Nonparametric test can be used

I think we will need a little more information before a recommendation can be made. How many replicates did you do with each species category? What is the specific nature of the percentage data? For example, did you collect proportion of time each individual fed, or the percentage of individuals that fed per tank or group? There are some consideration before you jump into a linear regression with proportion data. You can read a little bit about it here as a start. https://stats.stackexchange.com/questions/284843/percentage-as-dependent-variable-in-multiple-linear-regression

Good luck!

Kindly check http://www.ebd.csic.es/documents/240051/0/A+simple+statistical+guide+when+data+are+constrained+due+to+practical+or+ethical+reasons.pdf/f818b09d-0042-4d4c-9849-567fc0ca0782

Also check http://biolozi.bio.bg.ac.rs/attachments/article/1250/Predavanje%202%20-%20Dizajniranje%20bihevioralnog%20eksperimenta%20i%20analiza%20podataka.pdf

The information is not sufficient to decide which test can be used if same hypothetical data ll be made available then it ll be easy.

To Mr.

Blake Carlton Jones Mubashir Ali Rather and Chinaza Godswill Awuchi

this is detail of my research, My replications are 8 so i have three treatments first treatment Bellamya only, second only semisulcospira and third Bellamya and Semisulcospira. Each box consist 10 individuals but in mix treatment have 20 individuals (1:1). No food available as I want to know the activity during winter (snow condition). I found some characters such as burring themselves, climbing, crawling and grooming other snails shell. my observation once a week as i do not want disturb the condition during observation such as removing snow activity frequently. could you suggest it? Thank you so much, very appreciate with your answer –

To do a statistical analysis of an animal's behavior, one can do Applied Animal Ethology, which involves using knowledge and methods from fundamental ethology to study animal behaviors expressed in various situations.

Non Parametric tests will suffice.

It is very simple and easy if you need any help contact me on 9906594488.

I think Sir, it will always depend on what kind of animal behavior will be your parameter to identify the statistical test to be used. Thank you.

Behavioural data most of the times follows N-P test

Dear colleague

with the info you provided I suggest to do analysis in each week of observation, unless you are interested in time effect. every activy could be analized individually. Simple anova is possible if you have a normal distribution. check the animals density per box, also look the Petit method or animal behavior.

regards,

Redimio