The likelihood that the difference is due to chance - USMLE Forums
USMLE Forums Logo
USMLE Forums         Your Reliable USMLE Online Community     Members     Posts
Home
USMLE Articles
USMLE News
USMLE Polls
USMLE Books
USMLE Apps
Go Back   USMLE Forums > USMLE Step 1 Forum

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


Reply
 
Thread Tools Search this Thread Display Modes
  #1  
Old 08-19-2011
USMLE Forums Veteran
 
Steps History: Not yet
Posts: 296
Threads: 41
Thanked 167 Times in 102 Posts
Reputation: 177
Stats The likelihood that the difference is due to chance

two experimental drugs are being researched for the treatment of CHF

patients getting drug X have a cardiac index of 2.5L/m with a 95% confidence interval between 1.5 to 3.5.
patients getting drug y have cardiac index with 95% confidence interval between 0.7 to 3.7. a teat of significance of difference shows a p-value of 0.1. which of the following is the likelihood that the difference in mean cardiac index of patients getting drugs x and y is due to chance:
a.0%
b.2.5%
c.5%
d.7.5%
e10%
f.66.7%
g.95%

i would really appreciate a concise explanation as i have no idea what they are talking about,i have read statistics in FA
thanks
Reply With Quote Quick reply to this message



  #2  
Old 08-20-2011
bebix's Avatar
USMLE Forums Master
 
Steps History: 1+CK+CS+3
Posts: 1,357
Threads: 194
Thanked 3,262 Times in 881 Posts
Reputation: 3272
Default

which of the following is the likelihood that the difference in mean cardiac index of patients getting drugs x and y is due to chance:

This is the same as saying "we fail to reject H0"

H0 = difference =0
HA = difference =! 0

So, you preset an alpha value of 5% (0.05)
Then you did the test statistics under the null hypothesis and you got a p value = 0.10 (10%) = post-test

Conclusions = you fail to reject the null hypothesis

When the p-value is < 0.05, reject the null hypothesis. With such a low probability for the p-value, there is little likelihood that the observed difference between the sample mean and hypothesized mean is due to chance - it must be do to some program, process change, intervention or other effect.

When the p-value is > 0.05, fail to reject the null hypothesis. There is a high probability for the p-value that the observed difference between the sample mean and the hypothesized mean is so small that it must be do to chance involved in sampling error.

In this case, the answer is 10%
Reply With Quote Quick reply to this message
The above post was thanked by:
dinosaur108 (08-30-2011)
  #3  
Old 08-20-2011
drnrpatel's Avatar
Guest
 
Steps History: 1+CK+CS
Posts: 441
Threads: 153
Thanked 396 Times in 211 Posts
Reputation: 419
Default

Here difference in mean is due to chance is null hypothesis.

Now, p = probability of making alpha error or Rejecting the null hypothesis when it was true. = 0.1 = 10%

When p>0.05 we do not reject null hypothesis.

So yeah difference was definitely due to chance and 10% chance were there we may have rejected it. And also 10% is likelihood of difference is due to chance!!!!!!
Reply With Quote Quick reply to this message
 
  #4  
Old 08-20-2011
bebix's Avatar
USMLE Forums Master
 
Steps History: 1+CK+CS+3
Posts: 1,357
Threads: 194
Thanked 3,262 Times in 881 Posts
Reputation: 3272
Default

Quote:
Originally Posted by drnrpatel View Post
Here difference in mean is due to chance is null hypothesis.

Now, p = probability of making alpha error or Rejecting the null hypothesis when it was true. = 0.1 = 10%

When p>0.05 we do not reject null hypothesis.

So yeah difference was definitely due to chance and 10% chance were there we may have rejected it. And also 10% is likelihood of difference is due to chance!!!!!!
Be carful though.

the probability of a Type I error here is equal to alpha = 0.05
alpha is NOT the same as p-value (in this case 0.10)

Reply With Quote Quick reply to this message
  #5  
Old 08-20-2011
USMLE Forums Veteran
 
Steps History: Not yet
Posts: 296
Threads: 41
Thanked 167 Times in 102 Posts
Reputation: 177
Default thanks

thanks for the explanations
Reply With Quote Quick reply to this message
  #6  
Old 08-25-2011
USMLE Forums Guru
 
Steps History: Not yet
Posts: 324
Threads: 70
Thanked 238 Times in 144 Posts
Reputation: 248
Warning!

This is NBME form 3 question..
Posting NBME qs is not allowed.
Reply With Quote Quick reply to this message
  #7  
Old 08-30-2011
USMLE Forums Addict
 
Steps History: 1+CK+CS+3
Posts: 117
Threads: 6
Thanked 78 Times in 35 Posts
Reputation: 88
Default

Quote:
Originally Posted by bebix View Post
Be carful though.

the probability of a Type I error here is equal to alpha = 0.05
alpha is NOT the same as p-value (in this case 0.10)

Alfa crieterion is hypothesized - .05
alfa error is p value - .1 in this case
Reply With Quote Quick reply to this message
  #8  
Old 08-30-2011
bebix's Avatar
USMLE Forums Master
 
Steps History: 1+CK+CS+3
Posts: 1,357
Threads: 194
Thanked 3,262 Times in 881 Posts
Reputation: 3272
Default

Quote:
Originally Posted by dinosaur108 View Post
Alfa crieterion is hypothesized - .05
alfa error is p value - .1 in this case
no, they are not the same.

alpha is the probability of a type I error. In this case is a PREtest probability equal to 0.05 (something that an investigator can manipulate BEFORE do the actual experiment)
(probability alpha = Probability to reject the Null given the Null is true)

p-value is a POSTtest probability...you cant change this...is is the probability that you get with the data that you have in that particular time! (in this case 0.1)

The interpretations are completely different.

