Research Culture: The SAFE Labs Handbook as a tool for improving lab culture
- Institute of Neuroscience, Consiglio Nazionale delle Ricerche (CNR), Italy
- Centre for Chromosome Biology, School of Biological and Chemical Sciences, University of Galway, Ireland
- Department of Psychiatry and Gerald Choa Neuroscience Institute, The Chinese University of Hong Kong, Hong Kong
- Institut Jules Bordet, Université libre de Bruxelles, Belgium
- Institute of Biology and Bernstein Center for Computational Neuroscience Berlin, Humboldt University of Berlin, Germany
- Neuroscience Department, Scuola Internazionale Superiore di Studi Avanzati, Italy
- Biologie Intégrative des Organismes Marins (BIOM), CNRS and Sorbonne Université, France
- Department of Psychiatry, Amsterdam UMC, University of Amsterdam, Netherlands
- Center for Cancer Research, Medical University of Vienna, Austria
- Institut de Neurobiologie de la Méditerranée (INMED), INSERM and Aix Marseille Université, France
- Institute of Neuroscience, Consiglio Nazionale delle Ricerche, Italy
- Center for Neuroscience and Cognitive Systems, Italian Institute of Technology, Italy
- Cell and Developmental Biology Department, University College London, United Kingdom
Figures
Figure 1
The commitments in the SAFE Labs Handbook.
The SAFE Labs Handbook logo (left) and summary of its commitments (right), arranged by category. Matrix indicates (black) whether each commitment should be publicly documented, internally documented, or established. The complete handbook, along with explanations of each commitment and example text, can be found here: https://safelabs.info/home/safe-labs-handbook/.
Figure 2 with 1 supplement
Survey: demographics of respondents.
(a) Number of respondents to the SAFE Labs Handbook survey, sorted by employment role. (b) Fraction of respondents as a function of seniority in each role from (a, color). Black line indicates the average across all respondents. (c) Number of respondents sorted by research field. (d) As in (c), but by country of employment. For all panels, n=105 group leaders, 57 postdocs, 46 PhD students, and 16 staff.
Figure 2—figure supplement 1
2024 SAFE Labs Workshop demographics and feedback.
(a) Preference score for proposed SAFE Labs workshop topics, averaged across all applicants, sorted from the least preferred to the favourite. Only topics and formats with high rating (red dotted line) were chosen. Error bars mean ± s.e.m, n=21 (b) As in (a), but for the proposed formats of discussion at the workshop. (c) Percentage of selected workshop applicants (n=9) as a function of country of employment. (d) As in (c) for research field. (e) Evaluation of the workshop outcome, averaged across participants: range from 1 (below expectation) to 5 (exceeded expectations). (f) Evaluation of workshop feedback statements, averaged across participants: range from 1 (strongly disagree) to 5 (strongly agree). All participants indicated a high rating (i.e. above the red dotted line). (g) Rating of the workshop discussion topics, averaged across participants: range from 1 (poor) to 5 (excellent). (h) Fraction of participants in favour of publicly documenting vs internally documenting each SAFE Lab Handbook commitment. Divisive commitments (within dotted interval) were re-assessed in the public survey (n=12). For panels e–g, n=13.
Figure 3 with 2 supplements
Survey: different response from group leaders and lab members.
(a) Distribution of mean importance score across commitments, for all survey respondents (black), and for each career stage: group leader (blue), postdoc (orange), PhD student (purple), and staff (yellow). Violin plot shows median (white dot) and range. All career stages rated the handbook commitments as significantly important (i.e. above the grey dotted line, one-sample t-test, ** = p < 0.001). The importance scores from group leaders were significantly lower than lab members (Mann–Whitney U test, ** = p < 0.001). (b) Mean importance score for each commitment and career stage, sorted in ascending order based on group leaders’ answers. (c) As in (a), but for the implementation rate. All career stages reported low levels of implementation, significantly below 50% in most cases (grey dotted line, one-sample t-test, ** = p < 0.001). The implementation rate indicated by group leaders was significantly higher than lab members (Mann–Whitney U test, * = p < 0.05). (d) As in (b), but for the implementation rate. (e) Relationship between mean importance score (x-axis) and mean implementation rate (y-axis) for each commitment as scored by group leaders. The significant correlation was captured by a robust linear fit (black, dotted 95% confidence intervals, P<0.001) (f) As in (e), but for postdocs (P>0.05). (g) As in (e), but for PhD students (P>0.05). (h) As in (e), but for staff (P>0.05). (i) Relationship between difference in importance score between group leaders and lab members (x-axis), and the global implementation rate across all groups (y-axis). The significant correlation was captured by a robust linear fit (black, dotted 95% confidence intervals, P<0.001).
