Darwin's dog identification survey!

[Two Best Dogs!]

Right spot? Not sure Here's a survey that is also contributing to research on dogs! It's only going for a few more weeks.

 

It's a "can you identify the breeds in a mix" type survey. I'll just post the details below:

 

About The Project

Darwin’s Dogs, a project of the Broad Institute of Harvard and MIT, and U-Mass Medical School, has been conducting ancestry research on thousands of dogs, analyzing DNA saliva samples to tell us the genetic history of our canine companions.

In conjunction with the International Association of Animal Behavior Consultants (IAABC), the team is launching a “citizen science” survey. The data collected will tell us a lot about how people perceive breeds and mixes, how different breed mixes affect the appearance of a dog, how well people can guess their canine friends’ ancestries from looks alone, and how that might affect our lives with dogs. This information is vital to many future projects throughout the scientific world.

 

How To Participate

The survey will introduce you to 31 mutts. You’ll be asked to guess the 3 breeds contributing the largest percentage of DNA for each dog.

• Only the answers to the “practice survey” will be given at this time. This is real science, and we need to be sure everyone’s giving their best guesses without influence or hints from others.
• At the end of the research period of 60 days, each participant will be sent a certificate of participation and the answers to what makes up each of these wonderful mutts.

Begin Survey Now >

Thank you for your contributions to the citizen science that makes Darwin’s Dogs one of the most important research initiatives today!

 

Here's some pics from the practice survey!

 

 

 

contributing to research! the more people who give it the best shot, the more data points they can analyse on how bad we all are at identifying dogs XD

[PossumCorner]

 

 "This information is vital to many future projects throughout the scientific world."

 

I'm just not making the connection between the quote and the good point Thistle made about showing how bad we are at identifying dogs (especially in another country where the Standards of the Breeds differ slightly, plus breeds we are not contact-familiar with are present).

 

[RuralPug]

Can't resist! 

It will be fascinating to see how well I've guessed when the results come out. Did myself proud on the practice! 

[DogsAndTheMob]

I can’t  find kelpies on the drop-down menu!

[Two Best Dogs!]

Yes it was a small set of breed options? I figure they’re going very American esque because of location. Kinda like how one of the practice runs was a bloodhound x but because I’ve never met one I thought “beagle”. 

[westiemum]

Well how bad was I at that !!    I just see the same breeds regardless!  

[Maddy]

3 hours ago, DogsAndTheMob said:

I can’t  find kelpies on the drop-down menu!

There was one dog that definitely looked part kelpie to me but kelpie wasn't on the list. A lot of these mixes must be more common US mixes because they don't look like any of the common mixes we have in pounds.

[Selkie]

It would very much depend on country. Kelpie, ACD, JRT and SBT crosses seem to make up 90% of crossbreds in the pounds where I am.

[sandgrubber]

23 hours ago, DogsAndTheMob said:

I can’t  find kelpies on the drop-down menu!

I had the same problem... I think it was the 3rd dog that looked kelpie.  I gave up because of how often I couldn't find the breed I wanted to choose.

[sandgrubber]

23 hours ago, Thistle the dog said:

Yes it was a small set of breed options? I figure they’re going very American esque because of location. Kinda like how one of the practice runs was a bloodhound x but because I’ve never met one I thought “beagle”. 

I didn't find the selection very typical of dogs you'd find in a US pound.  Not a lot of pit bull or Chihuahua.

[Two Best Dogs!]

I’m meaning the drop down list of breeds to choose from, not the pictures themselves. 

[ish]

I couldn’t get past that the dogs we were guessing were made up of at least 10% of each of 3 different breeds - to me that could still leave  a lot of dog! If it’s 70% lab and 20% beagle you wouldnt expect to see many if any traits from the 3rd breed, or even the second. 

 

Also ageee that Kelpie was a necessity in the breed list and gave up

[Asher]

I couldn’t see if they were including a sample of ‘control’ dogs of KNOWN breed or ancestry (I.e. they know for sure what the breed or combination is prior to genetic testing and surveying people) which would be necessary if they were to draw any scientifically valid conclusions... Plus I agree that the breed list is very limited and biased. All a bit pointless...

[RuralPug]

3 hours ago, Asher said:

I couldn’t see if they were including a sample of ‘control’ dogs of KNOWN breed or ancestry (I.e. they know for sure what the breed or combination is prior to genetic testing and surveying people) which would be necessary if they were to draw any scientifically valid conclusions... Plus I agree that the breed list is very limited and biased. All a bit pointless...

As I read it, every dog had already been tested and definitely had AT least three breeds identified. The point of the survey was to see how well people guess breed make up.

Yes, this will definitely be affected by the breeds that you most commonly see. Kelpies obviously did not feature in any of the dogs tested (OR the Kelpie DNA is simply not available to these particular testers, therefore did not appear in their results!)

Since they do know the breeds that their DNA tests indicated, they no doubt included all of those breeds in their drop down list, plus some extra breeds that (I'm just assuming this) that they thought would be recognised by the average survey taker ( with a US bias of course, quite often it seems to me that US survey makers are totally ignorant that the rest of the world exists!)