RESULTS: How much can I improve my listening in 6 weeks?

TLDR: Listened a lot. Measurably improved at JLPT listening despite listening to mostly comedy stuff. Speaking also felt like it improved. Numbers are in the main post.

INTRO
During the summer I spent six weeks focusing on improving my listening ability. Listening ability is often something that people struggle to improve and is one of the least reliably measurable skills when learning a foreign language. I wanted to see if anything can be done in a short period of time to markedly improve one’s listening.

IN THEORY…
As listening is so hard to measure there is far less research into general methods for improving listening comprehension compared with reading comprehension. There are three things which are provably useful for increasing listening comprehension though. First, repetition. Second, more time spent listening to level appropriate material. Third, aided listening. Repeated listening improves comprehension to a certain degree for the obvious reason that one can notice more on repeated listens. More time spent listening is, again, intuitive at least in theory. Aided listening (this is listening where comprehension is aided via scripts, images, contextual information, etc) works as you already can roughly follow what is happening so you can more easily associate sounds with meaning. With these things in mind I laid out a plan.

MY LEARNING BACKGROUND (optional read)

A brief note on my learning background. I have been learning Japanese for the best part of three years now, around two and a half years at the start of the summer, and have lived in Japan for about two years. I am by no means a speedrunner but my progress has been solid and I am now around the lower advanced levels as a learner. I have mostly been a reading focused learner and use minimal SRS, probably about half the time I have been learning Japanese I have used no SRS. My reading is slightly ahead of my listening, I would say.

IN PRACTICE
Of the three proven factors above I personally cannot endure repeated listens. Time on task is a far more important factor so making my listening enjoyable is the priority. I am already at a level where I can comprehend things so lack of comprehensibility influencing enjoyment is not a massive factor, assuming the material is level appropriate. As such, I just tried to listen as much as possible to whatever I wanted to in the moment. I also read when I felt like it but deliberately limited my reading when possible. In practice this meant that I spent the majority of my time listening to comedy podcasts, watching variety shows, and a chunk of random more casual youtube videos. I also watched a handful of dramas, news related things, and movies but not much compared to comedy stuff. In terms of aided listening, I used subtitles when available.

I didn’t record everything perfectly (I live in Japan so random reading, speaking, and listening happened) but rough statistics across the six week period are as follows:

Podcast listening: 82 hours
Easy listening: 87 hours
Difficult listening: 13 hours
TOTAL listening: 182 hours
AVG daily listening: 4.3 hours

Easy reading: 7.5 hours
Difficult reading: 11 hours
TOTAL reading: 18.5 hours
AVG daily reading: 0.4 hours

Bunpro: 3.5 hours
Anki (vocab): 12 hours
Textbooks: 5 hours
TOTAL study: 20.5 hours
AVG daily study: 0.5 hours

Intensive conversation: 23 hours
AVG daily: 0.5 hours

Translation work: 3.5 hours

TOTAL recorded: 246.5 hours
AVG weekly: 41 hours
AVG daily: 5.9 hours

RESULTS
I will skip the details on this but measuring listening comprehension is difficult and at non-beginner levels you’d want to test across a wide range of domains, including domains that would typically have low comprehension (legal language, academic language, etc). There was no good way to concisely do this. My initial workaround was to listen to various things and record what percentage of words and sentences I couldn’t comprehend however this proved fruitless as comprehension is not a yes/no skill so everything I tested would end up as seemingly very high comprehension as I could follow it pretty well, especially given context and images as in the case of video based media. Traditional comprehension testing therefore turned out to be the quickest and easiest “least bad” option.

So I took a JLPT N1 practice paper. I took a couple of breaks as I did it at work and had other stuff to do however I would say that my ability that day was average. I then retook the same test 6 weeks later and I took a completely different practice paper as well. For the retake I was having a rather low ability day and took one break. For the fresh test I was having a reasonably good day and took one break.

Results
First test: 65%
Retake: 76%
New test: 81%

INTERPRETATION
As you can see my result went up a noticeable amount and the new test confirmed that this wasn’t a fluke or simply me remembering the questions (which I consciously did not barring one question). This is clearly not a rigourous test or result and other factors may have been at play (I could have gotten better at taking the test itself for some reason). Regardless, I was surprised myself how significant the improvement was.

The more interesting part of this result is that my listening was absolutely not focused on JLPT style material at all. I was concerned that my vocabulary would not expand in JLPT relevant areas, and I honestly don’t think it did, however my general ability to catch and process the language and also to infer meaning quickly when I am confronted with an unknown probably improved and carried me here. I did learn new vocab but it was almost all slang and it all remains at a shallow level of understanding. The questions I struggled with the most were ones set in a business environment as I’m just not as used to that kind of speech and language.

ANECDOTAL RESULTS
Anecdotally my ability improved a lot in conversation and my confidence when speaking also went up. Interestingly when I stopped doing so much daily listening I felt me ability drop again, like I was out of the groove. Although subjectively my listening improved slightly I am happier about the improvements to my speaking ability than my listening. Also anecdotally, my reading ability felt like it improved as well despite doing minimal reading. By the end my reading speed felt slightly slower however my fluency of reading felt better.

