Function of a Word or Phrase Practice Test

Economic models often predict that raising the minimum wage reduces employment by increasing labor costs. The simplest competitive framework anticipates fewer low-wage jobs when wages rise. However, several recent natural experiments complicate that prediction. Studies comparing bordering counties with different wage floors found minimal job losses in some sectors, while other analyses observed small reductions concentrated in specific age groups. A subsequent meta-analysis argues that methodological choices—such as how regions are matched—drive much of the variation in results. The debate now centers less on whether effects exist and more on their size and distribution across workers and industries. Consequently, policy inferences hinge on local labor dynamics.

Debates over algorithmic fairness often assume that better data will eliminate disparities across groups. Improved sampling can indeed reduce noise, and clear labeling can limit biased proxies. However, when base rates differ between groups, no single classifier can satisfy all commonly used fairness criteria at once. This mathematical incompatibility, demonstrated in multiple proofs, means that developers must choose which notion of fairness aligns with the system's purpose—equalizing false positive rates, for instance, or calibrating predicted risks. Explicitly articulating that choice enables regulators and the public to evaluate trade-offs, turning a vague demand for fairness into a transparent policy decision. Absent such clarity, debates stall.