Documents
Resources
Learning Center
Upload
Plans & pricing Sign in
Sign Out

Exercises

VIEWS: 0 PAGES: 22

									 EXERCISES
• POP QUIZ
• POOLING
• LOGISTIC REGRESSION
POP QUIZ
  Exercises on
Pooling Exchangeable
  Matched Strata
  Exercises on
Logistic Regression
For Matched Data
          Answers to Exercises on
           Pooling Exchangeable
              Matched Sets
1.    How many concordant pairs are there where both pair members are exposed? «1»
2.    How many concordant pairs are there where both members are unexposed? «1»
3.    How many discordant pairs are there where the case is exposed and the control is unexposed? «2»
4.    How many discordant pairs are there where case is unexposed and the control is exposed? «1»
5.    This table is called «McNemar's» table.
6.    What is the estimated mOR for these data? «2»
7.    What type of matched analysis is being used with this table? «unpooled»
8.    What is the estimated mOR from these data (use the formula for the mOR in the box below)? «2.5»
9.    What type of matched analysis is being used here? «pooled»
10.   Which type of analysis should be preferred for these matched data (where smoking status is the
         only matched variable), pooled or unpooled? «pooled»
11.   W= «2»? 12. X = «1»? 13. Y =«0»? 14. Z= «2»?
15.   mOR (unpooled) = «undefined».
16.   mOR (pooled) = «2.5».
17.   Which of the following helps explain why the pooled mOR estimate should be preferred to
       the unpooled mOR?
       a. The pooled mOR's are equal whereas the unpooled mOR's are different.
       b.The unpooled mOR's assume that exchangeable matched pairs are not unique.
       c.The pooled mOR's assume that exchangeable matched pairs are unique.
        «a»
        Answers to Exercises on
         Logistic Regression for
            Matched Data

1.   Which of the logistic models shown below is appropriate for analyzing these data?
     «b»
2.   What method of estimation should be used to fit the appropriate model for these
     data? «Conditional MLE»
3.   What is the adjusted odds ratio for the effect of SMK on cervical cancer outcome?
     «exp(1.4361)»
4.   Is number of lifetime sexual partners a significant predictor of cervical cancer
     outcome, controlling for the matching variables and the other non-matching
     variables? «yes»
5.     Are the matching variables being controlled? «yes»
6.   A 95% CI for the (adjusted) effect of SMK is given by the expression
        «exp[1.4361 +- 1.96(.3167)]»

								
To top