Hacker News Reader

I don't know much about statistics, but is the sample size of 1000 people surveyed here significant enough to be drawing conclusions from? It feels small to me, would anyone with a stronger background in this area care to comment?

View on HN

1000 is enough for this case.

Depends on how "accurate" you want to measure a ratio. Here the ratio is the percentage of Americans that support net neutrality.

In this case, sample size is 1000, estimated ratio is 76%, the 95% confidence interval is 76% +/- 2.65%, which means if you repeat this survey again, you have a 95% chance that new estimated ratio is within 76% +/- 2.65%.

Edit: 99% confidence interval is 76% +/- 3.48%

Well, it depends on where the 1000 people are drawn from as well. Your assumptions only hold if for both the original survey and the repeat, the pool of people being surveyed are an accurate sample of all of the United States.

Good point. On the other hand, the confidence interval I provided is correct as long as the original survey and the repeat are sampled in the same way (regardless whether it is un-biased sampling or not)

Reply on news.ycombinator.com