Good scientific practice
or a few things we should keep in mind
General principles of scientific practice
Two major guidelines to consider :
- EMBL’s IP 51
• DFG: Recommendations of the Commission on
Professional Self Regulation in Science, Jan. 1998
• DFG: Sachbeihilfen mit Leitfaden für
Abschlussbereichte und Regeln guter
wissenschaftlicher Praxis
• MPI: Good practice 2000
Copies of these documents are on the JGS 2010 web-page.
April, 21st, 2010
Wolfram Meyer-Klaucke
In submitting a proposal to the DFG, you agree to:
1.
adhere to the rules of good scientific practice. In cases of scientific misconduct,
the DFG may impose sanctions. Scientific misconduct is defined as the intentional
and grossly negligent statements of falsehoods in a scientific context, the violation
of intellectual property rights or impeding another person’s research work. The
circumstances of each case will be considered on an individual basis.
Depending on the nature and extent of the misconduct exposed, the DFG may:
•
issue a written reprimand to the person involved;
•
exclude those found responsible from the right to apply for DFG funds for a period
of one to eight years, depending on the severity of the scientific misconduct;
•
revoke funding decisions (completely of partially revoke approvals and demand the
return of authorized funds or the repayment of funds spent);
•
demand that those concerned either retract the publications containing false data,
correct the false data (by publishing an erratum) or include a reference regarding
the DFG’s retraction of funds in the relevant publication;
•
exclude those found responsible from acting as a reviewer or from membership on
DFG committees;
•
deny those responsible the right to vote in DFG elections.
April, 21st, 2010
Wolfram Meyer-Klaucke
DFG: All institutions have to implement 8 rules:
1. Day to day scientific practice
2. Relations with colleagues and cooperation
3. Publication of results
4. Cooperation and leadership within working
groups
5. Guidance of junior scientists
6. Securing and storing of primary data
7. Scientific publications
8. Appointing an ombudsperson
April, 21st, 2010
Wolfram Meyer-Klaucke
Day to day scientific practice
•
precise observance of discipline-specific procedures
for acquiring and selecting data,
•
reliable securing and storage of primary data; clear
and comprehensible documentation of all important
results,
•
the rule of systematic scepticism: openness for doubt,
even about one’s own results and about the results of
one’s own group,
•
a realisation of tacit, axiomatic assumptions:
watchfulness for any “wishful thinking” motivated by
self-interest or other extrinsic factors including
economic, political, religious or moral considerations;
systematic alertness for any possible misinterpretation
or over-interpretation, including overgeneralisation.
April, 21st, 2010
Wolfram Meyer-Klaucke
Relations with colleagues and cooperation
•
•
•
•
•
no hindrance of the scientific work of
competitors, for example by delaying reviews
or by passing on scientific results which have
been acquired in confidence,
active promotion of junior scientists’ scientific
qualifications,
openness to criticism and doubt expressed by
other scientists and team colleagues,
careful, non-self-interested and unprejudiced
assessment of colleagues; avoidance of bias
when appropriate, declaration of both
commercial and scientific conflicts of interest.
April, 21st, 2010
Wolfram Meyer-Klaucke
Publication of results
•
•
•
•
publication on principle of results obtained through public funding
(principle of the public availability of the results of basic research),
publication also of falsified hypotheses in an appropriate manner
and admission of mistakes (principle of a science culture open to
the possibility of error),
strict honesty in the recognition and appropriate consideration of
the contributions of predecessors, competitors and colleagues
(principle of recognition)
distribution to academic researchers of any materials (e.g. cells,
DNA constructs) used in published experiments. Deposition of
sequence, structural, and other relevant information in the
appropriate public database on publication.
April, 21st, 2010
Wolfram Meyer-Klaucke
Cooperation and leadership within working groups
The head of each EMBL Unit, in concert with the Scientific Director, is
responsible for a proper organisation which ensures clear
allocation, (…) monitoring, conflict resolution and quality control
and guarantees that these tasks can in fact be undertaken
effectively.
Reproducibility. The more surprising, but also the more desirable a
result is, the more important it is – as far as is possible with
justifiable expense or effort – that the route to that result be
confirmed within the research group before the results are passed
on to the outside.
Leadership
•
Full knowledge of all relevant circumstances
•
expertise in the field, presence and a broad perspective
•
If no longer possible
–
delegated in such a way that the leadership division remains
manageable and effective
April, 21st, 2010
Wolfram Meyer-Klaucke
Guidance of junior scientists
•
•
training and furthering of junior scientists
guiding them in the observance of the principles of
good scientific practice.
These rules and regulations are integrated in the
training provided by the EMBL PhD programme
and need to be rigorously followed.
April, 21st, 2010
Wolfram Meyer-Klaucke
Securing and storing of primary data
Primary data must, as far as possible, be stored
for at least ten years from the date of
publication on durable, secure carriers in the
institutes or research establishments in which
they arose
Access to the data has to be granted to persons
with a justifiable interest or, when appropriate,
in a public data repository
Full and adequate records must be kept for a
minimum period of ten years as a source of
reference, should the published results be
called into question by others
April, 21st, 2010
Wolfram Meyer-Klaucke
Scientific publications
Publications must adequately describe the results and the methods
used, and give full and correct credit for own and third-party
preparatory work
Any findings which support or call into question the results presented
should equally be made known
Several originators
•
co-authorship requires significant contribution through the design
of the studies or experiments, through critical and previously
unpublished materials or procedures, through working out,
analysing or interpreting the data and through preparing the
manuscript
•
These persons also having agreed to its publication
•
Authors always bear joint responsibility for the content
•
Group leader carries ultimate responsibility
•
All publications that include group members as authors must
therefore be approved by the group leader
April, 21st, 2010
Wolfram Meyer-Klaucke
Appointing an ombudsperson
A person who becomes aware of any significant
indication that scientific misconduct within the
meaning of the catalogue of misconduct (Appendix
1) has occurred has the responsibility of notifying
the Director or the Coordinator of the EMBL Unit
concerned, who in turn must notify the DirectorGeneral in writing.
If the Unit Coordinator is himself/herself implicated in
the enquiry the Director-General should be notified
directly.
April, 21st, 2010
Wolfram Meyer-Klaucke
Can we prevent willful misconduct?
We have to try as hard as possible!
Measures:
• Good lab books
• Reproducible results (check!)
• Retain skepticism
• Discuss surprising results with your colleagues
–
–
•
•
•
•
•
•
•
•
Results are surprising as well, if they fit your expectation to 100%
Check statistics (Tailored numbers can be detected (preference for certain digits))
Check quality of publications prior to starting a collaboration (read publications, do not
trust impact factors)
How many publications can a person write and be responsible of per annum?
Store (raw) data
Explain data analysis in the publication. Some examples:
“Based on these considerations … were varied in the quantitative …analysis”
“Anyhow, typical error margins in … are on the order of …”
“These results are consistent with, but not proof of, the possibility that “
“reveals … in both samples”
 Focus on the long term pay-off
April, 21st, 2010
Wolfram Meyer-Klaucke