<p>DwightEisenhower is convinced:
</p>
<p>Random probably isnt a helpful term here, since it has multiple meanings (ranging from formal definitions in statistics to slangy ones in everyday speech). I think what were debating here is the predictability of a dependent variable (admit vs. deny) based on independent variables (such as grades and test scores). </p>
<p>We can probably agree that knowing someones grades and test scores increases our ability to predict we know that someone with Bs and a 2000 SAT probably wont make it, where a valedictorian with a perfect SAT has a much better shot. If we add in even more variables, the quality of our predictions will go up. Its likely that our predictions will be particularly good with many of the people at the low end of the pool for long-shot applicants with weak stats, we can predict that theyll all be rejected, and well be right well over 95% of the time which is pretty good in predicting any human decision process.</p>
<p>Where it will get much harder is with the substantial number of more-than-qualified applicants, most of whom will be rejected because Harvard doesnt have room for them all. I think the debate is how much those decisions are predictable based on applicant characteristics (assuming we knew both the relevant characteristics and how Harvard values them), and how much other exogenous factors creep in and out of the process. </p>
<p>Admissions decisions arise from a social psychological decision process social in the sense that multiple people interact with one another in making the decisions, and psychological in the sense that human emotion and cognition are involved. Experience and research tell us that such processes are subject to a variety of external, serendipitous factors, many of which the decision-makers dont realize. </p>
<p>One of many examples appeared recently in the New York Times in the form of decision fatigue. (<a href=“http://www.nytimes.com/2011/08/21/magazine/do-you-suffer-from-decision-fatigue.html?pagewanted=all[/url]”>http://www.nytimes.com/2011/08/21/magazine/do-you-suffer-from-decision-fatigue.html?pagewanted=all</a>). Psychologists studied the parole decisions of three-judge panels at a prison. The judges were professionals who took their work seriously. But the best predictor of parole decisions was not characteristics of the candidates, but the time of day the case was reviewed. If you came up early in the day when the panel was fresh, you had an excellent chance of parole. Late in the day, you were out of luck. [Decision fatigue] routinely warps the judgment of everyone, executive and nonexecutive, rich and poor in fact, it can take a special toll on the poor. Yet few people are even aware of it, and researchers are only beginning to understand why it happens and how to counteract it. </p>
<p>So, its likely that your chances of admission are affected by how late in the day your case comes up, and how tired the admissions officers are when they get to you. And thats only one of many examples of factors beyond the individual case that can influence the decisions. (If I wanted to make a long post even longer, I could go into detail on many other equally significant factors that make decision-making less rational than we like to think it is.)</p>
<p>If youre accepted, its nice to think the process was systematic and you absolutely deserved it. Thats how my Harvard son feels, and Im happy for him. But there are a lot more applicants who get rejected at H and the other elite schools, and many of them go into a funk because theyre convinced that theres something wrong with them, and there must be something they did wrong. I think its important for them to understand that they may have done everything right, but theres significant serendipity in the process.</p>