The word “stereotype” began life in the late 18th century as French printer’s jargon: stéréotype, from Greek stereos (“solid”) plus type. It named a solid metal plate cast from a mold of movable type so the same page could be printed over and over without resetting every letter. By the mid-1800s the term had stretched to any image or phrase repeated without change. In 1922 Walter Lippmann borrowed it for the “pictures in our heads” about social groups, and the modern meaning was born.
Fast-forward a century and the academic orthodoxy insists those pictures are almost always toxic, inaccurate, and powered by pure bias. Rutgers psychologist Lee Jussim and a small army of inconvenient data keep refusing to play along. Stereotype accuracy is one of the largest and most replicable effects in social psychology. Correlations routinely sit around 0.5; only about 5 percent of findings in the entire field clear that bar. Gender and age stereotypes perform especially well. National-character stereotypes largely miss. Political stereotypes track reality directionally, though people reliably inflate how extreme the other side actually is.
This is not fringe speculation. Multiple labs, objective criteria, and repeated replications deliver the same pattern. Yet the preferred sermon remains that noticing group averages is itself a moral failing. Reality keeps scoring higher than the sermon. Accuracy does not license treating every individual as a walking average, nor does it bless crude insults. It does mean the reflexive claim that stereotypes are “inaccurate and driven by bias!!!!” is itself the oversimplified, poorly calibrated belief. People notice regularities. Sometimes the regularities are real. Pretending otherwise does not erase the distributions; it only makes the conversation dumber and the field less honest.
