If we retain the null hypothesis about the difference between two population means using the .05 criterion of significance, what is the probability that we have made a Type I error?

When participants can be matched, the matched t test is preferable to the t test with independent samples because it:

The mean of 16 subjects on an attitude scale after receiving a persuasive message is 38.4. The mean for the same subjects before receiving the message was 32.6. If the appropriate standard error of the difference is 9.7 and the critical value needed to reject the null hypothesis is 2.145, the researcher should conclude that:

the message makes no change in peoples attitudes at all.

the message changes peoples attitudes.

there is insufficient evidence to say that the message changes peoples attitudes.

none of the above.

The mean of 100 subjects on an attitude scale after receiving a persuasive message is 38.4. The mean for the same subjects before receiving the message was 32.6. If the standard error of the mean of the differences is 1.5 and the critical value needed to reject the null hypothesis is 1.99, the researcher should conclude that:

the message does not change peoples attitudes.

the message changes peoples attitudes.

there is insufficient evidence to say that the message changes peoples attitudes.

none of the above

If a matched t test is carried out and the 95% confidence interval is computed, for which of the following intervals should the researcher retain the null hypothesis of zero change?

11.6 to .79

23.1 to 6.8

Both of the above

Neither of the above

For the matched t test, the degrees of freedom are:

	the number of scores.
	the number of pairs.
	the number of scores minus 1.
	the number of pairs minus 1.