Psych Exam#2 (2)

Front 1

Two defining features of single-subject designs:

Back 1

The behavior of each participant is observed before and after treatment.

Repeated observations are made until the behavior is judged stable.

Front 2

large-group designs.

Back 2

Typically 40-200 individuals are randomly assigned to a treatment or control group.Obviously, those in the treatment group get treatment (something thought to improve behavior) and those in the control group don’t get the treatment.

Front 3

Why don’t applied behavior analysts use large-group research designs?

Back 3

Reason #1: Because they are interested in the behavior of individuals.

Reason #2: To use a large-group designs, you need a large number of individuals with the same diagnosis.

Reason #3: People don’t want to be assigned to the control group.

Reason #4: Inferential statistics are a useful set of tools, but study after study has shown that psychologists (even those who publish their work in the most prestigious journals) are not very good statisticians

Reason #5: Single-subject designs encourage the researcher to present their individual data graphically and let the consumer decide if it was successful.

Front 4

internal validity

Back 4

Internal validity: Assuming that behavior changed after the intervention, if we can say that the intervention caused the behavior change, then the experiment has internal validity.

Front 5

3 main single subject design:

Back 5

Comparison Design (A-B design)

Reversal Design (A-B-A design)

Internal validity may be establishedMultiple-Baseline

DesignInternal validity may be established

Front 6

#Step 1: Get some baseline data.

Back 6

Rule #1: Do not introduce the treatment if behavior is improving.

Rule #2: If behavior is getting worse, then you should introduce the treatment.

Rule #3: Attempt to minimize “bounce” (i.e., between-session variability) before introducing the treatment.

Front 7

#Step 2: Get some treatment data

Back 7

Rule #1: Better to have too much treatment data than to not have enough.Learning takes time, so be patient before giving up on the treatment.If the treatment works, you will need to convince a skeptical audience that it has lasting effects on behavior

Front 8

#Step 3:

Back 8

Look for convincing evidence of a treatment effect.

Front 9

Step 1: Are the final three observations in the baseline & treatment conditions divided?

Back 9

(If the behavior is not divided, you are done – you don’t have a convincing behavior change.)

Front 10

Step 2: Are the data in the baseline and treatment phases stable?

Back 10

A stable baseline suggests that if we do nothing, behavior will not change.Stable treatment data suggest that the behavior will remain like this as long as treatment remains in place.

Front 11

Step 3: If the baseline and treatment phases are not stable, this may not be a problem.

Back 11

It depends on the direction in which the data are trending.

Front 12

Step 4:

Back 12

Did the experimenter use a design that has internal validity.

Front 13

Which of the single-subject designs have good internal validity?

Back 13

Reversal

Multiple Baseline

NOT Comparison