Poll of Iraq
September 15, 2003
Zogby International released a survey on September 10, 2003 that it calls the 1st Scientific Survey of Iraq. The methods reported by Zogby for conducting this poll says the interviews were done in public places such as shopping areas and coffee houses. Interviews were done in only four cities, excluding Baghdad. The Zogby methods statement claims the 600 adults were chosen at random, and goes on to call this a scientific poll.
What ever merit the poll may have, the methodology describes what is called a convenience sample. A scientific poll would have given every adult in Iraq a known chance of selection so that the chance, or probability, of selection for every adult in the country also could be calculated by the researcher.
A convenience sample gives no chance of selection to most of the population. Furthermore, the probability of selecting an adult in the areas where the Zogby interviewers went is therefore unknown.
Pollsters usually select samples at random so that the judgment of the interviewer about who to include does not bias or distort the results. The probabilities of selection become part of estimating the results for all adults in the population. They also are used to calculate the error due to sampling.
When this is not done the results may or may not reflect the opinions of the entire public. Also, there is no basis for computing the margin of error in the results. The value of the data is simply unknown.
For more information about this and other polling issues, contact the NCPP Polling Review Board Members.