Question

1. Why do we say that we "fail to reject" the null instead of saying "accept"?

2. When selecting a significance level, why should we consider the consequences of a type 1 error?

3. Based on the QQ plot, does the variable age pass the normality assumption?

4. If we set our alpha at 0.05, that means

a) we have a 5% chance of making a type 1 error b)we have a 5% chance of being correct c) 5% of our sample will follow the null hypothesis d) we allow 5% of our sample to not be random

5) Based on the QQ plot, does the varibale RCT pass the normailty assumption?

6. We want to show that reaction time in students is greater than 175 milliseconds. What is the correct hypothesis?

HO: m>175, HA: m<175

HO:m<175,HA:m>175

Ho:P=175,Ha:P<175

Ho:P<175,Ha:P>175

7. If we havve non-normal data, what should we do?

a) Use a nonparametric test b)ignore it and run the test as is c) throw out the data d) none of the above

8. Why is having a random sample important?

Answer 1

1. We say "fail to reject" the hypothesis because it is assumed to be true until we reject it. Hence, we cannot accept the null hypothesis.
2. Type 1 error is rejection of a true null hypothesis. Hence, when selecting a significance level we should know about the confidence level that we are rejecting the null hypothesis on.
3. The QQ plot is not shown.
4. Significance level means a) We have a 5% chance of making a type 1 error
5. QQ plot is not shown.
6. The correct hypothesis is:

Ho: m<175
Ha: m>175

7. If we have a non-normal data, we can a) use a non parametric test which are the alternatives to parametric tests

8. Random sample is important to remove certain bias from the experiment which would affect the validity