## Abstract

This chapter reviews several common statistical tools. Chi-square is the appropriate inferential test to use to compare most data from two or more groups, when the data to be analyzed consist of two or more distinct outcomes that can be classified by rates, proportions, or frequencies. Fisher's Exact Test and Yates' Correction are utilized in certain cases when the Chi-square contingency table has two rows and two columns. Analysis of Variance (ANOVA) is used for statistical inference to determine whether it is likely that at least two groups differ from each other, when parametric numeric data from three or more groups are compared. When the researcher wishes to model outcomes and predict the value of dependent variable Y for any single or set of independent variables, regression techniques should be employed. Simple regression permits determination of a regression line that minimizes the squared deviation along the y-axis between each individual data point, and the value for the point that would be predicted by the regression line at any individual value of X. Various multiple regression techniques exist to permit the modeling of outcomes when considering the impact upon a dependent variable of two or more independent variables. The exact multiple regression test to choose depends upon the nature of the data and various judgments that should be discussed with a biostatistician facile with regression methods.

Original language | English |
---|---|

Title of host publication | Doing Research in Emergency and Acute Care |

Subtitle of host publication | Making Order Out of Chaos |

Publisher | John Wiley and Sons Ltd |

Pages | 213-222 |

Number of pages | 10 |

ISBN (Electronic) | 9781118643440 |

ISBN (Print) | 9781118643488 |

DOIs | |

State | Published - Oct 6 2015 |

## Keywords

- Analysis of variance
- Chi-square
- Multiple regression
- Simple regression