A capsule manufacturing company is having difficulty filling capsules to the required 100 milligrams. Suppose a business researcher randomly selects 80 capsules, weighs each of them and computes its mean. By chance, the researcher selects capsules that have been filled adequately and that is how he gets the mean weight of 100 mg, which falls in the "fail to reject" region. The decision is to fail to reject the null hypothesis even though population mean is NOT actually 100 mg. Which kind of error has the researcher done in this case?