Question

A researcher conducted a survey on a sample of 2000 college-students in order to examine the relationship between academic achievement and procrastination of studying. Using several indicators and plots of the descriptive statistics, found that in the data were many outlier observations. However, he decided to proceed with the calculation of the correlation coefficient considering the large sample of his research significantly reduces the effect of outliers.

Is this correct? Yes or No.

I believe is correct because the sample is large.

Answer 1

Any help?? The question is tricky because i do not know the real number of the outliers.. For example the sample size is large but if the half value are outliers then it will be a problem... What do you think?

Answer 2

Under the given conditions I think it is clear that a valid line of best fit could be constructed.

Answer 3

@kropot72 Thank you.. So we agree.. It is true..They did well..

But what means "many" outliers?? If they are too many it will not be a problem even if the sample size is large?

Answer 4

A very large number of outliers relative to the sample size would not prevent a valid line of best fit being drawn with an approximately equal number above and below the line.

Answer 5

Yeaa... I didnt think about it... You are right!!! The conclusion is that a large sample is always "powrful"... Thano you for your explanation!!!

Answer 6

You're welcome :)