The P value of a test comparing two drugs is 0.01, what is the inference? (LQ)
|A||Probability that drug A is better then drug B when in truth it is not is< 1 %|
|B||Probability that drug A is better then drug B when in truth it is not is 1 %|
|C||Probability that drug A is better then drug B when in truth it is not is > 1 %|
|D||No Probability that drug A is better then drug B|
Ref: High Yield Biostatistics, Pg: 34-36& Park PSM 19th ed. 705
Two types of errors can be made in the test of hypothesis:
Type I Error :- That is the rejection of null hypothesis in spite being true
Also called as False positive error i.e. in reality there is no difference but the study has concluded a difference.
The probability of committing type I error is P value. Hence please remember it as a rule: Any question based on p-value, you only have to look at that option that describes a type I error or a false positive error. Any option beginning with type II error or β or false negative error or the power of the study can never be the answer to this question.
Type II Error, Beta Error or False negative error: Accepting null hypothesis in spite being false. This also means that in reality there is a difference but the study has failed to detect that difference.
Power (1-Beta)is defined as the probability of rejecting the null hypothesis when it is false. Normally accepted value for power is 80%.
Levels of type I error are usually kept at or under 5%, the measure of type I error is also called the ‘p’ value. Thus a ‘p’ value < .05 means the probability of getting a difference as extreme as; or more, given the fact that in reality no difference exists is less than 5%, in other words, the difference because of chance; is less than 5%.