Research Day 4 (06/07/21)
first day of reliability coding
Behavioral coding projects need to make sure that people are coding reliably. In order to check this 10% of the data is coded and compared by 2 people to see how similar their scores are to see if there needs to be any changes or clarifications on the coding scheme.
This morning I had a meeting to discuss my first comparison with a phd researcher, and I'm happy to say our scores were good for a first attempt at 0.63 and 0.86 - anything above 0.7 is good with the best being closest to 1. After going back over trials to improve them and doing some more trials - i got this up to 0.64 and 0.97!
So altogether today I did 2 more new trials and re did 2 trials after I received feedback from them.
I'm also receiving a laptop today which will suit the software better so hopefully with that and some more practise I will get faster at this new coding!