A model cannot repair an incorrect unit, duplicated key, or impossible value. Validation should fail near the cause, before plausible-looking output is produced.

readr, readxl, dplyr, and janitor are recommended; the live example stays in base R.

validation = schema + keys + ranges + missingness + provenance

1. Step 1Define every variable and unit.
2. Step 2Check assumptions and information timing.
3. Step 3Calculate, verify, and interpret in context.

Variable

A constructed five-row cleaning example.

Model choice

Validation turns substantive expectations into executable checks.

Calculation

Two inflation values are missing; complete inflation rows leave three observations; 99 is flagged as NA pending source verification.

Interpretation

Replacing a suspect value with missingness is safer than inventing a correction.

Reasonableness check

The threshold above 20% is a domain rule for this teaching context, not a universal law.

Replacing a suspect value with missingness is safer than inventing a correction. Limitation: The threshold above 20% is a domain rule for this teaching context, not a universal law.

The constructed table contains two missing inflation cells and one impossible rate.

Run the self-contained starter code in the protected STATLAB R runner. The code prints an auditable result and avoids network or unrestricted file access.

Compare your result with the definition, units, and model assumptions—not only with a target number.

Complete these without looking at the selected solutions. More than one defensible program may exist.

These are compact solution routes. Confirm dimensions, units, and any changed modelling choices.

Treat cleaning as an auditable chain of schema, type, missingness, range, and key checks.

Name the assumption, evidence you would seek, and how the recommendation might change.

Reach 70% to complete the chapter. Explanations appear after submission.