Question

b. Suppose you want to test a hypothesis about a proportion similar to what you've just done, but you don't know whether to use a two-tailed or a one-tailed test. You do know, however, that you have to have a Type I error rate of 0.05. You absent-mindedly take a look at your data results before forming $H_A$, and you notice $p > p_0$. So you decide to go with a one-tailed $H_A$. Why might this be considered cheating? Be very clear when explaining your reasons.

Name: b. Suppose you want to test a hypothesis about a proportion similar to what you've just done, but you don't know whether to use a two-tailed or a one-tailed test. You do know, however, that you have to have a Type I error rate of 0.05. You absent-mindedly take a look at your data results before forming HA, and you notice p > p0. So you decide to go with a one-tailed HA. Why might this be considered cheating? Be very clear when explaining your reasons.
Uploaded: 2023-06-15T18:57:56-08:00
Duration: 1 min 53 s
Channel: Adi S

          b. Suppose you want to test a hypothesis about a proportion similar to what you've just done, but you don't know whether to use a two-tailed or a one-tailed test. You do know, however, that you have to have a Type I error rate of 0.05. You absent-mindedly take a look at your data results before forming $H_A$, and you notice $p > p_0$. So you decide to go with a one-tailed $H_A$. Why might this be considered cheating? Be very clear when explaining your reasons.

b. Suppose you want to test a hypothesis about a proportion similar to what you've just done, but you don't know whether to use a two-tailed or a one-tailed test. You do know, however, that you have to have a Type I error rate of 0.05. You absent-mindedly take a look at your data results before forming HA, and you notice p > p0. So you decide to go with a one-tailed HA. Why might this be considered cheating? Be very clear when explaining your reasons.

Added by Sandra W.

Elementary Statistics a Step by Step Approach

Allan G. Bluman 9th Edition

Instant Answer

Solved by Expert Adi S

06/15/2023

Step 1

Step 1: The researcher is looking at the data before forming the alternative hypothesis. Show more…

Show all steps

Thanks for your feedback!

b. Suppose you want to test a hypothesis about a proportion similar to what you've just done, but you don't know whether to use a two-tailed or a one-tailed test. You do know, however, that you have to have a Type I error rate of 0.05. You absent-mindedly take a look at your data results before forming $H_A$, and you notice $p > p_0$. So you decide to go with a one-tailed $H_A$. Why might this be considered cheating? Be very clear when explaining your reasons.