Duolingo

I hate Rosetta stone. It's fine for learning some things, but I don't think you can really learn a language from it. You need more interaction/input than it can provide. Also, I have never met anyone who's purchased it and learned much of anything from it (though some of the blame for that is on the people who bought it thinking it could magically turn them into a native speaker with little work on their part).

I don't know about Duolingo, so I can't say. I'm very much of the belief that immersion is the best way to achieve fluency, but also, there have been studies done that showed that immersion + some kind of classroom/formal instruction is better than immersion alone, at least for adult learners. I wish I could cite the specific studies, but I read that while in grad school for foreign lang. education a few years ago, and I'm not about to try to hunt any of it down. Teaching Language in Context by Omaggio Hadley was the main text we used, and it almost certainly cites some of those studies.

I agree so much. Basically: there's no one magic bullet for learning a language,

Here's what leapt out at me in that summary:

• The main factor for higher effectiveness was the motivation of the participants, with people studying for travel gaining the most and people studying for personal interest gaining the least.
• Another factor for higher effectiveness was the initial level of knowledge of Spanish with beginners gaining the most and more advanced learners gaining the least.

That pretty much confirms what I've always thought: These products are most helpful for someone starting from scratch and show diminishing returns. And there's no substitute for motivation. Better materials or methods might make things easier for you, but if you're not serious about learning a language, you won't learn it no matter what you use.

(Also: it's only one study. A big problem with claims like these is that companies are under no obligation to disclose studies which demonstrate a lack of effectiveness. That is, if there are a dozen studies showing that Rosetta Stone is better than Duolingo and only one show the opposite, well, no prizes for which study is going to be listed in Duolingo's promotional literature.)

Very well said, thank you.

From my long ago studies of language acquistion, it seems that for any method, motivation is the key, as you say. Age too--helps to be under 12! But in that case, you probably just go out and start talking.

But John Grego is a very honorable guy, so I trust this study statistically, but as you say, what if there were others?

I am trying to learn Spanish at Duolingo. I used to take a beginners Spanish course at a community college a few years ago. What I prefer at Duolingo is the word "practise weakest words" section. I learn more words that way than at the college course with my flash card system, but there I had the feeling that I did learn grammar better and it actually helped to have some things explained rather than just read the tips at the bottom of a lecture/skill.

So far I do enjoy it, though and plan to get some basic French and Italian that way as well.

Oh and at the end of the study they said something about only measuring the people who stuck with Duolingo over the period of time with college students and Rosetta Stone users and that only the most motivated stayed on it, while even less motivated students might need a college credit and unlike unmotivated Duolingo students weren't able to drop out and thus maybe test worse because of lack of motivation. Motivation is a big key I guess.

I think they would have to do more studies to come to a conclusion about methods.

Totally agreed about the importance of motivation.

Bueno suerte con tus estudias españoles!

Gracias. :)

The 'practice weakest words' bit is great, although currently it isn't flexible enough: you can't for example define a group of your own to practice, say you can't pick a set of verbs whose forms you don't know so well from the vocabulary list and practice them, only if they appear in the same lesson or are marked the weakest (which often doesn't coincide with how you actually feel about particular words).

The thing that annoys me the most in Duolingo is that there are often not enough different phrases in a lesson, and if it isn't going too well, and you have to repeat it, you might end up memorising the phrases rather than the thing they are supposed to illustrate, just to get to the end of the level lesson.

Also, there isn't much incentive to translate a lot: no badges or titles or things like that. I used to be among the top Es-En translators there for some time, but ultimately stopped doing it because of this.

EDIT: the best part about Duo, though, is that it gets better, I registered there during the beta, and the project has gone a LONG way since then.

I am not far enough into the program to have verb forms yet, so I can't say anything about that. And I do get what you are saying about the phrases, though for me that doesn't seem to be a huge problem at this point, I think. Words are used in several different phrases and you have to know them both ways for each, so I at least feel like I actually learn the word after a while and not just the phrases and for some words the images created by the sometimes silly phrases are really helpful at this point. I'll have to see how I feel about this in a year. That study mentioned up there does say that it is better for beginners and the further you get with your skills and knowledge, the less useful it is as a teaching tool.

I do have the feeling the program repeats my weakest words more often than the ones I do know a bit better, but also sometimes tests the better ones in between. So I guess if it were just a phrase I memorized, I would forget about it over time and weaken the word so I'll have to study that again? I will see if this becomes an issue in the future.

I only do the translations when they are required for mastering the lesson. I don't know if I would feel different about this if there were badges or something to motivate me to do more. They are not my favourite part of the program. That is a good point, though. I hadn't thought of that. Maybe they will introduce something like that in the future?

