Reasoning behind Duolingo's course structure?
I'm curious to know more about the structure of the language programs at Duolingo if anyone would care to shed some light :)
I'm fairly proficient in French (wife is from Quebec, it's her family's first language), so I'm using the platform to brush up a bit on vocab. I find it quite useful for this purpose.
However, if I were learning French "from scratch" without any previous experience, I'm not sure it would be that effective - but maybe I'm wrong?
What is the basis of the program structure? Is it following any language pedagogy or based on research of any kind? It strikes me as pretty random.
I think mostly they deferred to the hoped-for good sense of the contributors. In certain instances, the results of this are widely lauded — Russian and Czech tend to get good reviews for structuring, although you'll also come across people with a very different opinion — and sometimes they're not, with even current contributors lamenting bizarre decisions made by previous contributor teams. Norwegian tends to get high marks for comprehensiveness and enjoyableness, but I'm less sure on what people think of it's structure per se. On their fourth or so go-round, it's probably pretty reasonable. I think it, like Dutch, makes a point to introduce the bulk of the grammar by the "end of the middle" of the tree, giving the change to apply it more naturally through the more advanced vocab units at the bottom of the tree. Other trees seem to do this much less (a decision that might date to an era when they hoped to move units around more freely to maximize learner retention through A/B testing) making the units more atomized, and it's pretty clearly to their detriment.
Japanese was designed to accord with the requirements for the first level Japanese certification test, although I'm unaware of others being as tied to an external criterion, other than a desire to get students to an A1 level of competence. Russian was designed to incorporate an intro to essentially the full sweep of Russian grammar in conjunction with a carefully constructed vocabulary selection based on frequency lists and a few select thematic units that also served as a medium to introduce more bookish general structures. My impression is that some of the original courses gave much shorter shrift to more advanced grammar topics, c.f. Luis's somewhat famous dictum along the lines of "we're not going to drill you on subjunctive."
The most common critique I've come across regarding the structure of the courses is that they don't introduce immediately useful phrases (useful to a tourist, for example) right from the get-go. I personally don't find that critique very reasonable. Tourist language phrasebooks are widely available for those seeking information along those lines, and their objective is pretty distinct from Duolingo's. My biggest critique on structure for most trees is probably the highly atomized grammar. For example, I don't think the Catalan course has many (if any) object pronouns outside the two object pronoun skills. That's some pretty basic stuff that then isn't getting drilled more "naturally" anywhere else in the tree (and those skills are even in the top half), to say nothing of the different verb tenses and moods that are confined to their respective skills. The courses that have more divergent grammar seem to have thought about course order somewhat more carefully. Beyond Russian and Czech, people also seem to quite like Turkish; Hebrew I think is reasonably ok; and the critiques of Hungarian and Swahili tend to focus on other matters. Some of these still have their grammar quite broken up, but because there's simply so much of it, I think that's more to be expected.
Good question. I'm a beginner who has finished the French tree. Duolingo has helped me with spelling, and I am gaining the ability to answer accurately more often when strenghening skills on Duolingo. I can also read and understand a bit of written French, i.e. sub-titles on French language television. I am married to a native French speaker from Quebec, and I can't say that it has helped me any with actually speaking the language in an everyday context.
As a comparison, two weeks ago I started a classroom French course at the Alliance Française. The curriculum follows the CEFR guidelines. The instruction is immersive, thematic and contextual. In just 2 weeks with only 8 hours of classroom instruction I am starting to connect French with everyday life and express basic ideas. Duolingo has already taught me most of the vocabulary that we cover in the classroom, but it's only now through classroom training that I'm starting to be able to communicate about the world around me.
This morning while practicing French on Duolingo I successfully translated the sentence "Two times as many women as men work at home". I'm proud of the fact that Duolingo has taught me to be able to translate that sentence from French into English, but I'm not sure of the value of being able to correctly produce a stand alone sentence out of context through rote repetition. I'm not able to see how to connect that sentence or even an underlying structure into real world use. That doesn't mean there isn't a connection, it's just that I can't see it through the Duolingo approach.
