Replication and Evidence Factors in Observational Studies

Observational studies can show that smoking causes lung cancer, even without randomized experiments. This is possible due to two or more associations being susceptible to unmeasured biases, but not to the same biases. Replication and evidence factors in observational studies can help increase insensitivity to unmeasured biases.

Format: Hardback
Length: 276 pages
Publication date: 31 March 2021
Publisher: Taylor & Francis Ltd

Observational studies cannot establish causation without randomized experiments, but our understanding of smoking's link to lung cancer is not flawed. How is this possible? If observed associations do not indicate causal effects, how can a sequence of such associations be decisive?

Two or more associations may each be vulnerable to unmeasured biases, but not to the same biases. An observational study has two evidence factors if it provides two comparisons susceptible to different biases that can be combined as if from independent studies of different data by different investigators, despite using the same data twice. If the two factors concur, they may exhibit greater insensitivity to unmeasured biases than either factor exhibits on its own.

Replication and Evidence Factors in Observational Studies includes four parts:

A concise introduction to causal inference, making the book self-contained.

Practical examples of evidence factors from the health and social sciences with analyses in R.

The theory of evidence factors.

Study design with evidence factors.

A companion R package, evident, is available from CRAN.

Weight: 671g
Dimension: 234 x 156 (mm)
ISBN-13: 9780367483883