| چکیده انگلیسی مقاله |
Data is considered as an asset in organizations and its quality is an important principle to achieve productivity. For optimal management these organizational assets need a kind of governance, so that with its help, the data align with the goals of our constructive leadership. The purpose of this research is to identify the dimensions of data quality for data governance. To achieve this goal, a two-stage qualitative approach was used. In the first stage of the meta-combination method by searching for the key and keywords of data quality, Data management, data governance and data quality management in Air&Dec databases, Science Direct, Google Scholar, Springer, IEEE and ACM were conducted between the years (1995-2022) and 268 related articles were identified and in a detailed study and evaluation, 62 approved articles were decided. By reviewing and studying these articles fully, 8 concepts were identified. After applying the opinions of professors and three data experts, finally 55 components were extracted for the research question. In the second stage, fuzzy Delphi method was used to get the opinion of experts. For this purpose, the items needed to design a fuzzy Delphi questionnaire were provided from the output of metacombination, and this process was continued until the experts’ opinions on the answers to the questions reached a consensus. 21 experts who had at least one research paper in the field of data quality were selected and finally 14 completed questionnaires were returned. In response to the questions of data quality dimensions for data governance, there are 46 indicators: completeness, timing, communication, accessibility, compliance with laws and standards, confidentiality, interpretability, redundancy (ability to add), reputation and reliability, ability traceability, value (value), simplicity, update, concept, regularity, linkability, referential integrity, uniqueness, purposefulness, auditability, accuracy, comparability, consistency, commonality, completeness, metadata compliance, acceptability, validity, conciseness, applicability (usability), believability, comprehensibility, reliability, reasonableness, consistency, retrievability, reproducibility, ability to display null values, appropriateness, clarity, added value, comprehensiveness, extensibility, patternability and variability were identified. More than 80% of the indicators obtained from the extracombination were accepted by the experts, so our system and data-driven businesses can use these indicators as a priority for measuring the quality of their data. |