Correlations

?
What do we do on a typical experiment?
manipulate the independent variable while holding all other conditions constant
1 of 36
What are the observed changes in the dependent variable caused by??
the manipulation of the independent variables
2 of 36
It isnt possible to do what?
Manipulate the variables of interest
3 of 36
It can be useful to see what?
Extent variables co-relate
4 of 36
What is correlation?
A causes B because A covaries with B
5 of 36
but correlation is not?
Causation
6 of 36
Correlations are useful in their own right why?
Highlight possibilities, Null effects may allow to discount theories, usually we need an experiment to find out why the correlation exists
7 of 36
What is correlational design?
measure 2 or more variables and determine the extent to which they are reated to eachother
8 of 36
What is the null hypothesis?
No relationships between variables
9 of 36
What is the alternative hypothesis?
There is a relationship between variables
10 of 36
What is covariance?
Mean of the produce of the deviation from the mean
11 of 36
What is the equation?
Cov = E (mean - x) (mean-y)/N
12 of 36
What are the good properties of covariance?
X and y go up, the covariance is large and positive, X goes up and y goes down, the covariance is large and negative, x and y have no relation so the covariance is o
13 of 36
What are the less good properties?
X and y vary a lot the covariance is large
14 of 36
What is Pearson's coefficient of correlation?
Based on covariance, by dividing by the standard deviations of all the variables
15 of 36
What is it?
The independent of overall variables it is independnet of the overall variability
16 of 36
What are the properties?
Only dependent on how well related the variables are not how much they carry, values between -1 and +1
17 of 36
What is a -1 correlation?
Negative
18 of 36
What is a +1?
Positive
19 of 36
What is a 0?
No linear relationships
20 of 36
What is r?
It can also be used for experimental data
21 of 36
What is Spearman's rank?
Coefficient of correlation related to Pearson's R
22 of 36
What does a non parametric view suggest?
Does not assume that data are evenly spaced or normally distributed
23 of 36
What is it used for?
Ordinal data
24 of 36
How does it work?
By the first ranking the data and then working out Pearson R on the ranks
25 of 36
Where is it found?
Graphs/Legacy dialog/ scatter/ dot
26 of 36
What are multiple pair wise correlations?
Run correlations on more than one variables at once
27 of 36
What did the variance account for?
expressed as a % or a fraction
28 of 36
What is it calculated by?
Squaring pearson's R, conflict explains a lot of variability in contrast but not all of it
29 of 36
What is Partial correlation?
A correlation between 2 variables A and B might be explained by C
30 of 36
What does a partial correlation measure?
The part of the correlation that is not related to variable C
31 of 36
What is variable C refered to as?
Partialled out or held constant
32 of 36
What is a simple correlation called ?
Zero order connection
33 of 36
What is a first order correlation?
Partials out a variable constant
34 of 36
What is second order correlation?
Holds two variables constant
35 of 36
where is it found?
Select/analyse/correlate/partial
36 of 36

Other cards in this set

Card 2

Front

What are the observed changes in the dependent variable caused by??

Back

the manipulation of the independent variables

Card 3

Front

It isnt possible to do what?

Back

Preview of the front of card 3

Card 4

Front

It can be useful to see what?

Back

Preview of the front of card 4

Card 5

Front

What is correlation?

Back

Preview of the front of card 5
View more cards

Comments

No comments have yet been made

Similar Psychology resources:

See all Psychology resources »See all Correlations resources »