05 Experimental Evaluation of Interventions / 05.06 Use Alternating Treatments Design Flashcards Preview

Test Cards COPY > 05 Experimental Evaluation of Interventions / 05.06 Use Alternating Treatments Design > Flashcards

Flashcards in 05 Experimental Evaluation of Interventions / 05.06 Use Alternating Treatments Design Deck (18)
Loading flashcards...
1

You want to compare two prompting procedures for teaching spelling. You might use
a concurrent schedules design.
a simultaneous treatments design.
a multielement design.
(all of the others)

a multielement design.


You could use an multielement/alternating treatments design. You could not use a concurrent schedules or simultaneous treatments design because two prompting procedures cannot be available to a student at the same time. (Bailey & Burch, 2002, pp. 182-186; Poling et al., 1995, pp. 94-96; Tawney & Gast, 1984, pp. 312-340)

2

With an alternating treatments/multielement design,
early termination of the study is often not fatal.
a treatment effect can occur quickly.
treatment effects do not have to be reversed.
(all of the others)

all others

With an alternating treatments design, the experimental condition changes every day, every session, or even within a session. This early exposure to all conditions means that early termination of the session may not be fatal to the experiment if there is enough data to show a contrast between the two (or more) conditions. Also, a treatment effect could be evident early and it does not have to be reversed to demonstrate experimental control. (Bailey & Burch, 2002, pp. 182-186; Cooper, Heron, & Heward, 2007, pp. 195-196; Poling et al., 1995, p. 96; Tawney & Gast, 1984, pp. 312-313)

3

In a changing criterion design, a larger magnitude change
provides more evidence of experimental control than a small change.
is never advised because it contradicts good instructional practice.
is critical to establishing stable responding.
(all of the others)

provides more evidence of experimental control than a small change.


Although too large of a criterion change may not be good instructional practice, it provides more evidence of experimental control than a small change. (Cooper, Heron, & Heward, 1987, p. 220; Cooper, Heron, & Heward, 2007, pp. 221-223)

4

The alternating treatments design controls for
attrition.
maturation.
sequence effects.
(all of the others)

all others


he rapid alternation of conditions characterized by the alternating treatments design controls for several threats to validity. These include maturation (changes that occur in the subject during the experiment), sequence effects (a condition run for several sessions influences the next condition), attrition (loss of subjects) to some degree, and data instability (variability obscures effects due to overlapping data points). (Bailey & Burch, 2002, pp. 182-186; Cooper, Heron, & Heward, 2007, pp. 195-196; Poling et al., 1995, p. 96; Tawney & Gast, 1984, pp. 91-93, 313)

5

In a changing criterion design,
each phase must be long enough to establish stable responding.
the length of each phase should vary.
behaviors that are slower to change require longer phases.
(all of the others)

all others

If the length of each phase were the same, changes in the rate of behavior might approximate a single trend, thus obscuring the trend and level as a function of the contingency. Behaviors that are slower to change require longer phases to give the behavior a chance to respond to the contingency. Similarly, each phase must be long enough to establish stable responding. (Cooper, Heron, & Heward, 1987, pp. 219-220; Cooper, Heron, & Heward, 2007, pp. 221-223)

6

Which experimental conditions can be found in a changing criterion design?
reinforcement contingent upon 10 responses per minute, baseline, reinforcement contingent upon 10 responses per minute
reinforcement contingent upon 10 responses per minute, 20 responses per minute, then 30 responses per minute
baseline until a criterion rate of behavior is achieved, then reinforcement is contingent upon 10 responses per minute
baseline for 4 days then reinf. contingent upon 10 resp/min for S1; baseline for 8 days then reinf. contingent upon 10 resp/min for S2


reinforcement contingent upon 10 responses per minute, 20 responses per minute, then 30 responses per minute

LTERNATING TREATMENTS (simultaneous treatments, multielement multiple schedule)--rapid alternation between two or more treatments
MULTIPLE BASELINE--staggered implementation of the intervention in a step-wise fashion across subjects, settings, or behaviors
REVERSAL (withdrawal)--alternation between baseline and a particular intervention
CHANGING CRITERION--reinforcement or punishment is contingent upon a certain level of behavior, which changes in a step-wise fashion
(Cooper, Heron, & Heward, 1987, pp. 163-219; Cooper, Heron, & Heward, 2007, pp. 176-224)

7

The alternating treatments design controls for
generalization effects.
implementation issues.
efficiency effects.
sequence effects.

sequence effects

The rapid alternation of conditions characterized by the alternating treatments design controls for several threats to validity. These include maturation (changes that occur in the subject during the experiment), data instability (variability obscures effects due to overlapping data points), sequence effects (a condition run for several sessions influences the next condition), and to some degree, attrition (loss of subjects). (Bailey & Burch, 2002, pp. 182-186; Cooper, Heron, & Heward, 2007, pp. 188, 195-196; Poling et al., 1995, p. 96; Tawney & Gast, 1984, pp. 91-93, 313)

8

In a changing criterion design,
the length of each phase should be the same.
each phase must be long enough to establish stable responding.
behaviors that are slower to change require shorter phases.
(all of the others)

each phase must be long enough to establish stable responding.

9

"A functional relation is demonstrated when rate changes correspond to the requirements of the contingencies" characterizes which design?
multiple baseline
reversal
alternating treatments
changing criterion

changing criterion

10

Analogy: Both interventions available is to switching interventions as
alternating treatments is to multielement.
multielement is to alternating treatments.
concurrent schedules is to alternating treatments.
simultaneous treatments is to concurrent schedules.

concurrent schedules is to alternating treatments.


