Statistics Clinic

The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.

— John W. Tukey (the first of six “basics” against statistician’s hubrises) in “Sunset Salvo”, American Statistician, 40, 72–76 (1986).

Links to various statistics resources that you might find useful can be found at http://stats-clinic-resources.wikidot.com/

What do we offer

We offer a free statistical consulting service to researchers in the university through our clinic. Our idea is to keep the threshold to participation as low as possible: all researchers are invited to attend, regardless of their fields. We aim to provide anything from statistical advice to new methodology development. We will also be interested in discussing potential academic collaborations and possible joint projects with our clients.

new banner

We can:

  • Help our clients understand how to formulate their statistical problems in a more precise and meaningful way.
  • Provide advice on different kinds of statistical models that can be used in different circumstances and being wary of common pitfalls.
  • Point our clients to books, papers and software packages that might help them with their work.
  • Provide advice on experimental design before data is collected (this is likely to result in a more successful route to addressing questions).
  • Show clients how to apply appropriate statistical tools to highlight the most useful information from the data.
  • Put our clients in contact with other researchers in Bristol and beyond, who specialise in their particular type of problem.

How to participate

The Statistics clinic is OPEN – but only for remote consultation at present. Do not come to the School of Mathematics to visit the Statistics Clinic. Instead, please email a short description of the statistical problem you seek help with, to statistics-clinic@bristol.ac.uk.
This should be not more than 3 or 4 sentences in length, it will be used only to assign your request to a suitable consultant. We will ask for more detail as necessary, and get in touch with you electronically in due course. Please clearly state your email address, phone number, and, if you have one your Skype ID.

How we normally organise the clinic – temporarily suspended

Simply turn up at one of our drop-in sessions – there is no booking system. We will deal with all queries within the session; we do not have resources to do follow-up work after the session. Prepare for the meeting by collecting your thoughts – come prepared to give a 2 minute oral description of your scientific problem, in accessible language. Give (only) enough background to your problem to set the context of the question. Bring a laptop if you need a computer to explain the stage you have reached in your analysis, but be aware that we may not be expert in the particular software you have used. Graphs displaying the data are more useful than huge tables. Ideally, install R on your laptop, and bring along your data already loaded into R.

Who can participate?

All researchers in the university – research students, research and academic staff. Research students and post-docs must inform their advisers if attending; in fact we encourage students to come along with their advisers. Sorry, we are not able to help taught-course students.

What we expect in return

We hope that sometimes our discussions might evolve into genuine scientific collaboration, but in many cases a small acknowledgement in your paper of our assistance (if there is any) will be fine.

picture

Who are we?

Academic and research staff and research students in the Institute for Statistical Science in the School of Mathematics.

Do our clients find our advice helpful?

We hope so; there is some feedback at http://www.maths.bris.ac.uk/~mapjg/ClinicFeedback.html

How to contact us

You can email statistics-clinic@bristol.ac.uk, for example if you are unsure if your query is appropriate.

Where else can you get advice on your data problems?

If you are looking for consultation on other data related matters (not necessarily statistical) or your question is statistical, but the timeslot we offer is not suitable, it may also be worth getting in touch with ‘Ask JGI’ (ask-jgi@bristol.ac.uk). This is a general data science support service offered by the University through the Jean Golding Institute. The clinic and Ask JGI work closely together so if you are unsure who to turn to feel free to contact either and you will be directed to the right place.

 

Windows image

 

 

css.php