• by Joe Mandese  @mp_joemandese, November 22, 2005

Nielsen Media Research's practices have been the focus of civil suits, Congressional hearings, and an inquiry by the Federal Trade Commission, but changes in another of its business methods--the way it compensates people in its national TV ratings sample--may raise the ire of another federal agency: the Internal Revenue Service. As part of a plan to dramatically increase the amount of money it pays some especially difficult "Nielsen households," the TV ratings firm has begun making payments to individual members of those households, MDN has learned. The reason: In some cases, the new payments will exceed the $600-per-year threshold the IRS requires for reporting income.

"The last thing we want to do is send out a 1099 and require people to declare it as income," acknowledges a senior Nielsen executive, referring to the IRS form the government requires companies to send to part-time employees or freelancers informing them of liabilities associated with reported income that had no taxes withheld.

It's not the additional paperwork Nielsen is trying to avoid, say Nielsen executives, but the fact that they fear reporting the income would do the opposite of what the payments are supposed to do in the first place: provide an incentive for people to cooperate with Nielsen's ratings methods.

In fact, the higher payment levels are earmarked for members of the most difficult households to measure--especially minority households, or households with very large family sizes that are prone to "fault," or those that may be taken out of the sample for technical reasons. In addition to the higher cash incentives, Nielsen also is providing personal coaching for individual members of these households. Between the two steps, it hopes to reduce fault levels and get better representation in its national TV ratings sample.

By sending money to individuals within households, instead of a single member--usually the head of the household--Nielsen is not violating any IRS laws, says IRS spokesman Anthony Burke, noting that the $600 reporting rule only applies to income received by individuals. But Nielsen's incentives--usually cash payments, although Nielsen is also experimenting with "cash cards"--have always represented something of a gray economy for people living in the roughly 8,000 households in its national people meter sample, as well as the tens of thousands participating in its local market surveys. But whereas in the past Nielsen would send regular installments of dollar bills, Nielsen executives say they now use "bigger denominations."

Nielsen executives said they already have raised cash payments "significantly" for the most difficult households in their sample, but that beginning in February Nielsen will begin testing incentives for some key demographic groups that would increase payments "50 to 100 times" over what they had been receiving in the past.

Nielsen executives declined to specify exactly how much they are paying the highest-incentive households because they fear that making the information public would raise issues among Nielsen households being paid lower amounts. But the higher monetary levels are leading Nielsen to conduct two types of research on how they might impact TV ratings.

In the first phase, which Nielsen is conducting now through February 2005, Nielsen is trying to gauge whether higher cash incentives influence the media available to Nielsen households--especially their acquisition of new media technologies such as digital video recorders, computers, DVDs, or other equipment the money could be used to purchase, as well as subscriptions to premium television services that might also influence how people living in those households watch television.

The second phase of research, which will be conducted early next year, will investigate whether higher cash incentives affect people's use of time and consequently, how they watch television.

"The question is, what they do with that money?" ponders a Nielsen executive, adding: "Do they use it to go out to dinner or to the movies when they would otherwise be watching television? If so, that could influence viewing behavior."