NCMP validation principle

The main principle behind the validation of NCMP data is to push the onus for data quality onto the data provider.

This then encourages the provider to take ownership for the quality of their data and avoids costly workarounds to clean data centrally.

Therefore all data which the providers enter onto the NCMP system will be used even if there is concern around the data quality of some records.

These data quality issues will be dealt with in two ways:

• adding a caution flag to the underlying dataset so other users of the data can make a decision on whether to include these data items in their analyses
• using the data quality note which will accompany the national report to make users aware of data quality issues.

The only time when a record will be excluded from the analysis carried out by NHS Digital will be when it is adversely impacting on national level results. This principle is in line with other data collections within NHS Digital. 

1. Pushes the onus for data quality onto the data provider.

2. Naming providers who appear to have poor quality data should encourage them to improve the quality of their data in future submissions

3. Reduces resource needed for validation centrally and allows for earlier publication.

4. Avoids the potential for conflict if a record is removed from the national report that the data provider insists is correct.

5. Alerts users of the data to data quality issues and allows them to make an informed decision on whether to include or exclude specific records.

1. Includes some data in the dataset which appear to be outliers. However, this will be mitigated against by adding data quality caution flags to the underlying data.