CARLHAYDEN
Atlas Icon
Posts: 10,638
Political Matrix E: 1.38, S: -0.51
|
|
« on: June 22, 2007, 04:25:57 AM » |
|
Like many others I have noted that there have been many problems with the Zogby Interactive polls in the past.
I want at this time to note that Zogby has invested the time and money to make this system work reasonably well.
Now one of the rather simplistic criticisms of ZI has been that the respondents are unrepresentative. Well, with respect to matters such as partisan makeup, age, gender, income, occupation, education, residency, etc. this can be handled by weighting.
A more difficult area is that the respondents in ZI are more knowledgeable and prone to strong opinions relative to respondents in conventional telephone surveys. This is both difficult to adjust for, and actually a potential strength. It is a good (abeit imperfect) indicator of the direction in which opinion is moving.
Other benefits of ZI are that it is quicker, cheaper and includes vastly larger samples.
Given the problems with telephone surveys (the response rate is really pretty low), it may well be that telephone surveys may go the way of the face to face surveys which we employed from the late thirties to the mid seventies.
In conclusion, John may have made one really smart investment, and have the last laugh in the survey research community.
|