Answer to last post

[cynthmillsm]

Hi everyone,

In answer to your reply to my post:

First, I did keep records of time of day of census. It did not comply
with what was recommended in the WHO report, but I think it worked. I
and the others counted mostly in mid-day, instead of dawn and/or dusk
when dogs tend to be active foraging. I think it worked better in
Rarotonga because there dogs are mostly associated with owners and
households. They don't forage much anymore (they did when I was there
8 years ago.) They tended to be close to home when I counted and came
out as I walked to challenge me as an interloper. They were nice
enough, and sometimes walked along with me, which tended to bring out
more dogs from neighboring households.

One reason I believe this timing worked better was that the dogs have
been driven underground, to some extent, because there is a 'dog man'
who shoots dogs. I can only attest to this by my own experience
(although I could show you pictures from years past versus current
pictures) that show lots of dogs downtown versus now there are lots of
chickens and cats (a sort of urban release) that come to beg food from
you at the sidewalk restaurants.)

In regards to your recommending a 'capture-recapture' methodology. I
intended to do this, and had used the same method in a census I did on
Bora Bora. Esther Honey tattoos dogs after neutering, but these marks
were not visible, and there I was doing a pre-campaign count (more or
less.) I did my count using a sort of transect count--walking a
kilometer down a road in multiple sites and repeating the counts. I
took digital photos of dogs and could remember them pretty well and
used re-sightings as re-captures. (Unfortunately we were not invited
back there and the project wasn't continued.)

I intended to do the same in Rarotonga. But I found that since I was
doing most of the counting I couldn't depend on my memory or digital
photos efficiently enough to get what I would consider to be a
dependable re-sighting. Again, there were collars and tattoos, but
these were not consistent or visible enough, and the dogs are owned
and marked often so that collars couldn't be depended on to be unique
identifiiers. For that reason I ended up using the informal survey.

I did ask specifically about the areas--that is, I defined the areas
when I asked. I would say how many dogs live on this cul de sac, on
this road, in this compound of houses, etc. If the person seemed
precise and to know, I accepted the number; if they seemed vague or
unsure, I didn't. I guess that would have to introduce bias--I was
only getting information from people who liked the dogs enough to
notice them and knew enough about Esther Honey Foundation to trust me
as a representative to tell us the true number of dogs. So I selected
the data on my own judgement as to what was dependable and what
wasn't. And from that I created a figure that I used as a index
proportion.

I did ask other census-ers to ask about 'hidden' dogs, but in the end
I had a hard time judging their numbers and didn't use them. It is a
hard concept, I agree. I am wondering if it could be included in a
survey and if there are means to confirm or double check the numbers
as there are other surveys.

At any rate, thank you for your input on all this. I'm sorry it took
so long to get back to you. (I wrote an answer earlier but made a
mistake in posting it and no one saw it.)

Cheers,

Cynthia Mills