Question

A study was conducted looking at the association between the number of polyps in the colon and mean grams of fiber consumed per day based on a standard food frequency questionnaire. The Pearson correlation coefficient was 0.9 with a p-value of 0.003 and the Spearman correlation coefficient was 0.28 with a p-value of 0.049. Which form of correlation would be more appropriate to report and why?

Answer 1

Answer:

C) The Spearman correlation results should be reported because at least one of the variables does not meet the distribution assumption required to use Pearson correlation.

Explanation:

The following multiple choice responses are provided to complete the question:

A) The Pearson correlation results should be reported because it shows a stronger correlation with a smaller p-value (more significant).

B) The Pearson correlation results should be reported because the two variables are normally distributed.

C) The Spearman correlation results should be reported because at least one of the variables does not meet the distribution assumption required to use Pearson correlation.

D) The Spearman correlation results should be reported because the p-value is closer to 0.0556.

Further Explanation:

A count variable is discrete because it consists of non-negative integers. The number of polyps variable is therefore a count variable and will most likely not be normally distributed.  Normality of variables is one of the assumptions required to use Pearson correlation, however, Spearman's correlation does not rest upon an assumption of normality. Therefore, the Spearman correlation would be more appropriate to report because at least one of the variables does not meet the distribution assumption required to use Pearson correlation.