That's not how controlling for a variable works. You don't "rule out people" - the question you need to answer is whether a college education makes people more

That's not how controlling for a variable works. You don't "rule out people" - the question you need to answer is whether a college education makes people more successful than would be predicted by IQ alone, and if so, by how much? Is your contention also that SAT/ACT are vanity measures? The correlation between IQ and ACT is very strong, or in other words they are both strongly g-loaded and are good measures for general intelligence. http://iapsych.com/iqmr/koening2008.pdf http://iapsych.com/iqmr/koening2008.pdf We also have ample evidence that general intelligence correlates with economic success: http://sci-hub.tw/10.1016/j.intell.2006.09.004 http://sci-hub.tw/10.1016/j.intell.2006.09.004 Vanity measure, my foot. Given that institutions of higher education aggressively select for high-scoring SAT/ACT individuals, and thus high IQ individuals, the burden of proof is on them to show that a degree results in significant socioeconomic success that cannot be explained by the preexisting psychometric attributes of the college bound population.