When will I be able to write a conversation in Swedish?

I've started to take Duolingo Swedish lessons in January. I don't practice daily because of the work I do and sometimes I don't practice for weeks (which is a shame, but it's just the way it is).

I reached Level 8 and I often repeat lessons - especially the verbs because I think those are really important to know.

As I don't have anyone to write in Swedish with I really can't tell how good I am. But I am pretty sure that I had to look up sooo many words if I wanted to.

Tell me about your experiences!

Any tips?

April 5, 2016


I've been practicing since January or February I think. I'm at 32% Swedish fluency now as I practice every day and as thoroughly as I can, but I can tell you that I still have to go a long way before I can hold a detailed ongoing conversation for longer than a few sentences. I'd even go as far and say that you're probably still going to have problems with somewhat advanced conversations past the basics (stuff like talking about how you are or the weather), even when you're past 50%.

Learning a language is certainly not easy, 50% for example might sound like a lot, but if you think about it, it really isn't. Technically, it means you can hold half a conversation (of course not literally, it always depends on what and you're talking about and who you're talking to). Learning a new language takes a ton of practice and learning and is not something that can be done within a month or two. Just be patient and keep at it, you'll see yourself improve over time, even though it might take a few months! :)

My personal advice would be to practice as regularly as possible, even if it's just one lesson or revision on the train/bus/etc., or when you have some spare time left. I don't know what you work as, so I don't know how much free time you actually have, but as little as 10 minutes per day can really help! Not being active for a few weeks can throw you back and learning and practicing more regularly helps your brain memorizing structures, grammar and vocabulary. :)


April 5, 2016

Thank you for your long and interesting answer. You are right, regular practice helps. I guess another reason is that I'm better at Spanish so I tend to do Spanish lessons because they are just easier for me and make me feel more confident whereas in Swedish I have to work much harder to achieve the same goal. I just try to be patient. After all, I don't have a deadline set until which I have to learn the language, so I'm quite free.

April 6, 2016

Are those 32% a number that Duolingo reports to you, or something that you have computed yourself somehow? I started in December but am not seeing such a number, unlike for French.

April 5, 2016

It's the number that Duolingo reports to me. Here, take a look at this screenshot, it informs me about my progress on the right:

April 5, 2016


April 5, 2016

Wow! I don't have it((( ok Duolingo guys, stop your A/B testing and give this feature to everyone!!! Or at least to me!)))

April 9, 2016

Thank you for the screenshot! You must have slipped into a different category for their A/B testing, or somesuch. Though how one would A/B test a read-only feature I am at a loss to explain.

April 5, 2016

A/B testing? I'm confused, what's that?

April 6, 2016

It's a randomised trial with two options, used to determine (a) whether the choice of option affects some outcome and (b) in what way the options affect the outcome.

For example, let's say that you're trying to sell lemonade. Your basic recipe calls for 100g of sugar, but you suspect that your friends might like it better if you have more sugar. So you create an alternative option in which you add 150g of sugar to your lemonade. You then produce two batches of lemonade, A and B. You then pour the lemonade into cups, put stickers at the bottom of each cup, and label the stickers with A and B. You cover the stickers so that your friends can't see them. Now (and this is important) you RANDOMLY hand the cups to your friends; for example, you flip a coin for each friend to determine whether they get an A cup of a B cup. (That's what makes it a randomised trial.) You then make notes: whenever a friend asks you for a second cup, you check the sticker and record a "plus" for that type. Alternatively (or in addition), you collect all cups and check which cups had at least 25% of the drink still in them, and record a "minus" for that.

If you do that enough, you'll collect statistics on which version of your lemonade is more popular with your friends, and you can use that knowledge to give them more popular lemonade.

Most major web sites, including Duolingo, make heavy use of A/B testing (Google used to run dozens of such experiments at the same time; they probably still do). The idea is slightly more sophisticated, though: they have a "default" version of the web site (A), and a `modified' version (B). Unlike with the lemonade, the B version is often shown to only a small fraction of the visitors. Duolingo then counts whether people who get the B version have substantially different interaction behaviour than people with the A version (e.g., whether they visit more frequently, do more exercises, gain more XP etc.). The Duolingo managers then use this data to decide whether the changes from version B should be rolled out to all or at least more visitors.

Of course, now that you know that you're part of a "B" bin AND know how A/B testing works, this knowledge actually influences the test. Scientifically speaking, this is a problem, but Duolingo will probably not be quite so particular about being 100% scientifically pure. ;-)

April 6, 2016

Woah, thanks for the detailed answer! I didn't know about that, that sounds pretty cool. Actually, I can't say if my interaction with the site would be any different if the percentage counter wasn't there, but I can say for sure that it really adds to the motivation to learn. You know, doing one or two extra lessons to get that one extra percent. ;)

Maybe I should ask some friends if they also have the Counter or not, you actually caught my interest now, haha!

April 7, 2016

For writing maybe you should try the website lang-8. There you can write a text in Swedish and afterwards native speakers will correct what you have written. In exchange you should correct other users' posts in your native language. I think it's a very good tool because native speakers can easily correct your errors and also make the sentences you have written sound more natural.

April 5, 2016

That sounds really cool, I'll have to try that out, too. Thanks!

April 5, 2016

Writing a conversation isn't too hard once you finish the tree. Speaking one, on the other hand..

April 5, 2016

Holding a spoken conversation with another person is something that you can't learn from an online program. You just have to do it to become more comfortable with it, but that counts for every language you're learning. It's good if you have friends who are native speakers, so you can practice with somebody to improve your speaking abilities.

April 6, 2016

It will be a loooong way until I finish my tree. But nevertheless, you are right. By then I will have learned a lot of vocabulary and grammar to be able to write

April 6, 2016
Learn Swedish in just 5 minutes a day. For free.