CORRELATIONS

Question 1

scatter diagrams can be…

Answer 1

Strong positive, Weak positive, No correlation, Strong negative, Weak negative, Perfect positive, Perfect negative.
(degree of correlation)

Question 2

Correlation meaning

Answer 2

A relationship between two factors/variables.
It is a statistical technique used to measure or quantify the strength of relationship between two variables.
Two measurements taken from same set of participants.

Question 3

Important to remember:

Answer 3

Do not use experimental terminology when talking about correlations, correlations assess relationship between variables, there are NO IVs involved, nothing is manipulated.

Question 4

Analysis of correlation:
Scattergraph

Answer 4

A graph that shows correlation between two sets of data/ co-variables by plotting dots to represent each pair of scores.
A line of best fit is then drawn through dots to show trend of data.

Question 5

positive correlation

Answer 5

two variables increase together, not necessarily at same rate.

Question 6

negative correlation

Answer 6

as one variable increases, other decreases-not necessarily at same rate.

Question 7

no correlation

Answer 7

no correlation between two sets of data

Question 8

correlation coefficient

Answer 8

A number falls between +1 and -1.
Calculated using Spearman’s Rho.

Question 9

Correlation coefficients

Answer 9

Some are written as a positive or a negative correlation with a plus or minus sign before the coefficient (number)
It tells us how closely the co-variables are related.
e.g -0.76 is just as closely correlated as +0.76, its just that -0.76 means as one variable increases the other decreases (negative correlation) and +0.76 means both increase together (positive)

Question 10

+1 coefficient=

Answer 10

perfect positive correlation

Question 11

-1 coefficient=

Answer 11

perfect negative correlation

Question 12

0 correlation=

Answer 12

no correlation

Question 13

Examples of coefficients and their meanings

Answer 13

+0.8= strong positive
+0.5= moderate positive
+0.3= weak positive
-0.3= weak negative
-0.5= moderate negative
-0.8= strong negative

Question 14

Directional correlational hypotheses

Answer 14

(One tailed)
Predicts direction result will go in.
-‘positive’ or ‘negative’ RELATIONSHIP.

Question 15

Non directional correlational hypotheses

Answer 15

(Two tailed)
Predicts a change but not a direction.
‘there will be a RELATIONSHIP between…’

Question 16

Null correlational hypotheses

Answer 16

(No change predicted)
‘There will be NO RELATIONSHIP’

Question 17

Correlations cannot show cause and effect…

Answer 17

Never use words; cause, effect, difference.
Only ever say; relationship.

Question 18

When sketching a scattergraph:

Answer 18

Must add a Title.
Have an x and y axis labelled.
Must mention if axis is rounded, e.g ‘data rounded to 1000s’

Question 19

Strength of correlations

Answer 19

Quantitative data provides info of patterns and trends, if a correlational study indicates presence or absence of a relationship further investigation can be justified using an experiment.

Question 20

Strength of correlations

Answer 20

Correlational studies can be conducted on variables that can be measured but not manipulated, can be used to research topics that are sensitive or otherwise would be unethical- no deliberate manipulation of variables is required.

Question 21

Strength of correlations

Answer 21

Procedures can be repeated, findings can be confirmed to check reliability.
(like experiments)

Question 22

Weakness of correlations

Answer 22

Correlation is not and cannot be taken to imply causation, even if there is a strong association between 2 variables, we cannot assume one causes another.

Question 23

Weakness of correlations

Answer 23

Correlations show linear relationships but do NOT reflect curvilinear ones (if change in variables is not constant) such as effect of stress on exam performance.

Question 24

Weakness of correlations

Answer 24

Lacks validity.
Findings are not always valid as they don’t reflect the full picture.

Question 25

Weakness of correlations

Answer 25

There may be intervening variables (third variable) that can explain the co-variables being studied are linked. Correlation doesn’t show more than 2 variables.

Question 26

Comparison of experiments and studies using correlational analysis

Answer 26

Experiments determine cause + effect because we can observe effect of IV and DV
The procedures can be replicated easily, findings can be confirmed.

Correlations cannot find cause and effect.
There may be other unknown variables that can explain wh co-variables being studied are linked.

Question 27

Comparison of experiments and studies using correlational analysis

Answer 27

Correlations can be used when its not possible to manipulate variables.
If no correlation can be found, a causal relationship can be ruled out.

Experiments are often artificial, a contrived situation where ps may not have acted naturally.
Lack ecological validity as we cannot generalise from experimental setting to real world setting.

Question 28

Reliability in correlations

Answer 28

In correlational analysis we are concerned about consistency of measures used, findings of correlational analysis will be reliable only if measures of both dependent variables are consistent.
For some correlations, using scientific scales (volume, time etc…) the measures will be highly reliable.

Question 29

Validity in correlations

Answer 29

To be valid a correlational analysis must be reliable but other factors affect validity too.
As with reliability findings of correlational analysis will only be valid is both measures of variables measure real phenomena in effective ways.
To achieve this, DV’s must be defined and measured effectively.