Dr Ilona Lipp


Open Science: A Necessity, Not a Trend

Why Open Science Practices Are Important For Upholding Research Integrity

August 2, 2024


The theory

Reflecting the core principles of scientific philosophy and ethics, integrity and transparency are foundational values upon which research institutions are built and are therefore indispensable. Open Science is often seen as independent of these values and argued for as an end in itself. Common arguments for open science include a moral obligation to taxpayers and the importance of openness as a value. However, what is often overlooked is that some Open Science practices are fundamental for conducting transparent, responsible research, and consequently for sustaining the scientific integrity at the heart of our scientific identity.
 

All staff, both scientific and non-scientific, at research institutions are required to adhere to mandatory principles and guidelines for Good Scientific Practice. For this article, I am focusing on the guidelines published by the DFG (German Research Foundation) and ALLEA (European Federation of Academies of Sciences and Humanities). The DFG guidelines are available here and the ALLEA guidelines are available here. These guidelines are publicly accessible, and often research institutions have their own adapted versions published on their websites.

The practice

In the real world (as I have experienced it), these guidelines seem not to have too much relevance to the daily lives of researchers. I wonder to what extent researchers are even aware of their existence or see them as more than abstract formalities created by a guideline-loving administration. 

In my experience, new PhD students (or any new scientific staff) do not spend their first day reviewing these guidelines and contemplating their application. Instead, they tend to follow the practices established by senior scientists in their research environment. In this way, research practices—both good and bad—are 'heritable' and passed from PIs to students. (See this fascinating simulation on that.)

Moreover, even if researchers conscientiously read the guidelines, the practical meaning of some principles is not always clearly defined, leading to many gray areas. For instance, ALLEA states that 'withholding research data or results without justification' violates good research practice. While the justification for withholding information may be clear in some cases, in others it might be subject to subjective judgment. Similarly, the DFG guidelines state that 'what constitutes a genuine and identifiable contribution [regarding authorship guidelines] must be evaluated on a case-by-case basis and depends on the subject area.' This ambiguity leaves considerable room for interpretation.


Open Science Practices and Their Alignment with Good Scientific Practice


Despite the room for interpretation, the ideology behind the GSP guidelines is clear: they aim to safeguard scientific integrity and ensure public confidence in research. These guidelines are regularly updated to reflect ongoing developments, such as recent trends in AI. I was surprised to see how multifaceted they have become; even aspects of workplace culture, such as leadership (see DFG Guideline 4), have been incorporated into Good Scientific Practice. Moreover, while Open Science proponents are sometimes criticized for promoting diversity issues as part of a 'woke agenda,' considering gender equality and diversity in staff selection and development is specifically mentioned in DFG Guideline 3. 


While Good Scientific Practice and Open Science are separate concepts, they share underlying motivations and objectives. I argue that Open Science practices offer practical solutions to many challenges outlined in Good Scientific Practice guidelines. Below, I explore how specific Open Science practices align with Good Scientific Practice, referencing relevant DFG guideline numbers and ALLEA subsections

Navigating Current Research with Open Access Literature

  • Good Scientific Practice Requirement: According to DFG Guideline 9, researchers are required to incorporate current research into project planning, a process often obstructed by paywalls that restrict access to essential literature. ALLEA (Section 2.3) asks researchers to publish results in an open manner.
  • Open Science Solution: Preprints and Open Access Publications offer instant and free (at least for the reader) access to scientific literature. Coupled with AI tools like Elicit, which help summarize literature, researchers can more effectively and comprehensively consider the current state of research.


Enhancing Documentation and Transparency

  • Good Scientific Practice Requirement: According to DFG Guideline 12, researchers are required to document all information relevant to the production of a research result clearly enough for the result to be reviewed and assessed. Given the sophistication and interdisciplinary nature of modern research, conventional methods sections in papers—often subject to word limits—can make it difficult to evaluate the thoroughness of the methods or identify potential problems.
  • Open Science Solution: Implement Open Protocols (e.g., protocols.io) to provide detailed methodological descriptions that surpass typical reporting standards. This approach not only allows reviewers to examine crucial details that may impact the results but also aids in facilitating replication studies and the independent validation of findings.


Promoting Comprehensive Reporting of Results

  • Good Scientific Practice Requirement: According to DFG Guideline 12, researchers must report all results, including those that do not support their hypotheses. Similarly, ALLEA list 'withholding research results' as scientific misconduct. However, the current academic system, which values publications in 'high-impact' journals as a measure of success, discourages this practice. These journals typically favor studies that yield clear, exciting results that support a strong hypothesis, presented with a clear storyline.
  • Open Science Solution: Registered Reports and Preregistration aim to offer solutions to this challenge. Registered Reports involve submitting your hypothesis and methods to a journal for approval before data collection begins, ensuring that the decision to publish is based on the research question and methodology rather than the results. Preregistration is a lower level implementation of the same idea, allowing researchers to publicly register their study plan and hypothesis in advance, to be able to demonstrate that they have not 'cheated', however, no guarantee of journal publication is associated with that.


Sharing Research Resources

  • Good Scientific Practice Requirement: According to DFG Guideline 13, researchers are expected to make their results, data, materials, and software (including source code) available when possible and reasonable. ALLEA (Sections 2.5) also mentions making data access as open as possible. However, traditional written publications remain the dominant form of research output, which can limit the sharing of actual research resources.
  • Open Science Solution: Open Science Infrastructure offers platforms for sharing software, data, and protocols, and for obtaining Digital Object Identifiers (DOIs). Using Persistent Identifiers (PIDs) in metadata additionally facilitates linkage to specific projects or researchers.


Reforming Research Evaluation

  • Good Scientific Practice Requirement: According to DFG Guideline 5, a multidimensional and qualitative performance assessment should be conducted. ALLEA (Section 2.1) specifically states that research institutions and organisations should reward open and reproducible practices when hiring and promoting researchers. Currently, hiring decisions are often based on the number of papers and the impact factor of the journals in which they were published in. This information is easily accessible. However, to evaluate researchers qualitatively and comprehensively, their multidimensional outputs also need to be accessible, and evaluation criteria need to be defined. 
  • Open Science Solution: Here, Open Science infrastructure solutions to publish papers, data, protocols, software, teaching materials, and public engagement online are highly relevant. Tightly linked with the Open Science movement are research evaluation reform movements, such as DORA and CoARA that are working on practical solutions for multidimensional evaluations.


Conclusion

If we want to stick to Good Scientific Practice guidelines, we will eventually have to fall back on tools and practices provided by the Open Science movement. Therefore, the Open Science movement should not just be seen as making science more 'open' in general. Its fundamental role in ensuring scientific integrity and increasing trust in science must also be recognized.