Hypothesis testing

I have an assignment due in a little while and I would like to double check my answers... it's basically true and false of m/c

1. The probability of committing a type 2 error increases as alpha increases:

2. If type 1 error for a given test is to be decreased, then for a fixed sample n, the type 2 error test will increase.

3. If a decision maker is concerned that the chance of making type 2 error is too large, one option that will help reduce the risk is to reduce the significance level:
False (number of samples must be increased)

4. When the decision maker has control over the null and alternative hypotheses, the alternative hypothesis should be the "research" hypothesis.
5. It is possible to appropriately test a null and alternative hypotheses using the test statistic approach and reach a different conclusion that would be reached is the p-value approach was used.

6. When a hypothesis test is two-tailed, the level of significance must be twice as large as when the test is one tailed.

Please let me know if you think that some of my answers are wrong...thank you!



TS Contributor
I think #4 is true. The null hypothesis is usually a statement about the absence of an effect or difference or correlation, etc. Research usually tries to show the presence of some difference, effect, or correlation.