Prep for USMLEPrep for USMLE Forum
   Forum    Step 1  Step 2 CK Step 2 CS Step 3  Match  IMGs Resources Search






Previous Topic | Next Topic  Statistics Question 




 
Kaplan Qbank USMLE



Author15 Posts
  #1

A researcher is examining the effects of a new drug on cognitive performance in people living under stressfull conditions versus those living in relatively stress-free conditions. He selects a "high stress" group that is matched in other charasteristics to an equal number of subjects in a "low stress" group. Half of the people in each group, recieve the experimental drug, half do not. After 1 hour, the performance on a set of cognitive tasks is mesuared in all of the patients. To rpoperly interpret the data concerning the drug's efectiveness, the researcher should use"

CHi squared test

paired test

ANOVA

linear regression

Pearson Correlation

Could anybody tell me whats the best answer and explain which are the variables you would be comparing?. My answer does not match the answer is given by the book. Thank you


  #2

ANOVA?

  #3

Actually, Pearson Correlation is given as an answer but as far i as know, you can only use correlation when both variables are measuare in an intervale scale. I dont know which intervals i could apply here. Honestly i dont know how could i arrange these variables to use this correlation statistic test.

  #4

Pearson Correlation uses 2 intervals, 0 nominal variables. In the question the variables used were high stress group vs low stress group.

ANOVA analyzes 3 groups.


  #5

Thanks for helping me our here. Even though the book says it is correlation the statistic test to be used, i think its not the appropiated test. Would you tell me which are the 3 groups and the interval and nominal variables to apply ANOVA?




  #6

Let me try to help you

1. Identify the data a)nomial i.e. not divisible into units like here high stress and low stress and then those given drugs and those not given (so we have 2 nominal variables here stress and drug. b)interval : any data in units like height and weight blood pressure .In this example the cognitive test scoring ( so 1 interval variable the cognitive test)

if you have only nominal variable (whatever the no.) do Chi Square, if you have only interval variable go for Correlation. One nomianl and one interval go for T test but if more than 2 variables (atleast one of each) go for ANOVA


___________________
In Life there are neither rewards nor punishments, just consequences.

  #7

Thank you for your post. I defenitly think Anova - 2 ways is the better test to apply here. 2 nominal way and 1 interval. Dont know why they give correlation as an answer. Anyway now i cleared up my mind.

Thank you


  #8

Hi, I think Pearson Correlation is the right answer, because you can not just nominate a person and say he is under "high stress" or "low stress", you have to do some measurement possibly by questionaire, which involves interval; same reason for the drug effect evaluation, "the performance on a set of cognitive tasks is mesuared in all of the patients", also interval.

  #9

well stress if measured on some intreval scale couldnot be divided into high and low stress group cause interval scale usually are not grouped .

___________________
In Life there are neither rewards nor punishments, just consequences.

  #10

well does pearson correlation not measure interval data, and in this case there re 2 nominal data, hence chi square must be the ans

plzzz correct if u think i am wrong


  #11

How about a Paired T test ? There is 1 Nominal data- High stress Vs Low Stress. Then theres 1 interval data- perfomance on the cognitive test. And the q does say that the groups are "matched" confused

  #12

yes the answer is ANOVA

  #13

Guys...if we look at the 'data set' we have here- we are actually comparing 2 numerical scores (aka performance on Cognitive test) between the groups so I'd go with a Pearson Correlation here.

___________________
Courage does not always ROAR. Sometimes courage is the quiet voice at the end of the day saying, "I will try again tomorrow" - Mary Anne Radmacher

  #14

its is pearson

___________________
If you think you can You can! If you think you cant you are right again!!

  #15

msyamp is right. its pearson.

msyamp, im very much impressed with ur explanations. my exam is on jan 18. when is urs??

give me ur mail id.








You don't have permission to post.




Login or Register to post messages in this topic





















Contact | Leaders | Disclaimer | Privacy

Copyright @ Prep for USMLE. All rights reserved.