tiia wrote:I'd say that input based methods may work well with beginner stages of rather closely related languages. But what about more grammar-heavy ones?
If someone tells you "the best thing for my English/Spanish/French was to watch videos" I wonder how they would cope with Hungarian, Finnish or Basque etc, where grammar is way more crucial to understand even basic sentences. Or what about the influence of a different script?
So far I probably know only one person, who may have succesfully be learning also unrelated languages by mostly watching videos. (But still practicing at the language cafe.) I definitely know way more people, who learned a language with textbooks or a mix of materials. The same goes for people, who were succesful by going to classes (+ doing something outside the classes)...
Yes, that's very true. But unfortunately (and partially logically), the market is heavily obsessed with Spanish for anglophones. But the things that might work in that situation get overgeneralised and will set anyone else, such as a francophone learning Hungarian or a Czech native learning Japanese, up for a failure.
And then there's the question how to measure the results. If you mainly test people on vocabulary, then SRS will probably get you good results, but the best learning methods for beginning stages and more advanced stages of learning are most likely to be different. Some mistakes may be acceptable for beginners but not for advanced learners. So long term studies would be really interesting in order to see whether a method would be able to get you to the higher levels or not. But that would probably just lead to either people not following the method or dropping out of the study. Considering the dropout rate of ... one would need thousands of participants to get a decent number through the program to have any data to actually compare the methods.
Yes, the long term and the result measurements are big issues. For example, I saw an article, that claimed to compare extensive and intensive reading. They were using a sophisticated looking method on vocab retention, which really fooled a lot of people into thinking "this is exact science, and anyone disagreeing about extensive reading being useless is dumb". But the problem was the whole experiement being skewed. The author, perhaps in well meant search for objective comparison, created an experiment that went against the whole point of extensive reading. It was a similar number of hours spent either intensively or extensively reading. Of course the extensive reading was bound to fail, it's in the name of the method
You need to set the experiment right, you need to do long run experiments, you need to evaluate the results right. It's not easy, and no wonder most people seem to fail. I have yet to see a single article, that would methodologically be as sound as average articles in medicine.
All that in a world, where we do not get access to even such basic information as the amount of people taking and passing official exams! Exams like DELF/DALF, TCF, Goethe Zertifikat, and so on are organised or co-organised by public founded, or partially public founded and surely public supported organisations. Nevertheless, you never get an answer even to such simple questions like:
-how many people took exam X last year?
-what were the average grades?
-were they mostly school goers or was there a significant % of independents?
-did people with some native languages do better than people with others?
-were the results different in various age groups?
-do people in general do better at this or that part?
This would be an excellent source of information on language learning, I would love to read research based on that. Much rather than the usual dumb quantifications of word retention. It would be a much better sample for surveys, than the usual worthless (for our purposes) sample of average college class goers.
But nope, we don't even get the % of pass/fails info
If the dropout rate is so significant one can also ask, why everything has to be as efficient as possible, if as a consequence too many people cannot follow the method. A less efficient method that keeps people studying may even lead to better results, if it means that students will be less likely to drop out.
(I still think only listening for hundreds of hours without doing anything else doesn't make sense for most learners.)
That's a very good point, but the word "dropout" needs to be used well. Duolingo uses it in their marketing material (which they generously call "research" in their PR), to label critics as drop outs, who surely would have had awesome results otherwise.
In case of Dreaming Spanish, I see an unpleasant tendency of the zealots to do a similar thing. In their world, you failed because you dropped out (even if it is after hundreds of hours), or because you did the logical thing of also normally studying.
Cainntear wrote:rowanexer wrote:luke wrote:rowanexer wrote:There seem to be a lot of people on reddit who are convinced that it is "settled science".
You may want to let them know that those in know know that "settled science" is pseudoscience.
Well, perhaps they are thinking of a scientific consensus?
I don't read a lot of research for this. And they are perhaps looking for some sort of research paper explicitly comparing different methods to find out which is the best. I have no idea if such a thing exists.
No, they genuinely believe in infallible science, which is a really dangerous thing. People who don't understand that science is nothing more the current best explanation of observed data are prone to view the scientific process of constantly re-evaluating itself as somehow disproving science: "science proved wrong again," as many creationists would put it.
I reckon that's a big part of why people don't accept language teaching as a scientific discipline: there are no provable facts in it, so they think in can't be science, but science never proves anything anyway! Did Newton "prove" Newtonian mechanics? Did his theories of gravity become useless once universal gravitation was understood? And even if it you don't think it's a science, that doesn't reduce the value of taking a
scientific approach to it.
The thing is, that people in some fields are used to considering these questions (and more) all the time: "is this good science" "how is this biased" "is the sample well chosen" "is the sample big enough to balance out some variables" "is this a good way to measure this" "what are the limits of these results" "how are these results aplicable" "what else has been published and how do these pieces of the puzzle relate". In medicine, where a lot of things are easier measurable than in language learning, we are used to following the results and also doubting them, to think critically about them.
When I see discussions about LL research, they are usually extremely childish. Like "you don't agree with this SCIENCE? You are just a moron that knows nothing. This article is the truth!". And I am afraid it is not just the random amateurs on the internet. When discussing it with a few people working in the field of language teaching and using "science" based arguments, they were behaving the same way, perhaps just with more polite words. It's really sad, because they were supposed to be university educated people, but it didn't show.