Hi all, I'm starting a PhD in biology and my problem is that I know nothing about statistics (I don't even know how to do a student's test).
Could someone please suggest a book, or even better a tutorial on the internet, from which I can learn the necessary stuff? I would ask people in the lab but I'm rather ashamed I don't have a clue about statistical analysis of data.
Thanks!

hmm, I don't know what a student's test is either and I teach statistics :$

The best way to understand statistcs is to clearly understand what you want to do with your data. Are you comparing groups? if so how many groups? or are you looking to predict an outcome with other variables? Or are you looking to understand what themes are key in a questionnaire?

Once you have VERY clear hypotheses about what you are looking to find out you can then go to a stats book to see which test applies. I would recommend Andy Field's 2009 book. It is very clear and easy to understand and gives you a little chart so you know which test to use in which circumstance.

hi Su
I recommend:
Andy field's discovering Statistics and SPSS Ed.2 or 3
SPSS: Version 16 Statistics without Anguish by Coakes and Sheridan
and other books depend on what analysis you're going to do.
But I love these two books.

The one by Coakes and Sheridan is very short and easy but very little explanation.

You can find lots of tutorials online, usually ppt files loaded by professors. just google t-tests and you will definitely find something. But the problem is although you may be able to learn how to do the test, you may not understand FULLY why you have done so (the process) etc.

If I were you, I would eat humble pie and ask my labmates, make friends with them and ask them.
If you want to ask silly quesitons, now is the time.
Learning something by ourselves is one thing, having people explain it to us is another.

Lots of students dont know about statistical analysis, there is nothing to be ashamed about. In fact most of us start from scratch; for example for my masters I did F test without knowing why I did it, I just did it because everybody did it but didnt understand why. I did reasonably well for my undergraduate Biostatistics module but that was to get a good score, only for exam purposes. I just learnt without understanding properly. So I didnt really know. See?

Now I am doing my best to learn it up properly in case during our phd viva, somebody asks us why did we do this and that (about statistical analysis), then we would be able to answer confidently.

good luck and dont give up
satchi

thank you so much sneaks and satchi for your quick answers and recommendations! I 'll definitely try the book.

@satchi: thanks for the encouragement, it really helps to know other people might have had similar problems too.:-)
@sneaks: well, I know what I have to do, the thing is I don't know how. (for example, I now have to compare the means of 3 groups and see if there is a significant difference. I guess that sounds simple, still I don't know how to do it.) I guess I'll have to get the book and read it thoroughly.

another question: assuming I read the theory, and find the right test to analyse my data. How do I actually do the analysis then? Are there specific softwares you use?

hi Su
testing means of 3 groups I think you are talking about ANOVA (but I dont know your research design).

In the past I used online stats programs; like GraphPad and here you can find great links
http://statpages.org/

You can also download PAST here http://folk.uio.no/ohammer/past/download.html

Other programmes like SPSS, MatLab, SAS etc.

Check what your statistic software your university provides; again, ask your labmates (ask your Seniors) what they are using. Best person to go to is your supervisor, see what he is using and learn from there.

Before you run your tests you have to check your data (normality assumptions etc) so again its better quickly grab a book/grab somebody/teacher etc. and start learning soon.

:-) satchi

Su, I sent you a private message about that Discovering Statistics book. It is good, so I hope you find it useful.

thanks satchi!!:-)

Hi Su,

if you have an iPod/iphone with an itunes account, theres a wholes series of lectures from Berkley College- Its under the 'iTunes U' tab.

Im planning on watching them when I get a little closer to analysis.

Good luck!

Andy field is great but I also like to use these books:

Pallant 2007 'survival manual'
Howitt 2008 'intro to statistics in psychology'
Brace 2009 'spss for psychologists'

your best bet is not to rely on one book.
:-)

hmm - I have noticed some quite handy guides on youtube that take you through step by step how to run various tests on SPSS.

As i say the chart in Andy field helps to narrow down what test. It sounds like you need to do an ANOVA of sorts, but depends on whether you have independent groups (male OR female) or you are giving the same group multiple conditions (time 1, time 2) or both!

There is no easy answer, reading the ANOVA chapters in andy field should help you narrow down which one is useful.

Hi Su
Being a biologist and unbelieveably slow to pick up stats I found there to be one book that has quite literally saved my sanity and stopped me pulling out both my eyebrows and the hair from my head (I tend to pull at them when I'm stressed, bad I know :-) ). The Andy Field book is good but I found it a bit psychology heavy and I'm much more reliant on applicable examples to help me understand things so the book I use is called Choosing and Using Statistics: A guide for Biologists. It's about a fifth of the size of most stats books and is a brilliant introduction to what tests you will need for most biology experiments.

thank you all for your interest and advice! :-)