This preview shows page 1. Sign up to view the full content.
Unformatted text preview: periment Variables: Response Explanatory Confounding Biases: Selection Non-Response Response Sample Surveys: Estimating a population proportion 95% conservative margin of error: 1 n Experimental Design Terms: Control Placebo Randomization Blinding Using results from a sample to estimate what is happening in a population UNCERTAINTY our `answer' may not be `right' but the process we used is fairly accurate: PROBABILITY Random Variables
Discrete Probability Model via table; E(X) and s.d.(X)
Value of X Probability Continuous Probability Model via density curve E(X) = balancing point
... ... x1 p1 x2 p2 x3 p3 n P ( X = k ) = p k (1 - p ) n - k k E ( X ) = = np s.d .( X ) = = np (1 - p ) Binomial(n...
View Full Document
- Fall '08