Z-Test in Medical Studies

Most homework questions can be answered by a visit to the internet.

Z-test - Wikipedia, the free encyclopedia

Yes Lyn, I have been to the wikipedia site and will continue searching, until I get my mind around it. Thank you .

Z Testing can show up in different forms, but all are based on the Normal (bell) Curve. However, to answer the question you asked:

First, you need to determine your required Confidence. Typically, this is either 90% or 95%. Let's use 95%. What this means is that your Null Hypothesis says that the calculated z statistic will be in the 95% group and, if the z statistic is outside of that 95%, then you reject the Null Hypothesis.

2-tailed test:

95% of the Normal population lie between -1.96 and +1.96 standard deviations from the mean. If your calculated z statistic is between -1.96 and +1.96, or -1.96 ≤ z ≤ +1.96, then you fail to reject the Null Hypothesis. Otherwise, the z is in the other 5% and you reject the Null Hypothesis:

1-tailed test:

95% of the Normal population lies on one side of a value, either above or below, depending on how your problem is defined. If 95% must be above, then your calculated z statistic must be greater than -1.64 standard deviations below the mean. (z ≥ -1.64) If 95% must be below, then your calculated z statistic must be less than +1.64 standard deviations above the mean. (z ≤ +1.64) If the z statistic does not fit the given tail requirement, then reject the Null Hypothesis.

A good on-line statistics reference is the NIST handbook:

http://www.itl.nist.gov/div898/handbook/index.htm

Your reply is most helpful. I'm starting to make some head way. Thank you so much.