ABSTRACT
Accurately transcribing data from the CRF to the database is essential. Errors in transcription are usually due to typographical errors (typos) or illegibility of the values on the CRF. Companies aim to reduce transcription errors using one of these methods:
• Double data entry with third-party reconciliation of differences • Double data entry with a second person resolving differences • OCR as —rst entry with one or more subsequent entry or review passes • Single entry with extensive data checking
Even after transcription errors are reduced to a minimum, there remains some variation in what an “accurate” transcription means. Does it mean that the data is an exact duplicate of the values found on the CRF, or are there deviations or variations permitted?
