Then we can try to merge the data sets. The result will be a mess, in this example because we are matching two data sets by nonsense columns, like time_left while time_left could be totaly different number for each data set. That means the data will be larger because there be more unique combinations of matching column values.
> data =merge(reviews, solutions, all=TRUE)>head(data) id start stop time_left solution_id reviewer_id accept problem_id subject_id answer11130409511913040951692343NANANA15629 B211304095698130409575817543271NANA<NA>32130409511913040951832329NANANA26925 C421304095188130409520623064221NANA<NA>53130409512713040951462366NANANA3422 C631304095276130409532021925281NANA<NA>