<p>annasdad;</p>
<p>You obviously never read the study. the D&K study is probably the most misquoted study of all times. It is also probably one of the most poorly designed studies. </p>
<p>There is simply no college by college analysis, just crude groupings. In their original study, D&K actually admitted that if you used school selectivity as opposed to SAT scores there actually was substantial effect of attending a more selective school everything else being equal. They even found that tuition correlated with income. The more you pay the more you earn. They had a tough time explaining that one. </p>
<p>For a more detailed critique of the methodology, this much better designed study shoots a hole through the D&K study. It demonstrated that taking merit money from a less selective college did not compensate for the loss of income over the long run from attending a more selective college. (See note 27)</p>
<p><a href=“Page not found | Harvard Kennedy School”>Page not found | Harvard Kennedy School;
<p>
</p>
<p>In summary, most respondents listed their own college (where they enrolled) and one less selective college which they did not attend.</p>
<p>There were very few respondents that did the reverse, i.e. list their own college and one more selective college which they did not attend. Among the most selective colleges the numbers were virtually zero. </p>
<p>With such a strong selection bias, there is simply no way to draw conclusions on students that aren’t even part of the study.</p>