I am capturing the time it takes for employees to work an item from a queue. I want to find the time it should take to work an item. Is there a formula in statistic that will eliminate data points which are out of the range when calculating the mean? I am using a two standard deviation. Hope this makes sense.How do I find the true average (mean) given a set of data?
If you want to eliminate outliers (data which is far away from the mean), you can do this be using a 5 number summary.
As a general rule of thumb, outliers are values which are:
a. Higher than 1.5(Q3-Q1) + Q3
b. Lower than Q1 - 1.5(Q3-Q1)
So you could use that rule of thumb, and get rid of points that are very far from the mean to create a new data set. With this data set you could recalculate a mean and standard deviaition.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment