Member Login



Auto-login for future visits

Join or Renew Today!

Membership Benefits:

Close Button



Sam Williams, PhD | Nov 18, 2006


Now I’m into my third month back in the UK I am finally getting used to the notion of being indoors virtually all the time! This adaptation is of course helped by the fact that the UK is getting colder and that sitting outside just to have lunch in natural daylight almost leads to a reduced core body temperature.

All the observations I made during the season have now been transformed into rows and columns of numbers. This is data! The thing is, now I have to do something with my data, so I’ve ventured into the university library and have begun reading up on statistics. I can assure you it is as bad as you imagine!

But! What am I trying to show with my data? And why do I really need to get involved with statistics? Well… On Bonaire I collected information on numerous habitat characteristics at over 70 locations. The kind of things I measured include: the height of the vegetation, the number of plant species, the distance from nest cliffs or human disturbance as well as recording the bird species that were found at each of those locations. From this I can ask questions like:

Do the parrots prefer areas with tall trees?
Are the parrots found in areas only with high varieties of plant species or are they associated with particular plant/tree species (eg. food trees)?
Do the parrots stay close to nest sites or avoid human disturbance?
Are there habitat types where the parrots are found where there are also reduced numbers of competitor bird species? And in these areas is the parrot’s reproductive success higher?

By looking at the sites the parrots go to and the sites where the parrots are not found then it’s possible to build up a picture of the parrots preferences. This is important for the conservation of the parrot and in fact the whole ecology because this information can be used in land use management and protection.

The statistics I’m trying to understand allow me to put large amounts of the data I have into a computer programme and then ask it to show me which of the many variables are the most important. Then I can go on compare the most important groups of variables with the information on where the parrots go. So maybe height of trees is important for the Amazons on Bonaire but human disturbance is not… Hopefully soon I’ll have a better idea.