Surveys.

The New York Times recently published an article titled "What's the Matter With Polling?" It tells about the "increasing unreliability of election and other polling in the United States." One of the many reasons given for the unreliability is the increased reliance on "online polling" and apparent growing agreement by professionals that it is "impossible to calculate a margin of error on such surveys."

Why is "online polling" creating such problems? The mathematical formula to confirm that a fairly small number of persons randomly selected to be surveyed can give a good estimate of the result for an actual decision by a very large number of people still is considered mostly valid. There is a catch: The sample must be a "representative sample." If the poll only counts the votes of the people who volunteered to take the survey, or answer a specific question, it is not a true random selection of the entire group and, therefore, not "representative" of the entire group.

The Nation Magazine and the Pew Research Center include the unfortunate additional fact that the design goals of some surveys are affected, sometimes intentionally, by the existing beliefs or personal goals of survey's designers.

Last year, The Florida Bar stated that its Vision 2016 survey indicated "substantial support" by the members of the Bar for the concept of reciprocity. The survey was an online poll. Later, after a huge negative reaction, the Bar changed the explanation by saying it really meant there was substantial support by "those who responded to the survey." The Board of Governors recognized, after extensive heavy criticism, that in fact there was very substantial opposition. As a result, the board unanimously rejected consideration of the concept of reciprocity.

A recent new Bar survey has now been released. It also was conducted online, with only 35 percent of Bar members who were asked to respond electing, for whatever reason, to answer the survey. Of interest on this subject, the Bar reports that two of the top three primary reasons that survey participant's opinion about the Bar has become more negative over the past two years are "Opposition to Reciprocity" and "Disagree With Not Pursuing Reciprocity."

Bar leadership should ask what caused the apparently incorrect report of and interpretation of the survey last year and to what extent are other conclusions in that report and other survey reports this year subject to the same type errors? Are we spending time...