When four out of five dentists recommend a certain toothpaste, you have to take the claim seriously. But how do you evaluate a claim like that for Nutro Natural Choice dog food which advertises “8 out of 10 dogs prefer Natural Choice Lamb & Rice Formula*”?
*MOUSE PRINT: The asterisk refers to a fine print footnote that says, “Based on paired preference tests performed 10/2008.”
When Mouse Print* asked the company in an email to explain exactly how the test was done (Who interviewed the dogs or did they fill out a written questionnaire? Did you have a translator for foreign breeds? <grin>) the company declined to specify the methodology but rather invited a call to customer service. Upon doing so, the representative spoke only in generalities off the top of her head without real knowledge of the protocol. And, being transferred to a supervisor only resulted in being deposited into voicemail.
Assuming the test was conducted in a legitimate manner, their own data do not support the claim they made that 8 out 10 dogs prefer their Nutro. When tested against Science Diet, only 67.3% preferred Nutro, not 80% or more. Likewise only 73.3% preferred it over Natural Balance brand. In no example given in the ad did eight out of ten dogs tested prefer Nutro.
Certainly, Nutro did well in the tests, just not as well as they advertise. Hey, Nutro, why even make a claim that your own data contradict?
I’m guessing they’re referring to the comparison against EUKANUBA of 76.7% which they rounded up to 80 and then ignored the other two results.
Our dog eats cat turds and drinks from the toilet when he gets a chance although at all times he is provided with Science Diet dog food and clean filted water. Maybe we should try Nutra! 😉
I’m thinking they literally had 10 dogs. Otherwise they would have reduced the ratio to 4 out of 5.
My dog prefers whatever happens to be in his bowl. Does that mean if I feed him Natural Choice eight times and Science Diet twice, he prefers Natural Choice?
Why would this be a difficult test to conduct? You put out two bowls of food and observe which one the dog eats more of. You use enough dogs to be statistically significant. Not sure if you need x number of each particular breed, or a random sampling of all pet breeds. When 77% prefer Natural Choice and 23% prefer Eukanuba, that is, for all intents and purposes, 8 out of 10. It is perfectly reasonable and scientific, and it seems more honest than most advertising. You could repeat the test yourself in conjunction with your local animal shelter, and see if your results are consistent with their results–that would make for a more interesting post.
Honestly, I think this study may be more objective than four out of five dentists recommending a toothpaste, since you have no idea what the dentists motives are. Dogs don’t have any competing interests–if they prefer one brand of dog food over another, it is because they really actually like that brand better.
We were trying to decide which dog food to buy. Fish vs. chicken vs. beef vs. rice…so many choices! One bag claimed that it was the taste dogs preferred with an asterisk. The small print said “when served wet compared to xxx brands served dry”.
Poster ‘Z’ wonders why such a test would be difficult to conduct. The question is how can anyone assume that the test was conducted honestly with a large enough sample size? How were the test subjects selected? Were they already on diet of the preferred dog food? Were the bowls an equal distance? Was there even an equivalent amount of food in each bowl? The variables are as endless as “Z’s” assumptions.
At the very least test procedures should be readily available so the results can be verified by others. One has to suspect that the reason the study’s protocol is not available is because the testers are not confident that the results are not duplicable.
@ Richard B: Were the bowls at equal distance? There are two bowls. How is the concept of ‘equal business’ possible? To put it differently, equal to what? Or how do you put to bowls not at an equal distance?
@ Jasper.
I am assuming Mr. Richard B. meant that the bowls are an equal distance from the dog. If Company A put their food 1 foot closer to the dog than Company B’s food, then the dog will probably go to Company A’s.
If you give my dog two bowls of dog food he will eat all of the food in both bowls and then look around for more.
Regarding Richard B.’s comments: My point is that it would not be difficult to conduct this sort of study in a fair and scientific manner. Whether they actually did or not, I don’t know. If they were trying to be fair, the experimental design would obviously address the kinds of questions you raise. Random sample of dogs that haven’t been trained to eat your brand, putting the same amount in each bowl and spacing them evenly apart. If they weren’t trying to be fair, well, it is marketing afterall. They aren’t making any claims about health benefits, or anything beyond doggie preferences. I don’t think most people will blindly follow the outcome of this study, but they will give their dog what they think is best for them.
Actually, the “four out of five dentists surveyed prefer Brand X” is complete rot as well.
Consider: My company manufactures and markets Brand X mouthwash, a bargain product made from recycled battery acid and transmission fluid. I ask 30 dentists their opinion of Brand X. 26 indignantly tell me they’d sooner drink raw sewage. The other four allow that, faced with the choice, they’d probably choose Brand X over the sewage–provided they could dilute it first by pouring it into an Olympic swimming pool.
Having collected this valuable feedback, I then proceed to ignore all but one of my sewage-drinking respondents, leaving five dentists in all. I can then quite honestly report that “four of five dentists surveyed recommend Brand X!”
Wanna buy some mouthwash? I can give you a great price…
It’s just another corporate marketing scam. I have a 30 lb cat that would eat ALL 10 bowls of dog food.