A blank cell may mean a question was not answered, a measurement was unavailable, or the field did not apply. Treating all of those cases as zero can change the meaning of a calculation.
Decide how missing values are represented and explain the choice. Check whether a summary excludes them or replaces them with another value. The reason information is missing can matter as much as the amount missing.
Look at a few absent values in their original context before choosing a rule. Preserve distinctions that affect the question, and avoid filling gaps merely to make the table look complete. Honest incompleteness can be more informative than a tidy but misleading number.
- Ask why the value is absent.
- Document how calculations handle missing values.
- Avoid silently replacing every blank with zero.
Imagine a practical setting.
Consider a temperature record with a gap during an unavailable reading. Substituting zero would describe a physical observation that never happened.
Before the next experiment.
A column name can hide an assumption. Check units, missing values, and the counting rule before combining observations that only appear to be alike.
Follow a related question
Identify unnecessary transfers or repeated work.
Build with a little more restraintIdentify the claims that materially matter.
A plausible answer can still be wrongKeep learning
Related background to continue exploring this subject.
NIST: statistical engineering NIST: the International System of Units
