Biostat - p value - USMLE Forums
 USMLE Forums         Your Reliable USMLE Online Community     Members     Posts
 Home USMLE Articles USMLE News USMLE Polls USMLE Books USMLE Apps
 USMLE Forums Biostat - p value
 Register FAQs Members List Search Today's Posts Mark Forums Read

 USMLE Step 1 Forum USMLE Step 1 Discussion Forum: Let's talk about anything related to USMLE Step 1 exam

#1
06-19-2012
 USMLE Forums Guru Steps History: 1+CK+CS Posts: 394 Threads: 40 Thanked 218 Times in 99 Posts Reputation: 228
Biostat - p value

A report of a clinical trial of new drug for HSV2 vs a placebo noted that new drug gave a higher proportion of success then placebo. Report ende with a statement:chi-sq = 4.72, p=<0.05. So, we may conclude that

A) fewer than 1 in 20 will fail to benefit from the drug
B) the chance hay an individual patient will fail to benefit is less than 0.05
C) if the drug were effective, the probability of the reported finding is less than 0.05
D) if the drug wer ineffective, the probability of the reported finding is less than 0.05
E) null hypothesis is false

Do explain.. How to improve Bio-stat, it one of my weak subject?
__________________
The Journey of thousand miles begins with a single step.

#2
06-19-2012
 USMLE Forums Addict Steps History: 1 + CS Posts: 193 Threads: 29 Thanked 134 Times in 71 Posts Reputation: 144

where did u get this question?
E] is an answer for sure, 100%,, since the p is less than .05, u have to reject the null, which means it's false

but also D] could be correct,,, if the findings were due to an [alpha error],,, the probability of that is also less than .05

so??
__________________
i'm so bad i make medicine sick,
Imma show you how great i am
#3
06-19-2012
 USMLE Forums Addict Steps History: 1 + CS Posts: 193 Threads: 29 Thanked 134 Times in 71 Posts Reputation: 144

on second thought, he said nothing about results if they were more or less than [p], se we cant decide to accept or reject the null, i guess its only [D] left
__________________
i'm so bad i make medicine sick,
Imma show you how great i am

#4
06-19-2012
 USMLE Forums Master Steps History: 1+CK+CS Posts: 1,039 Threads: 189 Thanked 570 Times in 323 Posts Reputation: 580

Quote:
 Originally Posted by Rookie A report of a clinical trial of new drug for HSV2 vs a placebo noted that new drug gave a higher proportion of success then placebo. Report ende with a statement:chi-sq = 4.72, p=<0.05. So, we may conclude that A) fewer than 1 in 20 will fail to benefit from the drug B) the chance hay an individual patient will fail to benefit is less than 0.05 C) if the drug were effective, the probability of the reported finding is less than 0.05 D) if the drug wer ineffective, the probability of the reported finding is less than 0.05 E) null hypothesis is false Do explain.. How to improve Bio-stat, it one of my weak subject?
D) if the drug were ineffective, the probability of the reported finding is less than 0.05

Ok, here's my approach.
I would ignore chi-sq and pay attention to p-value only. It's < 0.05, which = type I error = if we reject the null hypothesis there is <0.05 chance we will be wrong = if we say that there is a difference btw effectiveness of placebo and the drug, there is <0.05 chance that the drug is actually no better than placebo.
The wording of the answers is bizarre to me..
#5
06-19-2012
 USMLE Forums Master Steps History: 1+CK+CS+3 Posts: 791 Threads: 76 Thanked 673 Times in 317 Posts Reputation: 691

my answer is D.....p value less than 0.05 means we reject null hypothesis and the probability of error is <0.05.

and reg option E...we cannot say that null hypothesis is true of false. you just reject or fail to reject it. Rejection does not necessarily mean its false....!!!!
#6
06-19-2012
 USMLE Forums Scout Steps History: 1+CK+CS Posts: 40 Threads: 1 Thanked 51 Times in 16 Posts Reputation: 61

yeah it's D
it can't be E because you can't be 100% sure that the null hypothesis is false, even if the p-value<0.05, there's still a 0.05 chance that it's true !
#7
06-20-2012
 USMLE Forums Guru Steps History: 1+CK+CS Posts: 394 Threads: 40 Thanked 218 Times in 99 Posts Reputation: 228

Nice Explaination casandra. It's from Kaplan notes review I couldn't understand the Explaination
Quote:
 Originally Posted by Casandra I would ignore chi-sq and pay attention to p-value only. It's < 0.05, which = type I error = if we reject the null hypothesis there is <0.05 chance we will be wrong = if we say that there is a difference btw effectiveness of placebo and the drug, there is <0.05 chance that the drug is actually no better than placebo. The wording of the answers is bizarre to me..
Points to remember

Look at p-value

<0.05 - reject null hypothesis chance of type I error- probability of error if the drug is not better than placebo is same as p-value i.e, in this case 0.05

>0.05- do not reject the null hypothesis Remember you never accept null hypothesis nor term it as true or false in that case you are likely to encounter type II error which is not related p-value. Type II error=1-statistical power(do not confuse power with p-value, power is given as % unlike p-value)

__________________
The Journey of thousand miles begins with a single step.
 The above post was thanked by: Casandra (06-20-2012)

Message:
Options

Register Now

In order to be able to post messages on the USMLE Forums forums, you must first register.
User Name:
Medical School
Choose "---" if you don't want to tell. AMG for US & Canadian medical schools. IMG for all other medical schools.
 AMG IMG ---
USMLE Steps History
What steps finished! Example: 1+CK+CS+3 = Passed Step 1, Step 2 CK, Step 2 CS, and Step 3.

Choose "---" if you don't want to tell.

 Not yet Step 1 Only CK Only CS Only 1 + CK 1 + CS 1+CK+CS CK+CS 1+CK+CS+3 ---
Favorite USMLE Books
 What USMLE books you really think are useful. Leave blank if you don't want to tell.
Location
 Where you live. Leave blank if you don't want to tell.

Log-in

Human Verification

In order to verify that you are a human and not a spam bot, please enter the answer into the following box below based on the instructions contained in the graphic.