CC Image Source
Cause is a construct.
Can’t manipulate variables like ADHD
random error :-//
correlation occurs – so
measure coefficient from the data.
People fill out what is in their head,
not on the page
…data is unreliable
Scaling messes with distribution where
Self report inventory is completely useless –
perfectionism instrument? Effectively useless
Check internal & external
Who am I going to convince?
There’s no plausible alternative…
averages out feedback,
Throw it out.
(impossible to eliminate culture)
People want to predict and control
at the end of the day.
Confirm self reports:
Interpretive study would give better