12 terms

# 5. hypothesis testing errors and power

###### PLAY
8 steps of hypothesis testing
1. determine purpose of test 2. state the hypotheses (and decide one-tailed vs. two-tailed test) 3. set the significance (alpha) level 4. compute the test statistic 5. determine the degrees of freedom of freedom (if necessary) 6. identify the critical value and find the rejection region 7. make a statistical decision 8. make a clinical/substantive conclusion.
null hypothesis
hypothesis being tested
alternative hypothesis
research hypothesis
when we write the null and alternative hypotheses always use the...parameter (remember we are testing a sample statistic to...info about the population)
population, infer
one tailed test
directional. interested in whether the mean of group 1 is significantly larger or smaller than the mean of group 2
two-tailed test
nondirectional. interested in whether the mean of group 1 is significantly different than the mean of group 2. looking for a difference without specifying whether the difference is higher or lower
significance level
level of willingness to make an error in rejecting the null hypothesis.
Type 1 error
the null hypothesis is rejected when it is true. the probability to make this kind of error equals alpha. is a false positive.
type 2 error
the null hypothesis is not rejected when it is false. probability to make this kind of error is beta. is a false negative.
Power
=1-beta. probability of rejecting the null hypothesis when it is false. the probability of making a correct decision to reject the null
factors that affect power
alpha level, effect size, sample size
effect size
simple way to quantify the difference between groups (i.e. magnitude). a measure of the effectiveness of the treatment.