S W 318 Study Guide - Midterm Guide: Control Variable, Confidence Interval, Chi-Squared Distribution
Document Summary
In class stats review: when we can assume the curve is normal, we know two things about it. We know all the values underneath the curve (z scores) and we know the mean. These are both important in order to compute statistics about a distribution. It is impossible to prove or disprove the research hypothesis, we can at best estimate the likelihood that it is true or false. Alpha needs to be greater than p: alpha is so small because it is the number predicting that we make a type one error which would be to reject a true null. We want there to little chance that we say there is a relationship when there is not: the value of alpha is associated to our confidence interval. We are 95 percent certain that what we are saying will test out (confidence interval) but there is a 5% or . 05 chance that we are wrong.
