listing the scores of the participants tested in each of the conditions of the ex- periment and then summarizing these scores with descriptive statistics such as the mean and standard deviation.

Data Analysis of Incomplete repeated measures design

Each participant provides one score in each condition, but it is still relatively straightforward to summarize the scores for each condition. Once all the scores for each condition have been listed together, means and standard deviations can be computed to describe performance in each condition.

Data Analysis of Repeated measures design

You first must compute a score for each participant in each condition before you begin to summarize and describe the results. This additional step is necessary because each participant is tested in each condition more than once in a complete design.

Differential transfer

Arises when performance in one condition differs depending on the condition that precedes it. Researchers can overcome the potential problem of practice effects in repeated measures designs by using appropriate techniques to balance practice effects. There is a much more serious potential problem that can arise in repeated measures designs that is known as differential transfer

\
*Therefore, when differential transfer could occur, researchers should choose an independent groups design.*

\
One way to determine if differential transfer has occurred in an incomplete repeated measures design is to examine the results for the first ordinal position only