In his Nov. 27 op-ed piece "Who Needs Those Polls?" Michael J. Berlin criticized political, marketing and policy surveys for narrowing the range of consumer products, candidates for office, public policies and media content. He said that all of the above are "predetermined by marketing data" derived from polls and "damaging to the public interest because they limit our choices to the lowest common denominator."

"A grass-roots trend has begun," he wrote, to refuse cooperation with poll-takers, and more alarmingly, "there is increasing evidence" that respondents are lying in their answers. Approving of both practices, Berlin closed by counseling others who agree that polls "restrict vital options" to reinforce these trends.

Berlin is mostly mistaken in claiming that polling has narrowed our options. In consumer product/service marketing, for example, Berlin is apparently unaware that the trend during the past decade has been toward segmentation of consumers into a larger numbers of discrete groupings, each with its own tastes and preferences. This naturally leads to a greater number of products -- not fewer aimed at Berlin's "lowest common denominator." With the proliferation of specialty periodicals and the growth of cable TV, the choices in print and electronic media information and programming are also greater today than ever.

Further, important non-marketing surveys -- usually sponsored by government agencies or disinterested nonprofit organizations -- are used to help allocate scarce resources and shape public policies, the U.S. Census being the best-known example. Would Berlin advise noncooperation with these types of surveys too?

By making a blanket appeal for the public to "say no" or prevaricate to poll-takers, Berlin did a disservice to the many legitimate researchers and policy makers who depend on being able to collect accurate statistical information from surveys. -- Sid Groeneman