glasskerfu

2021-01-10

A company is marketing a new product they say works better than the traditional test tube. There is so much interest in the product that 30 different labs around the world are testing the claim that this product is actually better. If each study uses an alpha level (alpha) of .10, and if the null hypothesis is true (that the test tube isn't any better that the traditional one), how many of the hypothesis tests would we expect to incorrectly find statistical significance (that is, conclude that the new test tube is better, when it actually isn't)?

Macsen Nixon

Step 1
When a researcher incorrectly rejects the null hypothesis or incorrectly finds the statistical significance, then this type of error is termed as type 1 error. Type 1 error is the rejection of true null hypothesis.
Step 2
The probability of committing a type I error is represented by our alpha level $\left(\alpha \right)$, which is the p-value below which we reject the null hypothesis. A p-value of 0.10 indicates that we are willing to accept a 10% chance that we are incorrectly rejecting the null hypothesis.
Hence we expect 10% of the hypothesis tests to incorrectly find statistical significance.

Do you have a similar question?