John Christie
2007-May-15 17:36 UTC
[R] efficient way of aggregating two variables at once
Hi, I'm working out d' and therefore I'd like to aggregate two variables at once (hits and false alarms). In the raw data frame is subject, stimulus_presence, and response information. I aggregate this into a table now currently by separating out the target_presence=true data and working out a hit rate and then workout out the false alarm rate from the target_presence=false data. Then I cbind the "x" column from the false alarm aggregate to the hit rate data. Then I can also append a dprime column if I wish. I am just inquiring as to whether there is some facility I am missing for aggregating the variables simultaneously or just generating the d' in one step.