Measuring Data Quality of Theses and Dissertations in the Data Preparation Stage of Registration Systems

Abstract

Today, academic research plays a very influential role in the economic development of countries. These researches are often recorded and disseminated in the form and structure of theses and dissertations in scientific institutes. The better the quality of this data in the systems that collect and distribute it, the more it can be used and exploited by organizations and businesses. Therefore, providing this data requires proper monitoring to put the output of the recording and dissemination process in good condition. This paper offers a framework for evaluating theses and dissertation data quality. In the framework, the data inconsistency coding structure is introduced and presented in Word and PDF files and in the form of metadata (bibliographic information). The approaches presented in data quality methodologies (TDQM and DWQ) are also used to provide solutions to improve data quality in the provisioning phase. At this stage, approaches such as owner attribution to data or process, root cause analysis, process control, and continuous monitoring are considered. The focus group method determines the operational strategies for quality improvement. Finally, process-oriented techniques, such as quality control checklists and image processing, and data-driven approaches, such as data cleansing, are localized and developed in this section to improve the quality of theses/dissertation documents. The provided improvement solutions were categorized into two different groups. Guiding the user in the "Theses/Dissertations" registration process is identified as a process-driven category. On the other hand, introducing a specific format for "Theses/Dissertations" files and resolving the quality issues of PDF files were among the data-driven solutions.

Keywords

• data quality
• information quality
• quality control
• thesis/dissertations registration system
• incompatible analysis
• data preparation