correlations

What is a correlational technique?

A non-experimental method used to measure the strength + direction between two co-variables.

What are co-variables?

The 2 variables being measured in a correlational study.

How do correlational studies differ from experiments?

Correlations measure relationships between co-variables, whereas experiments test cause + effect between IV & DV.

Why are correlational studies useful?

Identify rships that can be further investigated (don’t show cause effect but show association between v’s).

Types of correlation?

• +ve correlation - as 1 variable increases the other does too. E.g; height + shoe size.

• -ve correlation - as 1 variable increases, the other decreases. E.g; schl absences + GCSE grades.

• 0 correlation - no correlation between co-variables. E.g; rainfall in Wales + no. Of ppl who’ve read Lord of Rings.

Correlational coefficient: What is a correlational coefficient?

A numerical measure of the strength + direction/nature (+ve/-ve) of a correlation.

What is the range of a correlation coefficient?

-1.0 → +1.0

• -1.0 = perfect -ve correlation.

• +1.0 = perfect +ve correlation.

• The closer the no. Is to +1 or -1 the stronger the correlation.

What does a coefficient close to 0 represent?

A weak or no correlation.

Scattergrams: What is a scattergram?

• A graph showing the rship between 2 co-variables using plotted points.

• Shows direction + strength of correlation.

What does an upward trend on a scattergram show?

+ve correlation.

What does a downward trend on a scattergram show?

-ve correlation.

What does no clear pattern on a scattergram show?

0 correlation.

Application: Give a real-world example of correlation leading to further research?

The rship between cigarette smoking + lung cancer (+ve correlation found initially → cause effect rship discovered).

What are the strengths of correlational techniques?

• Useful for preliminary research - measures rship between co-variables → helps identify patterns + generate testable hypotheses → useful starting point for further investigation.

• Ethical + practical - variables=measured not manipulated, useful when experiments would be unethical → reduces ethical concerns.

• Can use secondary data - alleviates concern over informed consent as info alrdy in public domain (e.g; gov statistics) → access to large data sets.

What are the weaknesses of correlational techniques?

• Cannot establish cause + effect - only shows association between co-variables; could be due to 3rd variable problem → reduces validity of conclusions.

• Limited to linear rships - mainly detects linear rships, may fail to identify curvilinear ships (e.g; temp + aggression) → can oversimplify rships between v’s.