Tutorials
Home >

# Validation of Continuous Variables

Continuous variables are numeric variables; variables that would make sense for you to add, subtract, multiply or divide. You can count, order and measure continuous data. Some examples of continuous variables are:

• age
• height
• weight
• temperature
• the amount of sugar in an orange
• the time required to run a mile.

## Typical Validation Procedures

Typical validation procedures for continuous variables are checking the minimum value, the maximum value, the mean, and the median.  Let’s look at an example.

Example: Suppose you are working with hospital data and one of the variables you are interested in is a patient’s length of time living at their current residence.  So, the first thing you do is find the minimum and maximum values of length of time at their current address.

• Min:  .05 years
• Max: 47 years

This seems reasonable; some people will have lived at their current address for a very long time, and others may have just moved into their home.

## What If It Doesn't Seem Reasonable?

What if you ran the same check as the above example and instead of getting those results, you got the following results:

 Length at Residence (yrs) Lowest Values Highest Values -100 -100 -100 -100 -98 19 19 22 31 47

Does this result make sense? You might immediately be concerned that the lowest values are negative.  How can someone have lived at their current address for a negative amount of time?  It appears something may have happened to the data, and this should be looked into to see if it can be fixed, and the real data recovered.

rev. 29-Aug-2016