Reddit reviews A Frequency Dictionary of Japanese (Routledge Frequency Dictionaries)

This is a step up from the other list to the extent that it is sourced from a modern Japanese corpus and mostly is composed of real words, but this still is not very good.

The biggest problem is that the corpus is a single source. A good language frequency dictionary or list will use many sources as its corpus. Newspapers, novels, tv and movie dialogue, and so on. In short, language from fiction and non-fiction writing and spoken dialogue from broadcast and film. The failure of using Wikipedia as a corpus is clear when you see that 放送 is ranked with higher frequency than 行く, even though 行く is one of the most commonly occurring verbs in the Japanese language.

Another problem is that it uses kanji for common words rarely written in kanji such as 成る、有る、居る. What is remarkable is that 居る（いる) is marked as the #1,340th most common word in Japanese. This is ridiculous!

If you want a real and useful Japanese frequency dictionary, here it is - the 5,000 most frequently used Japanese words. Yes, it costs about $45 but at least you are getting something useful. Its corpus is compiled from books, newspapers, official documents, web pages, and spoken dialogue for a comprehensive sample of the Japanese language.

I promise everyone - you are allowed to follow Rule #4.

I got the Routledge Frequency Dictionary of Japanese. I love it a lot.

Note that there are plenty of free frequency lists online, but I was never able to find one that wasn't mixed up with grammar structures, conjugated verbs and the like.

Routledge have just published one of these. Anyone know if it's any good?

There are various JLPT vocab books out there. There is also A Frequency Dictionary of Japanese. If you’re going to learn from a word list, it might be effective to just learn the first few thousand words that way, and then learning words directly from native materials, focusing on a specific genre or author (narrow reading).

The above is merely armchair theorizing on my part, because I did not use these books to learn vocabulary; I simply read as much as I could, for example from older readers such as Modern Japanese: A Basic Reader.

Japanska via Duolingo har en väldigt konstig learning curve. Den går från enklare än nybörjare till meningar med kanji utan att ens försöka förklara hur en mening eller kana fungerar i en mening. Rekomenderar att man inte kör med duolingo om man ska lära sig japanska.

Om slutmålet är att man ska kunna prata, läsa och skriva japanska så är det jag rekommenderar:

1. Att man investerar i en arbetsbok, tex Genki 1 (obs dyr), Japansese From Zero! 1 eller Minna no Nihongo (obs dyr), de ger dig en bra grammatisk grund att stå på. Böcker kan också finnas i en sjörövarbukt i din närhet.
   
2. Samtidigt som du använder dig av Anki för att köra igenom Remembering the Kanji.
   
3. Antingen medans du går igenom Remembering the Kanji eller efter så börjar du lära dig japanska ord genom att lägga in dem i en Anki kortlek. Finns en del färdiga kortlekar som redan innehåller alla ord från Japanese from Zero! eller från Genki.
   
   Detta är min strategi. Jag har kört igenom RTK på ca 3 månader och är nu inne på Japanse from Zero! 2, samtidigt som jag håller på och lär mig alla orden från A Frequency Dictionary of Japanese genom min kortlek som jag har postat här: LINK
   
   

• Denna videon förklarar rätt bra varför man ska använda sig av RTK om slutmålet är 100% förståelse av japanska.
  
• Nybörjarguide för Anki

For those interested in something with far fewer issues than this list, I can recommend A Frequency Dictionary of Japanese [non-ref] from Routledge. They use a larger corpus (~100M words) from a more representative sample of contemporary Japanese.

It also has the meaning of each word and usage notes (register and example sentences) It's kinda pricey, but it is a lot better than any of the online frequency lists I've seen - main difference being that this was prepared by language scholars with access to high-quality corpora prepared by other language scholars, not by someone on the internet parsing a random selection of public domain (or possibly illegal) books.

Different selection of words. I don't remember what exactly the source of 10k was, but as was said, it's rather newspaper-heavy and includes some outdated vocabulary like 日ソ, along with some incredibly infrequent words.

Nayr is based on this book, which uses a mix of written and oral material for its corpus and is purely frequency-based.

I kinda prefer Nayr's approach, but since I was already 6k words into Core10k when I found it, I haven't actually used it.