[R-sig-eco] Teaching statistics to ecology undergraduates

[Graham Smith]

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