rm chapter 14

The basic requirements for an alternating treatments design are

1. It must be possible for the researcher or clinician to switch randomly between treatment conditions and 2. The participants behavior must show an immediate response to the treatment being administered. There is not time for a response to evolve over a series of observations

There are three fundamental differences between single subject designs and traditional group designs

1. Single subject research is conducted with only one participant or occasionally a very small group. 2. A single subject design can be modified or completely changed in the middle of a study without seriously affecting the integrity of the design. 3. Single subject design require continuous assessment

A phase must consist of a minimum of

3 observations

Baseline phase

A series of baseline observation, identified by the letter A

There are several factors that can influence the decision concerning when and if a phase change is appropriate

Consideration involves changing from a baseline phase to a treatment phase

The four phases of ABAB design

a baseline phase, a treatment phase, a return to baseline phase, and a second treatment phase

Typically five or six observations are necessary to determine

a clear pattern

ABAB design is not appropriate for evaluating treatments that are expected to have

a permanent or long lasting effect

Treatment phase

a series of treatment observations, identified by the letter B

ABAB design also known as a reversal design

a single subject experimental design consisting of four phases *although the design is typically used to show that a treatment does have an effect, it also can provide convincing evidence that a treatment is not effective

The risk of multiple baseline designs is that

a treatment applied to one behavior may generalize and produce changes in the second behavior

When high variability exists in the data points

additional observations should be made

Primary strengths of a single subject design is

allow researchers to establish cause and effect relationships between treatments and behaviors using only a single participant. Also a single subject design is flexible

The first issue related to withdrawal of treatment focuses on the particiaptns response

although the researcher may return to baseline by removing the treatment, the participants behavior might not return to baseline *the credibility of the treatment effect is seriously compromised

Initiating a treatment when the participant is already showing a trend toward improvement can only result in

ambiguous results. If a treatment is started and the participant continues to improve, the researcher cannot determine whether the continued improvement is caused by the treatment or is simply the continuation of an established trend

Multiple baseline design

begins with two simultaneous baseline phases. A treatment phase is initiated for one of the baselines while baseline observations continue for the other. At a later time, the treatment initiated for the second baseline *requires only one phase change- from baseline to treatment

To qualify as a true experiment the graph must provide convincing evidence that the treatment has

caused a change in behavior

four characteristics of a single subject data that help determine whether there is a meaningful change between phases

change in average level, immediate change in level, change in trend, latency of change

The pattern within a phase can be described in terms of level or trend. In either case, however the critical factor is the

consistency of the pattern. Single subject research does not use statistical analysis to summarize or interpret the results, but depends on the visual appearance of the data in a graph

Dismantling design also called a component analysis design

consists of a series of phases in which each phase adds or subtracts on component of a complex treatment to determine how each component contributes to the overall treatment effectiveness

Changing criterion design

consists of a series of phases in which each phase is defied by a specific criterion that determines a target level of behavior. The criterion level is changed from one phase to the next. Evidence for a successful treatment effect is obtained when the participants level of behavior changes in accordance with the changing criterion levels

To demonstrate cause and effect relationships, single subject designs must demonstrated

convincingly that it is the treatment, noto coincidental extraneous variables causing the changes in behavior

The ABAB design is used in a variety of areas

demonstrating effectiveness of a treatment to improve social interaction in children with autism, a dietary treatment of OCD symptoms, and a token economy to increase adherence to an exercise program to improve airway clearance for children with cystic fibrosis

Strength of the multiple baseline design

eliminates the need for a reversal, or return to baseline phase and therefore is well suited for evaluating treatment effects that are permanent or long lasting

The goal of a single subject design is to

identify cause and effect relationships between variables

The researcher changes phases by

implementing or withdrawing a treatment and the goal is to show the pattern of behavior changes from one phase to the next

Causal interpretation depends

in large part on the reversal (return to baseline) that is a component of the design

Immediate change in level

initial response of the participant to the change. This involves comparing the last data point in one phase with the first data point in the following phase

Phase change

involves changing the conditions, usually by administering or stopping a treatment *purpose is to demonstrate that adding or removing a treatment produces a noticeable change in behavior

An advantage of the alternating treatment designs is that

it allows a fairly rapid comparison of two different treatments

Before it is possible to demonstrate a change in patterns

it is essential that the pattern within a phase be clearly identified

Researcher can simply wait

keep making observations and hope that the data will stabilize and reveal a clear pattern

Statistically significant results

means that the observed effect, whether large or small, is very unlikely to have occured by chance

Although single subject studies are experimental, their general methodology incorporates elements of

non experimental case studies and time series data

Treatment observations

observations made when a treatment is being administered

Baseline observations

observations made when no treatment is being administered.

