T-Test of a Probability %

From Q
Jump to: navigation, search

Where \hat{a} is a probability % from a ranking question, \mu is the mean probability %, n is the effective number of observations (note that this is different to the effective sample size where repeated observations exist for each respondent) and se_{a} is the standard error of \hat{a}, the test statistic is:

t=\frac{\hat{a}-\mu}{se_{a}},

where:

p = 2\Pr(t_{n-k+1} \ge |t|),
k is the number of alternatives.