At any rate the Mohs scale of question hardness was actually developed in the 1980s, (See Buzzer Issue 5 http:// guir.berkeley.edu/quiz-bowl/resources/newsletters/Buzzer5.html) though there was only a Delphi /Expert system process to classify questions, along with some question frequency analysis done by Dr. Robert Meredith (much like Eric Hilleman does today). The use of google or a google-like substance along with a log formula is an improvement that gets us out of the self-referential quizbowl mode. However, the particular formula can go negative for super-common words, and will change over time as the Google index becomes larger (without the fact really becoming easier or harder). A good measurement should remain constant over time (like the meter). The previous measure would not be approved by the SI people. I think a good metric here is required, and with some work (like controlling for the size of the google database), is possible. I don't know the size of google's database, but the word "the" shows up "2,560,000,000" times, so that is a lower bound. If we assign "B" to be the baseline measure, and then look at ratios, we could develop something a little bit more stable e.g. LOG (GoogleCount(B)/GoogleCount(X)) where X is the word in question and B is the baseline word or wordphrase (and preferably B >> any X we are likely to test). B should be large (it need not be "the"), but should be something common and unlikely to change relative position (e.g. "George Washington" ) Hits = 1,040,000 or "William Shakespeare" Hits = 406,000 but not "quiz bowl" 22,800 (not counting "quizbowl") I am open to what the Baseline word should be -- David M. Levinson Wondering Why wouldn't it be the Matthews-Levinson (or Levinson-Matthews) Scale ? "David+Levinson" 3130 Google Hits "David+M.+Levinson" 64 Google Hits Hoping that I become Lexiconical if not Canonical. --- In quizbowl_at_y..., castrioti <no_reply_at_y...> wrote: > > Fluidmosaic6 wrote: > > The Castrioti Scale of Question Hardness sounds authoritative, > > scientific and European. It's also a Chevy dealership around here, > > if you remove "Scale of", "Question", and "Hardness". > > (Really?) For fun, anyway, > > I guess my inclination is to call it the Levinson-Castrioti Scale of > Question Difficulty (trying to give credit where credit is due-- > DavidLevinson inspired the idea by trying to figure out whether Tonio > Kroger or Felix Krull was more difficult using google), yet the > equation 10-logN = difficulty (where N = hits on google), as far as I > know, is original. Plus, I think it somehow sounds more academic with > a hyphen --This, of course, only if DavidLevinson has no objection... > > --Wesley (not really trying to resurrect an old thread)
This archive was generated by hypermail 2.4.0: Sat 12 Feb 2022 12:30:46 AM EST EST