Good to hear they constantly keep working on the site, though, as there still seem to be some mistakes, even though it is out of beta. This is that annoys me the most. Aside from the random "I accept this as a typo, but not that, even though the first was clearly wrong and the second one was actually a typo" thing Duolingo has going on what I don't like is when my answer is technically correct but Duolingo won't accept it, as it hasn't been added to the database in that form yet.I am always a bit tempted to do the German lessons (my native language) just to see if I only feel like my answer should be correct or if they really do ignore correct but differently phrased answers. .

With respect to the errors in lessons and the misinterpretation of typos, Duolingo have streamlined the bug reports and most importantly each phrase has had a discussion thread for some time now: you can actually see if others have the same problem, or if there really is an error in the lesson, or it was you who made the mistake, or ask people for help.

I think these discussions and people who contribute to them actually make Duolingo's otherwise mediocrish language courses good.

My boss's husband did the study, and that totally impresses me. Haven't tried it yet, but will! Admit to not being much of a fan of RosettaStone. Or Pimsleur.

Whoa, your boss's husband? Small world!

John Grego, head of Statistics Dept. at University of South Carolina. He's really a fine person, so I don't question that this is honest research.

That's fantastic! Thank you for sharing!

But the point about not seeing opposing research is still valid.

I got the same email. If you click through to the article, you will see that it doesn't actually compare Duolingo and Rosetta Stone, although its author has also done commissioned research on Rosetta Stone's effectiveness. I have emailed Duolingo to ask for further information but have not received a response. The fact that Duolingo released the study with the "more effective than Rosetta Stone" when this was NOT proven is fairly repugnant.

Like the other commenters, I think that the single most important factor is motivation. Also, different people will, of course, learn best in different environments. Computer-based instruction might be superior to classroom instruction for someone who is shy. Some people learn like sponges and will thrive in an immersion environment, while others are more analytic and really want to work their way systematically through structured grammar lessons.

True, it is just a study of Duolingo.

Learning through translation is an inferior method in any case, so even if some translation program or syllabus is better than some other, it could still be irrelevant for real-life knowledge of a target language.

Native speakers, it seems, do not talk with "words", there are larger chunks of language that come into play and are used as "Lego pieces" to construct your speech from. The most important part of language learning is knowing those "standard/frequent word combinations" and having a pretty good inventory of them.
The use of translation on the other hand encourages a view that "words" are combined "freely" according to "the rules of grammar", more or less, and most of today's textbooks pretend that each "word" has a "translation" or "meaning", well, 2 or 3 of them at most - while in reality most frequent words' entries could run up to 50 various meanings and uses in learner's dictionaries.

In English, for example, the first (most frequent) 650 or so words have about 13 meanings on average, the next 1500 about 6, etc. - and you have to add to them various collocations in which the word participates.
The major conclusion one can draw from these statistics is that the main target in language learning is the acquisition of vocabulary, learning ALL meanings and uses of the most frequent words in the language (and about 3000 will give a learner ability to express pretty much everything, although in a limited way), and doing it in a way that teaches "grammar" (the structure of the phrase the word in a given sense is used in) and "collocations" (with which groups of other words it naturally combines in this sense) together with the word.

TRANSLATION methods
(a) habitually disregard this information
(b) substitute vocabulary learning with infantile "word-translation" pairs and usually pretend there is one meaning (like "hand" = part of human anatomy, disregarding about 50 other uses of it)
(c) pretend learning a language equals to learning 'grammar', which is taught not as a set of ideas/notions to convey, but as a set of rules and'or structures
(d) encourage very, very strongly (by the fact of referring to the leaner's mother tongue) to substitute collocations, verb patterns, etc. from his mother tongue for the patterns and collocations as they are in the target language -- while the learner remains unexposed to those.

Translation methods based on traditional grammatical approach are not and cannot be efficient. What is needed is a LEXICAL SYLLABUS (with grammar taught as a set of notions to convey rather than formal structures).

p.s. I had a chance to see how foreign learners of English, deprived of information about the natural "chunks" of English and substituting collocations and patterns from their mother tongue ended up inventing their own variety of Pidgin English. They got amazingly fluent in it (mutually encouraging themselves in their learning group), and incomprehensible to a native speaker or someone exposed to the real, natural English.
A native speaker stumbled and asked for clarifications several times per minute, which, paradoxically, created an illusion that the native speaker was a dummy, while someone really fluent had constantly to explain things to him in a sort of condescending way.

No, translation as a main method of language-learning goes to a garbage dump.

I wish there was a lime button for each cooment so an inane "SO MUCH THIS" post wasn't neccessary, but


SO MUCH THIS

Thank you for putting my thoughts into words!

This!

Not disagreeing with your comment, really, but have you actually used Duolingo? Because I am under the impression that you haven't.