There is a famous saying in statistics that \"Correlation does not imply causati

Question

There is a famous saying in statistics that "Correlation does not imply causation". Please address the following questions:

1) In your own words, what does this statement mean?

2) In statistics there are what are called 'spurious' correlations and 'lurking variables'. What do these terms mean? Does a spurious correlation necessarily have a lurking variable or could it just be a coincidence?

3) Finally, go to any website and find a spurious correlation. Do not choose one that another student has chosen. The spurious correlation can be implied or explicitedly stated by the authors. What are the variables being correlated? Why is the correlation spurious? Are there any lurking variables that could otherwise explain the correlation or is it simply a coincidence?

This website has some good examples: http://www.tylervigen.com/spurious-correlations

Explanation / Answer

Solution:-

1) We know that the statistical relationship between the two variables does not imply that the one variable is a cause of other. Actually correlation coefficient is the statistical measure and it does not imply causation at all time. If there is no any cause and effect relationship occurs between the dependent and independent variables and although it is showing high correlation coefficient or relationship between them, then this is a case of coincidence. In many cases we observed that there is no any causal relationship between the given two variables but there is a strong statistically relationship exist between these variables. Causation is when one of the events actually causes another for occurring. Two variables that are correlated often occur together but do not occur all times.

2) The spurious correlation is the false correlation between the two variables. This correlation shows that there is a correlation exists between the two variables but in reality there is no correlation between two variables. The spurious correlation is many times a result of the third variable or factor which is not present during the time of examination. This third variable or factor is called as the lurking variable. The lurking variable is not included as response or explanatory variable in the correlation study but it can be affect the interpretation of relationships between the response and explanatory or dependent and independent variables. The spurious correlation does not necessarily have a lurking variable and it could just be a coincidence.

3) Let us see the example of the spurious correlation between the two variables. It was observed that the worldwide non-commercial space launches correlates with the sociology doctorates awarded in the United States. It is seen that there is a high positive correlation exists between the given two variables. But actually there is no any actual relationship between these two variables. Also, there is no any lurking variable present for this spurious correlation. This correlation is simply a coincidence.

Related Questions

Navigate

• Browse (All)
• Browse T
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.