SPH 509 Chapter 11 Correlated-Groups and Single-Subject Designs
How many orders of presentation are necessary to counterbalance in an experiment with four experimental conditions?
24; There are 24 orders of counterbalanced presentation for the 4 experimental conditions. There are 4 conditions. The number of orders of presentation can be computed with a simple formula (N!, read N factorial). 4 X 3 X 2 X 1 = 24
Reversal designs are also called
ABA designs
Replication can be achieved in single-subject designs through
ABA designs, randomized time-series designs, multiple baseline designs
Which of the following is a major control for sequence effects?
Counterbalancing; This is the best control for potential sequence effects. By counterbalancing the order of presentation we avoid having any one condition unduly impinge upon any other in a systematic way. By counterbalancing the order of presentation, each condition or presentation appears an equal number of times, and appears before and after each other condition an equal number of times.
Which of the following is true for within-subjects designs?
Each participant serves as his or her own control
What is the major strength of the within-subjects design?
It guarantees that the participants in the various conditions are equivalent at the start of the study.
In within-subjects designs, the removal of individual differences would have what effect on the F-ratio?
The F will be larger; Removing the individual difference component from the error term will decrease the denominator and therefore increase the F-ratio.
The single-subject, randomized, time-series design is essentially an interrupted time-series design with one refinement. What is that refinement?
The assignment of the manipulation in the time-series is randomized.
The single-subject, randomized, time-series design
The single-subject, randomized, time-series design can be used when there is only a single behavior. Note that a multiple baseline design can be used only if you have more than one behavior.
Which of the following is true about single-subject designs?
They are variations of within-subject designs
Why do the authors of this text consider correlated-groups designs to be true experiments?
They meet the requirement of equivalence of groups.
What makes within-subjects designs more sensitive than between-subjects designs?
They reduce error variance
Why is a final A to B (i.e. back to the treatment) reversal sometimes carried out in reversal designs?
To reduce the number of participants needed and for ethical reasons
Compared with between-subjects designs, correlated-groups designs
are more sensitive to the effects of the independent variable
Matching-subjects designs
are sensitive to small experimental differences, just like within-subjects designs
Single-subject designs
are usually variations on time-series designs; single-subject designs are variations on time-series designs
Effects that are due to the influence of a particular condition on responses to conditions that follow are called
carry-over effects
In within-subjects designs, the unwanted effects due to the influence of one condition on the following conditions is called
carry-over effects
The controlled study of a single individual has become particularly important in
clinical psychology
A major control for sequence effects in within-subjects designs is
counterbalancing the order a presentation
Single-subject experiments
differ from case studies because independent variables are manipulated
It is important to note that carry-over effects may be
differential
In matched-subjects designs,
each participant has a matched participant in each condition
In matched-subjects designs
each participant is exposed to only one level of the independent variable
In randomized, between-subjects experimental designs, each participant is
exposed to only one level of the independent variable
Matching participants
is best when there are a small number of matching variables
In using a matched-subjects design, it is important to remember that matching on more than one continuous variable
is difficult to accomplish and will usually result in significant participant loss
The advantage of a within-subjects designs is that
it equates participants in conditions prior to experimental manipulation; Within-subjects designs create groups that could not be more equal, because the same people appear in each group.
Single-subject experimental designs are useful when we want to
limit our conclusions to a particular individual; single-subject designs are perfect when we are only interested in a single individual, which is why they are used so often in therapy situations, where the therapist wants to know if his or her intervention is benefiting the client.
The key to analyzing data from a matched-subjects designed is to
maintaining the information on the matching of participants all away through to the analyses; In a matched-subjects design, you will analyze the data as if they came from a within-subjects design, with all of the scores for a given group of matched subjects appearing on the same line. Therefore, you must remember who is matched with who and maintain that ordering throughout the analysis.
Compared to between-subjects designs, within-subjects designs are
more sensitive to effects of the independent variable
A researcher is using a within-subjects design. After condition two (there are a total of three conditions), the researcher notices that the participants are looking bored and restless. After the third condition, the researcher records a falling off of performance. This is a(n)
negative practice effect
That is the term for the unwanted enhancement of performance on subsequent conditions in within-subjects designs?
positive practice effect
Which of the following is NOT used in correlated-groups designs?
random-assignment to conditions
Which of the following is an appropriate statistical test for a within-subjects experiment with two experimental conditions and a dependent variable that produces score data?
repeated measures ANOVA
A within-subjects design is also referred to as a(n)
repeated-measures design; the terms within-subjects and repeated-measures are synonymous
A potential confounding factor that can occur in within-subjects designs if the researcher does not vary the order of presentation is called
sequence effects
Which of the following is a potential confounding factor in within-subjects designs, but not in a between-subjects design?
sequence effects
The main drawback of within-subjects designs is that
sequence effects can occur; sequence effects are always an issue when people appear in more than one condition
Matching participants on the variable of sex is relatively simple because
sex of participants has only two levels;The more possible values a variable has, the more difficult it is to match participants on that variable.
In the repeated-measures ANOVA, the within-groups sum of squares is split into the ____ and the ____.
subjects term; error term
In the hypothetical example given in the text of the use of an ABA reversal designed to reduce Betty's self-stimulatory behavior, the independent variable was _____ and the dependent variable was _____.
teachers attention; the self-stimulatory behavior; The therapist wanted to know if the teacher's attention (the independent variable) could be influencing the self-stimulatory behavior (the dependent variable).
In ABA reversal designs, the effects of the independent variable on the dependent variable are demonstrated if
the behavior changes in the predicted direction whenever the conditions are reversed; to demonstrate the effect of the independent variable on the dependent variable in a reversal design, the dependent variable must change in the predicted direction at each reversal.
Correlated-groups designs are generally more sensitive than between-subjects designs to
the effects of the independent variable; Correlated-groups designs are more sensitive to the effects of the independent variable on the dependent variable.
The "experimental analysis of behavior" refers to the developmental of methods for
the intensive, systematic, and controlled study of individual participants
A within-subjects design helps to ensure the equivalence of groups at the start of the experiment because
the participants in one group are the same participants as in the other groups
If strong carry-over effects are expected in an experiment
the within-subjects design is not recommended
In which design are all participants exposed to all experimental conditions?
within-subjects
What are the two types of designs used to introduce the correlation in correlated-groups designs?
within-subjects designs and matched-subjects designs
Single-subject designs are extensions of
within-subjects designs; single-subject designs require multiple measures on a single individual, thus making them a variation on within-subjects or repeated-measures designs.