1.   The probability of missing a cervical infection is possibly higher when using biopsy specimens because of inevitable sampling errors.

2.   In addition, it is likely that morphometric analysis in the transjugular biopsies is subject to some degree of sampling error, which may obscure this relationship.

3.   To limit sampling error, we took multiple biopsy specimens at standardised sites.

4.   Sampling error because of an insufficient number of biopsy specimens may have been responsible for the fact that this difference did not reach significance.

5.   As in the two other surveys, the margin of sampling error was plus or minus five percentage points.

6.   Because the New York City researchers did not choose the participants randomly, they could not estimate the degree of sampling error, as is done in polling.

7.   Both polls had a margin of sampling error of plus or minus three percentage points.

8.   Both polls had margins of sampling error of plus or minus three percentage points.

9.   Both polls have margins of sampling error of plus or minus three percentage points.