Thursday, August 20, 2009

Testimonials and disclaimers

A big chunk of the fraudulent advertising (diet, exercise, work-at-home, credit repair) prosecuted by the FTC involves a testimonial advertisement that mentions a number, e.g., "I lost 74 pounds wearing slimming insoles." Some consumers do not seem to understand that the results for the endorsers may not be typical, despite the disclaimer required by the FTC, "results not typical."

Now the FTC is moving to bar use of such testimonials unless an advertiser has evidence
"... that the endorser's experience is representative of what consumers will generally achieve, the advertisement should clearly and conspicuously disclose the generally expected performance in the depicted circumstances, and the advertiser must possess and rely on adequate substantiation for that representation."
In other words, advertisers can use testimonials only if they develop statistical evidence to support testimonial claims, like the kind of evidence required by the FDA for new drug applications. This could prove so costly that it could discourage the use of testimonials in advertisements in products that actually work, like Jenny Craig, whose website uses customized testimonials.

For a discussion of benefits and costs of these kinds of policies, see "Consumer Protection," an entry in the Encyclopedia of Social Science.

3 comments:

  1. So who is defining the minimum number of samples that are needed to develop the statistical evidence? 1, 2, 3, 4, ????

    ReplyDelete
  2. I suspect that it will be similar to the criterion FDA uses to test drugs: enough data to reject the null hypothesis of "no effect"

    ReplyDelete
  3. The good news is this will make it hard for scams to provide enough real results to use testimonials. I think good products like Jenny Craig will have no problem providing valid statistical results.

    The bad news is that new products that have a real winner will have to work hard to find alternative ways to market their product until they have enough valid data points to have someone tell of how their results are "are typical". - Shawn

    ReplyDelete