Share what you know with millions of people
Focus is the best place to turn what you know into remarkable content
0
Polls vs. Surveys: how does your definition of an 'adequate sample size' change? Does it?
In your opinion, what constitutes an adequate sample size on a survey? Do you adhere to the formula for a 'statistically significant sample' or do you arbitrarily set certain thresholds (ie. 100, 250, etc.)? Does this definition change for a [single] poll question? Why or why not?
Events
- Dos and Don'ts of Small Business Marketing May 29 @ 11 am PT
- Lead Nurturing 202: The Next Generation May 31 @ 11 am PT
- The Tricks to Paid Media June 6 @ 11 am PT
- Display Advertising for Brand Awareness June 20 @ 11 am PT
Related Activity
2 Answers
The calculation of sample's margin of error is the same for polls and surveys. The big difference is in sampling technique and the target population. When a poll is being conducted to predict the outcome of an election, the pollster needs a model that tells them who is a likely voter and who is not. These models are the proprietary difference between one pollster and another and they have a lot to do with the definition of a target population (screening questions), the weight given to responses, and eventual accuracy of the poll.
If you go to most any statistics book, you'll see the formulas for calculating the sample size needed for various accuracy levels. Without some statistics background, these can be confusing and daunting. What the accuracy level tells us is how good the data from the sample will be as a reflection of how our population (group of interest) as a whole feels.
There are several factors that affect the needed sample size. I have an article that addresses these -- and an excel spreadsheet that is a sample size calculator. http://www.greatbrook.com/survey_statistical_confidence.htm
As a rough rule of thumb with a decent size population, 100 responses are needed for marginal accuracy and 200 responses will provide a reasonable statistical accuracy. Those aren't arbitrary numbers; it's just the way the statistics work out.
Answer This Question