This preview shows page 1. Sign up to view the full content.
Unformatted text preview: n exam will improve students’ performance on the exam. She predicts that students who access review materials will do berer than students who do not access review materials. 10 9/19/13 As researchers, we open have a predicIon about a speciﬁc direcIon of eﬀects. So, you would expect that we would open be using one
tailed tests… In prac(ce, however, we almost always use two
tailed tests. If we happen to ﬁnd the opposite of what was predicted, then the two
tailed test allows us to accept our data as meaningful and staIsIcally signiﬁcant, and report on the ﬁndings. If we used a one
tailed test, and then found the opposite, we cannot reject the null hypothesis. (womp womp) Unless the author speciﬁcally states otherwise, you can assume that psychology research papers used two
tailed tests. Divide up your signiﬁcance between the tails. Instead of .05 in one tail, you have .025 in both tails. What does this do to your criIcal values? Divide up your signiﬁcance between the tails. Instead of .05 in one tail, you have .025 in both tails. What does this do to your criIcal values? 11 9/19/13 We have been focusing on a .05 level of signiﬁcance (which is .025 in each tail in a two
tailed test)… What about a more stringent level, sekng our p
value < .01? A researcher is interested in the eﬀect of environmental condiIons on the quality of sleep. In parIcular, the research wants to examine whether living in a polluted region has an eﬀect on individuals’ amount of sleep. A randomly selected person from the polluted region sleeps an average of 10.2 hours per night. Using a two
tailed test at the .05 signiﬁcance lev...
View Full
Document
 Winter '13
 bernard
 Psychology

Click to edit the document details