<p>In conducting a survey, the wording of the question matters. It matters a lot.
The following are not necessarily trivial variations of the same question:</p>
<ol>
<li>What are the 5 best colleges or universities in [country X] for undergraduate students to study IR?</li>
<li>What are the 5 best colleges or universities in [country X] for undergraduate students?</li>
<li>What are the 5 best colleges or universities in [country X] for undergraduate students to prepare for careers in IR?</li>
<li>What 5 colleges or universities in [country X] have the best undergraduate IR [programs/departments]?</li>
</ol>
<p>Number 1 apparently was the question asked in the survey. In my judgment, #1 is more similar to #2 than to #4. Therefore, a reasonable answer might include Harvard, Yale, Princeton and Stanford (the USNWR top 4). </p>
<p>So what’s #5? The USNWR #5 is MIT. Although MIT has some superb liberal arts departments, many respondents are likely to skip over it as a “technical school” (whether it offers good IR courses or not). Once they do that, some of them are likely to pick the first school they think of that they associate with “IR”. So, how about Georgetown?
Or, you work down the USNWR list until you come to the next highest-ranking school you associate with IR (or government … or something). Columbia? Chicago? Or you go for geographic diversity and throw in the highest-ranking southern school (Duke!) Or, toss in one or two public universities. (Berkeley! No, Michigan!) Or you ask, what are the top USNWR-ranked LACs? ( Williams! Swarthmore!)</p>
<p>I think the resulting list strikes many people as odd, because the question used to generate it was not well focused (if the intent was to distinguish excellent schools for the study of this field, not just replicate the USNWR top-N with some minor jiggerings). A small change to the wording of the question can elicit very different responses, even if the participants are informed and cooperative. </p>
<p>So, what question do we really want to ask?</p>