Correlation

Question 1

Two tailed correlation hypothesis

Answer 1

‘There will be a significant correlation between co-variables y and z’.
No direction given

Question 2

One tailed correlation hypothesis

Answer 2

‘There will be a significant positive/negative correlation between co-variables y and z’

Question 3

Null hypothesis

Answer 3

Can be retained if the alternative hypothesis is not supported by the evidence.
‘There will not be a significant correlation between co-variables y and z; any relationship will be due to chance factors’.

Question 4

Primary data

Answer 4

Data gathered directly from the participants by the researcher.

Question 5

Secondary data

Answer 5

Data that has already been gathered by someone other than the researcher.

Question 6

Difference between findings and conclusions.

Answer 6

Findings: raw data e.g. mode, median, mean, range, outliers).

Conclusions: broad inferences that you can make from that raw data e.g.direction of correlation (positive/negative/no correlation), strength of correlation.

Question 7

Inferential statistics for correlation

Answer 7

Positive correlation: a correlation that has a plus sign as part of its correlation co-efficient (e.g. +0.58).

Negative correlation: a correlation that has a minus sign as part of its correlation co-efficient (e.g. -0.72)

No correlation: a correlation with a co-efficient around 0 (e.g. +0.12 or -0.17)

Question 8

Correlation study in practice

Answer 8

1. Each participant in psychology class tested their reaction speed by tapping on their screen once a green light turns red.
2. Then each participant was asked to rate their tiredness on the Stanford Sleepiness Scale.
3. No correlation was found between the co-variables so they had no relationship.

Question 9

Disadvantages of correlation studies.

Answer 9

• They don’t tell us anything about cause and effect (e.g. if it was established there was a positive correlation between amount of violent TV people watched and level of aggression, this wouldn’t tell us what is causing what).
• The inferential statistical tests won’t always pick up on a relationship between two co-variables (e.g. a curvilinear relationship would produce a correlation co-efficient of 0 even though there’s a pattern to the relationship.
• Because correlation studies involve investigation of the statistical relationship between two co-variables, they don’t require the collection of any qualitative data which limits our ability to know what sits behind any relationship that might be found.

Question 10

Experiments vs correlations

Answer 10

• Experiments are looking to see cause and effect of the IV on the DV.
• Experiments study the difference between IV’s influence on the DV.
• Correlations are examining how 2 unmanipulated co-variables have an effect on eachother.
• Correlations study the relationship between two co-variables.