Experiments and statistics Flashcards
1
Q
What are the steps in experimental design?
A
- formulate hypothesis
- manipulate variables by translating hypotheses into treatment conditions
- administer treatment to groups or same p’s
- measure performance on a response measure
2
Q
Independent variable
A
treatment conditions
3
Q
dependent variable
A
Response measures
4
Q
3 types of independent variable
A
- Quantitative variables ( represent variation in amounT, E.G. AMOUNT OF DRUG)
- Qualitative variables ( represent variation in kind or type, E.G. TEACHING STRATEGY)
- classification ( variables represent characteristics that are intrinsic to thr subjects/participants, E.G. SEX, SPECIES, AGE)
5
Q
Nuisance variables
A
- potential independent variables which if left uncontrolled could cause systematic influence in the differnet treatment conditions
- if uncontrolled known as confounding variables
6
Q
dependent variable
A
- what you measure
- a good dependent variable should capture the hypothesised differences
7
Q
what’s one solution for systematic differences in experiments
A
random allocation - equal chance of participants being in either group
8
Q
completely randomised design (between subjects design)
A
- each subject randomly assigned to one of the treatment conditions
- helps to prevent confounds
9
Q
repeated measures / within subjects design
A
- subjects matched closely on some relevant characteristics
- common procedure to treat subjects as a “block” and participants in all treatment conditions of a IV
10
Q
research hypothesis
A
fairly general statement about the presumed nature of the world that inspires specific experiment
“physical exercise decreases dementia symptoms”
11
Q
statistical hypothesis
A
- precise statement about the parameters of distributions for different treatment populations
- null and alternative hypotheses
“mean dementia scores will be lower for the exercise group than for the No-Exercise group (more so than what we would expect to observe by chance)”
12
Q
The null hypothesis
A
- tests seek to accept or reject it
- 𝑯_𝟎: 𝝁_𝟏=𝝁_𝟐= 𝝁_𝟑
- this is the same as saying that no treatment effects are present in the population
13
Q
The alternative hypothesis
A
- if the treatment parameters do not satisfy the null hypothesis ( one or more of the differences between treatment means is greater than expected by chance) REJECT the null hypothesis
- parameters are not equal between treatment conditions
14
Q
what is the typical alpha value?
A
0.05
15
Q
when do you reject H0?
A
- if the p value is less than alpha its statistically significant and we reject H0
- otherwise fail to reject H0
