Lead a team or a unit

Team and unit managers provide the institutional support that staff need to define the team and unit archiving policy. The aim of secure backup and archiving is to guarantee the security of the data, and to facilitate access to it for all current and future project collaborators.

You must promote the implementation of a data management plan (DMP) for each project developed in the unit.

To take advantage of the digital revolution, to speed up research and harness the power of automatic analysis at scale while ensuring the transparency, reproducibility and societal usefulness of your work, the data and other digital objects created and used for research must be findable, accessible, interoperable and reusable (FAIR principles).

The final report and action plan of the European Commission’s expert group on data sharing is available here: bit.ly/turningFAIR

How to do it:

To meet the FAIR principles, we need to ensure that data is properly stored and backed up.

The idea is that data should be duplicated and stored in different places and on different media, applying the 3-2-1 rule:

• Keep 3 copies of the data,
• On 2 different media or technologies,
• 1 of which is off-site.

Many hosting companies offer solutions that meet these criteria: don’t hesitate to get in touch with the hosting company’s local digital services.

In practice, you need to:

• Determine the type of hosting : local servers or secure cloud,
• Understand the characteristics of these hosts (are they duplicated and backed up: frequency, location, security procedures, etc),
• Consider whether the proposed solution is consistent with the unit’s needs,
• Estimate the volume of data required,
• Organise and plan data back-up,
• Anticipate!
• Select the data to be backed up or deleted regularly,
• Define their retention period,
• Manage versions,
• Keep a record of the various data statuses and processing stages, so you can go back to an earlier stage if necessary.

For example, you can find advice and tools on the dedicated pages of the Inserm intranet: https://intranet.inserm.fr/rubrique/support-administratif/archiver/

• In particular, the following sheets:
  • Sorting and classifying your electronic data,
  • Organising your files in an electronic tree structure,
  • Memento on the conservation of electronic archives,
  • Rules for naming electronic files, etc.
• For each type of document, a management table (or archiving repository) sets out the administrative use-by dates and the final treatment (retention or disposal) to be applied.
• The research archive management repository proposes retention periods equal to the life cycle of the data. It is available on DoRANum and can be accessed online via a number of different entries,
• The electronic archiving system is or will be offered in all institutions. It is a tool for the permanent and secure storage of electronic data. Once integrated into an SAE, they can no longer be modified and retain their evidential value.

The support you need:

• First of all, think about the archive services. At Inserm: disc@inserm.fr and the network of archive correspondents deployed in the regional delegations and universities,
• The AURORE network, which brings together ESR archivists to address archiving issues and make tools and procedures available to the scientific community. For example, consideration of archive management in UMRs.

Obstacles / complexities:

• Formats: many file formats are not permanent archiving formats (to find out more, see the list of formats accepted by CINES for permanent archiving: https://facile.cines.fr/
• File naming: impossible to reread files (if the file name is too long, the access path is not recognised), difficulty of interoperability between MAC and PC (loss of the tree structure originally established, impossible to reread or even copy),
• The human time required for archiving,
• Taking digital pollution into account.

Outlook:

– Selecting data for better re-use,

– Pooling jobs and skills: scientific community/Archives/IST/DPO/IT department/legal department.

Quality storage and quality data management in accordance with the FAIR principles (see link to the dot: Defining an archiving policy) are essential if data is to be shared now or in the future, enabling it to be re-used to confirm results already obtained or to test new hypotheses.

You will find useful information on naming files and folders in the following resources:

• DoRANum: 5 naming rules for your documents. https://doranum.fr/stockage-archivage/comment-nommer-fichiers/
• https://ge.ch/archives/media/site_archives/files/imce/pdf/procedures/20151207_aeg_rm_nommage_v2.pdf
• https://cours.ebsi.umontreal.ca/sci6005/co/nommage.html

Working to produce research that is reliable and as reproducible as possible is an important aspect to instil in a research team. Encouraging greater transparency can help to achieve this objective. The previous pearl stresses the importance of sharing data, an approach that can nevertheless sometimes be complicated in the case of sensitive data. But improving transparency is not always complex and can involve very simple measures that can be encouraged. Before the research begins, recording the protocols in advance makes it easy to trace how the hypotheses were generated and how the methodological choices were made. Then, when the time comes to write the article, the use of reporting guidelines such as those of the EQUATOR network (https://www.equator-network.org/) ensures that no important element of the experimental approach is forgotten in the protocol. It seems very important for a team leader to promote these guidelines and ensure that they are systematically implemented within the team.

