Data Entry, Retrieval, and Information Accuracy
This domain tests how accurately information is captured, validated, stored, retrieved, and corrected. Expect questions on scan-versus-manual entry, validation rules, units, timestamps, audit trails, and retrieval methods.
Core rules and concepts
- Validate at the point of entry: required fields, format, range, check digit, and cross-field consistency, for example quantity against unit of measure.
- Prefer scanning over manual keying for identifiers. When manual entry is unavoidable, read back critical characters and use check digits where present.
- Enter corrections as new, linked records with a reason code; do not overwrite the original. The audit trail must show who changed what and when.
- Retrieval depends on stable keys: asset ID, work order number, serial number, or UUID. Search filters should be narrow and the result sorted by timestamp, with the latest validated record identified.
- Units and decimal separators are data. Mixing inches and millimetres or comma and point decimals creates silent errors that pass format checks.
Worked example
A technician must enter serial number "0O12-7B" by hand; the characters zero and capital O look alike. Correct approach: scan the label if possible. If not, enter it, then read it back character by character against the label, and confirm the check digit if the format provides one. Reasoning: a single substitution creates a record that looks valid but links to the wrong asset. The correction, if needed, is a new linked entry with a reason code, not a silent edit.
Traps that cost marks
- Overwriting the original record and destroying the audit trail.
- Assuming the first search result is the latest record.
- Mixing units or decimal conventions.
- Trusting a scan without confirming the field mapped correctly.
- Truncating long identifiers to fit a field.