11.1Hypothesis TestingDependent Groups

Before/after or matching link

Many statistical tests involve paired data samples that help us draw conclusions abt the difference of 2 groups

This freqly happens in before/after sits in which the same item is measured prior to and after a treatment

Ex= guided exercize 1 pg 453

Sometimes there is a matching link, a natural match for collecting the pairs even without a before/after feel

Feet Data

So What?

Advantages=

Using matched data can often reduce the danger of extra factors except for the 1 characteristic we want to measure (blocking)

Pairing data reduces variance, thus increasing the accuracy of our results.

See diamond middle of pg 453

d bar is the mean difference betw the paired/matched data…it’s a sample mean…an x bar

S sub d is the sample standard deviation

The remainder of this hypothesis test is already familiar to you, if you like t distributions