If, like me, you have a only a few hours (11 hours over 3 years in my case) to try and teach statistics to ecology undergraduates, how do you do it? Any introductory statistics text seems to assume, more time and more mathematical ability than in practice is available. Although, I emphasise graphical techniques and the use of confidence intervals, and how these might help understand the ecological process being looked at, I still spend a large chunk of precious time on hypothesis testing. The more I have been thinking about this, and the more I search for a suitable text book, the more I realise how hopelessly confusing the average text book is, with t-tests, anova, manova, ancova, OLS regression, poission regression, logistic regression GLM etc. Yes I know that all of these may well not appear in the average introductory text. I have reached the stage where I am wondering whether I should just teach GLM. This would give the students a single flexible method capable of tackling a wide range of ecological problems. It would also,I think, provide a better framework for approaching ecological questions than simple hypothesis testing. I admit, that this email is really just me thinking out loud, but does anyone who teaches statistics to ecologists, or indeed anyone at all really, have any views about how best to spend my 11 hours (which I may be able to increase 13 hours). I should point out that at the moment I also spend some of this time on good practice in data management, a bit on scientific method, and a bit on the importance of random sampling, but nothing really on experimental design. Graham
Teaching statistics to ecology undergraduates
3 messages · Carsten Dormann, Graham Smith
Dear Graham, 11 hours is short - there's no mistaking. I teach (among other things) a 6 day stats course for beginners, and find that I need the first 3 days to get the student to "think straight". I tried for a couple of years to teach "only" GLM, as you suggested. I "waste" one full day of explaining what a distribution is, what parameters of distributions are and on what ground to suspect data to be derived from a certain distribution. That would be at least 5 of your 11 hours. The next half day goes into explaining (and running examples) on likelihood and its maximisation. It is a good way to start, I find, and eventually students are very comfortable using glm rather than aov and friends. Using only GLM is clear (and Ben Bolker's book sets the right tone, albeit at a much too high level for beginners). At the same time, the learning curve is VERY steep. 30% of the participants fall by the wayside. Is that acceptable? Maybe it is me, not the GLM. However, I think you have to be very realistic about what you can achieve (and I have heard speaking highly of your courses, so I am sure you are doing the right things). Giving the students a "feeling" about what the idea of a "fit" is and what is behind comparisons of samples is rather independent of distributional assumptions and a very general point they can take away from a short course. Also, as you said, visualising the data, getting a feeling for it, is SO important, particular when a student has little idea what to expect from an experiment/observation. I my little 6 day course, I spend roughly 2 days on introducing R, distributions and (maximum) likelihood (half of this time the participants run examples). Another 2 days are devoted to multiple regression (going wildly through different distributions to make them comfortable with GLM) and issues such as collinearity and model selection. Then I throw in a day of design of experiments (randomised block, nested, split, survey design, stratification, sample size estimation) and run some simple (?) mixed models to illustrate the practical problems attached to DOE. The final two days we run largish examples (such as Harrell's Titanic data set), touch very superficially multivariate methods (PCA, CA and CCA) and end up with some miscellaneous issues such as randomisation and bootstrapping. If I had to reduce it to 11 hours: Unless the students are likely to do experiments (which seems to have fallen out of funding), I would ditch DOE and focus on GLM plus a few sexy but tricky examples. I love the Titanic study, because you can get the students to identify with the passengers. If that leads them to transfer their newly gained knowledge to the ecological work is a different question. If you additionally make the buy a good book (I always recommend Quinn & Keough, having myself been "raised" on Sokal and Rohlf and always hated it, because it never addressed my type of non-Gaussian problems) I think they should be set up for the next level. I shall stop now (and prepare some stats course next week), otherwise I would also have a word to say about Crawley's approach, which I find enchanting and confusing. Carsten
Graham Smith wrote:
If, like me, you have a only a few hours (11 hours over 3 years in my case) to try and teach statistics to ecology undergraduates, how do you do it? Any introductory statistics text seems to assume, more time and more mathematical ability than in practice is available. Although, I emphasise graphical techniques and the use of confidence intervals, and how these might help understand the ecological process being looked at, I still spend a large chunk of precious time on hypothesis testing. The more I have been thinking about this, and the more I search for a suitable text book, the more I realise how hopelessly confusing the average text book is, with t-tests, anova, manova, ancova, OLS regression, poission regression, logistic regression GLM etc. Yes I know that all of these may well not appear in the average introductory text. I have reached the stage where I am wondering whether I should just teach GLM. This would give the students a single flexible method capable of tackling a wide range of ecological problems. It would also,I think, provide a better framework for approaching ecological questions than simple hypothesis testing. I admit, that this email is really just me thinking out loud, but does anyone who teaches statistics to ecologists, or indeed anyone at all really, have any views about how best to spend my 11 hours (which I may be able to increase 13 hours). I should point out that at the moment I also spend some of this time on good practice in data management, a bit on scientific method, and a bit on the importance of random sampling, but nothing really on experimental design. Graham
_______________________________________________ R-sig-ecology mailing list R-sig-ecology at r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
Dr. Carsten F. Dormann Department of Computational Landscape Ecology Helmholtz Centre for Environmental Research UFZ Permoserstr. 15 04318 Leipzig Germany Tel: ++49(0)341 2351946 Fax: ++49(0)341 2351939 Email: carsten.dormann at ufz.de internet: http://www.ufz.de/index.php?de=4205
Carsten
11 hours is short - there's no mistaking. I teach (among other things) a 6 day stats course for beginners, and find that I need the first 3 days to get the student to "think straight".