Figure 3—figure supplement 1
Survey results were comparable across countries.
(a) Distribution of importance rating averaged across commitments for respondents sorted by the five most numerous employment countries: UK (blue), Italy (orange), France (yellow), Germany (purple), Belgium (green). All countries rated the commitments as significantly important (i.e. above grey dotted line, one-sample t-test, * = p < 0.01) but were not significantly different from each other (Kruskal-Wallis test, P>0.05) (b) Average importance score for each commitment, sorted as in Figure 4b, for the five main respondent countries. (c) As in (a), but for the implementation rate. Respondents from all countries reported similarly low levels of implementation (Kruskal-Wallis test, P>0.05), significantly below 50% (Gray dotted line, one-sample t-test, * = p < 0.01). (d) As in (c), but for implementation rate. (e) Relationship between mean importance and mean implementation rate for each commitment as scored by group leaders (n=20) from the UK. The significant correlation was captured by a robust linear fit (black, dotted 95% confidence intervals, P<0.001) (f) As in (e), but for Italy (blue, P<0.05, n=10). (g) As in (e), but for France (yellow, P<0.001, n=19). (h) As in (e), but for Germany (purple, P<0.001, n=9). (i) As in (e), but for Belgium (green, P<0.001, n=8).
Figure 3—figure supplement 2
Importance score split by seniority in career stage.
(a) Distribution of importance rating averaged across commitments for group leaders sorted by seniority in their career stage. (b) As in (a), but for postdocs. (c) As in (a) but for PhD students.
Figure 4 with 2 supplements
Support for the handbook is strong across all respondent groups.
(a) Relationship between mean importance rating of 10 commitments (subset chosen for the paid survey) by PhD students from the paid and unpaid survey. The significant correlation was captured by a robust linear fit (black, dotted 95% confidence intervals, P<0.001). (b) Importance scores for both paid and unpaid respondents were significantly above 3 (one sample t-test), and significantly higher for unpaid than paid (Mann–Whitney U test, P<0.01). n=46 voluntary and 64 paid PhD respondents. (c) Fraction of respondents from each survey (paid and unpaid) in favour of implementing the handbook in their environment. For all panels, n=105 group leaders, 57 postdocs, 46 PhD students, 64 paid PhD students and 16 staff.
Figure 4—figure supplement 1
Support for public documentation of selected commitments.
(a) Fraction of respondents in favour of publicly documenting vs internally documenting selected commitments. n=105 group leaders, 57 postdocs, 46 PhDs, and 16 staff.
Figure 4—figure supplement 2
Support for handbook implementation divided by subgroups.
(a) Fraction of respondents for the main survey in favour of implementing the handbook in their environment split by different countries. (b) Fraction of group leader respondents (n=105) for the main survey in favour of implementing the handbook in their environment, split by seniority. (c) As in (b) but for postdocs (n=57). (d) As in (b) but for PhD students (n=46).
Additional files
-
Supplementary file 1
SAFE Labs 2024 application form data.
- https://cdn.elifesciences.org/articles/108853/elife-108853-supp1-v1.xlsx
-
Supplementary file 2
SAFE Labs 2024 exit survey data.
- https://cdn.elifesciences.org/articles/108853/elife-108853-supp2-v1.pdf
-
Supplementary file 3
Handbook public survey data.
- https://cdn.elifesciences.org/articles/108853/elife-108853-supp3-v1.xlsx
-
Supplementary file 4
Prolific paid survey data.
- https://cdn.elifesciences.org/articles/108853/elife-108853-supp4-v1.xlsx
-
MDAR checklist
- https://cdn.elifesciences.org/articles/108853/elife-108853-mdarchecklist1-v1.pdf
Download links
A two-part list of links to download the article, or parts of the article, in various formats.
Downloads (link to download the article as PDF)
Open citations (links to open the citations from this article in various online reference manager services)
Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)
Research Culture: The SAFE Labs Handbook as a tool for improving lab culture
eLife 14:e108853.
https://doi.org/10.7554/eLife.108853
