In logistic regression, what is the minimum confidence level recommended for dependable results?

• A. 85%
• B. 90%
• C. 95%
• D. 99%

Answer: (B)

In logistic regression, confidence level tells us how sure we want to be about the interval estimates for the model’s parameters and for predicted probabilities. While 95% is standard and many results report that level, 90% is often viewed as the practical minimum that still offers respectable protection against random variation without requiring excessively large samples. Lower levels (like 85%) can give intervals that are too optimistic and narrow, risking overconfidence, while higher levels (such as 99%) produce much wider intervals that can be less informative unless you have a lot of data. So 90% strikes a sensible balance for dependable results in many applied scenarios.