Credibility in Data, CID, is directed at protagonists, stakeholders, and decision-makers, who either need to serve or to request credible (and health) data in order to meet authentic decisions.
Credibility is met when doses of 3 dimensions have been balanced:
Whenever people identify data issues by missing rules and/or violations of rules on syntax and meaning first thoughts are directed at Data Quality (DQ).
Biased data does not comply with the ideas of accuracy, completeness, consistency, etc. Biased data is caused by physical and logic deficits during operations.
Pragmatical root cause analyses are performed to remediate biased data.
DQ (clean[s]ing) rules (re-)active trust in data.
However, DQ that is based on integrity rules remains a sufficient DQ condition as long as there is no data competence team that governs rules and processes of DQ over time.
Thus, its deliverable is an organizational schema / grid / matrix of roles & responsibilities that ensures solid DQs with allocated actions (activities + deliverables).
DG can be seen as a
necessary DQ condition which makes DQ robust over time.
It’s not credible to claim
necessary DQ prerequisites from clients (= DG organization). It's also unfair to limit its deliverables to
sufficient DQ conditions (= Integrity). It's also not candid to expect the opposite situation for both.
A data stakeholder needs to verify the dosage of DG & DQ to equilibrate DG & DQ.
However, reciprocal deliverables to a biased quote DG/DQ alone will not lead to successful data-driven decision makings. The balance of DG and DQ is at the same time a duality between DG and DQ that activates a didactical and dialectical process of trust in
Data Competencies (DC).