Articles

Since the 2000s, the digitization of bibliometric infrastructure has accelerated at a rapid pace. In this article, we analyse how these developments in evaluative bibliometrics have contributed to an extension of indicator-based research evaluation towards data-driven research analytics. Drawing on empirical material from blogs and websites as well as from research and policy papers, we discuss how interoperability, scalability, and flexibility as material specificities of digital infrastructures generate new ways of data production and their assessment, which affect the possibilities of how academic performance can be understood and (e)valuated.

Non-traditional students in higher education drop out more often than traditional students. To understand this difference more clearly, this paper examines differences between the two groups with regard to sociodemographic characteristics, study contexts and life circumstances, as well as individual assessments of prospects of academic success and the costs and returns of studying. A decomposition analysis shows that student age and study context are two factors that have a particularly strong significance for the higher tendency of non-traditional students to drop out. This paper suggests ways in which theoretical models for explaining educational decisions could be adapted with regard to the special group of non-traditional students.

Collecting life history data is highly demanding and therefore prone to error since respondentsmust retrieve and provide extensive complex information. Research has shown that responseburden is an important factor influencing data quality. We examine whether increases indifferent measures of response burden in a (mixed-device) online survey lead to adverseeffects on the data quality and whether these effects vary by the type of device used (mobileversus non-mobile).