40.7 percent: making survey data quality provable

In short

In the GDQ Data Quality Benchmarking, Wave 2, German research agencies remove 40.7 percent of their records before and during the survey, sample suppliers 10.1 percent. The report calls this a striking divide and warns in the same breath against reading removal rates as a league table. What the number shows above all is that every agency cleans differently and hardly any can back its rate with rules, thresholds and case counts, although the quality standards of ADM, ASI, BVM and DGOF have required exactly this documentation since 2023. This whitepaper explains the benchmark figures with their definitions, presents the GDQ code frame with its 18 removal reasons, gives checks, default thresholds and the known false positives for seven of them, and closes with a one-page cleaning protocol for the method report and twelve self-audit questions.

Who it is for

Institutes & agencies · Corporate insights teams

Format

PDF, 14 pages, with worksheets and checklists

Contents 8 chapters

1. 01 The benchmark: what GDQ Wave 2 measured
2. 02 Why removal rates cannot be explained today
3. 03 The GDQ code frame: 18 reasons, one language
4. 04 The rulebook: seven codes you can check
5. 05 Flag instead of delete
6. 06 The one-page cleaning protocol
7. 07 Self-audit: twelve questions
8. 08 Where DataLion fits

Download the whitepaper