Sampling error
(go to Outline)
Sampling error is the difference between survey result and population value due to the random selection of individuals or households to include in the sample
Unlike bias, sampling error can be predicted, calculated, and accounted for. There are several measures of sampling error:
- Confidence intervals
- Standard error
- Coefficient of variance
- P values
- Others
You have probably heard of sampling error in newspaper reports without recognizing it:
... the survey was based on 570 interviews conducted by phone between March 20 and 31. The sampling error was plus or minus 4.5 percentage points.
A poll for Le Figaro newspaper showed Sarkozy, the nominee for President Jacques Chirac's governing party, drawing 28.5 percent of votes for the first round. Royal tallied 25 percent. But the margin of error in surveys of its size is plus or minus three percentage points, meaning that statistically, they are in a dead heat.
48 percent felt there may be too many guns in this country...The AP-Ipsos poll of 996 adults was conducted April 17-19 and has a margin of sampling error of plus or minus 3 percentage points.
These descriptions are reporting 95% confidence intervals.
