Graphs for Research Methods Exam 1

withdrawal/reversal

What type of SCRD is this?

withdrawal/reversal

What type of SCRD is this?

withdrawal/reversal

What type of SCRD is this?

repeated acquisition design (RAD)

What type of SCRD is this?

repeated acquisition design (RAD)

What type of SCRD is this?

multiple baseline design

What type of SCRD is this?

multiple baseline design

What type of SCRD is this?

multiple baseline design

What type of SCRD is this?

adapted alternating treatment design (AATD)

What type of SCRD is this?

adapted alternating treatment design (AATD)

What type of SCRD is this?

adapted alternating treatment design (AATD)

What type of SCRD is this?

adapted alternating treatment design (AATD)

What type of SCRD is this?

variability

What does this show?

trend

What does this show?

treatment infidelity

What does this show?

testing error

What does this show?

overlap

What does this show?

maturation (attrition)

What does this show?

level

What does this show?

intervening when change is likely

What does this show?

instrumentation

What does this show?

instrumentation

What does this show?

immediacy of effect

What does this show?

history

What does this show?

attrition

What does this show?

Dependent variable

The dependent variable is the outcome measured in an experiment or study, which is influenced by changes in the independent variable.

Are dependent variables irreversible or reversible?

Can be either

Are vocabulary words a reversible or irreversible behavior?

Irreversible

What is a reversible behavior?

A reversible behavior is a behavior that can be undone or reversed, meaning that it can return to its original state after a change or intervention has occurred.

What is an independent variable?

An independent variable is a factor or condition that is manipulated or changed in an experiment to observe its effect on a dependent variable. It is not influenced by other variables in the study. It is the intervention

Regarding visual analysis, what three things analyze within the condition?

Level, Trend, Variability

Regarding visual analysis, what three things analyze between the condition?

Immediacy of effect, overlap, consistency

What is level?

The height on the y axis. How high or low are the data points?

What are the ways to describe level?

-Low and stable

-Immediate and dramatic increase in level

-No change in level

Is the intervention working is there is a change in level between the baseline and intervention conditions?

YES

Trend

Has to do with the movement o the date whether up or down. You need at least three data points to determine the trend.

What are the ways to describe trend?

-Zero-celerating trend

-Increasing/ascending trend

-Decreasing/descending trend

-Variable increasing trend

Is the intervention effective if the trend is going in the direction that you want in the intervention conditon?

YES

What is the best type of trend to have in the baseline?

Zero-celerating

Variability

Has to do with the amount of up and down movement in the data. Are the data roaring waves or a calm sea?

What are some ways to describe variability?

-Low variability

-High variability

-Stability envelope

What is a stability envelope?

Looks at a range of data points on the y axis around a certain level and you can count if the points are inside (more stable )or outside (less stable) of the envelope.

Is it ideal to have a low variability int he baseline?

YES

Immediacy of effect

Look at whether there is an immediate change upon introduction of the intervention. There is a change in level by the THIRD session in the intervention condition compared to the last three sessions in the baseline condtion.

Is it ideal to have a dramatic and immediate change between baseline and intervention conditions?

YES - it shows that the intervention is working

Overlap

How many data points are in the same range between conditions

How do you calculate percent non-overlapping?

-Do not look or count any of the data points in the baseline condition

-Count the number of data points that do not overlap with of the points in the baseline condition

-Do non-overlapping divided by total number of points in the intervention (EX: 12/15 = 80%)

Consistency of data patterns across similar condtions

Shows that the intervention is effective

Is it ideal to have consistency across different people or phases?

YES

What is the main goal of interventions?

To show that there is a demonstration of effect

Does there need to be a huge change to be evidence of a demonstration of effect?

NO

What does sufficient demonstrations of effect show?

That there is a functional relation between the intervention and the dependent variable

How many demonstrations of effect are necessary to show a functional relation in a withdrawal design?

3

What are the different types of single case research designs?

-Multiple baseline designs

-Withdrawal/reversal designs

-Adapted alternating treatments designs

-Repeated Acquisition designs

Why are SCRD experimental research designs?

Because they have experimental control

What does experimental control show?

That an intervention is really having an effect on the dependent variable

Are designs with experimental control more or less reliable?

More

What types of SCRD demonstrate that the intervention worls?

Multiple baseline, withdrawal/reversal, and repeated acquisition design

What types of SCRD compares the interventions?

Adapted alternating treatments design, alternating treatments designs, repeated acquisition design, multitreatment design

Multiple baseline graph

Baseline

what the client can do prior to treatment and are important for experimental control. To tell trend of a baseline, we need 3-5 points

Do we want the baseline to be flat and low?

YES

Multiple baseline graphs

• Also called demonstration design

• Has different tiers

• Need a irreversible behavior

• Want stability and consistency

• Need three demonstrations of effect (change in three people, three behaviors, three settings…)

• Needs to be a staggered baseline

Withdrawal design/reversal

• Type of demonstration design

• Starts with the baseline, then has the intervention, then the intervention is “withdrawn”, and then you have another phase of intervention

• Need a reversible behavior (have to be able to take the behavior away so that you can withdraw the intervention)

• Have to have three demonstrations of effect

Adapted alternating treatments design (AATD)

• Is a comparable design (comparing two different interventions)

• A way to establish experimental control in AATD is to rapidly alternate between two interventions

• Need a irreversible behavior

• Use equivalent sets for each intervention condition (2)

• Randomize or counterbalance treatment schedule

• consider multi-treatment interference (b/c there is two interventions)

• Good to have a control set

Multi-treatment interference

One intervention may be helping to improve what you are teaching in the other intervention

You can check by using a control set

Repeated acquisiton design

• Can compare two independent variables OR demonstrate effectiveness of one independent variable

• Multiple control sets or a pretest/posttest

• Need to have at least 5 sets of pretests and posttests

• Need a irreversible behavior

• Multiple and equivalent behavior sets

Experimental control mechanisms

1. Baselines

2. Replication- within study, across studies

3. Withdrawal/reinstatement (in withdrawal designs)

4. Rapid alternations between treatments (in AATD and RAD)

External validity

• Generalizabiltiy to other clients

• You can generalize to similiar children/adults to the current study

• Replication is needed

Internal validity

• Certainty in the causal effect of your intervention on your dependent variable

Why is external validity no great for SCRD?

• Dealing with a much smaller amount of participants and you can increase external validity by replicating the design across more participants

Threats to internal validity

You might not be able to change the threat, but you can overcome or you might need to stay in baseline longer or start your experiment over again.

1. Maturation/testing

2. History

3. Intervening when change is likley

4. Mortality/attrition

5. Treatment infidelity

6. Instrumentation

Why is Maturation/testing a threat to internal validity?

• Maturation: Participants are getting older and learning the skills over time

• Testing: the participants are getting used to the test or are losing motivation

Why is history a threat to internal validity?

• Having outside therapy impacting your study

Why is intervening when change is likely a threat to internal validity?

• Don’t move to intervention from baseline if the data is already improving

• Wait until the baseline is stable again

Why is mortality/attrition a threat to internal validity?

• Participants drop out of the study

Why is treatment infidelity a threat to internal validity?

• Not doing your treatment as planned

Why is instrumentation a threat to internal validity?

• Not measuring your dependent variable as planned

Treatment infidelity

• Not following your treatment as planned

Reliabilty

How carefully you are measuring your dependent variable

Calculating reliability

• Look at the + and -

• Do the + over the - to calculate