Hypothesis of paired t-test

• Null hypothesis: NO significant difference between the means of two groups.
• Alternate hypothesis: Significant difference between two population mean and is caused due to sampling error or chance.

Assumptions of paired t-test

• Dependent variable is normally distributed.
• Observations are sampled independently.
• Dependent variable measured on ratio and interval.
• Independent variable consists of two related groups.

When to use paired t-test?

• When the same item or group is tested twice.
• Also called repeated measure t-test.
• Eg:

i) Before and after effect of a pharmaceutical treatment.

ii) Body temperature using two different thermometers.