Data access
Students struggle to access data
sources for their empirical projects.
Every academic supervisor will have found some of their mentees being surprised, overwhelmed, and in need of help with accessing data. The students’ disappointment rises when it turns out that their supervisor is not in charge of enabling access to their project’s data requirements. The problem, however, emerges somewhere else entirely: not in the middle of the research journey, but, indeed, at its very beginning: the framing of the research question.
Interesting, evidence-based research requires data. But the most interesting data is difficult - sometimes impossible - to generate or access. Seasoned researchers are mindful of that and carefully review their research questions in light of the accessible data in order not to overburden a project with promises that the data simply cannot keep. They know that all elements of their projects need to follow one and the same trajectory. It is often overlooked in this facet of (student) research that data access often has a finality about it that cannot be argued away or worked around efficiently. If data is inaccessible and the largest share of effort around the project is not focused on making the impossible possible and generating novel data access, this is immutable. A project that depends on inaccessible data, and that does not focus on changing that, will fail. What seems particularly problematic and avoidable in this case is that most risked project failures in this domain do not emerge from conscious boldness, but carelessness. This carelessness manifests from an underestimation of which data is necessary to answer the set question reliably and comprehensively. Often times, the research question is simply asking too much. In the book, we discuss this forward and backward integration of project design and its limitations in the chapters on rigour and research design.
Instead of explaining to students that supervisors are in no position to help them access the data their projects require, let’s rather teach them to generate research questions that yield valuable, detailed insight out of more accessible data.
Problems with data access usually cannot be easily fixed - but they can be prevented by carefully and conservative formulation and structuring of research questions. Research mentoring that aims at guiding students away from this issue puts great emphasis on requiring a research question - data access combination that is plausible. Project drafts should not be greenlit if they do not discuss how the existing data access can sufficiently generate a basis to answer the question - or it would need to focus most of its research effort on generating that access and be flagged as risky. If the project is already in a precarious state, research mentoring ought to focus on helping to adjust the research question toward a trajectory that actually is addressable. Much of success in this domain depends on a researcher’s humility.
Did you experience this kind of challenge in the past with your mentees? How do you generally try to help your students to avoid and or fix such a situation?