@rjon27 Ping made an error in his model here. It doesn't harm the answer choices, because his model is entailed by the correct model, but it is weaker. Most modifies the subject, large nurseries, and so the text is making a claim about a specific most subset of large nurseries. In that subset, most sell primarily to commercial growers and guarantee their plants. These are the same nurseries, not some separate > 50% subsets.

@Bgsolo "but" operates exactly the same as "and" in logic. The only difference is rhetoric, because it implies a contrast between the two proposition. So the difference is stop thinking about a difference between "and" and "but" in logic! If you see "but" think "that means AND". "Or", on the other hand, is a very different logical operator. "Or" between two propositions (A or B) usually means either A is true and B is false, A is false and B is true, or A and B are both true. This is the inclusive or. You do have to pay attention to the grammar, though, because there are some specific places where "or" means ONLY ONE of A and B can be true, but the passage will make that clear if you pay attention to it. There is a final rare case where "or" in natural language actually means "and", which is when it's part of a comparative claim. If you say A is better than B or C, it actually means two distinct claims: A is better than B and A is better than C.

I think this is an interesting passage because it's one of the few that doesn't contain an argument. It's a statement of a phenomenon and then a statement of a causal explanation of the phenomenon (in the conditional chain), but the conclusion is not explicitly stated.

@JoeSolana4 yes. that's a feature of the MSS (most strongly supported) type of question. I think if it was a MBT (must be true), E would fail.

Ping's analysis of answer choice A makes a common scoping error. He see's an argument about elections and thinks 'presidential elections'. The domain of the argument is all elections, precisely including those without debates, and likely many where debates might occur but be sparsely attended and not well covered. Across that domain, it is certainly not self-evident that skill in something that doesn't even occur (or occurs in front of only a handful of people) would make one more likely to win an election. It's correct that it's not the point of disagreement, of course, as neither passage discusses the likelihood of winning. They just discuss the utility of the debate, if it happens, for the people that watch it. This is not a nitpicking comment. It's here to remind people to watch out for this common error. Don't assume an argument about a given topic (here, political elections) is scoped to the most salient example of that topic in your memory (US presidential elections).

@HenryLehmann close. The point is that you do not know the distribution or characteristics of C, so it is POSSIBLE that A -> /C. The scoop analogy acts as if any >50% subset of B is a random sample of >50% of B. That is the ecological fallacy. It need not be a random sample. It COULD be a random sample. It could require A. It could require A be absent. We do not know.

@HenryLehmann my issue is with the use of the scoop analogy, as it suggests that "most" implies a random sample of a specific size that may or may not pick up an element from a very small subset. Most need not be a random sample. It can be a very specific subset that happens to be larger than 50%. A -> B -m-> C is invalid not because A might be tiny, but rather because A might be 49.999999% of B, and the C that do not contain any A are defined by the absence of A. Any particular subset containing most of the elements of a superset is not required to have the same general characteristics as that superset.

Kevin's explanation is good, except the opening sentence cannot be context, as context should be neutral. The opening sentence is a premise that "some" disagree with in the second sentence. The argument isn't strong, but P1 is that the supply chain will break unless we act NOW (to change the contracts). It supports the main conclusion, which is a normative judgement that a delay is irresponsible. The analogy is P2, and it also supports that normative judgement. C looks like a common wrong answer choice, but it is wrong for two reasons. (1) it incorrectly paraphrases the first sentence, as it misses the urgency, and (2) the thing it is incorrectly paraphrasing is a premise.

Choice B is interesting, because it is a generalized form of the contrapositive of the main conclusion. Obviously it's not the main conclusion, but I wonder if it would be considered weakly implied or merely suggested on the spectrum for a 'most strongly supported'. The specific example of 'politicians' who criticize opponents for their 'muddled manner' as being insincere provides some (weak) support for the general case that critics in general who criticize the manner (any manner) of presentation are insincere.

"This, however, is unlikely" is the main conclusion. "This" refers to Apatosaurus being able to gallop. Of the answer choices, obviously C is the best choice, but I wonder if one were to have as two answer choices (A) Some paleontologists are wrong and (B) It is unlikely that Apatosaurus was able to gallop, whether (B) or (A) would be correct. I would say (B), but i can see an argument that the structure of the conclusion is rejection of another claim, not affirmation of a counterclaim.

Different technical domains use this term differently, but in causal inference, spurious correlations are understood as structural relationships in the data and are distinct from chance (where random error happens to produce an apparent pattern). The correlation between bed sheet deaths and cheese consumption is a great example of a classic and common source of spurious correlations, time series data, where the confounding arises from shared dependence on calendar time. The confounding is mediated by latent (and often unmeasured) variables, but it has precisely the structure of confounding, not chance. One could hypothesize a number of latent variables. Economic conditions and age structure of the population are two possibilities, but whatever they are, they are indexed to time, and occur upstream of both variables of interest. Depending on your lens, you could invoke collider bias in the search for relationships that would sound ridiculous to the casual observer and make for good fodder on a website devoted to them.

