اندازه گیری داده ها در سیستم های اطلاعات پژوهشی: معیارهای ارزیابی کیفیت داده ها Data measurement in research information systems: metrics for the evaluation of data quality

Introduction The topic of ‘‘research databases and research information systems’’ is by no means new. In recent years, the introduction of research information systems at universities and research institutions has strongly increased in Germany and throughout Europe (DINI AG Research Information Systems 2015). Research information systems can provide universities and research institutions with a current overview of their research activities, collect information on their scientific activities, projects and publications and manage and integrate into their website. For researchers, they offer opportunities to collect, categorize and make use of research information, be that for publication lists, for the development of new projects, to reduce the effort required to produce reports, or in the external presentation of their research and scientific expertise. Data quality plays an important role in the usability and interpretation of institutionspecific data. The quality of data is however also a significant consideration for external data sources. University administrations and researchers have since the early 1990s begun to recognize the importance of quality of data that are electronically stored databases. A few years ago, almost all German universities and research institutes were interested in the topic of quality of data in their RIS—a development that has been since further progressed. The growing quantity of data and the increasing number of source systems are becoming serious problems for institutions. In order to keep control and gain the greatest possible benefit from such information not only infrastructure measures, but also measures for the observance and increase of the data security and data quality are necessary (Apel et al. 2015). Almost all institutions rate high data quality as an essential consideration for their information on research activities. But only a few invest the time and resources to maintain and improve this data quality. In most cases, a poor data basis in individual departments is either reluctantly tolerated, or in the worst case not even perceived (Apel et al. 2015). This paper presents a holistic view of procedures for guaranteeing data quality in RIS. The handling of large data sources are daily operations for these institutions. Since data errors such as missing values, duplicates, spelling mistakes, incorrect formatting and inconsistencies occur during the collection, transmission and integration of research information in different systems and can spread over different areas, it is necessary to recognize these errors early and to treat them efficiently (Azeroual and Abusoba 2017). If users were unable to access the information they needed to make decisions, the value of the data they used and their confidence in the RIS would decrease. It is already sufficient if a small error renders the data unusable throughout the institution. Here, the completeness, correctness, timeliness and consistency of the data play a decisive role. It is important to understand that there cannot be data quality—and thus no data quality management— without measurements of data quality. Beyond that, the aim of this paper is to investigate the data quality of the research information given in RIS and to measure the quality dimension, based on the literature, and to develop a data quality framework with the objective of monitoring and improving the data quality of RIS.