Test Statistics STAT 1010
Popular in Introduction to Statistics
Popular in Statistics
This 1 page Class Notes was uploaded by Susannah Gilmore on Monday April 11, 2016. The Class Notes belongs to STAT 1010 at University of Virginia taught by in Spring 2016. Since its upload, it has received 14 views. For similar materials see Introduction to Statistics in Statistics at University of Virginia.
Reviews for Test Statistics
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 04/11/16
Test Statistic measures how far the sample data diverge from what we would expect should the null hypothesis be true So, a test statistic with a high magnitude would show that the date is not consistent with H o P-values the probability (when H0 I true) that the test statistic takes a value that is as extreme or more extreme than what is observed A small p value means stronger evidence against Ho provided by the data, and you reject the null hypothesis Significance levels Denoted by ? Used as evidence for or against H 0 If a p value is smaller than ? , we reject H o If the p value is larger that? , we fail to rejectoH If a significance level is not given, always assume = .05 ? Rejection Region The range of values that if the test statistic falls in their range, there is enough evidence against the null hypothesis to make us reject it. Calculations Always remember that when you calculate the test statistics, excel will give the region to the left, so you may have to take the inverse. Steps to find z test statistic 1 Define the population mean and standard deviation (what the mean and standard deviation are supposed to be) 2 Define the sample mean and standard deviation 3 Define H0 and Ha. Ho: mean = population mean, Ha: mean ≠<> population mean 4 Do the formulat for z: 5 Steps to find the p value On excel, use Norm.s.dist(z test stastic,1) If the number is smaller than the significance level ? , then you would reject the null hypothesis Interpretation of alpha The significance level alpha is the long run proportion of times that the null hypothesis would be rejected if the null hypothesis was true after many repetitions of the experiment