So first you preset an alpha = 0.05....then you do the test statistic and you get a p-value...if this p-value is less than you preset alpha, you'll reject the null hypothesis (there is little likelihood that the observed difference between the sample mean and hypothesized mean is due to chance)

Reply With Quote Quick reply to this message
  #9  
Old 08-30-2011
USMLE Forums Addict
 
Steps History: 1+CK+CS+3
Posts: 117
Threads: 6
Thanked 78 Times in 35 Posts
Reputation: 88
Question

Quote:
Originally Posted by nativedoc2 View Post
two experimental drugs are being researched for the treatment of CHF

patients getting drug X have a cardiac index of 2.5L/m with a 95% confidence interval between 1.5 to 3.5.
patients getting drug y have cardiac index with 95% confidence interval between 0.7 to 3.7. a teat of significance of difference shows a p-value of 0.1. which of the following is the likelihood that the difference in mean cardiac index of patients getting drugs x and y is due to chance:
a.0%
b.2.5%
c.5%
d.7.5%
e10%
f.66.7%
g.95%

i would really appreciate a concise explanation as i have no idea what they are talking about,i have read statistics in FA
thanks
can anyone tell me what is the chance that study result don't match the effect in large population?
Reply With Quote Quick reply to this message
  #10  
Old 08-30-2011
bebix's Avatar
USMLE Forums Master
 
Steps History: 1+CK+CS+3
Posts: 1,357
Threads: 194
Thanked 3,262 Times in 881 Posts
Reputation: 3272
Default

Quote:
Originally Posted by dinosaur108 View Post
can anyone tell me what is the chance that study result don't match the effect in large population?
I dont understand your qs. (large population)

here we have a 95% CI...so, we are talking about population parameters ("mu" and not "mu hat")....we are 95% confident that the population parameter mu lies between the lower and the upper bound...
Reply With Quote Quick reply to this message
  #11  
Old 08-30-2011
USMLE Forums Addict
 
Steps History: 1+CK+CS+3
Posts: 117
Threads: 6
Thanked 78 Times in 35 Posts
Reputation: 88
Default

Quote:
Originally Posted by bebix View Post
I dont understand your qs. (large population)

here we have a 95% CI...so, we are talking about population parameters ("mu" and not "mu hat")....we are 95% confident that the population parameter mu lies between the lower and the upper bound...
likelihood that the difference in mean cardiac index of patients getting drugs x and y is due to chance is type 1 error = p = 10%

when there is failure to reject null hypothesis (P>5%) as here which is 10% chance of alfa error is 0 bcz we dont put that drug in practical life.............but what is the chance of beta error(
difference in mean cardiac index of patients may be due to drug and not due to chance)?
what is the power ?

Reply With Quote Quick reply to this message
  #12  
Old 08-30-2011
bebix's Avatar
USMLE Forums Master
 
Steps History: 1+CK+CS+3
Posts: 1,357
Threads: 194
Thanked 3,262 Times in 881 Posts
Reputation: 3272
Default

Quote:
Originally Posted by dinosaur108 View Post
likelihood that the difference in mean cardiac index of patients getting drugs x and y is due to chance is type 1 error = p = 10%

when there is failure to reject null hypothesis (P>5%) as here which is 10% chance of alfa error is 0 bcz we dont put that drug in practical life.............but what is the chance of beta error(
difference in mean cardiac index of patients may be due to drug and not due to chance)?
what is the power ?

Again. Type 1 error is not the same as p-value.

Type 1 error = reject the null given the null is true
Alpha = probability of a type 1 error

In this case: type 1 error (PRE-SET) = 0.05
p-value (POST)= 0.10

The term significance level (alpha) is used to refer to a pre-chosen probability and the term "P value" is used to indicate a probability that you calculate after a given study.

Beta again is something that you must decide before you run your experiment (since youŽll need that value and the alpha value to calculate the sample size). Typically we use beta = 0.1 (power = .9) or 0.2 (power = 0.8).

Here you can find more info:
http://www.statsdirect.com/help/basics/pval.htm

if you want, you can send me a pm.
Reply With Quote Quick reply to this message
The above post was thanked by:
dinosaur108 (08-31-2011), drnrpatel (08-30-2011)



Reply

Tags
Biostatistics-Epidemiology, Step-1-Questions

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the USMLE Forums forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:
Medical School
Choose "---" if you don't want to tell. AMG for US & Canadian medical schools. IMG for all other medical schools.
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.

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.



Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes


Similar Threads
Thread Thread Starter Forum Replies Last Post
Likelihood Ratio kalpana USMLE Step 1 Forum 2 02-17-2016 10:24 AM
The likelihood that he has the disease? thunderbolt USMLE Step 2 CK Forum 15 05-17-2011 09:28 AM
The likelihood that the patient is having cancer achistikbenny USMLE Step 1 Forum 4 06-29-2010 12:13 AM
Step 3 and Residency Matching Likelihood Steve13 USMLE Step 3 Forum 2 04-19-2010 11:08 AM
do u think I have chance? yellowsand IMG/FMG Forum 4 10-11-2009 11:15 AM

RSS Feed
Find Us on Facebook
vBulletin Security provided by vBSecurity v2.2.2 (Pro) - vBulletin Mods & Addons Copyright © 2017 DragonByte Technologies Ltd.

USMLE® & other trade marks belong to their respective owners, read full disclaimer
USMLE Forums created under Creative Commons 3.0 License. (2009-2014)