Contested statistics – part two

Firstly, my apologies for the irregularity of my posting. I should have known better than to promise a two-part post in a month that included both an inter-state conference AND Christmas! However here we are with time for one more post before Christmas, so on with the show!

In my last post, I argued that the number of race horses ending up the knackery is actually much less than animal rights groups such as the Coalition for the Protection of Racehorses claim. This is largely because of a mis-calculation in the number of thoroughbreds born, who do race – the estimate of only 30% eventually racing used by the CPR is actually something of an inversion of the true figure. Data analysis of statistics provided in the 2012/13 Australian Racing Fact Book showed that between 71-81% of thoroughbred foals born went on to be registered.[1] Once again it is worth pointing out that just because a horse is registered doesn’t mean it will race, however once a horse is registered it is accountable (in the statistical use of the word), and its career trajectory is traceable.

In order to counter the claims of such high numbers of horses ending up at the knackery, the racing industry has sponsored its own research. Unfortunately, the outcomes of this study, undertaken by Renee Geelen, are just as dubious as the numbers put forth by animal rights campaigners. I mentioned Geelen’s research in a previous post, where I was critical of her methodology, but given her work is being given credence, often presented to counter the figures used by the CPR, it is worth taking a closer look.

Originally published in Issue 117 of Racing and Breeding magazine and later re-interpreted for a blog post on the National Museum of Australia’s website, Geelen’s research traces the destination of 3224 horses in the racing stables of 37 trainers. [2] Here lies the first major flaw of this research; Geelen’s entire sample consisted of just 37 trainers, among a potential pool of 3,891 trainers across Australia. As a percentage, that works out to be less than .1 of a percent of the sample population. On the other hand, her study covers 3,224 horses, covering a much greater proportion of the racing population, estimated to be around 31,000 horses in any one season.

These figures reveal a significant bias in favour of metropolitan stables, which train a large number of horses. This bias is particularly problematic, as research indicates that poorly-performing horses exiting larger metropolitan stables are much less likely to be sent to the knackery, instead going to new trainers at smaller establishments. Conversely, horses from smaller stables were more likely to end up at auction or knackeries. [3] Therefore, not including smaller-scale operations significantly skews Geelen’s results.

In fact, of Geelen’s sample group of horses, only around a third (806) actually exited the industry, with the majority either going to a new trainer, still in work or going to stud. This means that the population of greatest interest when dealing with these contested statistics –  that is, horses actually exiting the industry – is much smaller than the study initially makes it appear. Geelen’s figures are outlined in the table below:

Still Racing Combined Results Total % of Retired
Different Trainer 662 21%
Still in Work/Spelling 1,015 31%
Exported 77 2%
Total 1,754 54%
Completed Racing Career
At Stud 664 21% 45%
Sold/Gifted as pleasure horse 450 14% 31%
Returned to Owner 205 6% 14%
Died/Euthanised by Vet 109 3% 7%
Unknown 19 0.6% 1.3%
Career in Racing 17 0.5% 1.2%
Knackery 6 0.2% 0.4%
Total 1,470
TOTAL 3,224

(Source: ‘What Happens to all those racehorses?,’ blog post by Renee Geelen dated 1 October 2014, People and the Environment Blog, National Museum of Australia)

By applying the above logic, we see that, of the number of horses in Geelen’s study actually exiting the industry, 0.74% go straight to the knackery, not the 0.2% claimed by Geelen. It is also worth bearing in mind the heavy bias in favour of metropolitan stables, as articulated above, indicating the true figure is probably somewhat higher.

In fact, the only peer-reviewed study that has been published on this issue found that 6.3% of racehorses exiting the industry went straight to the knackery. [4] This study, authored by Thompson, Hayek, Jones, Evans and McGreevy, and published in the Australian Veterinary Journal in August of this year, probably represents the most accurate figures when debating the number of racehorses ending up at the knackery. The authors present a clearly outlined methodology, and base their findings on a much larger and more representative sample group, which covered 3816 horses, trained by 377 trainers based in stables of all sizes, across Australia.

So if you want to get a reasonably accurate figure of the number of racehorses who go straight from racing to the knackery, and an understanding of the the way the landscape differs in this respect according to State or Territory, then the Thompson et al study is the one I recommend you read.  However, there are further issues when trying to calculate an exact number of thoroughbreds born within the racing industry who end up at the knackery. These include:

* The record-keeping practices of knackeries and abbotoirs, which do not record the type of horse slaughtered, or even necessarily how many horses are slaughtered; [5]

* Those thoroughbreds who are not registered (between 19% and 29% of the foals born each year [6]), and who therefore do not figure in industry statistics;

* The argument over when exactly an ex-racehorse can be defined as ending up at the knackery. The racing industry claim they are not responsible for these horses beyond their initial exit from racing, while others, such as the CPR, believe that, as these animals were bred for the sole purpose of racing, the industry is responsible for their welfare for the duration of their lives. And that is a whole other debate!

REFERENCES:

[1] Data extrapolated from pp. 30-40 of the 2012/2013 Australian Racing Fact Book by statistician Clinton Paine, personal communication.

[2] Renee Geelen, “The Internet Age of Misinformation,” Racing and Breeding (117):53, accessed 29 November, http://breedingracing.realviewdigital.com/?iid=102110#folio=52

[3] Thompson, PC, AR Hayek, B Jones, DL Evans and PD McGreevy. “Number, Causes and Destinations of Horses Leaving the Australian Thoroughbred and Standardbred Racing Industries.” Australian Veterinary Journal 92 (2014): 310.

[4] ibid, p. 308.

[5] Hayek, Ariella. “Epidemiology of Horses Leaving the Racing and Breeding Industries.” Bachelor of Science (Vet Science) thesis, University of Sydney, 2004.

[6] Data extrapolated from pp. 30-40 of the 2012/2013 Australian Racing Fact Book by statistician Clinton Paine, personal communication.

Advertisements

6 thoughts on “Contested statistics – part two

  1. Great critical reading of the research! It’ll be interesting to see whether private funding for research (such as from the racing industry) does grow, and whether it does affect the integrity of the research, as feared by some (such as Alec Costin, the scientist (partly) responsible for getting cattle out of the high country, who I was reading about the other day when looking up brumbies http://www.abc.net.au/news/2015-02-21/ecologist-snowy-mountains-conservation-suffering-brumbies/6162804).

    • It disturbs me that, while the research of Prof Paul McGreevy is dismissed out of hand due to perceptions of bias (because he supports banning the whip), flawed research such as Geelen’s is just uncritically accepted.

      I think the assumption that if you are not outright 100% pro-racing then you must be anti-racing is a major stumbling block to communication of accurate facts; I think this is the same issue being faced with the brumby debate – if you don’t support keeping every single brumby in Kosciusko then you must automatically want to exterminate them ALL.

      While people spend time quibbling about numbers and statistics, nothing to progress the issue is actually being done.

      • Or, for an even more depressing example, the way that squabbling over perceived agendas (and focusing on results that have been updated/reinterpreted, as good science does) has resulted in climate change doubt, meaning that basically nothing is being done…

  2. Pingback: Off the track | horsesfordiscourses

  3. Pingback: Controversy Corner | horsesfordiscourses

  4. Pingback: On horseracing | horsesfordiscourses

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s