[Harvard Admit 分享帖] Flex考题难度如何

RH99UID：4089109

1. 为什么要写这个帖子    2020年7月LSAT考了177分之后，我在上一个申请季录取了Harvard Law；但是因为个人发展的原因defer到了明年入学。我觉得这个考试determines unjustifiably great a part of our application and career，所以希望可以在有限的时间里写一些东西帮到大家。另一方面，上一个cycle的Flex Inflation影响了包括我在内的很多申请者，既然现在已经release了真题我就不妨看看自己和peers当时是怎样考的。
2. 我对考试的一些感受
        Moving into the digital era a decade after other graduate admissions tests tested the water, LSAT gained much controversy as high scorers almost doubled in 2020. Sixteen months after the first remote test, the LSAC released three of latest practice tests at a time this September. Whereas those tests resulted in a de facto score inflation, they are by themselves not as easy as we might have anticipated.
        Out of curiosity, I purchased the tests and I completed those tests one year after I took my Flex test. The following is how those tests seem to me:

• Difficulty Distribution. There are more easier questions and more harder questions coming, whereas the moderate questions decreased in number.
• Hard Questions. Almost doubled, some of the hard questions are much harder than their counterparts in past tests. Whereas some pain points are recurring (including blurred author attitudes, hidden connection in comparative passages, hard-to-justify answer choices), the test makers seem to leveled up their difficulty.
• Easy Questions. The latest release fails to explain the inflation, if not adding up another layer of cloud.
  

＊        ＊        ＊

A. Reading Comprehension (Hard Qs Go Harder, Easy Qs Go Easier)
1. Blurred Author Attitudes
        In 90.3, the author of passage B does not reveal very clearly his attitude. Only after careful reading do I find passage B critical of Ricks’ suspicion of “historical approaches to ethical issues,” and to advocate a neutral stand in historical approach. This ambiguity lies primarily in the rhetorical settings of the passages, instances of which include the last sentence of the second paragraph (“But … approaches.”) and the third sentence in the last paragraph (“Yet … itself.”). They are clearly revelations of author attitudes, but they are not clear revelations thereof.
        Ambiguity in author attitude is certainly not unprecedented. Previously, we encountered in the Louise Gluck's Poetry passage (48.2) a taciturn author. In that passage, the author left nothing explicit about her attitude, but we are forced to infer her “tacit endorsement” in the Q12 of that passage. In the Calvaria Major passage (69.4), test makers teste upon the word “semblance of rigor” to mask the author attitude. In the Evolutionary Psychology passage (64.3), we are in a similar conundrum, where the author takes the stand of “maybe yes, maybe no.”
        By no means, though, do those past tests come near to the level of difficulty in 90.3. Here, the ambiguity lies in the stem of the passage, and different understanding of those key sentences render largely different understandings of the rest of the passage. Instead of testing on the author attitude only on individual questions, Q15, Q16, Q18, and Q20 all have to do with it. Instead of leaving the possibility that we can “compare” different answer choices based on the general preoccupation, answering Q18 and Q20 hinges on a precise understanding of the author attitude.
2. Hidden Connection
        The new test makers also seem to exonerate us on certain points in 90.3. In passage A, Ricks makes it clear that “plagiarism being dishonest,” whereas Ewes (author of passage B) presented this idea as “an abiding sense” in the first paragraph, only to give us her observation on this issue in the last paragraph. There, she doubted the possibility that there is a consensus.
        This connection – hidden and elusive – is fortunately not tested, but connections of this kind are not untested. In the Fiction Genre passage (85.3), passage B mentions that “[w]e are free to read any text by any reading protocol we wish.” This echoes the expression in passage A that “he or she might read any narrative as a detective story”. This is tested in Q18 of that passage. Lurking patterns!
        What is left untouched in 90.3 emerges from the water in 91.2. There, Iowa Electronic Markets appeared twice, and passage B seems to complement passage A to tell us a full story, which is not usually happening in comparative passages. But what is more elusive is the main stand of the two authors. In passage A, the author mentions that markets (1) predict and (2) take in wisdom. In passage B, the author seems to negate the first point in the first sentence of the last paragraph, though only implicitly (in the form of “if…”). And she also grants the fact that markets take in wisdom, but not omniscience. Does it mean that she does not think markets are efficient? No, she seems to describe the mechanism not differently from the author of passage A. Does it mean the opposite? No, she said there is no “omniscience” or “infallibility” in markets. I would not prefer to take any anticipation before going to the questions.
        I also find recurring patterns in this second connection. In the Conscious Deliberation passage (86.3), we are called upon to distinguish rational arguments and clinical results of the same theme of free will in legal systems. In the Judicial Candor passage (82.3), the distinction is between different kinds of cost-benefit analysis as justifications of judicial candor. But these distinctions, though I believe to be among the hardest in the 80s, are clearer than in 91.2. Only the Historical Objectivity passage (63.3) would be regarded as coming close.
        In Q11 and Q14, we are called upon to make this fine distinction and to figure out the relation between the two passages. Q14, the harder of the two, has a hardly justifiable correct answer choice – what does highly efficient prediction have to do with thermostats? Not a connection I will build in seconds. It was only after the arduous process of elimination that I made it through. And here comes another layer of difficulty in the newly released tests.

