How 'bout them tastes?

[aquilinum.livejournal.com]

Sites like www.babynames.com and similar have "rate this name" features, and you can see the average star rating, accumulated over thousands of clicks.

I've gotta say, some of the decisions ASTOUND me, especially given the fact that these aren't isolated opinions; these are average opinions!


Pahana (a unisex name, or so we are told — despite meaning "little white brother") is ranked higher than Rosamund.

Raghnall beats out Felicity.

Quasar tops Valerie.


Is it just me, or are there any rating trends you just can't understand?

Comments

[alexandria-skye.livejournal.com]

i'm going to go all statistics student on you.
because the "rate this name" feature is strictly voluntary, the results of the survey are going to be decided by only users of babynames.com (or the single particular website you're on) that feel like voting. usually this means that you'll only get results from people with strong feelings about these names. and there's nothing that says that a user can't go and vote multiple times. keep in mind that sometimes, averages aren't the best representation of a center of a data set.
i know that there's some loopholes in this explaination, like clt and such, but that's probably why the results are weird.

ps-i like quasar better than valerie :)

[bornto-fly.livejournal.com]

The names aren't going to be voted on by the same number of people. Names like Pahana and Quasar and Raghnall are going to be viewed by fewer people than names like Madison or Aiden or Valerie. So while Madison and Aiden and Valerie are going to have, let's say, 100 people voting on them, 25 of whom love the name, 25 of whom hate it, and 50 of whom are totally indifferent, Quasar might only have 5 votes, and they might all be people who love it.

It's not a scientifically accurate polling method (sorry, I'm a political science major, so I'm a geek about these things), so it really means nothing. You'd have to see how many people actually voted on the name, and the actual randomness of the sample. For instance (using only America as an example...) names like Rose might be common in every ethnicity, where as names that are Arabic or Native American are more likely to be used ONLY by people in those cultures. So names like Rose are going to be viewed by a more diverse sample than, let's say, Pahana (which is Native American).