Part III
Method
11.1 What is to be tested
11.2 Five stages to pass
11.3 What to check at each stage
11.4 Reconciliation versus verification
11.5 Where each stage is handled in this text
11.6 Priorities for the survey
11.7 Variables to record
11.8 What must be pre-registered
12 Selection Bias: Decomposing the Distortions That Enter Observation
12.1 The problem
12.2 Conditioning on survival: a circularity in the framework
12.3 Selection into disclosure: correlated with the axis of interest
12.4 Age: a problem of domain, not of bias
12.5 Age, period and cohort are not identified
12.6 Retrospective narrative
12.7 Mistaking the unit: what a row denotes versus what its values denote
12.8 Selection of variables: the gaps are not random
12.9 The observation frame in practice
12.10 How to proceed
13 A Taxonomy of Verification Obstacles: Six States
13.1 Why a taxonomy is needed
13.2 Six categories
13.3 The taxonomy also serves as a test before starting