Question

1. Insofar as we must generalize from a sample to a population, the observed difference between the sample mean and the hypothesized population mean a) can't be interpreted at face value. b) might be due to variability or chance. c) might be real. d) is described by all of the above

2. The advantage of a one-tailed test is that it increases the likelihood of detecting a a) false null hypothesis. b) false null hypothesis in the direction of concern. c) true null hypothesis. d) true null hypothesis in the direction of concern.

3. When the rejection of a true null hypothesis has horrendous consequences, use a level of significance equal to a) .10 b) .05 c) .01 d) .001

4. If the null hypothesis is really false and we reject the hypothesis, we have made a a) mistake. b) correct decision. c) type I error. d) type II error.

Answer 1

1. d) is described by all of above

Note we must standardize the observed difference using std deviation to use it for inference so a) is correct. Then difference between sample mean and population mean can be due to chance so b) is correct. Finally, if std.dev is close to 1 then this difference of emans is actually useful without dividing by std dev.

2. Ans : b) false .... in direction of concern

This is so because for one-sided hypothesis we will have larger rejection area. compared to 2-sided test.

3. Ans : d) 0.001

if rejection of null is to be avoided then probability of type I error i.e. significance level must be kept as small as possible.

4. Ans : b) correct decision

When null is false, if we reject it then it is correct decision.