The author seemed to work this way: Ask about 200 people for their choices for 10 best Canadian songs, then tally the results and publish. I would have preferred a second step: send the initial tallies back to the same 200 people and them score each song on a 1 to 10 scale. Summarize those results and publish.
The problem with a smallish sample size like this is that if 10% are fans of, let's say, Teenage Head and even if 90% of other correspondents think Teenage Head were a talentless bunch of schmucks, Teenage Head still gets 2 songs in the top 100. On the other hand, an act like Sarah McLachlan probably had about a dozen songs that split the vote, and perhaps 90% of the council would expect her to be on the list, but not enough panelists focused on any one song, and she ended up with nothing. |
'The' Bands 
Finish the Lyric 
Figure Out the Disney Lyrics II
