Question

1,

How do Ludwig and Phillips interpret the failure of the Head Start evaluation to reject the null hypothesis that the program has no effect? Pick One

		They claim that the confidence limits on the impacts is too wide to draw any useful conclusions.
		They claim that the program probably has no impact for the average client, but that it could still be vitally important for key subpopulations.
		They conclude that Head Start does not work.
		They point out that the program is still very likely to pass a cost benefit test, even if you can't be statistically confident that there is an impact.

2,

What is the general spirit of statistical hypothesis testing? Pick One

		Can we show that it is unreasonable to believe the opposite of our research hypothesis?
		Can we prove that our research hypothesis is definitely correct?
		Can we find the parameter values that are most likely to be true?
		Can we show that one situation is more likely to be true than another situation?

3,

When you test the difference of two means, how would you state the null hypothesis ( µ_e and µ_c are two means you want to compare) Pick One

		H0: µe - µc = 0
		H0: µe - µc ≠ 0
		H0: µe = 0
		H0: µc = 0
		none of the above

Answer 1

1)

when we fail to reject the null hypothesis

it means there is not sufficient evidence to reject the claim(if claim in null hypothesis)

or there is not sufficient evidence to support the claim(if claim in alternate hypothesis)

option B)

They claim that the program probably has no impact for the average client, but that it could still be vitally important for key subpopulations.

2)

option A)

Can we show that it is unreasonable to believe the opposite of our research hypothesis?

3)

option A) is correct

H0: µe - µc = 0