Two sided alternative hypothesis The parameter is different from the null value

# Two sided alternative hypothesis the parameter is

This preview shows page 4 - 7 out of 7 pages.

Two-sided alternative hypothesis The parameter is different from the null value (it could be either smaller or larger). Type I error Occurs if H0 is rejected when H0 is true. Type II error Occurs if H0 is not rejected when H0 is STATS Page 4 Type II error Occurs if H0 is not rejected when H0 is false. Does a parameter have a changing value or a fixed value? Fixed, unchanging value Do we know the parameter? usually we do not know the parameter because we cannot measure every unit in the population Three examples of what a parameter can be 1. A summary characteristic of a population 2. A random situation 3. A comparison of different populations If we cannout find out the numerical value of a parameter, how do we use it? We use statistical methods to make a good guess at the parameter Define statistic, or sample statistic a number computed from a sample of values taken from a larger population When is a sample estimate or estimate used? when the statistic is used to estimate the unknown value of a population parameter Can multiple samples of a population vary? yes, the value of a sample statistic can change from sample to sample What is the procedure we use for making conclusions about population parameters on the basis of sample statistics Statistical inference What is a confidence interval an interval in which the researcher is fairly sure will cover the true, unknown value of the parameter, such as "between 53% and 59%" What is hypothesis testing used for? to reject a hypothesis about a population Which notion do you want to reject when you are trying to establish statistical significance? You want to reject the hypothesis that chance alone can explain the sample results What value is necessary in hypothesis testing, and what does it mean if this value is true? A null value is necessary, and if the null value is true, it means that nothing of interest is happening and that chance alone can explain the sample results What would the null value be for a weight loss clinic? that the average weight loss for the population of clinic patrons is 0. The clinic wants to show that the actual value is not zero, that average weight less is greater than 0. Two basic types of variables? Categorical and quantitative categorical -- man, woman, dog, cat quantitative, 10 pounds, 2 weeks The big five scenarios for: Population parameters Categorical Variables one population proportion = p difference in two population proportions = p - p₂ Big Five scenarios sample statisic ne popula on p opo on o p o a ility = p STATS Page 5 