stratification vs post-stratification vs multiple covariates

[Christy Ihlo]

Hello All,



I am new to using distance sampling and I am confused as to the best way to=
proceed with my data analysis. The more I read, the less clear things beco=
me so I thought I could request some feedback.



I am trying to determine density estimates for a variety of east African sa=
vannah animals based on data collected over the past year. There are reside=
nt species and migratory species. Density for migratory species should vary=
by season (wet vs dry) while the density of resident species should remain=
the same. However, detectability varies between the wet and dry season for=
all species. Addtionally, cluster size for some species could also influen=
ce detectability.



I thought it would be best to run separate analyses based on season for the=
migratory species, since detectability and density would vary, but I'm won=
dering if I can somehow stratify the data to return separate detection func=
tions and densities based on seasons. I would still need to account for the=
changes in detectability based on cluster size though. For the resident sp=
ecies, I'm struggling to understand how to incorporate changes in seasonal =
detectability plus clusters into an analysis - stratification, post-stratif=
ication, or multiple covariates - to return a single density estimate. I ha=
ve run analyses using all three for one species and for the most part, I ge=
t generally the same density estimate. However, I have no idea which method=
is the most appropriate. I would certainly appreciate any insight. Feel fr=
ee to email me directly at cm...@duke.edu<mailto:cm...@duke.edu> or christy.i=
h...@duke.edu<mailto:christ...@duke.edu>.



Thank you for tolerating what is probably a repetitive email for this lists=
erve!





Christy M. Ihlo
Master of Environmental Management Candidate
Concentration in Ecosystem Science and Conservation
Nicholas School of the Environment, Duke University
christ...@duke.edu
404.313.7786

--
distance-sampling list home page:
http://www.jiscmail.ac.uk/lists/distance-sampling.html
To unsubscribe, go to the home page and click "Join or leave the list",
or email jisc...@jiscmail.ac.uk with the following message body:
leave distance-sampling
Before posting to the list, read the list guidelines at
http://www.jiscmail.ac.uk/files/distance-sampling/
distance-sampling.welcome
Any questions, email the list owner at
distance-sam...@jiscmail.ac.uk
--