Estimation of time needed for TI-AB screening #1470
Unanswered
Hanedinges
asked this question in
Q&A
Replies: 1 comment 1 reply
-
You could check out the 4-step heuristic that Rens and I came up with,. Or the poster we made with it: |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
-
Dear all,
I wonder: how can we estimate the time (left) needed for ti-ab screening?
My team has proposed as a stopping rule: 100 irrelevant articles and estimated it would take 20 hours to reach to that point. How do I know how realistic this is?
And what to think of this: in a recent publication in which Rayyan and ASReview are being compared, it is adviced not to rely on one criterion but use 'combinations to improve performance E.g. screen min 50% and stop after 50 consecutive irrelevants'
50%!!! Hmm, I thought, let's estimate how much time it would take to screen another 22% (we've screened almost 3%) See table below:
<style> </style>According to this estimation it would take another 73 hours to screen another 22% and we do not know if the other stopping rule will be applicable by then (how can I formulate this correctly??).
That is a huge difference with 20 hours.
I realise in fact I have two questions: how can we estimate the time needed in case we use a simple single stopping rule of 100 irrelevant articles?
And what to think of the two criterion advice and the estimation I made based on a scenario in which we use two?
Hopefully there's a rule of thumb? I would be grateful if someone could help me get a clearer picture.
Cheers! Simone
Beta Was this translation helpful? Give feedback.
All reactions