Skip to content

Latest commit

 

History

History
69 lines (49 loc) · 2.87 KB

qstudy.md

File metadata and controls

69 lines (49 loc) · 2.87 KB
Raw
pagetitle title author date output
BBR Study Questions
Biostatistics for Biomedical Research Study Questions
Frank Harrell
2020-06-10
rmdformats::readthedown
usebookdown hightlight mathjax
false
true

Biostatistics for Biomedical Research

Chapter 10: Ordinary Linear Models

Section 10.1

  1. What are some problems with matching as an adjustment strategy?

Section 10.2

  1. Do we need the $t$-test, Wilcoxon test, and log-rank test?
  2. Why is a statistical model required for estimating the effect of going from a systolic blood pressure of 120mmHg to one of 140mmHg?

Section 10.5

  1. For a continuous response variable Y why are we often interested in estimating the long-term average and not an individual subject response?
  2. What is an indication for a strong predictive model, thinking about SSR?
  3. Why does the standard error for a predicted individual Y value almost never approach zero?
  4. If you knew that the relationship between x and y is linear, what would be the best distribution of x for estimating the linear regression equation and for hypothesis testing?
  5. If you did not know anything about the shape of the x-y relationship and wanted to model this smooth relationship, what would be the best study design?

Section 10.6

  1. What are the central problems in percentiling variables?

Section 10.7

  1. What essentially means the same whether discussing single-variable vs. multivariable regression?
  2. Are model and hypothesis test d.f. most related to numerator d.f. or denominator d.f.?
  3. Does rejecting the global test of no association imply that all the predictors are associated with Y?
  4. What are two principal ways we test whether one or more predictors are associated with Y after adjusting for other variables?

Section 10.8

  1. A linear model with a single binary predictor is equivalent to what?
  2. Why is added more variables and doing ANCOVA better than this?

Section 10.10

  1. A researcher wants to analyze the effect of race (4 levels) and sex on Y and wants to allow for race x sex interaction. How many model d.f. are there?
  2. Why is heterogeneity of treatment effect difficult to demonstrate?

Chapter 9 Sections 9.4-9.9

Section 9.6

  1. What about our ANOVA is consistent with the color image plot for the two continuous lead exposure predictors?

Section 9.8

  1. Define the IQR effect of ld73.

Chapter 13: Analysis of Covariance in Randomized Studies

Section 13.1

  1. Why does adjustment for important predictors improve power and precision in ordinary linear regression?

Sections 13.2-13.3

  1. Why does adjustment for predictors improve power in nonlinear models such as logistic and Cox models?

Section 13.6

  1. Why is an increase in the absolute risk reduction due to a treatment for higher risk patients when compared to lower risk patients not an example of heterogeneity of treatment effect?