This is generally good (specifically, the description of confounding is accurate in a way that overview courses often aren't), but misses a few fine points. (1) correlation in the data that is due to chance is not, in fact, a correlation relationship between the phenomena in the real world. So "Chance" is not a valid alternative explanation for a correlation. It's an alternative explanation for an apparent (illusory) correlation. (2) For spurious correlations (real correlations that are not due to a causal relationship), there is much more to it than reverse causality and confounding. Confounding is often thought of as a catch-all, but is a very specific type of spurious correlation. There are others. A crucial one is collider bias, where conditioning on a common downstream effect produces a spurious correlation.

I'm struggling to follow this portion of the curriculum because of how far it strays from a rigorous discussion of causal inference. Generally, in my prior learning, prediction is distinct from causal inference (because prediction doesn't care whether you have identified a true cause or some other relationship, just whether it can predict the result), but I think when Ping says hypotheses make predictions, he means "hypotheses are testable". That's sensible, and foundational to an inductive approach (vs. the deductive approach in the formal logic section). It is strange for it to appear 10 lessons deep in this section, though.

There is a grammatical error in the argument here. Produce is a mass noun. It needs a singular verb. This should read: "Most produce from California is exported to Brazil."

This explanation uses a flawed argument itself, the ecological fallacy.

I'm not a fan of treating sentences that are clearly about inductive logic as if they are conditionals. Question 1 summarizes data. It does not attempt to make a universal claim. Translating it to "Hunting Permited -> Deer Population has not increased" is an error, in my view. I would expect it's part of a larger passage that uses that data point to make an argument about causality, which then could be analyzed on its strength.

ugh. i hate that we're using a slash for negation instead of a turnstile, tilde, or even exclamation point. I'm going to confuse it for a slash between two premises.

This skill builder has an error, and repeats it for two examples: for questions 4 and 5. In both cases, there is a structure: subject + predicate verb + predicate object + CONJUNCTION + SECOND PREDICATE. In the video, this structure is recognized as a possible, but technically wrong way of understanding the passage. It isn't simply possible. It is the only correct analysis of the passage, for both questions 4 and 5, and failing to recognize the second predicate structure results in failure to understand the passage. For question 4: the first predicate is "finds that it is unable to lend its printed books". "that it is unable to lend its printed books" is the object. "and" is the conjunction, which is followed by a second predicate for the same subject "displays them only when requested for an exhibition." This second predicate can't be part of the object, because it does not make sense to say that "a library find that so displays them only when requested for an exhibition". "displays them" can't operate as an object for "finds". Similarly, for question 5, there are two predicates. The first is "realizes that it can't continue to bake its traditional bread" and the second is "switches to a recipe that uses cornmeal." "switches" can't be an object for "realizes". They bakery didn't realize that they switched. They realized that they can't continue with wheat flour, and they switched to a cornmeal recipe. Considering this all to be the object of 'realized' misses the meaning that the switch was a result of the realization. This is a basic scoping error, and, honestly, it is disappointing to see it not once but twice, and shortly after a discussion of the concept that a subject can have two predicates. I realize Ping is saying he's not rigorously teaching grammar, but parsing the difference between an object and a second predicate is fairly important to understanding a passage.

The first sentence is not context. It's another premise. P1: the purpose of the posters is to increase motivation to work productively. P2: Workers are already motiviated. C: Therefore the posters are not achieving their intended purpose.

This is a sloppy taxonomy if the purpose is a functional classification of grammar. Earlier, Ping distinguishes between the restrictive relative pronoun "that", calling it a modifier, as it "cuts down into the subset", and the demonstrative sentential referential "that". That distinction is useful and follows the function of those separate uses of "that". But here, he seems to think the noun modifying "which" of "Azedcorp, which currently owns a majority stake" is performing a pointing function instead of a modifying function. Certainly, both the restrictive clause "that" and nonrestrictive clause "which" do reference something, but they do it to MODIFY (in this case) the subject. If he wants to identify the functional difference for "that", he should also identify the functional difference for "which". When "which" functions as a strict pointing referential and does not modify the referent, THEN it should be categorized as a referential in this taxonomy. When it DOES modify the referent, it should be categorized as a modifier.

I see a fatal error in the way this discussion is presented. Ping is making assumptions about assumptions, in particular, a common one, where certain outcomes are assumed to be equiprobable. We cannot say, for example, that the assumption that a cat has the strength to knock over a trash can is just as likely to be true as false. It is just something we have no information about. Uncertainty doesn't make it a coin flip. This is the principle of indifference. It works for coins, but not for cats. I agree that the Trash can argument is weaker, but it's not because the assumptions are just as likely to be true as false; it's because it makes an absolute claim from circumstantial premises.