So much can go wrong with data quality. But more importantly this can often go un-noticed. Decisions can be made with a very poor evidence base or a lack of understanding of how the evidence available should be used effectively.
Reviews of data quality in international development Programmes have shown us that resources spent on designing and evolving a data quality strategy can be a very worthwhile investment. Most of all, don’t take data quality for granted! This relates to your internal data sources as much as external sources such as 'official' statistics.
Common causes of data quality problems
- Human errors in the collection, processing and compilation of ‘field’ data
- Definitional and methodological errors/misunderstanding
- Errors driven by inadequate resources
- Inadequate communication, coordination and teamwork
- Inadequate documentation (metadata)
- Optimism bias
- Data is or becomes unavailable in the required time period
- Over-reliance on the reliability of ‘official’ sources of secondary data
- Poor coverage/understanding of Gender and Inclusion (G&I) and sustainability (who is your data 'leaving behind' ?)
- Qualitative data is not collected in a structured and statistically sound way