<p>
</p>
<p>We’re using it to make predictions on future hypotheses. For example, we can use the sciences to make predictions on the relative projections of students with low IQs and those with high IQs. We can predict that students with higher IQs will get higher SAT scores and higher educational attainments in the end. We can also use the sciences to predict what families that they are more likely than not to come from. </p>
<p>We also use the data in order to make predictions on who succeeds and who doesn’t (based on initial admissions data). For example, on whether students with higher SAT scores will really do better than those with lower SAT scores in the applicant pool, or how much SAT IIs correlate with academic performance. Is this a science? Does it utilize the scientific method? in a way, it does. There’s the hypothesis, there’s the experimentation, there’s the data collection, and then there are the results and analysis, which may force one to modify the hypothesis. Of course one should be careful not to overgeneralize the results to apply to all colleges or all students.</p>
<p>obviously, it is necessary for the college to know this data, so that it selects the students who are most likely to succeed in the college (obviously as according to the college’s standards of success).</p>
<p>Science is really just a method of forming hypotheses and collecting data to support hypotheses, from which we can build a consistent system. </p>
<p>this is just the institutional level. On the societal level, we can make hypotheses on which students are most likely to succeed in what types of environments and why this is so. </p>
<p>Is this largely independent of biology? Not necessarily. I actually suspect that brain scans will be able to find better neural correlates with success than the models we have now (of g, IQ tests, and so forth). but as humans are biological organisms that obey natural laws, we can make predictions on their behaviors.</p>
<p>Obviously, correlation does not imply causation unless all factors are ruled out (a major issue that the social sciences have to deal with, in a very complex system). Nonetheless, many social scientists resort to statistical methods that help sort out the causative factors in their correlation matrices. (and then use their hypotheses under different conditions, to see if their hypotheses continue to be consistent in other systems). Obviously the social sciences do have their flaws, but they’re far better than choosing from people at random. certainly, an institution like HYPSM selects people out of its applicant pool, which is already highly self-select from the beginning. Such institutions have normative values (such as being a “leader” in some field), both primary and secondary (secondary including a diverse student body for example), and they have to collect data (factual data) in order to best conform with their normative values. And really, most students at MIT are FAR better fits there than a randomly selected population of several thousand people in the United States. Why? Because institutions like MIT rely on data in order to predict “success”, in the way it operationalizes “success”. It’s social science.</p>