Density comparison

[riccardo...@gmail.com]

Good morning everyone, here I am again with a doubt about the ds results.

I am doing a study on the densities of a migratory species in two protected areas using a thermal imaging camera.

I am wondering if once the densities of the two areas are obtained, by entering the study area as a covariate in the analysis, the values are also compared with each other to assess if there is a statistically significant difference. Or if there is any way through the package to do this. If not, would Pearson's correlation test be a good way to make this comparison?

And finally, would comparing the two density values with each other make sense? Or would it be a superfluous analysis?

Thank you very much and have a nice day

Translated with www.DeepL.com/Translator (free version)

[Eric Rexstad]

Riccardo

Making a statistical comparison of densities between two areas (or two points in time for the same area) is a common analysis to perform.  It is understood that two areas are unlikely to have exactly the same density of the species in question; but the question is whether the two densities differ by an amount unlikely to have arisen by chance.

Traditional ways to make such comparisons are presented in Section 3.6.5 of Buckland et al. (2001) Introduction to distance sampling (Estimating change in density). The book includes formulas, but those formulas have not been incorporated into R code.

A few years ago, I wrote an R function to carry out the analyses from Section 3.6.5. It was extended to take a non-parameteric approach, based upon bootstrapping to make the comparison. The R code and an example of its use is available on our website of case studies:

http://examples.distancesampling.org/differences/differences.html

Detecting density estimate differences Using the bootstrap to derive the sampling distribution of pairwise differences in estimated density. examples.distancesampling.org

---

From: distance...@googlegroups.com <distance...@googlegroups.com> on behalf of riccardo...@gmail.com <riccardo...@gmail.com>
Sent: 10 May 2023 15:22
To: distance-sampling <distance...@googlegroups.com>
Subject: [distance-sampling] Density comparison

 

--
You received this message because you are subscribed to the Google Groups "distance-sampling" group.
To unsubscribe from this group and stop receiving emails from it, send an email to distance-sampl...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/distance-sampling/b7991640-1b66-4c9f-ad53-68872ff7d03dn%40googlegroups.com.

[riccardo...@gmail.com]

Hi Eric,
Wow! The function is perfect for the type of analysis I need to carry out. I will try to use it right away.

Congratulations and thank you.

Il giorno mercoledì 10 maggio 2023 alle 16:31:06 UTC+2 Eric Rexstad ha scritto:

Riccardo

Making a statistical comparison of densities between two areas (or two points in time for the same area) is a common analysis to perform.  It is understood that two areas are unlikely to have exactly the same density of the species in question; but the question is whether the two densities differ by an amount unlikely to have arisen by chance.

Traditional ways to make such comparisons are presented in Section 3.6.5 of Buckland et al. (2001) Introduction to distance sampling (Estimating change in density). The book includes formulas, but those formulas have not been incorporated into R code.

A few years ago, I wrote an R function to carry out the analyses from Section 3.6.5. It was extended to take a non-parameteric approach, based upon bootstrapping to make the comparison. The R code and an example of its use is available on our website of case studies:

http://examples.distancesampling.org/differences/differences.html

Detecting density estimate differences

Using the bootstrap to derive the sampling distribution of pairwise differences in estimated density.

examples.distancesampling.org

---

From: distance...@googlegroups.com <distance...@googlegroups.com> on behalf of riccardo...@gmail.com <riccardo...@gmail.com>
Sent: 10 May 2023 15:22
To: distance-sampling <distance...@googlegroups.com>
Subject: [distance-sampling] Density comparison

 

Good morning everyone, here I am again with a doubt about the ds results.

I am doing a study on the densities of a migratory species in two protected areas using a thermal imaging camera.

I am wondering if once the densities of the two areas are obtained, by entering the study area as a covariate in the analysis, the values are also compared with each other to assess if there is a statistically significant difference. Or if there is any way through the package to do this. If not, would Pearson's correlation test be a good way to make this comparison?

And finally, would comparing the two density values with each other make sense? Or would it be a superfluous analysis?

Thank you very much and have a nice day