Prediction, verification, and replication

Replication

"essence of believability" - repeat the experiment by reintroducing the independent variable to see if there is a functional relationship -Did something else cause observed effect ? -Reliabiliy.

Considerations

-# of data points plotted within a condition -# of variables changed between adjacent conditions -Level stability & change level within and between conditions -trend conditions, trend stability and changes in trend within and between conditions -% of data points in one condition that fall within range of data plotted in adjacent conditions

overlap

-% of data points from 1 phase that overlaps with range of data from previous phase. -Best little trend -Greater overlap the less confident you are with results -% of overlap data: the proportion of data in one phase that overlaps with another phase

Four steps

-Do baseline data document a predictable patterns - Do data within each phase allow documentation of predict pattern -Do data across phases document basis effect? -Do data across phases document experimental control?

Steady state strategy

-Forms the basis of baseline logic -eliminate or control extraneous influences on behavior -stable pattern of responding

Consistency of separation

-Greater consistency of separation between two conditions (no overlap) larger the demonstration of functional relationship

steps for visual analysis

-Is baseline adequate -Are there sufficient data within a phase to document a pattern -is there a basic effect between two phases -Functional relationship documented by the full data set

Immediacy of Effect

-Magnitude of change in level trend or variability -Look at last 3-5 data points in 1 phase and then first 3-5 data points in next phase -Do we predict immediate change or gradual change

Verification

-Termination or withdrawal of the independent variable ---> obtain verification of predicted baseline level

number of data points used to establish separation

-The more points documenting separation to larger the demonstration of a functional relationship

Visual Analysis

-Tradition of ABA (practice settings, single case research design) -Lack of consensus surrounding the statistical analysis of single case research design

Level

-Value on vertical axis around a series of data points coverage -mean of data within a phase

Affirmation

-Will the independent variable change the behavior? -If the independent variable were not applied then we can predict baseline would continue "If a then B" scenario

Benefits of Vis. Anal.

1. Can be used to evaluate individual or group data 2. Regularly collect, graph and analyze data. 3. Allows us to make data based decisions (individualized) 4. Permits individualization 5. Allows for serendipitous findings 6. Conservative method.

Six variables

1. Level 2.Trend 3. Variability 4. Overlap 5. immediacy of effect 6. Consistency across similar phases ** 7th consideration multiple baseline (stability is non-intervened series when effect demonstrated in one series)

considerations alternated treatment design/ Multielement

1. Magnitude of separation 2. Consistency of separation 3. number of data points used to establish separation

Baseline logic

1. Prediction (affirmation) 2. Verification 3. Replication

The greater the variability

The less trustworthy predictable outcomes

Trend

The slope of best fit straight line describing data within a phase -3 to 5 data points within a phase -The greater the trend, the more variability, the more difficult it is to describe data

Prediction

- Anticipated outcome -Consistency of data patterns -How many measurements are needed

Magnitude of separation

- Greater the difference between two conditions, the larger the demonstration of a functional relationship