Agreement on the Facts, Disagreement on the Reading

Every pair of texts you're asked to compare has two layers stacked on top of each other. The data layer is whatever was actually measured, counted, or observed: a date, a score, a depth. The interpretation layer is the claim an author builds on top of that data: what caused it, who did it, what it means. Two authors can walk away from the identical data layer and build completely different interpretation layers on it. That's not a contradiction. That's just two people reading the same number and asking different questions about it.

To separate the layers, read each text and ask two questions separately: what number or observation does this text report, and what claim does the author attach to it. Line the two texts up side by side on each question. If the numbers match but the claims don't, you're looking at an interpretation split, and the right answer will describe them as agreeing on the finding but dividing over its significance, cause, or implication. If the numbers themselves don't match, something more basic is going on.

That something more basic is a dispute about method. If one author questions how the data was produced, not what it means, the two texts no longer share a data layer at all. They hold two different measurements, and the question is testing whether you notice that the ground has shifted. This matters for the text-structure and word-in-context questions nearby too: the same habit of separating what's reported from what's claimed about it is what lets you track an author's stance across a passage, not just across a pair.