The Randomization Module within REDCap allows researchers to randomly assign participants to specific groups.
Table of Contents
Enabling the Randomization Module
This module can be turned on by selecting the "Enable the randomization module" checkbox during project creation or after the fact by clicking the "Project Setup" tab and then pressing the "Modify Project Settings" push button.
Setting-up a Randomization Model
A randomization model defines how the subjects in the study are randomized. To start the process, click on the "Set up randomization" button under the "Project Setup" tab.
Randomization Setup
The randomization model is defined via the "Setup" tab and consists of three main steps:
- Define your randomization model
- Download template allocation tables (as Excel/CSV files)
- Upload your allocation table (CSV file)
Define your Randomization Model
The first main step in the setup process is to define your randomization model. In this step, you will define the type of randomization and how the randomization is applied to the subjects. There are two randomization types:
- Use stratified randomization?
- Randomize by group/site?
Use Stratified Randomization?
Stratified randomization ensures that different groups are balanced. The balance is specified in the allocation table. Increasing the number of stratification variables will lead to fewer subjects per stratum.
Randomize by Group/Site?
If this is a multiple site study, this option allows you to stratify the randomization by each group. When this option is checked, you are given the opportunity to select randomization by data access groups or by a REDCap drop-down list of the sites.
Choose your Randomization Field
This is the field that will receive the randomization group designation.
Download template allocation tables (as Excel/CSV files)
The second main step in the randomization setup is to download the allocation table template.
REDCap creates several example files to download that you may use as a structure when creating your own randomization table. For this example, I have stratified a hair color field that has 7 different values and I wish to allocate the participants into 5 groups. My allocation template downloaded from REDCap looks similar to the following:
| random_group | hair_color |
|---|
| 1 | 1 |
| 2 | 1 |
| 3 | 1 |
| 4 | 1 |
| 5 | 1 |
| 1 | 2 |
| 2 | 2 |
| 3 | 2 |
| 4 | 2 |
| 5 | 2 |
| 1 | 3 |
| 2 | 3 |
| 3 | 3 |
| 4 | 3 |
| 5 | 3 |
| ... | ... |
REDCap will randomize the participants according to this table. For example, the first participant that has a hair color of "Light Blond" (corresponding to the value of one) will be allocated to random group one. The second person with "Light Blond" will be allocated to random group two, etc. Note: Make sure the number of entries in the randomization table is large enough to accommodate all participants in each category.
In order to make the allocation schedule more difficult to predict, a temporary column can be added to the spreadsheet for randomizing the order. The Excel function "func()" creates a random value between 0 and 1. This number can be used as an easy way to populate a third column in the spreadsheet with a unique random value. If the spreadsheet is sorted by this column, the group and hair color values will be randomized. For example, after sorting, the spreadsheet may look similar to the following:
| random_group | hair_color | Random Value |
|---|
| 4 | 4 | 0.01050529 |
| 3 | 1 | 0.01424553 |
| 2 | 5 | 0.02642509 |
| 2 | 4 | 0.03923845 |
| 3 | 5 | 0.03996002 |
| 5 | 3 | 0.04541500 |
| 5 | 6 | 0.04785563 |
| 4 | 2 | 0.05979234 |
| 5 | 5 | 0.06626952 |
| 1 | 4 | 0.06678536 |
| 2 | 7 | 0.06803015 |
| 5 | 2 | 0.06874587 |
| 3 | 2 | 0.07215219 |
| 3 | 3 | 0.07640087 |
| 2 | 1 | 0.07756467 |
| ... | ... | ... |
After sorting the table by the random value, the column with the rand() function should be removed before to saving the spreadsheet as a CSV file and prior to uploading it as the allocation table into REDCap.
Upload your Allocation Table (CSV file)
The third and final main step in the randomization setup is to upload the allocation table.
Two allocation tables will eventually need to be defined, one for the development stage and a different one for the production stage.
Randomization User Rights
User Rights need to be granted to an individual to enable randomization privileges.
The fields that have been specified for randomization will now display a "Randomize" button in the data entry screen. This will allow you to generate a random value according to the randomization allocation table.
Pressing the "Randomize" button gives you an opportunity to double check the required values that will impact the definition of the resulting randomization group.
After updating or providing missing values, a dialog of the random group that was assigned is displayed.
In addition, after the value of the random group is assigned, the random group field becomes read-only and the value can not be changed.
The strata fields you have chosen also become read-only and can not be changed.
A listing of the records that have been randomized is maintained. To examine those records, click on the Randomization link in the Applications section of the Navigation Pane.
A table displaying the allocated records and the random group assigned are shown in a table similar to one below:
At this point the random groups are populated and ready to use.
