[Excerpt] The first major prediction of the theory is that an increase in the extrinsic rewards for learning will cause student effort and achievement to increase. The primary extrinsic reward for achievement in high school is a higher probability of completing college. Thus the extrinsic rewards for learning in high school depend on the size of the payoff to college and on how contingent college admissions decisions are on achievement in high school. Time series data suggests that changes in college selectivity and payoff may have contributed to the ups and downs of student achievement during the postwar period. The college payoff and test scores rose during the 1950s and peaked during the 1960s. Both series then declined during the 1970s, bottomed out around 1980, and then rose during the 1980s. The doubling of the payoff to college during the 1980s appears to have contributed to more homework being assigned and done and a big increase in the proportion of students taking rigorous mathematics and science courses. Analysis of cross section data also supports a causal link. High school students living in communities where the payoff to college is large tend to take a heavier academic course load and are more likely to go to college.