Comparison of Visual and DNA Breed Identification of Dogs and Inter-Observer Reliability

| | | | | |
by ()
More Information...

Short Description:

Visual inspection remains the most common method of identifying dog breeds. The authors of this study surveyed over 900 individuals connected with dog-related professions or activities to determine how frequently they were able to correctly identify the breed of 20 different dogs. The results showed significant differences between visual and DNA identification as well as little agreement among participants. Advocates addressing breed-specific bans will be interested in these results, which the authors say raise questions about the “ability to implement laws and private restrictions pertaining to dogs based on breed composition.”

[Abstract excerpted from original source.]

“Until the recent advent of DNA analysis of breed composition, identification of dogs of unknown parentage was done visually, and visual identification is still the most common method of breed identification. We were interested in how often visual identification of dogs by people, assumed to be knowledgeable about dogs, matched DNA breed identification and how often these people agreed with each other (inter-observer reliability). Over 900 participants who engaged in dog related professions and activities viewed one-minute, color video-clips of 20 dogs of unknown parentage and were asked to identify the dogs’ predominant breeds. For 14 of the dogs, fewer than 50% of the respondents visually identified breeds of dogs that matched DNA identification. Agreement among respondents was also very poor. Krippendorf’s alpha was used to examine the reliability of the most predominant breed (selected across all dogs identified as mixed breeds) for all respondents, yielding alpha=0.23. For only 7 of the 20 dogs was there agreement among more than 50% of the respondents regarding the most predominant breed of a mixed breed and in 3 of these cases the most commonly agreed upon visual identification was not identified by DNA analysis.”

Along with the attached paper is a commentary by the National Canine Research Council and a poster illustrating the study’s results.

Spot Check Number: 2312
Sponsor: College of Veterinary Medicine, Western University of Health Sciences; Oregon State University; & Others
Researcher/Author: Victoria L. Voith et al.
Animal Type: Dogs, Companion Animals
Research Method: In Person Interview/Survey, Print Survey
Geographic Region: United States Regional
Number of Participants: 923
Year Conducted: 2013
Note that this research study is based at least in part on experiments on animals. HRC does not condone or endorse any animal research; we post this item (and others like it) with the hope that these findings can assist advocates in their work to help animals. For a description of how we select items for the database, please click here.
File Attachments: You must be logged in to access attachments (see login and registration links above)

Looking for full text articles?

If the full text of an article is not available, click here for other options.

How do we select database articles?

Want to know how we choose the articles that we share? Click to read about our process.


Did you find this research helpful in your work for animals? If so, please consider a donation to the Humane Research Council to help us with the costs of maintaining, expanding, and improving