It is useful to know that you had 6 days
Using only GLM is clear (and Ben Bolker's book sets the right tone, albeit at a much too high level for beginners). At the same time, the learning curve is VERY steep. 30% of the participants fall by the wayside. Is that acceptable? Maybe it is me, not the GLM.
I suspect that regardless of what you do, a percentage are just never going to get it.
However, I think you have to be very realistic about what you can achieve (and I have heard speaking highly of your courses, so I am sure you are doing the right things).
My guess is that you are speaking of the Highstat courses, which I'm not actually involved with, I'm asking about the basic undergrad intro to stats lectures.
Giving the students a "feeling" about what the idea of a "fit" is and what is behind comparisons of samples is rather independent of distributional assumptions and a very general point they can take away from a short course.
My main aim, I think, is that they gain some understanding of statistical thinking rather than specific techniques, which most will forget almost immediately.
Also, as you said, visualising the data, getting a feeling for it, is SO important, particular when a student has little idea what to expect from an experiment/observation.
Yes, I strongly agree with this.
If I had to reduce it to 11 hours: Unless the students are likely to do experiments (which seems to have fallen out of funding), I would ditch DOE and focus on GLM plus a few sexy but tricky examples. I love the Titanic study, because you can get the students to identify with the passengers. If that leads them to transfer their newly gained knowledge to the ecological work is a different question.
In my case, the stats is part of broader module, which focusses on a woodland ecology study, where the students collect vegetation data, and carry out some soil analysis. I introduce the stats, they then complete a stats exercise, I give them feedback on this, and they are then meant to apply the stats to help explain the vegetation characteristics found in the two woods they sampled.
the buy a good book (I always recommend Quinn & Keough,
Yes, a wonderful book.
I shall stop now (and prepare some stats course next week), otherwise I would also have a word to say about Crawley's approach, which I find enchanting and confusing.
:-) Graham
Carsten Graham Smith wrote:
If, like me, you have a only a few hours (11 hours over 3 years in my case) to try and teach statistics to ecology undergraduates, how do you do it? Any introductory statistics text seems to assume, more time and more mathematical ability than in practice is available. Although, I emphasise graphical techniques and the use of confidence intervals, and how these might help understand the ecological process being looked at, I still spend a large chunk of precious time on hypothesis testing. The more I have been thinking about this, and the more I search for a suitable text book, the more I realise how hopelessly confusing the average text book is, with t-tests, anova, manova, ancova, OLS regression, poission regression, logistic regression GLM etc. Yes I know that all of these may well not appear in the average introductory text. I have reached the stage where I am wondering whether I should just teach GLM. This would give the students a single flexible method capable of tackling a wide range of ecological problems. It would also,I think, provide a better framework for approaching ecological questions than simple hypothesis testing. I admit, that this email is really just me thinking out loud, but does anyone who teaches statistics to ecologists, or indeed anyone at all really, have any views about how best to spend my 11 hours (which I may be able to increase 13 hours). I should point out that at the moment I also spend some of this time on good practice in data management, a bit on scientific method, and a bit on the importance of random sampling, but nothing really on experimental design. Graham
_______________________________________________ R-sig-ecology mailing list R-sig-ecology at r-project.org https://stat.ethz.ch/mailman/listinfo/r-sig-ecology
-- Graham M Smith graham.smith at myotis.co.uk Station Cottage, Station Road Binegar, Somerset BA3 4UQ