A poll was conducted to determine the level of support within a large suburban school district for the construction of a new football stadium. Of the 655 people surveyed, 57% said they would support spending the money to construct the stadium. The poll had a margin of error of 4%. Which is the correct interpretation for this margin of error? There is most likely no more than a 4% difference between the proportion of support in the sample and the true proportion of support in the district.

If we conducted repeated samples from this population, the results will vary by no more than 4% from the current sample result of 57%.

If the poll had sampled more people, the proportion of district residents who would have said they supported the stadium would have been about 4% higher.

About 4% of the school district residents polled refused to respond to the question.

There is most likely more than a 4% difference between the proportion of support in the sample and the true proportion of support in the district.