Basic question of Hypothesis and Type I and II errors

#1
Hi,
I'm new to statistics. I have some basic question on hypothesis and Type-I and II errors. I'm posting two scenarios on the same topic below. Need your help to get some knowledge on the 2nd scenario

Scenario-1:
Let us say Null Hypothesis: Ho: I don't have covid

But due to government terms and conditions, I'm giving my blood sample for testing. Based on the test results, below is the type-I and type-II errors

Type-I error: Rejecting the true null hypothesis
Test result says - I have covid but in reality I don't have. This is False positive

Type-II error: Accepting the False null hypothesis
Test result says - I don't have covid but in reality I have. This is False negative

I'm pretty clear about this scenario

Scenario-2:
This is the one which I'm not able to understand. Please help

Let us say Null Hypothesis: Ho: I have covid (Just opposite to Scenario-1)

But due to government terms and conditions, I'm giving my blood sample for testing. Based on the test results, below is the type-I and type-II errors

Type-I error: Rejecting the true null hypothesis
Test result says - I don't have covid but in reality I have. Can we call this as False positive ? Logically speaking this is False negative rite ? Please advise

Type-II error: Accepting the False null hypothesis
Test result says - I have covid but in reality I don't have. Can we call this as False negative ? Logically speaking this is False positive rite ? Please advise


As per Hypothesis rule, Type-I error is called False positive and Type-II error is called False negative. That is the reason Scenario-2 is confusing for me. Please advise
 

Miner

TS Contributor
#2
Scenario 2 is not a valid scenario. The null hypothesis should be the status quo, while the alternate hypothesis is a deviation from the status quo.

The only time that scenario 2 would be valid would be if you were testing for a cure. Status quo is a sick patient. Deviation to status quo is a cured patient.
 

Miner

TS Contributor
#4
In reality, not having an illness is considered the status quo while an illness is a change. Therefore, no illness would be the null hypothesis and illness would be the alternate. It would not be appropriate the state it the other way.

  • The null hypothesis is a statement that assumes there is no relationship between two variables, no association between two groups or no change in the current situation — hence ‘null’. It is denoted by H0.
  • The alternate hypothesis is the opposite of the null hypothesis because it assumes that there is some relationship between two variables or there is some change in the current situation — hence ‘alternate’. It is denoted by Ha or H1.