Correlations Flashcards Preview

Psychology G541/G544a > Correlations > Flashcards

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

What are the pros and cons of median measures?

+ Can be used with any type of data
+ Easy to calculate
- Not useful is there are various values that meet the criteria
- isn't useful for small sets of data