Look at the number of respondents.
Knowing that you seem to reflexively disagree with just about every thing I way, here is an explanation from another site. I am also providing the url for your edification.
There is a nice table at the site:
http://resolutions.co.nz/sample_sizes.htm.
I wasn't reflexively disagreeing with you. I don't understand how MoE is calculated, and assumed I was missing something less obvious than what I missed.
Thanks for the link. I've bookmarked it! 
There's a simple way of calculating it: 0.98/square root(n). (n is the sample size.)
