Tests for the Difference of Two Proportions

AP Statistics· difficulty 3/5

For two-sample inference about proportions.

Why pool for the test but not the CI?

A
$The null assumes p₁ = p₂; pooling gives best estimate of the common proportion.$
check_circle
B
$Pooling violates assumptions for CIs.$
C
$Pooling reduces df.$
D
$Unpooled is always more conservative.$

Explanation

Under H₀, the proportions are equal, so we use pooled p̂_p. The CI does not assume equality.

Want 10 more like this — adaptive to your weak spots?

Related questions