Flashcards in experimental repeated measures design Deck (19)
Basic Repeated Measures Design
Unit of observation (e.g., participant; branches of an organisation; suburb; state) is measured multiple times on the dependent variable.
Focus on how the dependent variable changes
over time from an intervention. For example:
(1) self-report of well-being over 3 time points
(2) record once a week for 5 weeks the duration of time that children in sandpit fought over toys
over conditions – e.g., three conditions of attention – full attention, distraction, divided attention. This is the focus of this lecture.
over space (aka locations) – not covered in this course
Single-Factor Repeated-Measures Design
One treatment factor
Data is based on repeated observations of the same participants
We are interested in how much variability in the dependent variable/s there is within a group of participants
Advantages of Repeated-Measures Over Conditions
Researchers don’t have to worry about balancing individual differences across conditions of the experiment because all participants are measured in each condition
These designs require fewer participants
They are convenient and efficient
Repeated measures designs are more sensitive.
A “sensitive” experiment is one that can detect the effect of the independent variable even when that effect is small.
Repeated measures designs are more sensitive because “error variation” is reduced.
Because the same people participate in each condition, the variability in responses due to different people is reduced (compared to independent groups).
Disadvantage of Repeated Measures Designs
The main disadvantage of repeated measures designs is practice effects.
Practice effects arise because people change as they are repeatedly tested.
As participants complete the dependent variable measures after each condition, they may get better with practice, or they may become tired or bored.
Practice effects become a confounding variable if not controlled.
Example: Suppose a researcher compares two different study methods, A and B.
Condition A: Participants read a text passage and use a highlighter to indicate the important points. Participants then take a test of this material.
Condition B: Participants read 10 pages of similar text, and make up sample test questions and answers. Participants then take a test of this material.
Suppose all participants in this experiment first experience Condition A (use a highlighter while reading) and then Condition B (write sample questions and answers).
Suppose the results are that participants perform better on the test of Condition A text material than on the test of Condition B material.
Can we conclude that using a highlighter is a better study method than writing sample questions and answers?
No. Two possibilities exist.
The study method in Condition A (highlighting) is better than writing test questions and answers, or
by the time they did Condition B, participants were tired or bored.
It’s impossible to know which conclusion is correct.
This is a confounding due to practice effects; the internal validity of this experiment is threatened.
Practice effects must be balanced, or averaged across the conditions of the experiment.
Counterbalancing the order of the conditions makes sure that the practice effects are distributed equally across the conditions of the experiment
Counterbalancing the study conditions:
half of the participants do Condition A first, then Condition B,
the remaining participants do Condition B first, then Condition A.
In this way, both Conditions A and B have the same amount of practice effects.
Practice effects can’t be eliminated, but they can be balanced, or averaged, across the conditions of an experiment.
There are two types of Repeated Measures over-conditions design:
The purpose of each type of design is to counterbalance practice effects.
Each repeated measures design uses different procedures for balancing practice effects across the conditions of the experiment.
Practice effects are balanced within each participant in the complete design.
Each participant experiences each condition of the experiment several times (or hundreds of times), using different orders each time.
A complete repeated measures design is most often used when each condition is brief (e.g., simple judgments about stimuli
Two methods for generating orders of the conditions in the complete design are block randomization and ABBA counterbalancing.
A “block” represents all conditions of the experiment (e.g., 4 conditions, A B C D).
A random order of the block is generated (e.g., ACBD)
Thus, a participant would first do condition A, then C, then B, then D.
A new random order would be generated for each time the participant completes the conditions of the experiment.
Block randomization balances practice effects only when the conditions are presented many times.
Many administrations of the conditions are needed to balance, or average, practice effects across the conditions of the experiment.
Block randomization is not useful when the conditions are presented only a few times to each participant.
A different method is needed when conditions are presented only a few times to each participant.
Present the conditions of the experiment in one random sequence (e.g., D A B C), followed by the opposite of that sequence (C B A D).
If the conditions are presented again, generate another random order of the conditions, followed by the opposite sequence
ABBA counterbalancing should not be used when anticipation effects can occur.
Anticipation effects occur when participants develop expectations about which condition will appear next in a sequence.
Participants’ responses may become influenced by their expectations rather than by the conditions of the independent variable.
If anticipation effects are likely, use block randomization.
Each participant experiences each condition of the experiment exactly once (rather than many times, as in the complete design).
Practice effects are balanced across participants in the incomplete design rather than within subjects, as in the complete design general rule for balancing practice effects in the incomplete design is that each condition of the experiment (e.g., A, B, C) must appear in each ordinal position (1st, 2nd, 3rd) equally often.
If this rule is followed, practice effects will be balanced and will not confound the experiment
Techniques for balancing practice effects in the incomplete design include all possible orders and selected orders.
All possible orders: The preferred technique for balancing practice effects in an incomplete design that has four or fewer conditions of the independent variable