Encouraging an open science policy means promoting, as far as possible, the unhindered dissemination of all research data, i.e. methods and source codes, results and publications. Within your team or unit, you can encourage this policy of openness:

– by ensuring that research staff have the necessary training and resources to participate fully in open science practices,

– by encouraging the systematic submission of publications to HAL,

– by identifying and dealing with the financial and human costs of open science: do not forget, for example, to take into account the additional time that will be required for permanent and traceable annotation of data,

– by creating positive incentives to encourage and reward open science practices: recognition, visibility of open science practices, inclusion in evaluation and promotion files,

– by facilitating all initiatives that make open science the norm.

You will find practical resources to support you on the « Ouvrir la Science » website (https://www.ouvrirlascience.fr/livret-science-ouverte-pour-les-directeurs-dunites-2/).

The 2^nd National Plan for Open Science (2021-2024) is also available on this site:  https://www.ouvrirlascience.fr/second-national-plan-for-open-science-2021-2024/. Its aim is to provide France with a coherent and dynamic policy in the field of open science.

Quality research requires good working conditions: https://www.nature.com/articles/s41562-022-01508-2

Being vigilant about the quality of the working environment also means preventing professional and psycho-social risks. Preventing occupational hazards starts with an awareness of the risks and hazards in the laboratory, and in particular the organisation of the induction of new members of the team/unit.

It also means paying attention to psycho-social risks (stress, ill-being, harassment, burn-out, etc.), which can have an impact on staff health, laboratory operations and, ultimately, the quality of the research climate and work.

You will find all the regulatory information, videos and practical information sheets on the institutional websites. For example, visit https://pro.inserm.fr/rubrique/prevenir-accompagner-et-agir/sensibilisation-aux-risques.

Bear in mind that regulations are no substitute for a critical mind, and that you can always ask for advice or call on the services of various people (staff representatives, social workers, prevention assistants, prevention advisers, the special health and safety committee (CSHSCT) and the national health and safety committee (CHSCT), as well as the social monitoring unit).

As scientific training does not take sufficient account of the importance of human relations and management, we would like to remind you of a few simple and basic rules that may be useful:

• words are probably the most obvious signs of recognition of the work of a member of the team. A thank you, encouragement, congratulations. On a day-to-day basis, these are the details that count and that motivate people to invest a little more,
• it’s also about giving a team member a sense of responsibility, getting him or her involved in a project because a certain skill has been recognised or because he or she has expressed an interest in a particular area, confirming him or her in his or her role or inviting him or her to take on other challenges,
• giving the opportunity to evolve, to take on a new dimension, to envisage new professional developments in the long term,
• draw inspiration from tools such as CRediT (Contributor Roles Taxonomy) and, for example, include the names of engineers and technicians on publications when they meet these criteria, as they are too often overlooked.

As Kaoru Ishikawa (theorist and precursor of quality management) wrote: “Quality control begins with training and ends with training”.

As a team manager, you must encourage everyone to take training to improve their skills and/or meet safety requirements. Supervision/management of your team is a key point on which most colleagues receive little or no training. A lack of supervision is often at the root of an unfavourable climate for quality research.

The LORIER programme offers various training courses as part of initial and continuing training aimed at promoting the development of ethical and responsible research at Inserm. A list of training courses will be available shortly.

Find out about promotion procedures at different institutions

• Inserm: https://pro.inserm.fr/rubriques/content/join-us/hr-excellence-in-research-en
• CNRS: https://carrieres.cnrs.fr/en/vous-accompagner/
• For each university concerned, see the corresponding websites.

Training: On a personal level, each team leader must prepare and train to conduct individual annual interviews. They should also be prepared and trained, if necessary, in the drafting of aptitude reports (Gaia type). At Inserm, a guide is available for team and unit managers . Its purpose is to specify the role and organisation of the professional interview for technical and engineering staff, both contract and permanent. The professional interview for contract staff has been defined in the Inserm charter listing the best practices to be observed in recruiting and monitoring the professional careers of contract staff.

