91Ó°ÊÓ

The margin of error is a key concept when interpreting results from polls and surveys. It represents the range within which we can be confident that the true value lies. This range is typically defined as a certain number of percentage points above and below the reported value.
For example, in the given exercise, the margin of error is 3%. This means that the true support for a candidate like Kay Hagan could be 3 percentage points higher or lower than the reported 47%. So, her true support could realistically be anywhere from 44% to 50%.
Understanding the margin of error helps us grasp how precise or uncertain a particular survey result is. The smaller the margin, the more precise the estimate and vice versa.

Voter polling is a method used to gauge the support levels for candidates before an election. Organizations like Gallup conduct these polls by surveying a sample of likely voters.
In the exercise, Gallup polled 982 likely voters for the North Carolina senate race. They reported that Kay Hagan had 47% support, and Thom Tillis had 46% support. These percentages represent the support levels in the sample and are used to infer the likely outcomes if the entire population of voters was surveyed.
While polling is a powerful tool, its accuracy depends on many factors, including sample size, the representativeness of the sample, and the margin of error.

Overlapping intervals refer to the situation where the confidence intervals of two or more estimates overlap. In voter polling, this can signify that the difference between candidates is not statistically significant.
For instance, Kay Hagan’s confidence interval ranges from 44% to 50%, while Thom Tillis’s ranges from 43% to 49%. The overlap in these intervals (from 44% to 49%) shows that the true support for both candidates could lie within this range.
This overlap indicates that, based on the polling data, it is unclear who has more support, hence the race is deemed 'too close to call'. This is because the true support for each candidate could potentially be the same or very close.

In statistics, a population parameter is a value that represents a characteristic of the entire population. In the context of voter polling, it refers to the actual level of support each candidate has among all voters.
The goal of polling is to estimate this population parameter using a sample. For example, if Gallup reports that Kay Hagan has 47% support, they are estimating the population parameter for the true support level.
The confidence interval then provides a range within which this true level of support likely falls. If we have a margin of error of 3%, the population parameter is expected to lie between 44% and 50% for Kay Hagan.