Selection bias
When people in one group differ from others before we observe them, so group comparisons can mislead us about discrimination or treatment effects.
Pre-market characteristics
Skills and traits people have before working (test scores, schooling, family background) that help explain later wage gaps.
AFQT
A cognitive test used as a pre-market skill measure; controlling for AFQT shrinks Black–White wage gaps in Neal & Johnson.
Neal & Johnson (1996)
After controlling for AFQT, most of the Black–White wage gap for young women and much of it for young men disappears.
Heckman (1998) ‘Detecting Discrimination’
Skill tests like AFQT are shaped by past discrimination, so shrinking gaps after skill controls does not prove discrimination is unimportant.
Roland Fryer’s work
Controlling for education and test scores makes many racial wage and employment gaps much smaller, so skill gaps are central today.
PIAAC
An adult skills survey showing that literacy, numeracy, and problem-solving strongly predict wages and employment and differ across groups.
Single-item performance measure
One overall performance rating; simple but noisy and more open to bias.
Multiple-item performance measure
Performance broken into several specific parts (quality, productivity, teamwork, etc.) and then combined into a score.
Subjective performance criteria
Performance judged by human ratings and impressions (e.g., overall rating, leadership potential).
Objective performance criteria
Performance based on numbers like sales, output, or error rates, with less room for judgment.
Interpreting subjective vs objective gaps
If racial gaps are large on subjective ratings but small on objective measures, that pattern suggests biased evaluations.
Interpreting gaps on complex tasks
If gaps are bigger on complex, objective tasks, that points more toward genuine skill differences, not just evaluation bias.
Cognitive loading
How mentally demanding a task is; higher cognitive loading tends to show larger gaps when skill differences exist.
Roth, Huffcutt & Bobko (2003)
Found modest ethnic gaps in job performance, larger on more cognitive or objective criteria.
McKay & McDaniel (2006)
Estimated about a 0.27 SD Black–White performance gap favoring whites, especially on complex tasks.
Labor market discrimination
Treating equally qualified workers differently in hiring, pay, or promotion because of race, gender, etc.
Audit study (hiring)
Field experiment using matched applicants or résumés that only differ by a race signal to test for discrimination.
Bertrand & Mullainathan (2004)
Found that white-sounding names got more callbacks than identical résumés with Black-sounding names, showing hiring discrimination.
Overseas Chinese
Ethnic Chinese in Southeast Asia who faced harsh discrimination but often dominated major parts of the economy.
Jewish economic history (global)
Repeated pattern of success, resentment, loss or expulsion, and renewed success, often in trade and finance.
Jews in the US
Arrived poor but became highly overrepresented in elite jobs and high-income despite some antisemitism.
Middleman minority
A minority group specializing in trade, small business, and lending that becomes successful but often resented.
Middleman transactor
A go-between who buys from producers and sells to others, common among middleman minorities.
