DrVirgo
Forum Hero
Topics: 1084
Posts: 3,472
|
What is the probability of making a type I error?
what is the prob. of making a type II error?
___________________
Our greatest glory is not in never falling, but in rising every time we fall.
|
| DrVirgo
Forum Hero
Topics: 1084
Posts: 3,472
|
Which is worse?
___________________
Our greatest glory is not in never falling, but in rising every time we fall.
|
| satyaking
Forum Junior
Topics: 5
Posts: 58
|
The probability of making type I erroris alpha,P<0.05. The probability of making type II error is
beta.
|
| satyaking
Forum Junior
Topics: 5
Posts: 58
|
The probability of making type I erroris alpha,P<0.05. The probability of making type II error is
beta.
|
| satyaking
Forum Junior
Topics: 5
Posts: 58
|
Type I error generally represents False negative,that is as P becomes less than 0.05 the false negative are going to be decreased.In the same way in Type II error the power of the test is denoted by 1- Beta.the power represents the true negatives where as beta represents the false positives.hence a balance should be struck so that there is a balance between them.but most clinical research studies P less than 0.01.hence as cut off point is very low the alpha error is decreased but there is a chance of increased beta error,that is false pos.. are increased.and vice versa occurs.hence balace should be struck between the two.
hope Iam clear about it
|
| jonathon
Forum Senior
Topics: 25
Posts: 62
|
I think alpha error is worse. It's like saying a drug does produce a clinically significant when in fact it does not..(Like a Lie). A type 2 error is like saying a drug does not work when it fact it does have therapeutic significance. ??? I think this is it..correct me if i'm wrong.
|
| DrVirgo
Forum Hero
Topics: 1084
Posts: 3,472
|
You all are right. The probability of making a type I Error is alpha or The probability of making a type II Error is Beta. Beta=1-Power.
Now whats the answer to this Q? and WHY???
8. A group of researchers mistakenly conclude from a poorly designed experiment that acetaminophen cures the common cold. They have committed which of the following?<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />
A. 1-β error
B. Alpha error
C. Beta error
D. Type I error
E. Type II error
___________________
Our greatest glory is not in never falling, but in rising every time we fall.
|
| satyaking
Forum Junior
Topics: 5
Posts: 58
|
Type II error,because they have would taken alpha very less hence more chance of type II error
|
| sturge_weber
Forum Guru
Topics: 77
Posts: 1,042
|
i think alpha error
|
| Geroo
Forum Guru
Topics: 114
Posts: 799
|
isn't typteI error= alpha error?or they are different?
|
| sarika
Forum Guru
Topics: 195
Posts: 1,200
|
ans:type 1 error
alpha is the probability of making type one error .
|
| DrVirgo
Forum Hero
Topics: 1084
Posts: 3,472
|
The correct answer is D. Type I errors occur when researchers reject the null hypothesis when they should not have. In other words, they conclude a significant result when in actuality it does not exist. This is a particularly dangerous error to make, as it could lead to the administration of an ineffective drug to patients in need of life-preserving treatment. <?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />
Thanks everyone.
To conclude:
Type I error is worse
alpha is the probability of making a type I Error.
Type II error = 1-Power
(Type II error (beta error) does NOT equal alpha!)
___________________
Our greatest glory is not in never falling, but in rising every time we fall.
|
| mjl1717
Forum Hero
Topics: 955
Posts: 5,451
|
Very good Dr. Virgo, because the definitions of both may appear as "double talk" to many". But in inactuality this is a significant part of biostat and epidemiolgy.
___________________
Smell the coffee! "Is That an Osler move??"
|
| leen
Forum Guru
Topics: 79
Posts: 294
|
thanks
|
| zubr
Forum Junior
Topics: 14
Posts: 61
|
Is Type I error the same as alpha error?
Where is this question coming from? I assume that is not from actual USMLE exam, or from NBME.
Alpha = I = False Positive (in reality it is false, but was mistakebly accepted by study as truth, a bad drug but shown as a good one = LIE)
ONLY if we reject H(0) but must not - alpha(I) error
ONLY Rejecting H(0) we could make alpha(I) error with chance 5% (5 from 100, 1 from 20) if p=0.05
Beta= II = False Negative (in reality it is truth, but was rejected, a good drug but shown as a bad one = MISTAKE)
ONLY if we accept H(0) but must not - beta(II) error
ONLY Accepting H(0) we could make beta(II) error with chance 1-Power (p=0.05 does not matter)
LIE(I) is worst then MISTAKE(II)
----------------------------------------------REALITY
----------------------------------TRUTH(+)------------FALSE(-)
STUDY SHOW TRUTH (+)---------TP-------------------FP(I)alpha
STUDY SHOW FALSE (-)-------FN(II)beta---------------TN
TP=POWER= 1 - beta
|
|
| |
| | | | | | | | | | | | | | | |