Communicating : Within the team or unit, the manager must distribute information on professional training as and when it is received for example, as well as information on internal competitions (https://www.inserm.fr/en/our-research/good-practices-at-inserm/).

In order to avoid internal conflicts, the manager must be able to explain the choices made regarding promotion according to a relevant schedule and objective criteria.

Evaluating research practices can be perceived in a negative light, as it is often difficult to question oneself. The challenge is to evaluate our own practices in order to improve them, not to systematically question individual researchers. The difficulties are systemic and the challenge is to approach them globally. It is important to recognise that mistakes can be made, to know how to correct them and to recognise that research is not perfect but can be improved.

Want research integrity? Stop the blame game. Helping every scientist to improve is more effective than ferreting out a few frauds. Macleod Nature 2021.

The key to fostering a culture of integrity and responsibility within your team(s) is based on a combination of approaches, including values, standards of practice and reliability. To implement these approaches, you need to understand the mechanisms that can turn scientists away from scientific integrity, accept that there are no ready-made remedies, and bolster the commitment of all concerned to ethical and responsible research. Here are a few suggestions:

Setting up actions to raise awareness towards scientific integrity by way of seminars on good research practices, posters, workshops and training courses….

Creating an organization and tools to prevent “questionable” research practices, for example by making good research practices an integral part of the organizational strategy, and by implementing a quality approach aiming to maintain scientific integrity (https://www.inserm.fr/en/our-research/good-practices-at-inserm/).

Promoting awareness campaigns. For example, you could use « reproducibiliTea » as a starting point for discussion on reproducibility issues. This is the case, for example, with « reproductibiliTea » in Bordeaux : https://neurocampus-graduateprogram.u-bordeaux.fr/Training-activities/ReproducibiliTea-Open-Science-Journal-Club.

You can also draw inspiration from Schocker F. et al, Mission impossible? A cultural change to support scientific integrity, Embo Reports 2021.

Even research teams and units with a long-standing approach to research integrity can benefit from periodic reflection on what they are doing and how they can improve.

This dot suggests a number of questions to explore practices within your team or unit:

• What regular initiatives have been implemented within your team/unit to encourage ethical and responsible research? (i) Journal Club, seminars, positive incentives for training, quality, discussions addressing research values and responsibilities, and standards of good practice, etc. (ii) information on rules for publication, communication, traceability and archiving of data, policy in favour of the publication of negative results etc.
• Is this approach known and easily accessible to everyone involved in the research work of your team or unit, whether on a temporary or longer-term basis?
• Are the people who can advise you clearly identified, and are their names accessible to all? Does your team benefit from the presence of one or more LORIER ambassadors, or have you identified resource persons within your administration office or elsewhere (university, doctoral school, etc.)? Which themes are concerned? Are they familiar to everyone, and do members of the organisation trust them enough to contact them? Does the analysis of a problem enable actions to be taken to avoid its recurrence?
• What aspects should be addressed in the areas of research covered by your team/unit? Clinical research or research involving human beings, tissues or cells, or research involving animals, for example?
• Are there any initiatives to train staff in:

o Experimental design?

o Compliance with regulations applicable to the particular field of activity?

o The production of reliable data?

o Traceability and management of research data?

o Data protection?

o Research involving humans, including clinical trials?

o Supervision and mentoring?

o The impact of the research on society (ethics, sustainable development, etc.)?

o An approach targeting on-going improvement (quality approach)?

• How do you ensure that the research activity in your team/unit meets the criteria for ethical and responsible research? How confident are you that all research integrity concerns are appropriately addressed in your team/unit? How confident are you that research staff, particularly at an early stage in their careers, know that they can raise concerns about research integrity without being stigmatized or suffering any prejudice?
• Do you feel that research work within your structure is carried out in a favourable climate for quality research?

To help you assess your own research practices in terms of integrity, you can use the «Self-evaluation tool» developed by the «UK Research integrity office»: https://ukrio.org/publications/concordat-self-assessment-tool/

Ambitious research means original, critical and disruptive research. As a backdrop, this dot suggests reflecting on Doug Altman’s phrase «We need less research, better research, and research done for the right reasons.» (https://www.bmj.com/content/308/6924/283).

