Science without stats

Would any of us argue that it is important for many of the experimental scientists to have a basic training in statistics? I admit, I haven’t done much statistics myself, but if I was working in horticulture or biomedicine, I think it would be a no brainer that management of data would require a rudimentary understanding of statistics. I have been fortunate to have learnt from my colleagues the utility and applicability of statistics, and I have recently had the pleasure of watching Hans Rosling documentaries promoting the significance of statistics in our everyday lives. From this reflection, I am convinced that I am unqualified to carry out an experiment of any kind!

So why is it that at my university, that many of the future scientists will know little statistics beyond what I know?

Below is a list of our science majors. I have separated the units which require statistics when chosen as a first major. I’ve also removed the Maths and Stats, Physics and Engineering Sciences major, as they do statistics in the mathematics unit MATH1002. I am just concentrating on “applied statistics”.

Require statistics

Do not require statistics

There are two other issues which I haven’t discussed because they might need more background than what I write here, but they are:

  1. A student who takes a major as a “second major”, does not need to take its “complementary units”, and so does not need to take statistics at all! So for example, a student might take a major in Commerce and a second major in Agricultural Science, will go on to be an Ag Scientist, but not know any statistics! (By the way, what determines a first and second major is simply an ordering the student chooses).
  2. Only four of the majors above (Applied Computing, Quantitative Methods, Biochemisty and Molecular Biology, Genetics) are required to take OUR statistics unit. The remainder take the unit offered by the School of Agricultural and Resource Economics, which by all accounts, has less statistics covered in it.

10 thoughts on “Science without stats”

  1. The new 2012 structure really leaves me banging my head against the wall (or at least it would, if I had to contend with it).

    The physics major from last year, and previous years, required at least the stats/probability of MATH1020 and MATH2209, as well as a required data analysis module in second year (which was actually more useful than anything from the maths units). I suspect the physics major will still include the data analysis module. So at least for the physicists, all is not entirely lost.

    It is sad that the quality and structure of undergraduate degrees at UWA has severely regressed.

    1. Usually, I would say yes! At UWA, there are two psychology majors:
      – Social and applied psychology (within the Bachelor of Arts)
      – Neuropsychology and Cognitive Science (within the Bachelor of Science)

      Unfortunately, both do not list statistics as a required unit. I would certainly advise taking statistics if you would be interested in the research side of psychology.

      1. I just can’t imagine how statistics can not be a core subject for Psychology. That boggles my mind.

      2. It seems that there is a third year unit taught by the psychologists that covers statistics.

      3. There are three psychology units with “a bit” of stats in them:
        PSYC2203. How much and how rigorous is not yet known. It might simply meaning collation of data in MSExcel and clicking the correlation button. I doubt that linear regression is taught here. The handbook entries of all units are quite vague, and we won’t know the content of these units until semester starts.

  2. When the units were being developed, I warned the committee over and over again that a powerful evil force (henceforth known as Voldemort) was intending to put up his own “Stats-lite” unit, swear blind that it was not really a Stats unit and therefore did not violate the rules against duplication of the New Courses, and then make it incompatible with the real Stats units (so that students forced to do the Stats-lite unit by their choice of major would not be able to do real Stats and therefore contribute financially to a non-Voldemort Faculty).

    Unfortunately Voldemort was ON the committee (come on down “conflict of interest”) and the rest of the committee, despite some valiant individual efforts, was too feeble to stand up to him.

    Sadly, our university is too pathetic to uphold its own rules.

    1. There might be some irony here which Voldemort failed to recognise. The university has a policy on duplication of unit content. So “Stats-lite” will need to be “Stats-Zero”. Only four of the science majors will have a credible background in statistics. I for one will be wathcing “Stats-lite” very closely, I might even attend a few lectures.

      1. The unit has been labelled incompatible with STAT1400 and STAT1520, so one can only assume that the promises made about it that it would not overlap and that it only contained very little Stats were false.

        Policies are irrelevant if nobody has the balls to enforce them.

  3. This is a serious business of wider significance. One of the things that most concerned the recent International Review of Mathematical Sciences in the UK was the fact that small statistics groups are seriously threatened at the moment. The panel considered that access to good statistics training and expertise is essential for all parts of science, and that given how poorly taught the subject is at school, it is vital that every university should provide access to such training and expertise. A couple of pages of the report were devoted to statistics. I don’t think they envisaged this kind of threat, though.

    My own discussion of the report is at . I was horrified to find that the link to the report on the EPSRC webpages no longer works; EPSRC have replaced it by their forthcoming response. I don’t know where to refer to for the original report now. (When I made that posting, the discussion was hijacked by another issue the report raised, improving the numbers of women in mathematics, and some of the important issues I wanted to raise may have gone unnoticed.)

    Incidentally, wish us luck here at QM: we are shortly interviewing for a professor of statistics, who hopefully will give strength and leadership to our somewhat embattled statistics group. (But it is difficult to appoint academic statisticians at present, since the rewards outside academia are significantly better.)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s