A quote from the below link..

The overwhelming empirical evidence of discriminatory results in grading suggests the existence of serious flaws in exams, particularly issue-spotting exams. In their article on grade normalization, Downs & Levit observe: "A vast amount of research in educational testing theory suggests that the preferred method of testing in law schools is one least recommended by professional educators. A single examination followed by a course grade prevents professors from giving students repeated feedback, which many theorists say is essential to deep learning. A one-shot examination highlights inaccuracies in evaluation that may result from student illness or personal troubles, or imbalances between student coverage and selective testing."

In his article '"Uncivil Procedure: Ranking Law Students Among Their Peers," Douglas Henderson claims that "[j]udged by the standards of established psychometric theory, the law school essay is neither precise nor accurate -- both of which are necessary foundations of validity." Researchers consider the examination process to be a misrepresentation of legal practice because it ignores more complex forms of thinking. Some researchers suggest learning theory rejects stringent time limits. Kissam notes that, "Multiple grade categories can be generated and explained to students most easily by establishing the final exam as a race and then observing the order in which contestants cross the finish line."

Adding to the internal flaws of examinations, discrepancies in grading are ubiquitous. As Henderson remarks, "The standards in grading law school essay exams vary between professors and between exams graded by a single professor. Little direct evidence is available to show how law professors evaluate examination answers." Henderson elaborates by suggesting that "Law school policy which permits the standards to vary from teacher to teacher causes its evaluation process to be grossly misleading to the public and arbitrarily discriminatory to its students."