While progress has been made, improvements are still needed: Research waste is still a scandal—an essay by Paul Glasziou and Iain Chalmers BMJ 2018 (https://www.bmj.com/content/363/bmj.k4645)

The value of more collaborative research is discussed in the editorial by Waheed-UI-Rahman, A Collaborative Student Research Efforts Provide a Solution to Research Wastage, BMJ 2018 (https://www.bmj.com/content/364/bmj.l795)

• Publicizing the rules of good practice in your discipline (see Research Personnel section),
• Talking about scientific integrity when welcoming new members to the team or unit, making it a cornerstone of the structure’s activities, on themes such as risk management, and encouraging team members to get involved in LORIER or other initiatives,
• Displaying the contact details of scientific integrity officers and other resource persons,
• Making available the scientific integrity posters produced by RIQ, DIS and Inserm’s DISC (https://lorier.inserm.fr/pratiquer/affiches-pour-votre-laboratoire/),
• Promoting the traceability of results by way of laboratory logbooks, relevant, documented data, good sample storage management, etc. : (https://pro.inserm.fr/rubrique/support-a-la-recherche/le-cahier-de-laboratoire-electronique-de-linserm/ce-quil-faut-savoir),
• Fostering a climate of trust within the team: accepting and encouraging the presentation of problems linked to experimental work and negative results at team meetings, so that everyone can discuss them without judgment,
• Implementing a continuous improvement approach (Plan Do Check Act): managing malfunctions to learn from mistakes or problems encountered, and using them as sources of change and opportunities for improvement,
• Discussing upcoming articles at team meetings. Establishing a table of contributions to ensure that rules on authorship and co-authorship are followed, thus avoiding future conflicts (https://coop-ist.cirad.fr/etre-auteur/definir-les-auteurs/4-etablir-l-ordre-des-coauteurs),

” Recommandations pour la conduite, la présentation, la rédaction et la publication des travaux de recherche soumis à des revues médicales “; Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals – ICMJE – http://www.icmje.org

• Accepting that staff members can say they do not know how to do what is asked of them, and encouraging the acquisition of the skills that are lacking. Valuing honesty,
• Setting up tutorials or mentorships for new members (support from technical referents and/or trusted people who can answer questions, including questions on scientific integrity),
• Using MDAR tools to improve transparency: MDAR framework (Materials Design Analysis Reporting) Framework for transparent reporting in the life sciences, Macleoda M et al, PNAS 2021 Vol. 118 No. 17 e2103238118 (https://www.pnas.org/content/118/17/e2103238118),,
• Setting up a “team DMP” i.e. a common data management method so that all team members use the same data management mode (https://anr.fr/fileadmin/documents/2019/ANR_DMP_Template_EN.pdf).

The absence of clear and precise rules, or the failure to communicate these rules, is responsible for considerable conflict within research teams and units. Establishing clear rules means explaining the rules in force in the laboratory in didactic manner, relating to laboratory notebooks, publication, supervision, conflicts of interest, communication, archiving, valorization, promotion, etc.To do this:

• Set up tools to encourage communication within your team(s): meetings, in-house seminars, social events (see https://www.ofis-france.fr/the-french-charter-of-ethics-for-research-professions/),
• Disseminate information and conclusions from meetings (via minutes),
• Organize information meetings,
• Organize separate scientific meetings at team and unit level,
• Propose committees and groups to brainstorm, monitor and listen to organizational issues such as information and communication,
• Ensure that staff can express their views on any subject,
• Encourage the registration of protocols on platforms, in line with international best practice recommendations (ex: Open Science Framework, https://osf.io/).

The best training is often delivered by example. You can also:

• Encourage team members to attend training courses on scientific integrity, quality, best practices (some MOOCS sites (https://www.fun-mooc.fr/fr/cours/integrite-scientifique-dans-les-metiers-de-la-recherche/), or: regional scientific and technical training units (https://urfist.univ-toulouse.fr/ressources/ethique-et-int%C3%A9grit%C3%A9-de-la-recherche)). This investment should be included in the evaluation file. You can communicate outside the team about the “honest practices” approach adopted within the team. Be proud of it.
• Offer “institutional” training courses: for example, including a “responsible research” theme in the “welcome days” for new members (researchers, team leaders).

Disseminate information on scientific integrity.

