Question 836:

1

No answer provided yet.A Type II error means failing to reject the Null Hypothesis when it is in fact false. For example, perhaps you want to compare two groups of employees on some measure of merit. The Null hypothesis is that there is no difference between the groups. Let's imagine that in fact there really is a difference (you'd find it if you measured all employees).

If you took a sample of employees and the difference you observed in the sample didn't give you a significant p-value (the one management said to use) then you can't reject the Null. But since we know there is a difference (in this case we're pretending we know) and you didn't reject the Null, you committed a Type II error.  So what can you do?

You can increase your alpha cut off to say .05 to .10, which is the same as reducing the level of significance from 95% to 90%, but you could also increase your sample size. So of the choices given, a and c would work (and b is phrased a bit weird but it's basically the same thing as a).  Typically the best thing to do reduce a chance of a Type II error is increase  your sample size. If it is not practical to increase your sample size then you're stuck with a and b. D is definitely not an option.