Download presentation
Presentation is loading. Please wait.
Published byTimothy Hicks Modified over 8 years ago
1
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.1 Counts and Proportions
2
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.2 Binary Data Often in medical and public health studies, our endpoint of interest is binary or dichotomous Examples disease vs. no disease response vs. no response death vs. no death Success vs. failure
3
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.3 Binary Data Only 2 possible responses Often, continuous endpoints are dichotomized into a binary endpoint For example, in a study of the effect of a drug on LDL levels, for each subject, the LDL measurement at the end of the study (a continuous measure) may be dichotomized into “response” vs. “no response” based on a cutpoint defining whether the LDL level has been reduced to acceptable, normal, or safe levels.
4
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.4 Binary Data Similar to hypothesis testing with continuous data, one may perform hypothesis tests on binary data: 1-sample test of a proportion H 0 : p=p 0 H A : p p 0 2-sample test comparing proportions H 0 : p 1 =p 2 H A : p 1 p 2
5
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.5 Binary Data Also, similar to continuous data, we may derive confidence intervals for A single proportion The difference between two proportions
6
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.6 Note We may use the CLT for binary data also (as the CLT applies to all distributions) But note that the CLT is an asymptotic result (as n ) Thus, we must be careful when n is small
7
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.7 Example Etoposide for the treatment of relapsed or progressed Kaposi’s Sarcoma Insert Etoposide.pdf
8
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.8 Binary Data Insert Binary1.pdf
9
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.9 Exact Confidence Intervals “Exact” confidence intervals for a binomial parameter are possible These do not rely on the normal approximation to the binomial (i.e., use of the CLT) Computationally very intensive (particularly for large N) May require special programming/software
10
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.10 Exact Confidence Intervals General Rule: Use exact confidence intervals whenever software is available and is feasible given the computing resources If N is large then it is OK to use normal approximation (as CLT kicks in) If N is small: the normal approximation may not be appropriate Use exact CIs if possible
11
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.11 Example The proportion of students against the Iraq war
12
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.12 Example Special Case: Observe 0 events or responses. How do we get a CI for the response rate when the variability is 0? Insert A5129_example.pdf
13
Introduction to Biostatistics, Harvard Extension School, Fall, 2005 © Scott Evans, Ph.D.13 Example CI for the difference between two proportions Insert HIV_HCV.pdf Comparing HIV+/HCV+ with HIV+/HCV- individuals with respect to depressive symptomatology
Similar presentations
© 2024 SlidePlayer.com Inc.
All rights reserved.