On this page
gpp_good
Pre-Collection Validation
Preventing bad data from entering the system at the source.
The most effective data management strategy is preventing errors before they occur. By designing smart, validated forms, you minimize the amount of post-collection data cleaning required.
1. Enforcing Constraints and Validation logic
When building your survey in the Form Builder or via XLSForm, vigorously apply validation criteria:
- Numerical Boundaries: Restrict integer or decimal questions to realistic ranges (e.g., household size must be between 1 and 30).
- Text Length Limits: Use regular expressions (regex) or character limits to ensure phone numbers or national IDs follow exact formats.
- Date Constraints: Prevent enumerators from selecting future dates for events that happened in the past (e.g., Date of Birth).
2. Using Skip Logic (Relevance)
Do not rely on enumerators to manually skip questions based on previous answers.
- Build Skip Logic (or relevance conditions) directly into the form. If a respondent is male, the form should automatically hide all questions related to pregnancy.
- This reduces survey fatigue, prevents contradictory data, and speeds up the interview process.
3. Mandatory Questions
Ensure that critical survey questions cannot be accidentally skipped.
- Mark essential identifiers (like Respondent Name, GPS Location, or Consent confirmation) as Required.
- The enumerator will not be able to finalize the form on the mobile app or web browser until all required fields are answered.
4. Testing the Instrument
Before deploying a form to a live enumerator team, thoroughly field-test the instrument.
- Create “dummy” submissions that intentionally try to break your validation rules.
- Ensure that the error messages displayed to the enumerator are clear and helpful, guiding them to provide the correct format.
Was this page helpful?