In the last decade, repositories have become vital for research data management. They support Open Science principles, such as the FAIR principles, and face challenges in an evolving publishing landscape. Quality assurance and measuring scientific output remain important topics, with ongoing discussions on methods and metrics. Research in data management highlights the importance of metadata in research data repositories, but few have studied how it is created and maintained. Low-quality metadata can negatively impact usability. Traditional approaches for assessing cataloging quality may not suit indexing repositories. Developing a trust model for metadata editing is crucial. Investigating a repository's size, as done by re3data, offers valuable insights. However, defining size in information science is challenging. Establishing indexing rules is necessary for quality control and harmonization with other repository attributes. This thesis analyzes the size property of metadata in the re3data repository through data analysis and a case study. It aims to refine the cataloging model and improve quality by exploring semantic concepts and proposing automated and intellectual measures. The research questions focus on the usage of size property, detecting quality factors, and addressing challenges in indexing. The thesis lays the groundwork for future automation possibilities and offers supplementary approaches.