‘Test’ vs ‘Assessment’: Words matter, and that bathwater is looking dirty enough to lose a baby!

Back in 2017 I was sitting in the audience of Mykel Smith’s presentation at the Australian Institute of Animal Management yearly conference ‘Recruiting: Are we getting the most suitable candidate?’. He spoke about staff recruitment, focusing on the difference in an organisations ability to select high-quality candidates using traditional Q&A interviews or more holistic assessments of candidate skillsets that are task-based. I was struck while watching Mykel speak by the similarities between the content he was presenting, and that which I would present a few hours later. Mykel concluded that one-time, high-pressure ‘tests’ are just as bad at predicting the future performance of job candidates as they are at predicting future behaviour in sheltered dogs.

In the world of HR, there’s been a marked shift away from traditional interviews towards performance-based assessments. This has been informed by the realisation that how well someone answers questions in a formal sit-down meeting doesn’t accurately reflect how they will behave when dealing with an upset person in public, or wrangling a feisty dog, or talking to an irate local resident over the phone. It simply makes sense to ask people to do the things we want them to do and then assess their performance, so that’s what we’ve started to do – assess, not verbally test.

The concept of ‘assess, don’t test’ is fundamentally (and in some ways practically) the same when we talk behaviour evaluation of sheltered dogs and cats. We can put them through a battery of tests, interview-style, in a strange environment that bears little resemblance to the life of a pet dog or cat. A significant body of research tells us this approach isn’t very good at predicting how they will behave in a home though, and a dive into non-human animal personality research gives us some pretty informative reasons why this is the case!

We can also take them out of their cages and interact with them in much the same way people interact with their own pet dogs, and simply ‘assess their performance’ while they do what pet dogs do. Take them for a walk around the block, see how they go playing with other dogs, do some basic training, see how they go during a vet visit, and just sit down and chill with them on a couch – these are all things that pet dogs do daily, so it makes sense to ‘assess, not test’ by seeing how they might perform at the job of ‘family pet’.

Given that this approach seems to intuitively make sense given the information we are trying to collect about these animals, I was a little stumped by the viewpoint ‘tests suck, so don’t bother assessing at all’ that sprung up following the publication of a paper by Gary Patronek and Janis Bradley ‘No better than flipping a coin: Reconsidering canine behavior evaluations in animal shelters’.

Given that the main message of the paper is to abandon battery-style tests in favour of spending time with sheltered dogs doing normal dog stuff (which I wholly support!), I’m guessing that the core of the issue, at least in academic circles where there was some fast and strong push-back against the idea of not assessing at all, was simply a poor choice of words by the authors.

Tests are evaluations, but not all evaluations are tests – just ask any Year 12 student who excels in mathematics throughout the year and then bombs their final exam!

As the use of battery-style tests has justifiably fallen out of favour, there’s been a cultural shift in the sheltering word against using the word ‘test’ that has out-paced changes in practice. The result being that the word ‘test’ has been replaced by ‘assessment’ and ‘evaluation’ in any and all contexts, even when the situation being described is actually a test in the truest sense of the word.

This has muddied the water of this conversation and opened the door for a lingering false dichotomy of ‘test or nothing’ – while some people have adjusted their language around battery-style tests and not much else, others have grabbed onto the idea that all evaluations in a shelter environment are invalid with both hands, and conclude that all sheltered dogs should be placed into adoptive homes regardless of how dangerously they behave while in care.

We’ve got to be careful not to throw out the baby with the bathwater in our haste to improve how we assess dogs in shelters. We absolutely should be assessing the behaviour of dogs (and cats!) who enter our sheltering system to determine how best to find them a loving home that suits their needs, and indeed, if that is really the best outcome for all involved. We just need to ‘up our game’ and use a more holistic, fairer, and systematic approach to doing so – just like we are doing more often with people.

While we do that, we also need to remember that clear and accurate communication is crucial to what we do so that we don’t go losing our baby in muddy water.

This week’s image is Reggie, also known as Sir Reginald McBiteyface. He’s covered in mud from pushing his kong wobbler around in the muddiest part of the yard and doing his best to avoid looking straight at me laughing at his brown nose!

Leave a Comment

Your email address will not be published.