Why is it a good to assess inter-rater reliability when multiple observers look at behavior?

a. High inter-rater reliability indicates greater accuracy, which can aid replication
b. Researchers can check inter-rater reliability rates to make sure all observers are meeting established standards
c. If the researcher detects problems (low inter-rater reliability), he knows he may have to retrain observers regarding what to be on the lookout for
d. All of the above are good reasons to assess inter-rater reliability
e. None of the above are good reasons to assess inter-rater reliability