Wednesday, May 13, 2009

Helping you evaluate data quality

If you're interested in evaluating the quality of statistical collections and products (eg. survey data or statistical tables) then check out the recently released ABS Data Quality Framework (cat. no 1520.0).

The Framework can assist user's to:
  • decide whether a dataset or statistical product is fit for purpose
  • assess the data quality of seemingly similar collections
  • interpret data

It can also assist those developing statistical collections to produce high quality outputs.

The Framework provides a multidimensional approach for assessing the quality of statistics, incorporating: Relevance, Timeliness, Accuracy, Coherence, Interpretability, Accessibility and Institutional Environment.

For each dimension, the paper describes what constitutes the dimension, how it may be evaluated, and suggests questions to be considered for the purpose of assessing the dimension. The paper also includes information to assist data users and producers to apply the framework.

No comments: