1
$\begingroup$

From my understanding, a paired t-test is used when samples are dependent of each other.

I'm having trouble deciding whether a paired t-test should be used when comparing the average population of a group of animals. Is this data actually independent?

I'm trying to decide if it is better for me to use a two-sample t-test to determine if there is actually a difference in the number of animals in one year compared to another. As the number of animals in one year doesn't necessarily have an effect of animals on another year, I assume that this is independent data and that a two-sample t-test should be used?

$\endgroup$
1
  • 1
    $\begingroup$ Pairing, as it is usually used, would occur here only if it was the exact same animals the two years (and you had the IDs). That is probably not the case? $\endgroup$ Apr 18, 2022 at 14:33

1 Answer 1

0
$\begingroup$

I think a paired t-test is definitely the way to go. For most wildlife animal populations, the abundance in one year will, to some degree, affect the abundance the next year. For example, if population 1 is 400 animals while population 2 is 20,000 animals, you'd expect the abundance of population 2 the next year to be bigger than the abundance of population 1.

Also, as a fellow biologist, always happy to see people diving more into statistics :)

$\endgroup$
1
  • 1
    $\begingroup$ Yes, I believe it is still appropriate. Although t-tests are commonly used when there is a direct intervention, they can just as easily be used to simply compare two sets of numerical values. $\endgroup$ Apr 18, 2022 at 15:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.