The speaker proposes a method in combating the recently dramatic rise in cheating that honor codes should be adopted, yet the evidence offered remains unconvincing to support the conclusion in different respects.
First of all, the speaker asserts that honor codes should be adopted on an assumption that GC holds or once held a similar situation with those universities, if not, the means preventing cheating might be not effective due to different backgrounds, for instance, the difficulty of the tests or the morality level of the students attending tests varied from school to school. Even if the evidence supports the honor codes' success in GC in a fully justifiable manner, we cannot deduce that similar means would be effective on such different backgrounds without filtering the possible interference. The speaker ignores the essential points in building his conclusion on uncertain assumptions.
Secondly, the evidence supporting honor codes’ success in GC is based on another assumption that the cheating rate is fully illustrated through the cases reported. Yet the cases reported from students might be fewer than those from teachers whereas the situation remains as before. Considering the relationship between students, the rule in achieving a most benefits makes it possible that some of them agreed on a certain deal that they cease reporting cases and all involve in the process of cheating as a team, the actual cheating rate would thus be even higher with fewer reports to the faculty members, and which, might in turn worsen the problem due to less attention from above. Besides, without actual data of the cases in the last 5 years we can't distinguish weather the change in the number of the reports remains apart of fluctuate or reveals a certain tendency. Hence, without enough notification of the relationship between cases reported and actual cheating rate, we can't draw a conclusion from such evidence that the means in prohibiting students from cheating is a triumph.
In addition, the speaker supports his estimation in the success of honor code by giving a survey. Yet the survey lacks the basic information such as sample poor and control, thus the actual will in conducting cheating is not justifiably revealed. Even if the survey is conducted in a fully convincing manner, we can't guarantee that the students from other universities act in a similar manner. The survey is hardly able to lend any rational evidence in the possibility of success in those universities with different backgrounds.
To sum up, the author supplied a seemingly beneficent suggestion to prohibit students from cheating, while his deduction stands on weak assumptions and the evidence remains little rational.