Skip to content

Data Analytics

Missing values

Missing values can mean unknown, unavailable, or not applicable rather than zero.

Naomi opens this week's delivery report. One order shows a delay of 0, and another has an empty delay cell. Why don't the 0 and the blank tell her the same thing?

OrderDelay (minutes)
D010
D02
D038
Start the lesson

4 minutes. No sign-up needed.

What you'll learn5 ideas
  • A recorded zero, like a delay of 0 minutes, means someone checked and found no delay.
  • A missing value means no usable value was recorded, so you can't tell what happened.
  • In the delivery example, replacing the missing delay with 0 treats an unknown delay as no delay and lowers the average from 6 to 4 minutes.
  • A value is unavailable when it existed but can't be obtained now, and not applicable when the field doesn't fit the case.
  • A missing-value marker, like an empty cell or -999, gets its meaning only from the file's guide.
Chapter 5: Data quality assessment6 lessons · 23 min

This lesson is part of the Data Analytics course.

Example plan3 lessons · 10 min

Missing valuesData quality assessment4 min
Duplicate recordsData quality assessment3 min
Inconsistent categoriesData quality assessment3 min

An example. Your plan follows your answers, so it can start later and end sooner.

Build my plan

Start free, no account needed.