CNN 2000ats Page 131
Exhibit A.1
Approximate Sampling Tolerances (at 95% Confidence) to
Use in Evaluating Percentage Results
Number of
Survey
Survey
Survey
Survey
People Asked
Percentage
Percentage Percentage Percentage
Survey
Question on
Result at
Result at
Result at
Result at
Percentage
Which Survey
10% or
20% or
30% or
40% or
Result at
Result Is Based
90%
80%
70%
60%
50%
2,000
1
2
2
2
2
1,500
2
2
2
2
3
1,000
2
2
3
3
3
900
2
3
3
3
3
800
2
3
3
3
3
700
2
3
3
4
4
600
2
3
4
4
4
500
3
4
4
4
4
400
3
4
4
5
5
300
3
5
5
6
6
200
4
6
6
7
7
100
6
8
9
10
10
50
8
11
13
14
14
Sampling tolerances also are involved in the comparison of results from differ-
ent parts of the sample (subgroup analysis) or from different surveys. Exhibit A.2
shows the percentage difference that must be obtained before a difference can be con-
sidered statistically significant. These figures too represent the 95% confidence
interval.
For example, suppose one group of 1,000 has a response of 34% "yes" to a ques-
tion, and an independent group of 500 has a response of 28% "yes" to the same ques-
tion, for an observed difference of 6 percentage points. According to the Exhibit, this
difference is subject to a potential sampling error of 5 percentage points. Since the
observed difference is greater than the sampling error, the observed difference is con-
sidered statistically significant.
124
