- Published
- 8 October 2026
- Last reviewed
- 8 October 2026
How to use these FAQs
Investigate missing and repeated records before changing data or publishing a report.
Select a question to expand its answer. The first answer is open initially; each answer can be opened or closed independently. These are original RDE practical explanations. Any example described as fictional is a teaching scenario.
Common questions
Should I delete every repeated row?
No. Determine what one row represents and whether the repeated values describe the same event. Two purchases by the same customer may be valid records. A fictional attendance export could also repeat one session because two systems were joined incorrectly. Identify the relevant key and investigate the source before applying a rule that removes useful information.
What does a blank value mean?
It may mean unknown, not applicable, not collected or a technical problem. Check the data definition and collection process before treating it as zero. For a fictional satisfaction survey, an unanswered question differs from a rating of zero. Keep the distinction visible in the analysis so that a missing response does not become an invented observation.
Can I fill missing values with an average?
Only when the method is appropriate to the analysis and its assumptions are understood. Do not apply it automatically to identifiers, categories or every numerical field. Record which values were replaced and why, preserve the original data and examine the effect on the result. If the method could materially change a decision, obtain suitable analytical review.
How do I investigate a sudden drop in records?
Check the reporting period, filters, source refresh, joins and collection changes before interpreting it as a real change in activity. Compare a few known cases with the source. A fictional attendance drop might result from a new export filter. Record the checks and any unresolved uncertainty instead of immediately presenting a story about behaviour.
How should missing records affect a percentage?
Define the numerator and denominator and explain which records are included. A completion percentage using all invited people differs from one using only those who responded. Choose the definition that answers the question and label it clearly. If the denominator is incomplete or uncertain, state that limitation before a reader relies on the figure.
What should accompany the final report?
Include the relevant period, source, definitions and important data-quality limits. Explain cleaning decisions and retain a reproducible record of the steps. Keep sensitive source data in an approved location. A clear report distinguishes observed results from estimates and helps another analyst understand what changed, rather than hiding correction work behind a polished chart.
Put one answer into practice
Choose the question closest to your current task. Write one action in your own notes, identify the information or permission you need and set a review point. Use the related resources below to develop the task further. No response is submitted or stored by this page.
Sources and scope
The sources below provide relevant context and terminology. The examples, practical suggestions and wording are original RDE teaching material. Apply your organisation’s policies and the guidance relevant to your situation. Reviewed 8 October 2026. Author: Jefferson Otonbara Imgbi.
Cover image: AI-generated illustrative scene created specifically for this resource.