Combining datasets responsibly

A join combines records by a specified matching rule. Names may be duplicated or inconsistent; a stable identifier can reduce accidental matches. Different units and definitions still require reconciliation. A successful technical join does not prove that the resulting rows describe comparable events or the intended population.

Check which records have no match and whether excluded records differ systematically. A programme can efficiently produce a misleading summary if its rules duplicate observations, mix units or discard an underrepresented group. Keep provenance, question scope and privacy constraints visible throughout the processing pipeline.

A trip-time table uses minutes and trip IDs. A second table has one distance per trip in kilometres, but C is missing.

1. Match on trip ID, not a traveller’s possibly repeated name.
2. A mean speed using only matched records must disclose that C was excluded.
3. If another distance source uses miles, convert to a documented common unit before comparing speeds.

Claim: A join that runs without errors guarantees a valid conclusion.

Correct matching, units, missingness and collection scope still determine whether the analysis is meaningful.

Explain why matching on a nonunique name can multiply rows.

Work with a partner. Make a prediction, compare your reasoning, then identify the rule that supports your answer.

Design a join audit with at least three reported quantities.

Does a larger joined file necessarily contain more independent observations?

Write your reasoning before comparing with a classmate. If your teacher assigned this task, submit it through My classes to save your work and receive feedback.