3. Marginal Answer Choices
        In 90.3, the word “current” in answer choice (E) of Q15 and “determination … traditional” in answer choice (A) of Q17 are hardly justifiable. They are only justified in a comparative sense (when compared to inferior answer choices.)
We have met them before. In the Dental Decay passage (62.3), we can hardly justify “archaeological evidence” as having to do with the primary concern in Q15. But in that Q, no other answer choices can be called as contenders. Whereas the same happens in Q15 of 90.3, (D) in Q17 poses serious challenge to the correct answer choice. More importantly, (D), with less information load, seems easier to justify than (A) on the first look.

4. Emphasis on Genuine Understanding
        In the Deep-Well Injection passage (90.1), we are to understand the spatial relationship between the layers underground. The complexity of this passage reached a climax at the third and the fourth paragraph. Whereby a process (the implications of some geological changes to the injection method) is described, we are called upon not only to identify it in the questions but to apply it. This is obvious in Q7 of this passage.
        There are, of course, recurring patterns. We are tested of our ability to understand the passage. And to understand, simply put, is not just going through the words. This is obvious in some science passages, like the Complex System passage (25.4) and the Land Restoration passage (65.4). In Q22 and Q27 of 65.4, we are also tested our ability to apply.

B. Logical Reasoning (Moderate Difficulty Overall with Some Hard Qs)
1. Identify the Conclusion and the Premise of the Argument
        90.2.12 (pygmy bear restoration) and 90.4.22 (medieval epistemology) are not MC questions, but to earn your point on any of both, the structure of the argument should be first identified. What the two questions have in common are (1) that the last sentence of the stimulus could be easily misunderstood as the Conclusion, not only because of their position but also due to their phrasing; (2) that the correct Conclusion is hidden and hard to identify; (3) that the points can be hardly earned if we miss the correct Conclusion.
        Previously, we encountered in 85.2.8 (Ebson’s campaign) a confusing stimulus and some of us may have a hard time in figuring out whether the second or the third is the Conclusion of the stimulus. But the difficulty of 90.2.12 and 90.4.22 surpassed that of 85.2.8 in that (1) the new questions are not Main Conclusion questions but Weaken and Sufficient Assumption questions each, so we may underestimate the complexity in argument structure; (2) the correct conclusion is well hidden and we are more likely to fall into the trap in the new tests.

2. Genuine Understanding
        90.2.18 is an example that understanding can be very important. Notice that such a level of understanding is not unparalleled, as least when 76.4.21 and 81.2.22 are take into consideration. I would also label 80.1.17 (deep temperate lakes) as a counterpart (or even a harder counterpart) of 90.2.19.

3. Strong Contenders
        In 90.2.17 (colonizing Mars), we find a strong contender, a sibling of which is found in other tests as correct answer choice. In 90.2.17, (B) is a strong contender. In fact, in a very similar question in 84.2.7 (geothermal power plants), (D) is the correct answer choice. However, (D) may be even harder to justify as a correct answer choice for the following reasons: (1) in the latter, depth is not explicitly connected in the stimulus with the availability of geothermal power plants; whereas (B) in the former question is clearly relevant. (2) 84.2.7 is a Sufficient Assumption question whereas 90.2.17 is a Necessary Assumption question, and an answer choice that bears little relationship with the stimulus can be harder to justify in a SA question.

C. Very Easy Qs
        First five questions of three LR sections in PT90 and PT91 are considerably easier than their counterparts in other past tests. They echo some early LSAT questions and some of the questions are so easy that they are nicely mirrored in previous tests.
        There are reasons to believe that the test makers lowered the difficulty of some questions in the RC sections:

• Structural Clarity. The structure of 90.2, 90.4, and 91.1 are very clear.
• Emphasis on the General. In both RC sections, some easy passages contain several questions that tests our grasp of the general disposition. Even the hard comparative 91.2 witness the test makers’ inclination to test on the general.
• Verbatim Inference. In some questions that dig into the details, we find the answer choice exceedingly easy to justify. Q5 and Q6 of 90.1, among other questions, are so easy that the correct answer choices almost paste the corresponding part of the passage verbatim. There are at least five question in PT91 share this characteristic.