Level

occurs when a series of measurements are all approximately the same magnitude. In a graph, the series of data points cluster around a horizontal line

Trend

occurs when the differences from one measurement to the next are consistently in the same direction and are approximately of the same magnitude

The decision to make a phase change is based on the

participants response. If the responses establish a clear pattern, then a change is appropriate

Single subject designs

research designs that use results from a single participant or subject to establish the existence of cause and effect relationships. To qualify as experiments, these designs must include manipulation of an independent variable and control of extraneous variables to prevent alternative explanations for research results.

Phase

series of observations of the same individual under the same conditions

First phase change- baseline to treatment

shows a clear change in pattern of behavior. We cannot conclude yet that the treatment has caused the change in behavior because some extraneous variable that changed coincidentally with the treatment might responsible for changing the behavior

Second phase- treatment to baseline

shows the participants behavior returning to the same level observed during the initial baseline phase. The reversal component is important because it begins to establish the causal relationship between the treatment and behavior. It now appears more likely that the treatment is responsible for the change

Final phase change-baseline back to treatment

shows the same treatment effect that was observed in the initial phase change. The component of the experiment, the second AB sequence, provides a replication of the first AB. this replication clinches the argument for a causal interpretation of the results

The visual inspection of single-subject data is very much a

subjective task and different observers interpret data in different ways. There are guidelines that focus attention on specific aspects of the data and help observers decide whether a phase change produced a real change in pattern

Change in average level

the average level of behavior during a phase provides a simple and understandable description of the behavior within a phase

Average a set of two (or more) observations

the averaging process tends to reduce the variability of the data points and produces a more stable set of data in which the pattern of behavior is easier to see

Disadvantages of single subject design

the cause and effect relationship is demonstrated only for one participant. Also the requirement for multiple, continuous observations. And the absence of statistical controls

The most compelling evidence for a causal relationship between the treatment criteria and the participants behavior occurs when

the data consistently and closely tracks the criteria level

Stability of a set of observations refers to

the degree to which the observations show a pattern of consistent level or consistent trend. Stable data may show minor variations form a perfectly consistent pattern, but the variations should be relatively small and the linear pattern relatively clear

There are two general strategies for conducting a dismantling design

the first is to begin with a full treatment phase, then remove components one by one to see the if the effectiveness of the treatment is reduced. The second strategy is to begin with a baseline phase, then add components one by one to see how each individual component contributes to the effectiveness of the total treatment package

Latency of change

the most convincing evidence for a difference between two phases occurs when the data show a large, immediate change in pattern. A delay between the time the phase is changed and the time behavior begins to change undermines the credibility of a cause and effect explanation

Techniques when data appears to be unstable

the researcher can simply wait, the researcher can average a set of two observations and look for patterns within the inconsistency

For the multiple baseline experiment, it is essential that

the treatment affect only the specific behavior to which it is applied. If both behaviors show a response to the initiation of treatment, the credibility if the treatment effect is undermined

Practical significance

the treatment effect is substantial and large enough to have practical application

Criteria for a successful multiple baseline experiment

there is a clear and immediate change in the pattern of behavior when the researcher switches from a baseline to a treatment phase and the design includes at least 2 demonstrations that behavior changes when the treatment is introduced

Alternating treatments design or discrete trials design

two treatment conditions are randomly alternated from one observation to the next. The result is a series of observations that represent a corresponding series of alternating treatment conditions

Look for patterns within the inconsistency

unstable data may be caused by extraneous variables; it is often possible to stabilize the data by identifying and controlling them

Single subject designs rely on the

visual effect of a graph to convince other that the treatment effects are real. Reliance on graphed results helps ensure that researchers report only results that are substantial, the treatment effects must be sufficiently large that they are obvious to a casual observe

The presentation and interpretation of results from a single subject experiment are based on

visual inspection of a simple graph of the data

Multiple baseline across subjects

when the initial baseline phases correspond to the same behavior for 2 seperate participants *it is also possible to conduct a multiple baseline study using two or more different behaviors for a single participant. The key to the single subject version of the design is that the different behaviors are independent and can be treated separately by focusing a treatment on one behavior at a time

Multiple baseline across situations

when the initial baseline phases correspond to the same behavior in two separate situations

Multiple baseline across behaviors

when the initial baseline phases correspond to two seperate behaviors for the same participant

Change in trend

when the trend observed in one phase is noticeably different from the trend in the previous phase, it is a clear indication of a difference between phases