Data collection and repositories Weekly Summary

DATA COLLECTION AND REPOSITORIES Data collection is a systematic process of gathering information to address research questions, test hypotheses, and support decision-making across disciplines such as public health, business analytics, and the social sciences. The selection of appropriate methods is influenced by research objectives, resource availability, and ethical considerations. Ensuring validity, reliability, and reproducibility is essential, as weaknesses in methodological design can introduce bias and undermine the credibility of findings (Creswell & Creswell, 2018) A variety of data collection methods are employed, including surveys, interviews, observation, sensor-based technologies, transactional data, web scraping, and experiments. Each approach presents distinct strengths and limitations. Surveys enable large-scale data collection and facilitate generalisation, although they are sensitive to questionnaire design and prone to response bias. Interviews provide rich qualitative insights but are time-intensive and may be affected by subjectivity (Dillman et al., 2014. Technological methods such as sensors and application programming interfaces (APIs) allow for continuous, real-time data generation at scale; however, they raise important concerns regarding privacy, ethics, and data governance (Mitchell, 2018) Data can be broadly categorised into quantitative and qualitative forms, as well as structured and unstructured formats. Quantitative data supports statistical analysis and generalisation, whereas qualitative data provides depth and contextual understanding of complex phenomena (Bryman, 2016). Structured data, typically stored in relational databases, is organised in predefined formats that allow efficient querying and analysis. In contrast, unstructured datasuch as text, images, and video offers richer insights but requires advanced analytical techniques, including machine learning, to extract meaningful patterns (Gandomi & Haider, 2015). Data repositories are centralised systems designed to store, manage, and retrieve data throughout its lifecycle. These include databases, data warehouses, data lakes, cloud storage platforms, institutional repositories, and version-control systems. Each serves a distinct purpose, from handling transactional operations to supporting large-scale analytics and long-term data preservation (Jagadish et al., 2014). The effectiveness of these repositories depends on key features such as accessibility, security, scalability, data integrity, and metadata support, which collectively ensure efficient use and responsible governance of data (Borgman, 2015). The relationship between data collection and repositories is inherently interdependent. Data must be carefully collected, cleaned, and stored before meaningful analysis can occur. Importantly, the quality of insights derived from repositories is directly dependent on the quality of the original data. Poor data collection practices, including biased sampling or inaccurate measurement, can significantly compromise analytical outcomes regardless of how advanced the storage infrastructure may be (Kitchin, 2014). A typical data lifecycle includes collection, validation, storage, analysis, and dissemination. These processes are widely applied across sectors such as healthcare, government, business, and research, where data-driven decision-making is increasingly central. Ultimately, effective data collection and robust repository systems are critical for generating reliable knowledge, fostering innovation, and supporting evidence-based practice. For additional information, please find the inserted video on the link below.