In spite of that, I do enjoy using Duolingo and look forward to my daily practice sessions. For me it's just one tool among many. From my perspective Duolingo serves as a good way to learn/practice vocabulary and learn spelling.
This sounds more like a critique of Duolingo's lack of translation into target languages than about the ordering of material (which, of course, is a 1000% valid critique).
Your third paragraph is confusing because you give the English sentence and said you translated it (which would mean into French I would think), but then said you were proud that Duolingo had taught you how to translate it from French. If you translated it into French, there are of course several generalizable points of knowledge you might have employed: the fact that French has a pretty unique preposition for structures like "at home" or at least the employing of the correct more-general preposition, the knowledge and correct use of third person plural plural verb conjugation, the fact that French has two separate words for the two main senses of "time," comparisons of quantity (two times more), not to mention all the basic vocab.
If you have simply committed the French sentence to memory, it'll serve as sort of a "mental storehouse" for these concepts with only a small amount of mental effort to reflect simply on what value is already contained there.
Of course if you meant that you recognized the French sentence and thereby recalled the English translation, I would concur that acquiring that ability probably isn't all that valuable, and it's probably best to change up your learning modalities somewhat, possibly focusing on something as close at hand as the reverse tree (or Spanish for French speakers) for your Duolingo time, which at worst will have you memorizing sentences in a language that will actually help you but will in all likelihood help you much more than that. If you find yourself actually having whole sentences committed to memory, it's probably an indication that you're spending too much time on one lesson. I'm happy to get several hundred XP from a single skill, but once I start seeing that the whole sentences are in my memory, I take that as a sign it's already past time to move on.
Duolingo talks a little about how and why it does what it does here.
Also, I've never learned Spanish before and I can already notice myself picking pieces of sentences in Spanish that I can understand. I don't only use Duolingo, but it's a major part of my daily routine now with learning Spanish c:
Well, in the process the Portuguese from Spanish course, I found I could pick up the language faster because of the similarities between the languages; however, I did talk quite a bit with friends in the language, watch movies, and expose myself to Portuguese. While there were ambiguities in the course from how to introduce myself, I felt I filled them in with my knowledge of Spanish, changing the Spanish words I would use for introductions to Portugues and in other spaces to fill what I saw as gaps. However, after taking a Portuguese course at my University for two weeks, I recognized some pronunciation gaffs and a unawareness about introductions, important words I thought made sense but didn't, and non-grammar-based Spanish differences.
I think reading the notes under the lessons is helpful (however, looking back at them I don't see cultural awareness or important words I would have wanted to know) , and doing Duolingo alone is not going to lead to strength in every language skill; I think a lot of my vocab, ability to change Spanish words to Portuguese, and my comprehension skills comes from the course. However, this does not completely speak to the order of the course like you asked.
For an individual who has little knowledge of the language (or any second language), I could see how language learning could be a little more difficult; wider or background knowledge can make duolingo a great additional resource to re-enforce a language. But, I see the components of duolingo in and of themselves not to be the end all be all for learning— like I said, I think it is useful for vocab and comprehension, and there is some grammar and listening exercises too, in addition to a little amount of speaking. Because I see that duolingo is already a resource that does not help every language skill evenly, I question if changing the course order will yield largely different outcomes.
But are there studies and resources on language pedagogy? Yeah. In my view of having learned (different amounts of) 3 languages, I prefer a structure based on using the language in practical circumstances (and immersion); for my university language courses, at least. But being as I see duolingo as a resource, and not everything, I take it for what it is to me — a resource for vocabulary, and a little grammar and reading — and utilize other resources to learn, taking into account what I see as gaps.
Elizadeux asks what would be an effective structure, which I think is a good question. And for me it is learning info in a way that I can utilize it. And that is where I see this going (in addition to consulting academic resources, if one wants to do that)