5 ms·

It's not the order, it's the frequency when analyzing a corpus such as all newspapers.

• csa 5y ago
  > It's not the order, it's the frequency when analyzing a corpus such as all newspapers. It’s definitely not that either unless you have a very specialized corpus. From https://www.manythings.org/kanji/top2000kanji.html https://www.manythings.org/kanji/top2000kanji.html: 書 - 35 場 - 64 男 - 104 議 - 285 選 - 187 調 - 263
  • eska 5y ago
    It should be obvious that the exact result depends on the corpus and year. The one you link is apparently from 2003. 35 and 25 are in the same ballpark, so the article's point about it being a very common character stands.
    • Zarel 5y ago
      You misread. 35 and 285 are far from the same ballpark, and it's not really possible to explain that with any corpus. It's much more likely that the author found a list of jōyō kanji sorted some other way, and thought it was sorted by frequency.
  • fstrthnscnd 5y ago
    jisho.org doesn't provide the same numbers for these kanjis: 議 : 25 https://jisho.org/search/議%20%23kanji https://jisho.org/search/議%20%23kanji 選 : 57 https://jisho.org/search/選%20%23kanji https://jisho.org/search/選%20%23kanji 調 : 87 https://jisho.org/search/調%20%23kanji https://jisho.org/search/調%20%23kanji and for instance 書 : 169 https://jisho.org/search/書%20%23kanji https://jisho.org/search/書%20%23kanji I cannot tell how accurate these numbers are, but jisho.org is using the free dictionary built by Jim Breen of Monach University (see the about page for details). It's a well known resource in the open source world.