Simultaneous treatments/concurrent schedules design-two contingencies are running at the same time. That is, both contingencies are available to the student simultaneously. For example, contingent upon a greeting and within the same session, have one person give edible reinforcement and another person give praise. A consistently higher response rate to one person relative to the other would indicate the preferred reinforcer. (Of course, counterbalancing is needed to control for confounding.) With the alternating treatments/multielement design, interventions are switched back and forth (alternated) rapidly. (Bailey & Burch, 2002, pp. 182-186; Poling et al., 1995, pp. 94-96; Tawney & Gast, 1984, pp. 312-340)

11

Alternating treatments design data indicating behavior at a particular level
do not necessarily reflect behavior at that level in the presence of the other condition.
do not necessarily reflect behavior at that level in the absence of the other condition.
also reflect behavior at that level in the absence of the other condition.
reflect a true effect if the other condition is a baseline condition.

do not necessarily reflect behavior at that level in the absence of the other condition.

Due to possible multiple treatment interference, one cannot be certain that an effect obtained when alternated with another condition would be the same as the effect obtained if it were the only condition. (Bailey & Burch, 2002, pp. 182-186; Cooper, Heron, & Heward, 2007, p. 196; Poling et al., 1995, p. 96; Tawney & Gast, 1984, pp. 312-313)

12

Which event would necessarily be fatal to the internal validity of your alternating treatments design experiment?
A baseline condition cannot be run.
A couple of subjects drop out.
The data are unreliable.
The experiment ends prematurely.

The data are unreliable.

Unreliable data are always a threat to validity. If the data are not believable, the claimed experimental effect is called into question. With respect to the other options, these problems are minimized by the rapid alternation of conditions. (Cooper, Heron, & Heward, 2007, p. 196; Johnston & Pennypacker, 1993b, pp. 142-143; Poling et al., 1995, p. 29)

13

The alternating treatments design controls for
instability.
stability.
multiple treatment interference.
(all of the others)

instability.

14

The alternating treatments design controls for
multiple treatment interference.
generalization effects.
stability.
maturation.


maturation

The rapid alternation of conditions characterized by the alternating treatments design controls for several threats to validity. These include maturation (changes that occur in the subject during the experiment), data instability (variability obscures effects due to overlapping data points), sequence effects (a condition run for several sessions influences the next condition), and to some degree, attrition (loss of subjects). (Bailey & Burch, 2002, pp. 182-186; Cooper, Heron, & Heward, 2007, pp. 195-196; Poling et al., 1995, p. 96; Tawney & Gast, 1984, pp. 91-93, 313)

15

An experimental design that employs multiple standards of responding as the basis for reinforcement or punishment is the
withdrawal design.
multiple baseline design.
alternating treatments design.
changing criterion design.

changing criterion design.

ALTERNATING TREATMENTS (simultaneous treatments, multielement multiple schedule)--rapid alternation between two or more treatments
MULTIPLE BASELINE--staggered implementation of the intervention in a step-wise fashion across subjects, settings, or behaviors
REVERSAL (withdrawal)--alternation between baseline and a particular intervention
CHANGING CRITERION--reinforcement or punishment is contingent upon a certain level of behavior, which changes in a step-wise fashion
(Cooper, Heron, & Heward, 1987, pp. 163-219; Cooper, Heron, & Heward, 2007, pp. 176-224)

16

In a changing criterion design, experimental control is demonstrated when
performance approximates the requirement specified in the contingency as the criterion is varied.
the criterion is altered several times and performance follows a gradual trend independent of condition transitions.
the criterion is altered several times and performance follows a gradual trend across condition transitions.
performance approximates the requirement specified in the objective, which changes as the student progresses.

performance approximates the requirement specified in the contingency as the criterion is varied.


In a changing criterion design, experimental control is demonstrated when the level of behavior abruptly changes to approximate the requirement specified in the contingency as the criterion is varied. Therefore, the criterion must change in a stepwise fashion. (Cooper, Heron, & Heward, 1987, pp. 219-220; Cooper, Heron, & Heward, 2007, pp. 176-224

17

With an alternating treatments design,
early termination of the study is often not fatal.
loss of a subject (attrition) is always fatal.
an initial baseline condition is required.
unstable data is always fatal.

early termination of the study is often not fatal.


With an alternating treatments design, the experimental condition changes every day, every session, or even within a session. As such, an early termination of the session may not be fatal to the experiment if there is enough data to show a contrast between the two (or more) conditions. Contrast this to most other designs in which experimental control isn't possible until several multi-session conditions are run. (Bailey & Burch, 2002, pp. 182-186; Cooper, Heron, & Heward, 2007, pp. 195-196; Poling et al., 1995, p. 96; Tawney & Gast, 1984, pp. 312-313)

18

Multielement design data show reduced aggression. At the end of the study you eliminate the ineffective condition.
You might see a change due to multiple treatment interference in the previous condition.
You will demonstrate experimental control if an abrupt change in level or trend occurs.
You might see a change attributable to baseline logic.
Multiple treatment interference reduces the probability of a change in behavior.

You might see a change due to multiple treatment interference in the previous condition.



Due to possible multiple treatment interference, one cannot be certain that an effect obtained when alternated with another condition would be the same as the effect obtained if it were the only condition. (Bailey & Burch, 2002, pp. 182-186; Cooper, Heron, & Heward, 2007, p. 196; Poling et al., 1995, p. 96; Tawney & Gast, 1984, pp. 312-313)

Decks in Test Cards COPY Class (41):