SPEC POINT 1.8

Question 1

What is Correlation?

Answer 1

When a change in one variable is accompanied by a change in another

Question 2

What is Causation?

Answer 2

When a change in one variable is caused by a change in another

Question 3

Describe how data might provide evidence for links between risk factors and illness and mortality:

Answer 3

1. Look for a general correlation between the risk factor and illness/mortality in graphs
2. The same applies to tables, look for coinciding patterns
3. Focus on trends, not individual data points - this may involve comparing one set of data from control groups to other groups
4. If error bars are present, non over-lapping error bars means that there’s more chance of a significant relationship
5. Look out for details of the study: well designed studies add confidence to conclusions
6. Data manipulations may be required, such as calculating percentage increases to support your conclusions

Question 4

Describe how data linking to risk factors to illness and mortality may be critiqued:

Answer 4

1. Correlation doesn’t confirm causation
2. Have other risk factors definitely been ruled out/controlled for?
3. Are there parts of the graph/data that don’t fit the trend?
4. Are the error bars overlapping in the region of correlation
5. Is there anything in the design of the study, such as sample size, unrepresentative sample or certain variables not controlled?
6. Could there be any bias related to who did the research or who is making the conclusions?