Assessing Normality (OPG)

Question

Topics: Assessing normality for a continuous (quantitative) variable.

Category: Probabilities and Distributions

JMP features: Analyze > Distribution, Normal Quantile Plot.

Open One-Page Guide (PDF)

Watch short video:

nurasbani · Answer

So, we do not need to calculate the residual before the normality test.

I ask this question because I found these steps (https://scholar.vt.edu/access/content/group/43c8db00-e78f-4dcd-826c-ac236fb59e24/STAT5606/AnovaJMP.htm:(

Analyze the residuals to check the assumption of normality

* In the Fit Y by X platform, select JMP: Save > Save Standardized
* The result appears in the data table as a new column/variable labeled "Y" Standardized by "X". These are the residuals.
* In the data table select JMP: Analyze > Distribution
* In the Analyze Distribution dialog window select the residuals column/variable.
* Study the histogram and stem-plot for symmetry.
* Note the skewness under JMP: Summary Statistics.
* Perform a Shapiro-Wilks test for normality by selecting JMP: Continuous Fit > Normal.
* Under Fitted Normal select JMP: Goodness of Fit.

Btw, now it's clear for me.

Thanks

mia_stephens · Answer

Glad to hear it, thanks!

unknown · Answer

I found this article easy to apply to my real life data, thank you.