This preview shows pages 1–2. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: I. Alpha: the critical region of a probability distribution We know that with a normal probability distribution, values far from the mean are less likely than values close to the mean. We need to establish a critical region in the ends of the distribution, which we define as unlikely. Usually, we decide that the outer 5% of the distribution is unlikely, and we call this value alpha here alpha =.05 (5% expressed as a proportion). II. Confidence intervals The goal of inferential statistics is to learn about the population, based on a sample statistic (also known as a point estimate). This involves uncertainty, so we can calculate a confidence interval around out point estimate, which limits our uncertainty. A confidence interval : is therefore an interval within which we are confident the actual population value lies. For example, when we hear about statistics such as polling percentages, they often come with a margin of error this is a confidence interval. Normally, we use a 95% Confidence Interval. This means we are 95% confident that our range includes the true population parameter, or that 95 times out of a hundred, this interval around our point estimate will include the population parameter. A 95% CI means we use an alpha of .05, with 5% risk of Type I error (erring by not including the population parameter)....
View Full
Document
 Spring '09
 Kupchick

Click to edit the document details