I am currently doing a project for my Bachelor degree, for which I study vigilance in birds. Obviously, I have to analyse the data statistically. However, I do not know where to begin since I do not know which test to chose.

I am interested in how vigilance varies following a disturbance. I recorded the following data during my field work :

Environmental parameters (such as weather, group size, temperature, ...)

Vigi_before (% of time spent vigilant per minute just before the bird is disturbed)

Duration_vigi_1 (Time spent vigilant during the first minute following the disturbance)

Freq_vigi_1 (Frequence of vigilance events during the first minute following the disturbance)

Duration_vigi_2 (Same as Duration_vigi_1, but during the SECOND minute following the disturbance)

Freq_vigi_2 (Frequence of vigilance events during the SECOND minute following the disturbance)

... And so on until Freq_vigi_4

Here is an overview of my table

enter image description here

EDIT : My sample size would be about 250 observations.

Here are my first data

enter image description here

My questions are the following :

  1. Is vigilance influenced by the weather (or temperature, bgroup size, etc ...) ?

  2. Is post- disturbance vigilance influenced by the time spent vigilant before disturbance ? (eg : Birds that are highly vigilant before the disturbance could be more vigilant after disturbance than those that are less vigilant before disturbance).

  3. How does post-disturbance vigilance vary in time ? (Eg : We could expect to see the time spent vigilant after the disturbance to decrease over time).

After reading the literature, I saw that most people use Generalised Linear models or Mixed models. However, I do not know if it could be really relevant in this case, since I do not really know where to begin.

My question for you is the following :

Which test(s) do I have to chose to answer to these different questions ? (Precision : I work with R)

  • $\begingroup$ Can you please add some more information? What is your sample size? Maybe you could include a (link to) (some part of) the data, and at least replace the image file of the data with a table with actual numbers (that is, as text, not as image). And: Start out with some simpåle methods, such as visualization, aybe show us some plots. $\endgroup$ – kjetil b halvorsen Feb 17 at 1:16
  • $\begingroup$ Done ! But what kind of plots do you want to see ? $\endgroup$ – Kalia Feb 17 at 10:38
  • $\begingroup$ You seem to have repeated measurement data—for each bird, %vigilance before, and after 1,2,3,4 minites. You could plot such profiles, uniting points corresponding to each bird with a line, for instance. $\endgroup$ – kjetil b halvorsen Feb 17 at 10:46
  • $\begingroup$ That's why I tried to do, but I have no idea which code I could use. Iw ould like to combine to create a plot with the amount of time spent vigilant 1,2,3 and 4 min after disturbance on the same graph, but I definitely do not know how to do it... Any advice ? $\endgroup$ – Kalia Feb 19 at 21:58
  • $\begingroup$ Furthermore, do you know which test I could run to investigate if there is a difference between the amount of time spent vigilant 1,2,3 and 4 min after disturbance..? $\endgroup$ – Kalia Feb 19 at 21:59

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.