Exercise

# Standardized linear regression

**Standardization** is a process that involves subtracting an individual value by the population mean and then dividing by the population standard deviation. In linear regression this results in predictors that have a **mean of 0** and a **standard deviation of 1**.

Standardization is useful when one of the variables has a very large scale, as this may lead to regression coefficients of a very small order of magnitude. Executing a standardized linear regression in R is very similar to executing an unstandardized linear regression but involves the extra step of standardizing the variables by using the `scale()`

function.

Instructions

**100 XP**

- Create a simple
**standardized**linear regression: use the`impact`

dataset, which is already loaded in your workspace, with`sym2`

as the**dependent variable**and`ic2`

as the**independent variable**. Assign this model to the variable`model_1_z`

and look at the output. Print a summary of the model to the console. - Extract the
**R-Squared**value from the summary of this model and use this to calculate the**correlation coefficient**for this regression model. Assign these values to`r_square_1`

and`corr_coef_1`

respectively. - Now create a
**standardized**multiple linear regression with`ic2`

and`vismem2`

as the**independent variables**and`sym2`

as the**dependent variable**. Assign this model to the variable`model_2_z`

and look at the output. Print a summary of the model to the console. - Extract the
**R-Squared**value from the summary of this model and use this to calculate the**correlation coefficient**for this regression model. Assign these values to`r_square_2`

and`corr_coef_2`

respectively.