Question

1. By now, you’ve been reminded to “check the assumptions that np≥5 and nq≥5" on many occasions. What are we actually checking when we check these assumptions? For which statistical tests/techniques does this assumption apply?

Answer 1

We know that,

Success/Failure Assumption: if we have 5 or more successes in a binomial experiment (n*p ≥ 10) and 5 or more failures (n*q ≥ 10), then you can use a normal distribution to approximate a binomial.

Hence,

According to this assumption,

There should be 5 or more successes as well as failures, then you can consider binomial distribution to approximately normal distribution.

This assumption is applied when we are testing a single population proportion.

Thank you.