Every pps has an equal chance of being allocated to any condition
1 of 22
What is the purpose?
To spread any potentially confounding differences between participants evenly across treatment conditions: ie: To minimise systematic differences other than the treatment
2 of 22
What does random allocation aid in?
Aids in isolating the causal effects of the treatment on the dependent variable
3 of 22
When is it importnat?
Where potential nuisance variables are difficult to identify
4 of 22
What is the procedure for random allocation?
May involve using a draw to allocate participants to conditions or using a random number generator
5 of 22
What is the research procedure of a quasi experiment?
Aims to establish causal effect between an independent variable and variations in a dependent variable
6 of 22
Where is there no?
Full control over the allocation of participants to the different levels of the independent variable
7 of 22
What is there a lack of?
Random allocation
8 of 22
What is the lack of random allocation a key threat to?
Internal validity, as it increases the risk that group may systematically differ for respect to factors other than independent variable
9 of 22
What is the nomenclature one group pre post test design?
Observation, a treatment, observation
10 of 22
What is non equivalent control group design?
O, a treatment, O, not randomly assigned, O O
11 of 22
What is an interupted time series design?
O, O, O, O, O, O, treatment, O, O, O, O, O not randomly assigned, O O O O O O O O O
12 of 22
What is systematic approaches to control for nuisance variables?
Blocking, Matching, Counterbalancing
13 of 22
What is blocking?
Experiment is arranged in blocks, within which a known nuisance variable is held constant (E.G. Experimenter; sex of participant; time of testing)
14 of 22
What is allowed to vary in a blocking experiment?
Independent factor
15 of 22
What is tested within each level of the independent variable?
It is tested equally often and participants can be randomly allocated to the levels of the independent variable (Randomised Block design)
16 of 22
What does blocking do?
Seperate the effect of a known nuisance variable from the effect of the variable of interest
17 of 22
What is a matching design?
Groups are matched/comparable with respect to specific individual differences (education, sex, age, pretreatment performance, smoking status)
18 of 22
What does this minimise?
The influence of this nuisance variable on the dependent variable
19 of 22
In a matched subject design, what does every participants do?
Compared with another pps who is matched according to a relevant nuiscance variable
20 of 22
When does counterbalancing occur?
In within subjects design, the confounding effects of nuisance variables, such as testing order can be countered by balancing them across the different levels of the independent variable
21 of 22
What is latin square designs?
Used when there are more than 2 conditions
22 of 22
Other cards in this set
Card 2
Front
What is the purpose?
Back
To spread any potentially confounding differences between participants evenly across treatment conditions: ie: To minimise systematic differences other than the treatment
Comments
No comments have yet been made