False positive and false negative in Statistics

False-positive or Type I error

The false-positive or type I error occurs when the researcher rejects the null hypothesis even though it is true. It means that despite the null hypothesis is true, the research fails to accept it. Such a problem arises when a researcher sets the value of significance level higher.

For example: 
Ho: There is no relationship between gender and mark obtained
H1: Males score higher marks in comparison to females. 

Suppose that the p-value is estimated as 0.12.
then,
Sig level
Accept or Reject Ho
Remark
0.01
Accept
0.05
Accept
0.1
Accept
0.2
Reject
False-positive
If a researcher sets the tabulated p-value at 20 percent, the researcher encounters a type I error. 
The false-negative or type II error occurs when the researcher fails to reject the null hypothesis even though it is false. It means that despite the null hypothesis is false, the research fails to reject it. Such a problem arises when a researcher sets the value of significance level lower. 

For example: 
Ho: There is no relationship between gender and mark obtained
H1: Females score higher marks in comparison to males.

Suppose that the p-value is estimated at 0.015
Sig level
Accept or Reject Ho
Remark
0.01
Accept
False-negative
0.05
Reject
0.1
Reject
0.2
Reject
If a researcher sets the tabulated p-value at 1 percent, the researcher encounters a type II error.

What is the possible solution?

The researcher shall set the tabulated p-value at 1 percent, 5 percent, and 10 percent, and compare it with the calculated p-value. In general practice, the calculated p-value less than 10 percent is preferable for rejection of the null hypothesis. But it depends upon the discretion of the researcher to set the rejection significance level.  

Post a Comment

Previous Post Next Post