Biostatistics 4, 5, & 6
About this set
Created by:
lcaskidsn12s on September 9, 2010
Log in to favorite or report as inappropriate.
Order by
10 terms
Sundanese | English |
|---|---|
| Predictor |  The action or event that we believe may explain an outcome. Often the control vs. treatment group in clinical trials. |
| Response | The action or event that we believe is related to the predictor. The variable we are measuring to look for an effect in clinical trials. |
| Type 1 Error | Wrongly rejecting the null hypothesis, i.e. finding there is a significant difference when there really isn't one. α is the probability of committing this error. |
| Type 2 Error | Wrongly failing to reject the null hypothesis, i.e. finding no significant difference when there really is one. β is the probability of committing this error. |
| Power | The probability of NOT committing at Type 2 Error. 1-β |
| Conditional Probabilities | Probabilities only pertaining to certain cases. Ex: α only applies if the null is true. β only applies if the alternative is true. |
| Effect of Sample Size and Relative Risk Factor on Power | Increasing sample size increases power Increasing relative risk factor (expected?) increases power. |
| Familywise Error Rate | Probability that at least one of a family of variants (genetic in our example) is declared associated with an outcome (disease status) given that none are truly associated. Much larger than individual alpha level. Close to 1 for study with millions of variants. |
| Bonferoni Method | Divide your desired α (0.05 by convention) by the number of tests run so that the α of each individual test is much lower. α (overall) / n = α (each test) |
| Correlation | A measure of the extent to which two factors vary together, and thus of how well either factor predicts the other. |
First Time Here?
Welcome to Quizlet, a fun, free place to study. Try these flashcards, find others to study, or make your own.