I have bred a couple of litters and have been involved in assessing puppy temperaments in a number of other litters.  These are my thoughts and observations.  A quote "Puppy Test" is in my opinion, often misleading. The problem with such tests is they are often only conducted once.  One observation of a particular characteristic or lack of such, is not necessarily a true test of that puppy.  In addition, the tester may not be experienced and the testing site may not be correct.

Let's say you are testing a litter of 6 at 6 weeks of age.  Puppies are awake, playing, the tester goes and picks one and removes it from the others to the secluded testing spot, hopefully some place where the puppy has never been.  Tester runs it through the tests, takes 15  or 20 minutes, returns puppy to sibs and gets another.  By the time you get to the, 4th, 5th or 6th puppy, they may well be getting tired and sleepy.  The later ones are not being equally evaluated if they are nearing nap time.  They may not react to something they would have if taken earlier.

It has been my observation, that the best knowledge of a litter will be obtained by a caring, observant breeder who handles the puppies every day. She can then tell you, who is the first at the gate, who pounces on the others the most, who is the first to the food bowl and who is the last, who is the first to come running when they hear "Puppy, puppy, puppy", who makes eye contact and who doesn't, who is the first to wake and the last to sleep, who is adventurous and who clings to the familiar,  who fights to have nails done and who submits readily, who handles a strange environment with ease and who whimpers at such, who screamed at his puppy shot and who took it with dignity and licked the vet.

I would much rather base my puppy selection on the above observations over several weeks, than one test or maybe 2 conducted at most twice.  In my experience of evaluating puppies, the breeder observations usually held true throughout the life of the dog, whereas a puppy test is a maybe.

As I said once before, if I were looking for a puppy to do agility with, the most important factors to me are  soundness of the parents physically and mentally.  Jumping is 80% of agility, I want proof the parents are both fluid jumpers.  Everything else can be taught, but a natural jumper is more born than trained.  Training and trialing require mental concentration and a steady dog.  Sound temperaments are much easier to deal with spooks, or aggressiveness. Good luck! :-) Mary Whorton

The study mentioned is not the first that indicated that the traditional puppy testing was not a good predictor of adult temperament.  It simply confirms results from earlier studies, using a larger sample size.  The military is working on an even more extensive puppy testing program for their own dogs, trying to improve predictability, but their results have yet to be released.  Last update I heard a few years ago indicated that they were looking at over 20 different traits.  In all honesty, it shouldn't surprise us that the testing isn't a great predictor...it simply gets back to that age-old question of nature vs nurture.  Too much can happen between testing and adulthood, which can bring out both the best and the worst in a dog.

There are soooo many variables that can alter results in puppy testing, many of which are very difficult to control.  Someone already mentioned that the last puppies tested, once they are getting tired, may test very different from the first puppies, when they are all fresh.  I've also found that the litter dynamics play a huge role in how puppies test.  Too many times, I've had puppies test as very quiet / submissive or very independent or assertive, only to find that removing them from the litter into new homes can cause dramatic changes in their personalities.  I've seen bullies suddenly become very mild-mannered once they aren't competing for resources.  I've seen the shy quiet puppies suddenly become holy terrors once they no longer have to deal with the highly assertive puppies.  Also, puppies develop at slightly different rates, so while 7 wks might be the best age to test most puppies, there will be those that probably would test better a few weeks later. Certain traits may be better tested at different ages than others, or perhaps tests should be repeated at specific ages to improve predictability.
Finally, how the breeder raises and handles puppies prior to testing has an impact.  My puppies are so used to being upside down or dangled by the time they are 7 weeks old, they accept if from just about anyone.  "Puppy puppy" already means food is imminent, so they respond to anyone calling them.  They hear lots of loud noises in the whelping box, so it is sometimes hard to test noise sensitivites without knowing what a unique sound might be for them.

Puppy tests can be a useful guide to evaluate puppy temperaments at a certain point in time.  I agree, though, the observations of a good breeder over a period of several weeks is far more valuable in selecting a puppy for a particular task.  When I have used puppy testing to help select a puppy from a distant breeder, I want to see *video* of the testing, so I can observe the puppies, not just see a written evaluation from the tester.  I may not agree with the tester's evaluation or might see something I find significant that
the tester completely missed.  But I still would rely very heavily on the breeder's observations, above and beyond the test.  I used to have a friend test all my litters for me.  It was extremely rare that she found anything out about the puppies that I hadn't already picked up on through daily observations.  Since she's moved to another state, I don't even bother anymore.  In our current litter of 8 wk old Pembroke corgi puppies, I think I have a pretty good idea about basic personalities, noise sensitivites, play drive, who takes initiative, etc.  And from the looks of it, I may have several good performance prospects among the group.  Of course, the puppy *I* am keeping is the first one to figure out how to escape from every box and pen she's been in so far!  I see trouble ahead <VBG>.Lyn Johnson DVM 

© 1999 Agility Ability and the noted authors of some of the individual listings.

Puppy Aptitude Test Developed by Joachim and Wendy Volhard.