SOAN 3120 Lecture Notes - Lecture 14: Normal Distribution, Sampling Distribution, Statistical Hypothesis Testing
Document Summary
Hypothesis tests for two sample means with small samples. For s(cid:373)all sa(cid:373)ples (cid:894)(cid:272)o(cid:373)(cid:271)i(cid:374)ed (cid:374)"s < (cid:1005)(cid:1004)(cid:1004)(cid:895) is too u(cid:374)relia(cid:271)le a(cid:374) esti(cid:373)ator of o so we (cid:272)a(cid:374)(cid:374)ot use standard normal distribution. The formula for computing t(obtained) is: 1. Calculate the denominator (n1 + n2 2) or (n1n2: 2. The same five-step process is followed for hypothesis testing. However, the t table requires us to calculate degrees of freedom: because we have two samples, we have to take both into consideration when calculating df. The probability of rejecting h0 is a function of four independent factors: 1. Size of the difference: value of alpha, use of one vs two tailed test, the size of the sample (n) As long as we work with random samples, we must conduct tests of significance: however, significance is not the same thing as importance. When working work large samples, even small differences may be significant.