Home Page Forums Darwin’s Ark Latest research

Viewing 1 post (of 1 total)
  • Author
  • #14519
    Avatar photo
    jessica hekman

    A quick report from a Darwin’s Ark researcher!

    I’m still working on different ways of using our survey data to evaluate whether different breeds really do have different personalities. (I’m a dog person so I do believe they do – my boss is a cat person and argues it’s all in our minds.)

    The latest is this new statistical technique I just discovered called Latent Class Analysis. Basically what it does is look through all the survey answers, and tell me if there are groups of dogs who tend to answer in particular ways – so, different “classes” of dogs. Our hope is that this will correspond to breeds or breed types, but we’ll see. I’m deep in getting this software to work right now.

    Today’s goal is just to find out how many classes of dogs we see in our data – two? Five? Ten? Only after figuring that out can I see which dogs fall into which class, and then see what the similarities of dogs in the same class are. (I’m guessing size and age will be really important in determining class.)

    Thanks SO MUCH to all of you who have filled out all the survey questions! It’s amazingly helpful to have lots and lots of data to play with here and for that I owe all of you.

    Jessica Hekman, DVM, PhD
    Postdoc, Karlsson Lab

Viewing 1 post (of 1 total)

You must be logged in to reply to this topic. Login