Shelter’s Edge

In the first installment of this two-part series, the ASPCA’s Trish McMillan Loehr, MSc, CPDT, Director, Applied Research and Behavior, covered questions to ask when building a full behavioral profile for any dog in your care. Today, she tells us what those assessment results really mean, from false positives to true negatives.

When assessing a dog for aggression (or in testing any hypothesis), there are four possible results:

1. The dog shows aggression during the assessment, and the dog also shows aggression in similar real-life situations (a true positive, in science-geek speak)
2. The dog doesn’t show aggression during the assessment and the dog does not show aggression in similar real-life situations (a true negative result)
3. The dog does not show aggression during the assessment, yet the dog does show aggression in similar real-life situations (a false negative, or type II error)
4. The dog shows aggression during the assessment, but does not show aggression in similar real-life situations (a false positive, or type I error)

Source: http://www.gilliganondata.com

The first two correct results are what we’re hoping to uncover with SAFER®, but behavior is not always such a tidy thing to predict, and incorrect handling can make false negative and false positive results more likely. How can having more information about each dog in your care help you avoid false negatives and false positives?

False negatives
If you are not looking at all of the pieces of the puzzle, you may end up with false negatives, and incorrectly assume that a dog has a low probability of displaying aggression when she actually has triggers that SAFER can’t uncover (such as predatory aggression toward cats). False negatives might also occur, for example, if a food guarding dog was too stressed (or full) to eat on the day you assessed her. If you are getting a full behavioral picture from many sources, you may find out that both the original owner and your kennel staff have seen this dog guard her food bowl. This is a behavior that can be modified if you know about it, and one that shouldn’t come as a surprise to the adopter.

False positives
You may also get false positives for aggression during a SAFER assessment. These may be caused by handler error (a classic mistake involves not giving dogs enough choice to move away if they want to). Dogs can be stressed in a shelter situation, and may show behavior that they don’t show in other contexts. Perhaps a dog growled at the helper dog during his assessment, but in taking a history you find out that he lived with several dogs peacefully in his old home, and your walkers say that he play-bows and wags at other dogs while out on walks. Knowing this history may help you decide if he’d be successfully placed with another dog.

So don’t stop at SAFER – if you carefully consider all of the pieces of each dog’s behavioral “puzzle,” you will be in a great position to make the best possible placement decisions for the dogs in your care.

If you’re interested in implementing a standardized assessment at your shelter, sign up for session one of our SAFER Canine Assessment webinar series with Dr. Emily Weiss.

Related links:
About ASPCA SAFER
Upcoming Free Webinar Series: SAFER Canine Assessment
About SAFER: Just One Piece of the Puzzle

Tags: ASPCA SAFER, Behavior & Enrichment, Saving Lives, Tools, Trish McMillan Loehr

You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.