In search of a word...

May 29, 2008 15:02

I need a word, which I doubt has ever been coined ( Read more... )

Leave a comment

Comments 10

whswhs May 29 2008, 05:36:25 UTC
Forming any latinate compound from "sample" doesn't actually work; it's a French root, not a Latin one. The Latin root from which it derives is exemplum. I don't think a Latinate form derived from that would be understood readily, though.

You could go with a French past participle, along the lines of resamplé. Or if you don't mind a phrasal noun, you could say "resampled population." Or if there's a single word that means specifically "a population that has been sampled," you could try to get something from that.

Unfortunately my Latin dictionary has concealed itself in the clutter of work and gaming. . . .

Reply


freetrav May 29 2008, 08:53:37 UTC
When you sample something, you divide it into two populations - the "sample", and "everything else" (for which I don't believe there is a specific word).

Analogously, when you sample the sample, you're dividing it into two populations, though we don't seem to have specific words for them. I propose that the first population - the sample of the sample - be called the "resample". Is this what you're looking to name? Or are you looking for the correspondent to "everything else", i.e., the "original sample"?

Reply


freetrav May 29 2008, 08:55:55 UTC
Actually, it occurs to me that the population that you're taking the sample from could be called the "focus" of your study; by analogy, then, the population that you take the resample from could be the "refocus".

Reply

By way of explanation glenbarnett May 30 2008, 05:50:54 UTC
"Resampled population" would be misleading in context, and so I won't be suggesting it ( ... )

Reply


glenbarnett May 30 2008, 05:52:16 UTC
Hmm, that was supposed to be a generic reply to the whole thread, not just freetrav's comment. I managed to stuff that up, sorry

Reply


late reply anonymous July 31 2008, 20:41:18 UTC
Although you have probably already resolved this, how about "resamplicant"? It has the right linguistic hooks to mean what you want, and sounds like something from Philip K Dick to boot.

Reply

Re: late reply glenbarnett August 6 2008, 07:31:11 UTC
It's certainly a possibility; I'll bring it up with my student.

I have further support for resamplicand, in that I have recently found the "icand" ending used on some other words in statistics. Yes, it's ugly, but no uglier than lots of other things in statistics.

Reply

Re: late reply glenbarnett September 11 2008, 08:56:56 UTC
My student has settled on resamplicant in her text.

Reply


Leave a comment

Up