Introduction
This article describes how to scan through a data file looking for data that likely need to be examined or corrected, and creates tables with cells that may be of interest highlighted.
Requirements
A Displayr document with a data set.
Please note these steps require a Displayr license.
Method
- Go to any of the following menu options:
- Anything > Report > Tables for Data Checking
- Table > Report > Tables for Data Checking
- A new folder will appear in the Reports tree named "Tables For Data Checking" and will contain two folders:
- Data with values needing checking - outputs that show tables with cells that should be reviewed marked in yellow.
-
Data with varying Base n - Tables where the Base n varies between cells.
Technical details:
This script examines all tables containing non-Text and non-Date data and:
- Identifies all tables containing cells with sample sizes of less than 30.
- Identifies all tables containing Don't Know responses.
- Identifies all tables containing blank labels.
- Identifies all tables containing options chosen by 5 or fewer people, or, less than 1% of respondents
- Identifies all empty tables
- Identifies numeric data containing outliers, where a variable contains values that are more than (less than) 3 standard deviations above (below) the mean
- Identifies questions where the Base n is not the same for all variables
Identified tables have the relevant cells highlighted in yellow. Data containing outliers will be shown as histograms.