Disc3

Understanding Polls - Confidence Interval

 

 Many newspapers, when reporting results of political polls, say that “with 95% confidence, the results are in error by no more than +/- 3 percentage points.” The typical sample size is about 1,500. This allowance for error is intended to cover both sampling variability and the effect of small biases

Assume that the poll (sample) indicates that just about 50% of likely voters favor a particular candidate. (Wilkinson et al., 1999).  

How large a +/- term is required for a 95% confidence interval for the population proportion?   

Solution

Using the formulae

Sample size =

1500=

Marginal error = +/-0.025

1. Would the +/- term be much different if 40% of likely voters in the sample favored the candidate?   

If 40% of likely voters in the sample favored the candidate, the marginal error will be: -

From the results arrived at, the term will not be much different if 40% of the voters favor the candidate. At this point, it is interesting to indicate that a candidate’s lead is greater than what we would expect from sampling error, or that a race is “a statistical tie (Bevington & Hill, 1969).

2. Why is the quoted +/- 0.03 larger than the +/- term you calculated in Question 1?

The quoted +/-0.03 is larger than the calculated +/- term due to rounding off error. The calculated term is 0.025 which can be rounded off to 0.03.

References

Wilkinson, L. and Task Force on Statistical Inference, APA Board of Scientific Affairs (1999) ‘Statistical Methods in Psychology Journals: Guidelines and Explanations’. American Psychologist, 54, 8, 594-604.

Data Reduction and Error Analysis for the Physical Sciences.’ Philip R. Bevington, McGraw Hill (1969).

http://astro.cornell.edu/academics/courses/astro3310/Books/Bevington_opt.pdf

Feelback

Initial posting reveals a clear understanding of all aspects of the discussion question; uses factual and relevant information; cites course readings, and demonstrates full development of concepts.