User distributions, success rate

[omoide]

Hello Everyone,

I just joined the WK community, and I would be interested in any census data and statistics on WK community.
I am interested to see how all the past and present WK members is distributed (age, country, WK level, “success rate”, quitting rate, university degree, etc).

For example, what percentage actually reach level N?
Or it seems obvious that users living in Japan will have more incentives to progress.

I think this information would help me decide on subscribing.

I found a similar request in the archived discussions :

https://community.wanikani.com/t/wanikani-statistics-user-distribution/5549

Thanks in advance

[Leebo]

Well, we have one stat for you.

https://community.wanikani.com/t/its-a-stat-22-people-make-it-to-level-60-every-month/30156

[rfindley]

I think the only one of those statistics that is actually collected anywhere is WK Level. Well, except some users that posted polls in the forums at various times.

[xyzbuster]

@hibou

For example, what percentage actually reach level N?

The data might be imperfect, as it apparently only looks at forum data, but this is probably the closest approximate we could get

Wanikani Users Distribution by Level

I’ve gathered all users information and created these graphs showing how many users are at what level. The number of all users was at the time of download around 28,000. [image] Number of people at each level. Drawing the data in log scale and ignoring the first 3 levels since they are biased reveals something interesting. [image] The number of users drops exponentially which might be another case of Zpif’s law Now for some some graphs how many users have given up meaning they are no l…

[rfindley]

The most recent data I have (extracted from the forums) is November 2017:

Lvl  Users
01   7849
02   4378
03   3563
04   2474
05   1444
06   1402
07   1223
08   1182
09   1097
10    989
11    948
12    746
13    587
14    516
15    415
16    386
17    404
18    419
19    343
20    340
21    347
22    334
23    264
24    213
25    225
26    193
27    197
28    153
29    123
30    142
31    143
32     90
33     93
34     94
35     70
36     69
37     91
38     83
39     77
40     75
41     72
42     58
43     47
44     44
45     34
46     25
47     38
48     31
49     33
50     42
51     63
52     29
53     39
54     30
55     37
56     30
57     20
58     30
59     18
60    358

[acm2010]

Do you have any success rates of other methods for comparison?

I think statistics will not tell you if WK is good or not, people would probably quit any method when they are tired of kanji

[omoide]

I find the replies already quite insightful, in fact

Regarding the statistical distribution: the correlation to known models (zpif, etc) may be improved if the raw data are corrected to compensate the account creation time, as well as the birthdate of the WK site.

We could play around the hypothesis that the monthly number of new participants (ie. lvl 1) is constant, and the minimum progress rate is 7days/level.

Getting raw data (which hopefully contains creation data of each account) could give some great stats.

[jprspereira]

@hibou

“success rate”, quitting rate

@hibou

I think this information would help me decide on subscribing.

If I had decided not to subscribe because a bunch of people give up before even paying for this wonderful product, I wouldn’t know 1700 kanji by now The product works, you just need to show up everyday for the next 1 to 2 years.

[Ninkastmin]

I hadn’t realized that in the last 5 weeks you learn 300 kanji. That’s a lot
Edit: that is the amount of kanji to learn at Max speed

[jprspereira]

@Ninkastmin

I hadn’t realized that in the last 5 weeks you learn 300 kanji. That’s a lot

I’d say the kanji itself is not a problem, but keeping up with the vocab lessons (30+ words/day). The amount of reviews you get are easily above 300. Yesterday I had 400 (and I still have 9 levels left).

At this point, one gets pretty good at learning kanji. At your level, I wouldn’t imagine myself learning 35 kanji in a day. Now it takes around 40 minutes or so and a good study environment

[acm2010]

@Ninkastmin

I hadn’t realized that in the last 5 weeks you learn 300 kanji. That’s a lot

Edit: that is the amount of kanji to learn at Max speed

The number of kanji per level are the same, but starting with level 46 all levels except 48 unlock 90%+ directly, so you can level up in 3.5 days if you want.

That is the time when you realize that “going faster” in the beginning is not a good thing to wish for

[anon6098980]

In recent years, an increasing number of WK users have been hired by their local Japanese embassy, their country’s embassy in Japan, their country’s ministry of foreign affairs, various Japanese companies such as Mitsubishi, Makita, etc., various tourism companies, universities and other private and public Japanese teaching schools.
By having a WK degree (level 60), you are almost guaranteed to land any job in Japan or Japanese companies abroad.
Also it has been proven that reaching level 60 increases your chances of travelling to Japan by 420%.

Disclaimer

I may or may not be getting this information from my rear end…

[sornvru]

I have the opposite problem. The vocab are usually easy and I struggle with learning all of the new kanji.

[Belthazar]

@anon6098980

Also it has been proven that reaching level 60 increases your chances of travelling to Japan by 420%.

I’m only level 16, and I’ve already had a 200% increase on visits to Japan compared to before I started on WaniKani, sooo…

[jprspereira]

@sornvru

I have the opposite problem. The vocab are usually easy and I struggle with learning all of the new kanji.

Yeah, I agree. I was talking specifically about the “fast levels” (46,47, 49+), since you can do them in 3 days and 10 hours at the maximum speed (compared to the usual levels where the maximum speed is 6 days 20h).

[Aragnos]

That gave me a good chuckle, even before reaching the Disclaimer part.

[Momoiro]

It took me longer than I care to admit to realize this wasn’t serious hahaha