Guide to select the right Hypothesis Test

What is the objective?

Choose one of the following options

Compare one sample set with a target
Compare two sample sets with each other
Compare more than two sample sets
Regression

Data Type for Input (X)

Data type for the input (X) may be Continuous or Mixed. Mixed Xs are not covered in this guide.

Continuous / Variable

Data Type for Output (Y)

Identify the data type for the Project Y

Continuous / Variable
Discrete / Attribute

Data Type for Output (Y)

Identify the data type for the Project Y

Continuous / Variable
Discrete / Attribute

Data Type for Input (X)

Identify the data type for the input (X)

Continuous / Variable
Discrete / Attribute

Number of Inputs (Xs)

One
More than one

Sample Size

Choose one of the following depending on the number of observations in each sample data set (or sample data sets)

Less than 10
Greater than or equal to 10

Sample Size

Choose one of the following depending on the number of observations in the sample data set

Less than 10
Greater than or equal to 10

Condition of Normal Distribution

Perform Anderson Darling Normality Test on both the sample data sets

Yes
No

Are external conditions same?

Determine if the hypothesis is done on same set of objects

Yes
No

Condition of Normal Distribution?

Perform Anderson Darling Normality Test

Yes
No

Condition of Normal Distribution?

Perform Anderson Darling Normality Test on all the sample data sets

Yes
No

Condition of Equality of Variances

Perform F Test to check if the sample variances are equal

Yes
No

Population Standard Deviation Known

Yes
No

Condition of Equality of Variances?

Perform Bartlett’s Test to check if the sample variances are equal

Yes
No

Does data have outliers?

Use Box Plots

Yes
No

Does data have outliers?

Use Box Plots

Yes
No

Hypothesis Test to use