1. Biases
• Information bias
– Bias arising from mis-measurement or erroneous
classification of study variables
1
2. Biases
• Selection bias
– Bias arising from procedures used to select study
subjects and from factors influencing participation
40
3. 3
Biases
• Confounding bias
– Sometimes understood as being a separate
phenomenon from the broader category of “bias”,
but we are considering it a form of bias
• Disagreement about whether confounding should be called bias relates to the fact that
confounding comes about because of true causal associations in the source population
(just not the one of interest), while other kinds of bias do not reflect true causal
associations in the source population
• – Approaches to handling confounding are different from
selection and information bias and thus it’s covered in a
separate module
5. 5
Biases
• Lumpers versus splitters
– We are lumpers – in this course we present the
fewest possible categories of bias
– Others take the approach of splitting biases into
fine-grained categories