Anytime you choose a measurement system, you must ensure that the measurement system you selected gives a valid representation of the behaviors you intend to target. When extending the baseline it is important to ensure that the data points are stable, level, and have at least 3 data points.
BCBAs should always consider practical and ethical concerns when selecting an experimental design. Practicality and ethics are highly important in influencing the success and validity of the design.
When combining experimental designs, the designs can affect the weakness in one design and spotlight the strengths in another design, to explore the additive or interactive effects of 2 or more IV are investigated. BCBAs should understand the strengths and weaknesses of each design to consider how to combine them to effectively work together
DRO, DRI, and DRA reversal techniques can all be used to demonstrate the effectiveness of contingent reinforcement. They all require a contingency for reinforcing a response that is different than the maladaptive target bx.
The 3 practial and ethical issues that a BCBA MUST identify about single-case designs when showing effectiveness are: the baseline trends, excessive variability in data, and duration of phases. When baseline trends are increasing or decreasing instead of stable responding it is not clear whether the IV was the cause for the change in DV. The frequency of the behavior is not one of the practical and ethical issues.
Multiple baseline design
- there is multiple baseline across subjects, across settings, and across behaviors.
- do not require a reversal effective treatment
- the sequential implementation of IV parallels practice of changing multiple behaviors in different settings and/or subjects,
- the concurrent measurement of multiple behaviors allows for direct monitoring of generalization of behavior change,
- it is relatively easy to conceptualize
- Bx should be functionally independent and should both change when the IV is applied, bx must be measured concurrently and have EO infulenced by same set of relevant variables, the IV shouldn't be applied to the next bx until the previous bx is changed maximally and a sufficient period of time passes, the length of baseline phases should vary significantly, the IV should be applied to the bx showing most stable level of baseline first, and the reversal phase in one or more tiers can demonstrate a functional relation.
- By demonstrating a reversal phase in one or more of the tiers of the multiple baseline design, the existence of a functional relation is strengthened. The reveral allows for a replication of results under similar conditions, which shows that the change in the IV can cause the effects in the DV.
- could potentially be practicality concerns. There is more evaluation of the independent variable's effectiveness than analysis of the behavior involoved in this design which requires considerable time and resource(i.e. training)
Delayed multiple baseline designs
- involve an initial baseline, or intervention, are begun for one behavior(or setting, or subject), and subsequent baselines for additional behaviors are begun in a staggered/delayed fashion.
- Some advantages include: no reveral is required, it can be used with limited resources, and new behaviors, settings or subjects can be added when available.
Changing criterion
- do no require a reversal
- there must be stable baseline responding before moving on to the first treatment phase where reinforcement/punishment is contingent on performing at a specific level
- is seen as a variation of the multiple baseline design. It is an experimental design that consists of successive and gradual changing criteia for reinforcement or punishment. It contains an initial baseline phase, followed by a series of treatment phases applied in a stepwise manner. Experimental control is demonstrated by the extend that the level of responding changes to fit to each new criterion.
- are appropriate to intervene upon behaviors that are already within the learner's repertoire. It can be used to gain success with increasing manding, decreasing cursing, and increasing reading time. It should not be used to shape a new behavior.
- Decreasing cigarette smoking is for a changing criterion design. This behavior is already within the learner's repertoire and this design can set gradually changing and successive criteria to provide punishment for the smoking behavior. It can slowly decrease the amount of cigarettes smoked to reach an end criterion of terminating smking behaviors entirely. #stepwise #endcriterion #mastery
Reveral and withdrawal designs
- it is common to combine reveral design with NCR, DRO and DRA interventions.
- It is the least common combination to use extinction in conjuction with a reveral design.
- you must have a baseline phase to show what the behaviors look like before you intervene,
- the design also must have at least 3 phases to show that the results occur again upon reversal/withdrawal.
- you must consider the most powerful design foe showing functional control-within subject design(reversal)
- have the potential to have ethical concerns. If an intervention is effective and then reversed or withdrawn, this can be counterproductive for intervening upon maladaptive bx. This can also inadvertently cause harm or undesirable side effects when returning to baseline levels as the client is no longer receiving effective treatment. They must never b used for behaviors that are dangerous.high-risk as it is not ethical to risk the safety of the client or others.
Alternating treatments design
- is based upon stimulus discrimiantion and allows for the rapid alternation and comparison of two or more different treatments' effects on the target behavior measured. It efficiently compare intervention effectiveness for multiple interventions.
- can be implemented with a mazimum of four treatment conditions, it is one its limitations of the design.
Multiple treatment reveral designs
- have disadvantages such as sequence effects, which are the effects on a subject's behavior that are the result of the subject's experience with a prior condition.
B-A-B design
- this variation of the A-B-A-B Design starts with application of the treatment(IV)
- to show reveral to the treatment condition