Class Notes (835,094)
United States (324,041)
Statistics (75)
STAT 301 (55)
Lecture

# Checking for normality, transforming non-normal data, sample variability

3 Pages
91 Views

Department
Statistics
Course
STAT 301
Professor
Brett Hunter
Semester
Fall

Description
14 September Checking for Normality Check histogram We can check a normal distribution assumption with the use of a normal probability plot (Q-Q plot) What is it? A scatterplot of the (normal score, observation) pairs A normal score is what we expect the value of the observation to be if it truly came from a normal distribution. What are we looking for? A substantial linear pattern suggests that population normality is plausible Curvature casts doubt on a normal assumption Points falling outside the “bands” are warnings of possible non-normality (when bands are used) Transforming non-normal data If we have non-normal data, we can try to transform the data so that it is approximately normal. Then we can apply statistical methods used for normal data to the transformed data. The common transformations are 1 x Log (x) (base 10 or base e, usually base e unless otherwise stated) √ x So if the data is not normal, try a transformation to see if the transformed data is normal. If it is, we can then use the normal tables to find probabilities involvi
More Less

Related notes for STAT 301
Me

OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Join to view

OR

By registering, I agree to the Terms and Privacy Policies
Just a few more details

So we can recommend you notes for your school.