For example, this answer makes the following statement:

What is true is that [vaccine effectiveness] in people aged 65 or older is 9%, with a 95% confidence interval that spans from -84% to 55%. They suggest interpreting these results with caution.

I want to ask a question like "What does it mean for a confidence interval to include negative values?" because I don't understand the statistics needed to make sense of answers like that. But that's not about a specific claim, so I don't know if it's on topic here. Can I ask a question that basically means "can someone explain an idea to me"?


No, this would not be on topic on either meta or the main site.

You may have a valid question on Cross Validated, a Stack Exchange site "for people interested in statistics, machine learning, data analysis, data mining, and data visualization". I suspect such a question has been asked before and you could likely find it with some searching.


As an alternative to asking a new question on this or another Exchange, you could simply add a comment to the answer to ask for clarification. Explaining uncommon or unusual terminology is definitely an improvement to the answer, so a comment is appropriate and likely to put that relevant information right in the formerly-confusing answer so that future readers will benefit as well.

  • 1
    I think this is the best answer. Regurgitating technical information may answer the question, but it doesn't help most readers. Most readers will need a little hand-holding and a "bottom line". If you can't or won't explain the technical information then please don't answer. Or at least make it a community wiki and invite others to improve it.
    – fredsbend Mod
    Nov 15 '18 at 21:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .