The Importance of Having Clear Questions
I’ve been struck recently by the importance of having clear questions or theories or hypotheses that one wishes to explore.
Otherwise simply fishing around in data we find statistically significant findings but we have no idea what they mean or how to interpret them (a bit like those correlations people use to demonstrate correlation does not equal causation).
This seems an ever bigger problem as more data is captured and computational power enables more ‘analysis’.
In fact today I was listening to a #BMJ podcast (https://soundcloud.com/bmjpodcasts/big-metadata ) which highlighted a hidden issue when clinical data is used for research – the context of the clinical data is typically absent. For example, the life expectancy of a middle-aged man with low white blood cell count at 3am is far worse than one taken at 3pm; and this is explained by the fact the patient and clinician are sufficiently concerned about the health of the man to order tests in the middle of the night, rather than a test taken in outpatients or working hours.
Reading this blog (https://www.health.org.uk/blog/measuring-humanity-marginalised-communities) made me think once more of that and the final message about rethinking our relationship with data and measurement to one that is genuinely focused on curiosity and learning seems really important.
QIClearn are pleased to announce a new Masterclass in Measurement for Improvement.