<br>
<blockquote> <p>OK, you've got 132 comments for USC, so that's definitely a large enough number to give you a good statistical sampling. With that number you are going to get a fairly high degree of accuracy on the survey data.<<</p> </blockquote>
<br>
<p>Nope. That could be counted on to give an accurate picture of the school ONLY if the 132 kids were carefully selected to be representative of the school population as a whole and they were then asked to write a review. The simple fact that they were motivated enough to write in on their own already introduces a selection bias as noted above. As a general rule, the only way that self-selected samples can be accurate is by accident.</p>