Student Surveys as Part of the Development of Student Data Analytics in Institutional Research
This section focuses on the challenges and opportunities concerning student data analytics as part of the institutional research: the practices of collecting, synthesizing, and analyzing student data to serve as evidence in university decision-making and planning, and also to fulﬁll mandatory reporting requirements and external assessment (cf. Klemenčič and Brennan 2013; Klemenčič et al. 2015). The institutional research on students is part of the larger process of expanding the function of institutional research from the basic reporting approach for statistical purposes, funding and accreditation and record keeping towards a bigger role in quality assurance, assessment of institutional performance, and, ultimately, also in strategic planning and development (Klemenčič et al. 2015). The development of institutional approaches to student data analytics towards the strategic approach requires several changes in terms of types of data collected, sources of data and data management systems (see Fig. 1).
As we move from reporting approach towards quality and strategic approaches, the range of student data collected expands: from the basic records on enrollments, academic progress and student proﬁle to student course evaluations, student approaches to studying and learning, student satisfaction with the learning environment (student services and facilities) as well as student learning outcomes and
Fig. 1 Approaches to student data analytics
employability. In reporting approach, data tend to be generated only from the institutional records provided by students upon registration and from student academic records. In quality enhancement approach, data is also sought directly from students with surveys on opinions, satisfaction and behavior, and possibly also through qualitative methods, such as interviews, focal groups or direct observation. In strategic approach institutions also gather data from external data sources to develop intelligence on international trends in student recruitment, and compare themselves to other institutions. They use new technologies which allow for data mining and data scraping to extract information from public data sets, social media and public blog posts. Another new source of data on student behavior comes with web analytics which track students' usage of university webpages.
Data on and from students typically presents one of the sources of university intelligence, and it varies from one institution to another to which extent this data is integrated into a central data warehouse and translated into “business intelligence” to inform decision-making, or is kept within warehouses of student registrars or quality assurance centers of teaching and learning units. Methods of data management—collection, storage and analysis—differ in the reporting, quality and strategic approaches. In reporting approach, different types of data are kept in individual data warehouses units (e.g. student registrars, units for quality assurance, student affairs, teaching and learning, international ofﬁce, etc.) and processes within that unit. Data is automated and processed with basic statistical tools. Standardized reports are prepared for internal or external use. In quality approach, data tends to be managed within different units, but a central data management system is put into place to conduct quality checks and prepare institution-wide reports. In strategic approach, institutions connect student data from various sources into one integrated institutional data warehouse, where it is linked also to other data on university operations (e.g. academic staff, ﬁnance, etc.). The advantage of such integrated systems is that student data—and other key institutional data—is available across the institution and processed in a timely and reliable manner through common data management software. In this way data is accessible across the institution for performance evaluation and strategy planning. For such data managements systems to work, universities ﬁrst of all need to build technical capacity, but—also and equally important—hire and train skilled analytic professionals who are able to turn data into evidence for decision-making (Klemenčič et al. 2015). Often in universities there already exist much reliable data and information, which is not put into use in decision-making, because it is not readily and easily accessible or because it is not sufﬁciently processed for use.