214k views
4 votes
Why wouldn't a researcher set the confidence level at .0001, since it's better than .05 and even better than .01?

2 Answers

4 votes

A researcher would not set the Type I level at .0001 because it is too small. It suggests that there is a 1 in 10,000 chance (a REALLY small risk) that you will reject the null hypothesis (that there are no differences) when it is actually true.

User Jontsnz
by
7.8k points
4 votes
  • Setting a confidence level at .0001 cannot give you what you want. That is because it is too small.

It shows that in every 10,000 opportunities given, one one chance is lucky enough.

That is a really small risk that will you will reject the null hypothesis

  • It is not right to select a very low statistical level in testing a null hypothesis such as α=0.0001α=0.0001

Conclusively, a low confidence interval as this indicate of how much uncertainty there is in an estimate of the true mean and as such, the narrower the interval, the more precise is our estimate.

Learn more from

User Maynull
by
7.2k points