Results that inspire confidence are results whose trustworthiness cannot be called into question. To achieve this:

• Ensure that results are reliable, honest, and respect ethical principles and deontology (https://www.inserm.fr/en/our-research/good-practices-at-inserm/),
• Promote the transparency of results and the sharing of raw data through open science (see the dot “Promoting open science”),
• Accept negative results (see the dot “Implementing a policy for valuing negative results”),
• Establish a climate of trust, helping to coordinate work within the team and the unit, assuming your responsibilities as a manager – these are actions that will combine to enable you to enhance the quality and therefore the impact of the results produced by your team or laboratory.

As Nobel Prize-winning economist Ronald H Coase wrote, “if you torture data long enough, it will admit to everything”. These inappropriate uses underpin the reproducibility crisis that the research world is facing. Resisting “torturing” the data can sometimes be difficult in this context (for instance, removing certain data, multiplying numbers of analyses, selecting the findings to be presented, and many other strategies). Our research environment, which overvalues originality, sometimes – more or less voluntarily – encourages this distortion of data. A few references will help you to gain a better understanding of this subject and to envisage valid solutions :

https://s4be.cochrane.org/blog/2021/06/25/what-is-data-dredging/

Andrade C. HARKing, cherry-picking, P-hacking, fishing expeditions, and data dredging and mining as questionable research practices. J Clin Psychiatry. 2021;82(1):20f13804. https://s4be.cochrane.org/blog/2021/06/25/what-is-data-dredging/ ; https://www.psychiatrist.com/jcp/assessment/harking-cherry-picking-p-hacking-fishing-expeditions-and-data-dredging-and-mining-as-questionable-research-practices/

https://www.psychiatrist.com/jcp/assessment/harking-cherry-picking-p-hacking-fishing-expeditions-and-data-dredging-and-mining-as-questionable-research-practices/

A video illustrating the risks associated with the proliferation of statistical tests (p-hacking) : https://www.youtube.com/watch?v=i60wwZDA1CI

Although they are an essential component of research work, most negative results are not published or disseminated. However, the fact that a result does not validate the hypothesis set out at the start does not mean that this result is unimportant and should disappear from the scientific community (Highlight negative results to improve science, Nature 2019).

Numerous initiatives have been launched to valorise negative results (https://www.datacc.org/bonnes-pratiques/diffuser-des-resultats-negatifs/les-donnees-negatives-la-partie-immergee-de-liceberg-des-publications-scientifiques/#part-6).

Do not hesitate to publish negative results, in the same way as positive results, in preprints (BioRxiv.org; medrxiv.org). In addition, several journals have implemented editorial policies to facilitate the publication of negative results. You will find useful information on the websites:

• Journal of Negative Results : http://www.jnr-eeb.org/index.php/jnr
• PlosOne: https://collections.plos.org/collection/missing-pieces/
• BMC research : https://bmcresnotes-biomedcentral-com.proxy.insermbiblio.inist.fr/articles/collections/negative-results?gclid=CjwKCAiA5IL-BRAzEiwA0lcWYsEyvxIRKKzpQJd34OFtQuorz_c1rFjtx0Naps4CCbLsBpOeVVNfIRoC3XEQAvD_BwE
• ACS Omega : https://acsopenscience.org/open-access-journals/omega/)
• F1000Research : https://f1000research.com/
• L’American Heart Association : https://www-ahajournals-org.proxy.insermbiblio.inist.fr/doi/10.1161/CIRCOUTCOMES.120.007448
• The All Results Journals: Biol : http://www.arjournals.com/index.php/Biol

Registered reports are a new publishing format that you can promote in your teams. With this format, peer review of the article occurs before the study is carried out. It takes into account the relevance of the research question and the quality of the methodology proposed. Acceptance in principle is then given, regardless of the outcome, provided the methodology planned has been respected. You can encourage the use of this format (https://www.cos.io/initiatives/registered-reports).

In the face of environmental challenges, it is the responsibility of team and unit managers to encourage the implementation of an environmentally-responsible research strategy. On the LORIER website, you can find information on how to raise awareness among staff and train them in environmental issues (see “Training in environmental issues” section of the “Research staff” section), on how to assess the current situation and share experiences, as well as practical advice to guide the actions to be implemented.