# Regression

#### Linear Regression Overview

Linear regression is commonly used to create a predictive model based on a given dataset. (For example, does a given set of independent variables do a good job of predicting a dependent variable?) Another common use is to quantify the strength of a relationship between a dependent variable and an independent variable(s).

#### Run Linear Regression

- Drag and drop desired variables into the corresponding dependent variable and independent variable fields (A)
- Click Run (B)

#### Linear Regression Example Output

In the example above, we are testing to see if our independent variables have a strong relationship with the total money spent. The coefficients (C) can be used to create a predictor model by making an equation. In our example, this would be:

Predicted total money spent = 75.04 + 0.05(Food Satisfaction Score) + 0.89(Time Spent at Airport) – 0.14(In-flight entertainment score)

In addition to this you may see various symbols next to your p-value(s) (D).

#### Linear Regression Significance Codes

#### Logistic/Ordered Logistic Regression Overview

Logistic regression is similar to linear regression, but is ideal in situations in which your dependent variable is binary or ordinal. Coefficients in the output table show coefficients that are the log of odds e^x. Coefficients will need to be transformed to get a probability that is easily interpreted.

#### Run Logistic/Ordered Logistic Regression

- Drag and drop desired variables into the corresponding dependent variable and independent variable fields (A)
- Click Run (B)