Data cleaning

Data cleaning enables you to selectively remove or restore records from appearing in the data download and analysis. This process is valuable for various reasons, such as removing test data or excluding records that appear to be from cheaters (e.g., completed surveys in unrealistically short times).

Records can be removed using either system-generated Record IDs or author-defined Unique pass-in IDs. Values for either variable can be found in the all data download.

Data cleaning can be accomplished by:

Pasting a list of IDs that you want to remove from the dataset, or
Pasting a list of IDs that you want to keep, thereby removing all others.

Follow these steps to remove a list of records:

Select Remove records from the dropdown menu.
Select Record IDs or Unique pass-in IDs.
Paste or type a list of IDs. Each ID should be on its own line.
Click Save. All the records entered are removed.
After performing data cleaning, the section will display the number of records that have been removed. To view all removed records, download the spreadsheet. You can remove additional records by clicking the Define records button again.

Alternatively, if you have a list of IDs that you would like to keep (while removing the other records), follow these steps:

Select Keep records from the dropdown menu.
Select Record IDs or Unique pass-in IDs.
Paste or type a list of IDs. Each ID should be on its own line.
Click Save.
After performing data cleaning, the section will display the number of records that have been removed. To view all removed records, download the spreadsheet. You can remove additional records by clicking the Define records button again.

To restore the removed records, simply click the Restore all records button and confirm in the dialog that appears.

Discover Help Guide

Data cleaning

Introduction

Removing records

Remove method

Keep method

Restore records