Question

Nesta is making a scatterplot of the digit spans (how many numbers you can remember and repeat back) for his psychology class, with the spans for digits the students hear on one axis and the span for digits the students read on the other. The association is strong, but he notices that one student has a visual digit span that is twice as long as anyone else. What statistical validity question is he raising?

Is the correlation statistically significant?

Is there a restriction of range?

Is the relationship curvilinear?

Could outliers be affecting the relationship?

Answer 1

Answer:

Could outliers be affecting the relationship?

Explanation:

In most practical circumstances an influence outlier decreases the value of a correlation coefficient and weakens the regression relationship, but it's also possible that in some circumstances an outlier may increase a correlation value and improve regression.

Notice that in the scenario it is mentioned that ''he notices that one student has a visual digit span that is twice as long as anyone else.'' , this will raise the question as to ''what is increasing the value of the correlation coefficient  (the span) between the 'digits the students hear' AND 'the digits the student read'