Discussion Document: Future of CANSSI

> français

August 2016

It’s time to start planning for CANSSI’s future. This discussion document outlines some possible options going forward. CANSSI is soliciting comments from statisticians across the country. We welcome your feedback on this very important issue. You can email any member of the Directorate or leave a comment in the box at the bottom of this page. Thank you for your interest in the future of CANSSI.


The Canadian Statistical Sciences Institute is currently funded through the three mathematical sciences institutes. Their NSERC awards include a component of funds in support of CANSSI. In the case of the Fields Institute their contribution also includes a component of their provincial funding.

Beginning with the next funding cycle, 2019–2024, CANSSI will apply to NSERC as an independent entity. This will require a host institution for CANSSI, which would typically be the institution of the Director, who is the PI on the grant proposal.

In planning for the future, and for a change of Director, we see two main options:

  • The host institution “follows the Director”; when the directorship changes, the incoming Director becomes the new PI and CANSSI is hosted at his/her institution.
  • The Director “follows the host institution”; part of the commitment of the host institution is a position for the Director, along with space and administrative support.

To date we have followed the first model, although the mathematical sciences institutes are hosting CANSSI for purposes of the NSERC funding. The institution supplying the time of the Director and the local infrastructure moved from Waterloo to Toronto when Nancy Reid became the Director. An advantage of this model is that the search for the next Director can possibly be more wide-ranging within Canada, as there is no expectation that the Director take up either a fixed term or a permanent appointment at a different location. Another potential advantage is that the virtual, rotating nature of CANSSI may help to keep statisticians across the country engaged. It would require a salary commitment (or teaching release) from the university, unless we were able to raise funds to support the Director’s salary.

The second model may be more stable for the longer term. As CANSSI activity continues to increase, it will become increasingly cumbersome to re-create the necessary administrative procedures at a new location. Visibility of CANSSI, and access to space, seems improved if a host institution makes a long-term commitment to the Institute. The second option also provides the possibility of hiring someone from outside Canada as Director. Access to provincial funds would likely be easier with a long-term commitment from a host institution. There is a risk that once an institution is established as the host, there may be a lack of engagement from units far from the host location. One way to mitigate this is to establish a number of Regional Nodes of CANSSI.

Other models in Canada: 

  • The Canadian Institute of Theoretical Astrophysics is an extra-departmental unit of the University of Toronto, with several salary lines. The Director holds a faculty position in the institute. There is also a national not-for-profit corporation, CITA Inc., associated with CITA.

The three mathematical sciences institutes are associated with their regions, although national (and international) in scientific scope.

  • PIMS is a consortium of its member universities, and is hosted at UBC. The Director and Deputy Director are faculty members of the mathematics department, and have to date been either appointed from within the department or been new external appointments.
  • Fields is a provincial not for profit corporation, with its own facilities. The Director is employed by Fields, and typically has an adjunct appointment in the Department of Mathematics at the University of Toronto. Fields has had Directors from outside U of T, from outside Canada, and from within U of T.
  • CRM is a research centre of the Université de Montréal. The Director has often been a faculty member there, but also has been seconded from other Quebec universities via a teaching buyout.

All these models are variations of the second model outlined in the previous section: “Director follows the host institution”.

Going forward

Under the first alternative, we would likely want to move to incorporate as a national not-for-profit reasonably quickly, so that we have a fixed identity, if not a fixed location.

Under the second alternative, where the Director follows the Institute, a first step would be to circulate a call for expressions of interest from institutions across Canada in becoming the host institution for CANSSI. Depending on the response to this call, we could consider as well circulating a call for expressions of interest in becoming a CANSSI regional node.

Comments are closed.