Knowledgebase

Cleaning Data Before Analysis Print

  • dataanalytics, data, errors, guide, howto, solution, zillionkinghost, hosting
  • 0

Preparation.

WHAT TO CHECK FIRST

Duplicates Missing values Obvious errors Inconsistent categories Values outside plausible ranges

WHAT DUPLICATES CAUSE

Overstated totals.

WHAT MISSING VALUES CAUSE

Understated totals, or excluded records you did not notice.

WHAT TO DO ABOUT MISSING VALUES

Establish whether the value is genuinely absent or simply unrecorded.

Those need different treatment.

WHAT TO DO ABOUT OUTLIERS

Investigate before removing.

An unusual value may be an error or the most interesting record you have.

WHAT NEVER TO DO

Delete records because they are inconvenient Alter figures to produce a tidier result

WHAT TO RECORD

What you cleaned and why.

WHY

So the analysis can be reproduced, and so you can defend it.

WHAT TO KEEP

The original, untouched.

Work on a copy, always.

HOW LONG THIS TAKES

Longer than the analysis itself. That is normal.


Was this answer helpful?
Back

Are you happy with your experience? Leave us a review on Trustpilot.


Trustpilot