Quantitative- small n designs

Question 1

Why is small n design used?

Answer 1

• when it is difficult to find potential subjects
• grouped data can lead to misleading results (when you group everyones data is shows a slope of learning when in fact it is a vertical slope as people suddenly understand)

Question 2

Small N designs in Applied Behavioural Research

and requirements

Answer 2

-Showing that the behaviour of a single individual changes as a result of the treatment

This requires:

• The target behaviour must be operationally defined in terms of easily recordable events (countable)
• Establish a baseline level of responding against which the effects of the programme can be assessed
• Introduce the treatment and continue monitoring

Question 3

3 types of small N designs

Answer 3

1. withdrawal (reversal designs- ABAB)
2. multiple baseline designs: used when a withdrawal design is not feasible)
3. Alternating treatments design: used when comparing the effectiveness of more than one treatment for the same individual

Question 4

drawing a small n design graph

Answer 4

X axis= time
Y axis= number of observations of events

time broken into A and B
A= baseline phase
B= treatment phase

Question 5

benefits of withdrawal case study:

Answer 5

ABAB design shows treatment work wasnt history or maturation

Question 6

Multiple baseline designs

Answer 6

Asthma study (eye direction, breathing in, facial positioning) 
Shows the treatment across different baselines- reason for them is you want to show the same treatment is working across different things.

Question 7

Multiple baselines across subjects

Answer 7

stuttering study
multiple children
multiple types of stuttering

Question 8

Evaluating small N designs: what about external validity?

Answer 8

you replicate

Question 9

Evaluating small N designs: what about statistical significance?

Answer 9

You can see on graph (can do stats if have 120 data points on axis but you don’t have to)

Question 10

Evaluating small N designs: what about interactions?

Answer 10

You can get interactions (works because of steve in a particular environment that interacts with treatment) you should do a long time period and multiple baseline test because multiple people have different environments

Question 11

Evaluating small N designs: what about studies that don’t rely on frequency of response?

Answer 11

Mostly frequency counts but may use reaction times, habituation to a stimulus.