A sample of 300 readers, from a population of 310,000 (InfoWorld subscribers) with 95% percent confidence yields an error of 5.7% in a worst case scenario (a score of 50%).
On a best case scenario, say a score of 95%, we get a margin of error of 2.5%.
Typically speaking, a confidence of 80% is considered acceptable, rather than the higher 95% I posit.
Also bear in mind, we're looking for a sample to reflect the population of InfoWorld subscribers, not the entire world.
Sean Dugan
Research Editor
For questions or comments, email the Forums Editor