<p>In the 2009 graduating class, 1.530 million students took the SAT, while a roughly equivalent number, 1.480 million, took the SAT. That’s about 3.3% more taking the SAT—an inconsequential difference. Here’s how they scored on the respective tests, using the official SAT-ACT Concordance Table as the basis for comparison:</p>
<p>SAT CR+M: # scoring (pct rank) // ACT composite: # scoring (pct rank)</p>
<p>SAT 1600: 1,192 (99+ %) // ACT 36: 638 (99+ %)
SAT 1540-1590: 5,969 (99+ %) // ACT 35: 3,450 (99+ %)
SAT 1490-1530: 11,611 (99%) // ACT 34: 7,533 (99+ %)
SAT 1440-1480: 20,238 (97-98%) // ACT 33: 12,082 (99%)
SAT 1400-1430: 28,139 (96-97%) // ACT 32: 17,558 (98%)
SAT 1360-1390: 37,733 (94-95%) // ACT 31: 23,981 (97%)
SAT 1330-1350: 29,556 (92-93%) // ACT 30: 31,457 (96%)
SAT 1290-1320: 48,279 (89-91%) // ACT 29: 37,830 (93%)
SAT 1250-1280: 58,853 (85-88%) // ACT 28: 47,716 (91%)</p>
<p>Thus at the very top of the scale, there are nearly twice as many students scoring a perfect 1600 on SAT CR+M as there are scoring a perfect 36 on the ACT. At almost every high-scoring level in the official SAT-ACT Concordance, there are significantly more students earning the SAT score than the corresponding ACT score. </p>
<p>Now in part this just reflects a flaw in the official Concordance, which privileges SAT scores at the top end by corresponding them with ACT scores that are, statistically speaking, more difficult to achieve. Going by percentile ranks, an ACT composite of 34 (99+ %) should correspond to an SAT CR+M of 1540-1590 (99+ %), or perhaps something like 1540-1570 (also 99+ %) —not 1490-1530 (99%), as the current Concordance has it. Similarly, an ACT composite of 33 (99%) should correspond to SAT CR+M of 1480-1520 (99%); an ACT composite of 32 (98%) should correspond to SAT CR+M of 1450-1470 (98%); and so on. Making those adjustments would equalize the numbers of top scorers in each test with the equivalent scorers in the other test. But it would also have the effect of indicating that an ACT score anywhere in the 28-34 range should count as the equivalent of a higher SAT score than is indicated in the current Concordance.</p>
<p>Bottom line: there’s just no basis for concluding that most people score better on the ACT than on the SAT (cf. post #14).</p>