The additional expense of a

probability sample can be justified only if the resulting information is of high quality.

With a pure

probability sample of 319 one could say with a ninety-five percent probability that the overall results have a sampling error of +/- 5 percentage points.

With a pure

probability sample of 659, one could say with a 95 percent probability that the overall results would have a sampling error of +/- 3.

With a pure

probability sample of 2,385 adults one could say with a ninety-five percent probability that the overall results have a sampling error of +/- 3 percentage points.

With a pure

probability sample of 1,197 employees and 630 retirees, one could say with a ninety-five percent probability that the overall results would have a sampling error of +/- 3 percentage points for employees and +/- 4 percentage points for retirees.

With a pure

probability sample of 1,390 one could say with a 95 percent probability that the overall results have a sampling error of +/- 4 percentage points.

With a pure

probability sample of 2,534 adults one could say with a 95% probability that the overall results have a sampling error of +/- 3 percentage points.

With a pure

probability sample of 1,302 adults one could say with a ninety-five percent probability that the overall results have a sampling error of +/-3.

The survey consisted of a random digit-dial

probability sample of 1,008 Americans.

The GfK OmniTel[R] survey is a random digit dial (RDD)

probability sample of Adults 18+ in all telephone households in the continental United States.

With a pure

probability sample of 2,302 adults one could say with a 95 percent probability that the overall results have a sampling error of +/- 2 percentage points.

With a pure

probability sample of 1,031, one could say with a ninety-five percent probability that the overall results would have a sampling error of +/- 3 percentage points.