Passage 1 (excerpt from an essay on remote work).
Defenders of remote work often argue that it makes workers more productive. The data, on average, supports them. But to stop there is to misread the argument. The serious case against remote work was never that individual workers produce less. It was that organizations as a whole — the mentorship of juniors, the slow accumulation of shared judgment, the unforced encounters that produce ideas no one was tasked with producing — slowly degrade in ways that aggregate productivity numbers do not capture. The aggregate is doing the work of obscuring a question we have not yet answered: what happens to an institution when the next generation has never been in the room?
Passage 2 (excerpt from an essay on standardized testing).
I have been a critic of standardized testing for two decades. So it surprises my friends when I argue against the recent push to eliminate the SAT outright. The case against the test has always been about its misuses: as a proxy for parental income, as a sorting device that masks itself as a measure of merit, as a cudgel against under-resourced schools. None of that is wrong. But the test does one thing that the alternatives proposed to replace it do not: it produces a measurement that is the same for a student in Mississippi and a student in Manhattan. Imperfect measurement is a stubborn good. The proposal to replace the test with admissions essays, letters of recommendation, and 'holistic review' substitutes a measurement we know is biased for a series of evaluations we know are MORE biased — and which can never be audited. This is not a defense of the SAT. It is an argument that the alternative is worse.
Passage 3 (excerpt from an essay on social media).
The argument that social media platforms are 'just neutral conduits' has a serious version and a foolish one. The foolish version says that platforms are like the phone company: they pass traffic, they don't shape it. This is obviously wrong — what shows up in your feed is a sorted, ranked, filtered selection determined by code the platform owns. The serious version concedes the sorting and says that platforms should be neutral about the CONTENT of what they sort, even if they cannot be neutral about the FACT of sorting. I do not agree with the serious version either, but I want to be honest about the trade-off. The price of demanding platforms intervene in content is that we will be governed, in our daily reading, by the values of whoever owns the platform. That price is real, and dismissing it is the cheap move. The expensive move is to argue that we should accept the cost anyway. That argument is what this essay tries to make.
Across Passages 1, 2, and 3, which technique BEST describes a common feature of all three authors' arguments?