Cut-offs are values that separate your classes.

Modified on Mon, 16 Nov, 2020 at 2:10 PM


Cut-offs allow you to transform a numeric variable into a categorical variable, separating your numeric data into two or more groups of values.

In practice

You can be interested by separating the values of the Body Mass Index (BMI) of your population into two groups: "less than 30" and "more or equal to 30". This could allow to compare the proportion of patients suffering from obesity with the proportion of patients not suffering from obesity, instead of comparing BMI values.
The transformation of numeric variables into categorical variables is not usually recommended because it could lower the power of your test. Nevertheless, it could make more sens to get a categorical variable than studying a numeric variable in some clinical contexts.

How to use it on EasyMedStat for testing two variables?

  1. Create a Numeric variable and a List or a Yes-No variable 
  2. Go to Statistics > Test 2 variables
  3. Select a Numeric variable you would like to compare between and a List or Yes-No variable to create groups of patients 
  4. Click on "Which tests can I do ?" 
  5. Click on "Find a statistical association between the 2 variables (Fisher, Chi2)"
  6. Under the Numeric variable, click on the "How many groups would you like to create ?" field and choose the desired numbers of groups

How to use it on EasyMedStat for multivariate analysis?

On multivariate analysis, the transformation of a dependent numeric variable into a categorical variable with two modalities (binary variable) is the only one allowed.  

  1. Create Numeric variables and/or List or Yes-No variables 
  2. Go to Statistics > Multivariate analysis
  3. Select a Numeric variable you would like to study as a dependent variable
  4. Click on "Define a cut-off"
  5. Chose one of the options proposed on the "choose" tab
  6. Chose the value(s) you want to use to define your cut-off

See also

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article