Question

how could measurement error contribute to sampling variability, but not to bias? Describe a situation where measurement error results in bias.

Answer 1

​​​​

Measurement bias results from poorly measuring the outcome you are measuring. For example:

• The survey interviewers asking about deaths were poorly trained and included deaths which occurred before the time period of interest.
  • This would lead to an overestimate of the mortality rate because deaths which should not be included are included.

• One survey team's portable machine to measure haemoglobin malfunctioned and was not checked, as should be done every day. It measured everyone's haemoglobin as 0.3 g/L too high.
  • This would lead to an underestimate of the prevalence of anaemia because the readings would overestimate the haemoglobin for everyone measured by that team.