Small N Design

Question 1

Examples of Narrative Case Studies

Answer 1

HM - Anterograde amnesia

Phineas Gage (pole through his frontal lobe)

Freud’s studies

The man who mistook his wife for a hat

Question 2

Narrative case study

Answer 2

Intense, richly detailed study of a single person

Initial basis for developing new theory

Question 3

Why do narrative case studies?

Answer 3

Can provide unique and valuable information

Source of ideas and hypotheses

Source for developing new therapy techniques

Allow study of rare phenomena

Side effects of treatment

Question 4

Limitations of narrative case studies

Answer 4

Alternative explanations often available

Relies heavily on anecdotes

Limited generalizability

Non-representative cases often presented

Many threats to validity

Question 5

Single case experimental design

Answer 5

Inferences made about the effect of an intervention by comparing different conditions presented to the same subject overtime

Question 6

Single case experimental design essential components

Answer 6

Baseline assessment (document behaviour prior to the intervention)

Continuous assessment (Observations made prior to and during the intervention)

Stability of performance (baseline behaviours must be stable in order to make proper evaluations)

Multiple phases (baseline and intervention has to be observed across many periods of time to show possible patterns)

Question 7

A-B single-case design

Answer 7

Quasi experimental design

Cannot exclude for low internal validity (maturation, history, regression to the mean)

Low external validity

Question 8

ABA Example Cocaine

Answer 8

Escalating monetary reinforcement for cocaine free urine tests, then stopped at $2000

Question 9

Problems with ABA(B)

Answer 9

Not all treatment-related behaviours may be reversible

Ethical problems associated with treatment withdrawal

Switching the treatment “on and off” may have undesirable consequences (e.g., lack of trust)

Question 10

Multiple Baseline Design

Answer 10

Treating many behaviours/problems in one setting or one behaviour across many settings

Avoids the ethical dilemma of removing treatment

Question 11

Changing-criterion design

Answer 11

Apply treatment over a series of trials

Each trial has a different threshold for defining a treatment response

E.g., reducing smoking

Question 12

Challenges for single-case designs

Answer 12

Difficult to establish a stable baseline

Hard to know when baseline has stabilized (due to little data)

Interpretation of treatment efficacy can be ambiguous without stats

Behaviour may be reactive to measurement and thus impossible to obtain a valid measurement

Question 13

Data analysis in single case design

Answer 13

Generally without stats

Usually visual inspection

Question 14

Criterion for visual inspection

Answer 14

1. MEAN
   - Reflects a shift in average performance from one phase to the next
2. LEVEL
   - Difference between last day of previous condition and first day of new condition
   - Behaviour assumes new rate that is clearly associated with new phase
3. SLOPE
   - Direction or magnitude change during each phase
4. LATENCY OF CHANGE
   - How quickly does the difference between the two conditions become apparent
   - Generally shorter latency

Question 15

Stability

Answer 15

Whole purpose of single case designs are to show control over the IV

Variability should be followed by stability

Question 16

Problems and considerations

Answer 16

• Unclear decision making rules
• Need for big effects
• Attrition
• Fatigue
• Data may be hard to interpret