A NOTE ON QUALITY OF ATTENTION (optional read)

I want to mention something that is very difficult to measure which is the quality of one’s attention when learning. I of course did not always have perfect attention when listening over those six weeks and sometimes material was a bit beyond me. The conclusion I personally think is valuable here is that although quality of attention is obviously important the actual time spent is even more important. You should aim for 100% attention but if you’re doing 6 hours a day then you’ll naturally have ups and downs. Better to show up and try even when you can’t perfectly focus. There is a lot to discuss here regarding extensive/intensive listening but I will leave that for another time.

CONCLUSION: WHAT IS COMPREHENSION ANYWAY?
In the way of concluding remarks I want to leave you with the question of what the hell is comprehension anyway? When I was a beginner I would often count my ability to roughly follow something as comprehending it. Now I’m a bit better I am quite strict that if I don’t know everything perfectly then it doesn’t count as comprehended. I’d even go so far as to say that if you couldn’t answer a regular high school level question like “Why did the speaker choose this word instead of that word?” then you don’t fully comprehend it. Of course that is a ridiculous measure in general when learning a language but as a mindset it really helps push me to think a bit deeper sometimes. What are your goals? What counts as comprehension for you? Those are the real questions when it comes to this topic.

Thank you for reading. I hope to do similar things in the future and I’ve learnt a lot about how I may better approach measuring and recording things if I were to repeat this kind of expirement again. I left a lot out of this post to avoid waffle so if you have a question or something then let me know.

15 Likes

I think comprehension you describe is too close to Laplace’s Demon level of understanding, but I can appreciate the effort.

You seem further along than me and I like your writing style, so a few questions.

  • What is “easy listening” for you and where do you find it?
  • Is your reading “consecutive” like listening to ongoing material / the same podcasts etc. or a bit more random?
  • How many new and/or unfamiliar Kanji do you come across each week?
  • Any notes on study patterns? I’ve got 4 days off work this week and today and was thinking of trying some new learning structures to see if I find something I like more :slight_smile:

(And to be clear, I’m less interested in JLPT stuff, I just want to be better at the language in general!)

3 Likes

For example, this video is full on easy (I know basically every word or phrase and don’t have to think at all to comprehend it - subtitles obviously help). And this video is just about in the easy category for me (there are words I have not seen much or words I have never heard but from the subtitles I can understand fairly straightforwardly - they are also explained in the video or clear from the context but I may need to pause to consider parts). My comprehension is very domain dependent so for example I am pretty bad with fantasy stuff however I can still follow it just fine even if words to do with magic and armour and stuff fly over my head fairly often.

I just browse or search youtube for stuff or I go on netflix and choose whatever looks good. If something is too difficult then I may turn it off. For books I go to the bookshop and browse and then normally buy the ebook later.

I like to have a few different books on the go at the same time. I will occasionally binge stuff though. Listening tends to be more casual and I’ll just watch whatever. I do have a handful of favourite podcasts though.

Depends on what and how much I read that week but I guess 0-10? I would say definitely not more than 10 unless I happened to have been seeking something out that specifically uses difficult kanji. In general no idea though. 90% of the time they are kanji that natives also are not so familiar with although sometimes there will be a glaring hole in my knowledge for some reason. My most recent memorable one was 閂 which I had actually seen in Hong Kong a year ago but didn’t know was also in Japanese. One person out of about ten that I asked could read it. Normally I forget them as soon as I see them. If it is important I will see it again. Definitely my knowledge could be deeper but generally kanji is a non-issue.

I may actually count how many new kanji I see this week and report back. Could be interesting.

If I have some days dedicated to Japanese then I try to start as soon as I wake up. Before I realise I’ll have done an hour of reading/listening before breakfast. When taking a break don’t switch to another language. Just do nothing or go for a walk or do some light chores or something. I personally think this gives your brain some space to sort out what you’ve been learning/inputting (there is research supporting this). I also like flipping between two things so I always know what to do if I get bored, normally flipping between reading and more casual listening. I would always advise to just do whatever gets you to spend more time with the language and that tends to be different for everyone.

Enjoy your days off!

4 Likes

I counted last week and the results were 2 unfamiliar kanji.

é–ľ, which I had seen before but never actually in the wild (always on lists or in the dictionary in passing). I asked 12 natives if they knew it (with no context) and 3 did.

And 塑, which I think I’d never seen before or if I had then I ignored it. I asked 7 natives if they knew it (with no context) and only 1 did.

I’d wager in context they’d probably have a far far easier time recognising them, of course.

Both these words came up in in this book which I read the entirety of during that week. I also started on some other books (maybe another 50k characters in total) however no new kanji came up in those. Some books I come across unknown kanji pretty frequently so it is kinda down to luck as well. For this book both kanji were used in technical terms.

I may count again at some point as it was kind of interesting to do! Thanks for the question

3 Likes

I really appreciate the reply! It’s a bit of a light at the end of the tunnel for me as I approach more brick walls in listening comprehension :slight_smile:

1 Like