Skip to ContentSkip to Navigation
Research Open Science Open Research Award

Preregistration of exploratory research: symptom level network analysis of internalizing and functional disorders

Urvi Saini (Psychiatry, UMCG)

Open Research objectives/practices

Objective: Make research reproducible and transparent by making as much information on methods available to the public as possible. Practice: Creating a preregistration of study aims and analysis plans.


For my first project in my PhD, I, together with my supervisors (Hanna van Loo, Tineke Oldehinkel, and Judith Rosmalen), designed a study to perform a symptom level network analysis on internalizing (major depressive disorder and generalized anxiety disorder) and functional disorders (chronic fatigue syndrome, fibromyalgia, irritable bowel syndrome). Specifically, we aimed to understand how the symptoms of these disorders connect and overlap. To perform this network analysis, we used the data from Lifelines, a multidisciplinary prospective population-based cohort study examining the health and health-related behaviors of over 167,000 persons living in the north of the Netherlands. This was an exploratory study because we did not have pre-defined hypotheses and were instead interested in seeing symptom connections visually with a network plot to generate possible questions on the diagnostic criteria of these disorders. Therefore, to ensure that our aims and analysis plan for this large data were clear, we preregistered our study on OSF. By doing so, we could confidently say that our work was honestly done.  


My supervisors and I decided to preregister the study because we wanted to be transparent and open about our study aims and methods. Having this clarity helped us plan for a more focused and well-designed study. It was also important for me to adapt honest scientific practices right from the start of my PhD.
Working with big data such as Lifelines, I wanted to make sure that our research intent was known prior to any data analysis. By preregistering our study, I made clear what our intentions were with the data and the specific information we would examine. Moreover, by outlining the methods and analysis plans in advance, I ensured that the research remained focused, unbiased, and accountable. I needed to be sure of not only how to analyze the data, but exactly what I was going to analyze. Therefore, by making a clear procedure for what variables I would study and the statistical approach I would take, I had a ‘recipe’ to return to in order to make sure I remained on track. I also wanted to make it clear that this study was exploratory, and the results would help ask more questions that could be answered in future research.

Lessons learned

The preregistration took time and effort because it required me to do thorough background research on which statistical methods to use and finalize exactly what I wanted to study (i.e., which specific symptoms I wanted to include). However, after it was done, the rest of my work went much quicker, and I was glad to have the predefined analysis plans. Completing a preregistration emphasized the importance of accountability and ethical considerations. I recognized that the preregistration was not a mere formality but a declaration of research integrity. It held me accountable for adhering to my research plan, and in the event of any deviations, it compelled me to transparently communicate the reasons behind them. I also recognized that deviations are okay, as sometimes things happen that cannot always be predicted. It is important to adhere to the outline, however it is not the end of the world if something changes. Transparency is key and if I report these deviations, no harm is done.

URLs, references and further information

Last modified:01 November 2023 11.03 a.m.