datatekorten

Datatekorten (Dutch: data shortages) is a term used to describe gaps or shortages in data that limit the reliability, scope, or timeliness of analytical work. It is used across research, public policy, and industry to refer to situations where available data do not adequately cover the phenomenon of interest, leading to uncertainties and potential bias in conclusions.

Causes of datatekorten include limited data collection or geographic and demographic coverage, privacy and regulatory constraints that restrict data sharing, fragmentation across organizations and systems, inconsistent data standards, and missing or historical data gaps. Selection bias and nonresponse can further worsen datatekorten when certain groups are underrepresented in the data.

The consequences of datatekorten can be significant. They reduce statistical power, distort estimates, hinder trend detection, and impair model performance. This can lead to biased policy decisions, unequal outcomes, and increased uncertainty in forecasts and evaluations, undermining confidence in research and strategy.

Examples of datatekorten occur in public health when surveillance data underrepresent marginalized groups, in economic analysis when small businesses are absent from administrative datasets, in climate records with intermittent measurements, or in regional planning where municipal reporting is incomplete.

Mitigation strategies include data integration and harmonization across sources, targeted data collection to fill gaps, and the use of imputation, robust statistics, and explicit uncertainty quantification. Privacy-preserving approaches such as federated learning or synthetic data can help expand usable data without compromising privacy. Where possible, establishing data-sharing agreements and thorough documentation of data quality is encouraged.

See also data availability, data quality, data gaps, and data governance.