Flashcards in Correlations Deck (14):

0

## What is the definition of a correlation?

### A relationship between 2 co-variables.

1

## What is correlation design?

###
PP provides data for both co-variables.

(NO IV OR DV)

2

## What is a test of correlation?

### Tell you whether there is a relationship and also the STRENGTH and DIRECTION of it.

3

## What are the differences between one tailed, two tailed and null correlation hypotheses?

###
ONE= there will be a significant positive/negative correlation between...

TWO= there will be a significant correlation between...

NULL= there will be no significant correlation between....

4

## What is a positive correlation?

### As the values of one co-variable increases, the values of the other co-variablea also increase.

5

## What is a negative correlation?

### As the value of one co-variable increases, the value of the other decreases.

6

## What is no correlation?

### There is no relationship between the co-variable, any relationship is due to chance.

7

## What is a correlation co-efficient?

###
Between -1 and +1.

Perfect negative= -1

Perfect positive= +1

No correlation=0

The more extreme the number, the stronger the co-effiecient.

8

## What are the strengths of correlation?

###
-gives precise info on degree of relationship between variables

-no manipulation of behaviour is required (can be used in situations where it is unethical/impractical)

-good starting point for further experimental research

9

## What are the weaknesses of correlations?

###
-No cause and effect can be inferred

-correlation can only be relied on if the way of collecting the data for the two co-variables is accurate

10

## What is the mean?

### Often known as an average, calculated my adding up all the scores and dividing by the number of PPs.

11

## What are the pros and cons of mean measurements?

###
+ Makes use of all the values, thus the most representative measure

- can be unrepresentative if there are extreme values

12

## What are the pros and cons of mode measurements?

###
+ Unaffected by extreme values

- Not as sensitive as not all values are reflected

- Ignores very large or very small outliers

13