Lecture12 - 6 June 2003 Biostatistics 6650-L12 1 Todays...

Info icon This preview shows pages 1–10. Sign up to view the full content.

View Full Document Right Arrow Icon
6 June 2003 Biostatistics 6650--L12 1
Image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
6 June 2003 Biostatistics 6650--L12 2 Today’s Schedule Contingency tables 2x2 tables RxC tables 1-way tables(Goodness of fit) McNemar’s test for paired(correlated) proportions
Image of page 2
6 June 2003 Biostatistics 6650--L12 3 2x2 Tables Recall the test for comparing two independent binomial proportions, p 1 and p 2 , using a normal approximation. Ho: p 1 = p 2 , Ha: p 1 = p 2 not using correction factors 1 2 obs 1 2 i i i i 1 2 1 2 1 2 ˆ ˆ p -p Z = , 1 1 ˆ ˆ pq( + ) n n ˆ p =x /n , i=1,2, x =number of successes in group i ˆ p=(x +x )/(n +n ), ˆ ˆ ˆ ˆ n pq 5 and n pq 5
Image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
6 June 2003 Biostatistics 6650--L12 4 2x2 Tables Another way to test the same hypothesis is by considering a 2x2 contingency table “Success” “Failure” Population 1 Population 2 This can be extended to test the equality of K proportions Ho: p 1 =p 2 =…=p K We can evaluate two types of hypotheses using 2x2 tables 1. Ho: homogeneity (all proportions equal) 2. Ho: independence of rows and columns
Image of page 4
6 June 2003 Biostatistics 6650--L12 5 2x2 Tables Now fill in the table with observed frequencies “Success” “Failure” marginal totals Population 1 x 1 n 1 -x 1 n 1 Population 2 x 2 n 2 -x 2 n 2 x 1 +x 2 N-(x 1 +x 2 ) N=n 1 +n 2 =Grand total marginal totals Under the null hypothesis, p 1 =p 2 =p, so we combine both groups to estimate the overall proportion of successes: 1 2 1 2 (x + x ) ˆ p= n + n
Image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
6 June 2003 Biostatistics 6650--L12 6 2x2 Tables Acute Renal Failure example(ARF): Hospitalized ARF patients requiring dialysis. Randomized trial comparing type of dialyzer membrane (cellulose vs polysulfone). Outcome is proportion(p) with recovery of renal function within 30 days. Two independent treatment groups with fixed n1 and n2 Recovery No recovery • Cellulose x 1 =19 n 1 -x 1 = 14 33=n 1 , =19/33=.58 Polysulfone x 2 =13 n 2 -x 2 = 20 33=n 2 , =13/33=.39 32 34 66=N Ho: p 1 =p 2 =p, Homogeneity hypothesis Row totals fixed. Column totals random. Mayo Proc 2000, 75:1141-47 1 ˆ p 2 ˆ p
Image of page 6
6 June 2003 Biostatistics 6650--L12 7 2x2 Tables Association of Gleason score and lymph node status in N=485 1990 Mayo prostatectomies--L4 N is fixed, n1 and n2 not fixed. Cross classify two factors. Node - Node + Gleason<7 340 31 371 Gleason 7+ 85 29 114 425 60 485=N Ho: row factor is independent of column factor I.e. that Pr(Gleason 7 and node +)=Pr(Gleason 7)Pr(Node +) Ha: is that the two factors are associated
Image of page 7

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
6 June 2003 Biostatistics 6650--L12 8 2x2 Tables Association of Gleason score and lymph node status in N=485 1990 Mayo prostatectomies--L4 Expected cell proportions under independence model: p ij =p i. p .j Node - Node + Gleason<7 .67 .09 .76 Gleason 7+ .21 .03 =.12(.24) .24 =114/485 .88 .12 1.00
Image of page 8
6 June 2003 Biostatistics 6650--L12 9 2x2 Tables: Notation It turns out that the statistical test used is identical for both the homogeneity and independence hypotheses The test statistic, X 2 obs , is based on observed and expected frequencies or percentages Let O ij = observed number of subjects in the i-th row and j-th column of the table.
Image of page 9

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Image of page 10
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern