Trouble understanding stats & differences between versions 1.0/2.0


#1

Hi there!

I’ve been studying Chinese actively since the start of 2017 when, coincidentally, I also decided to start using Skritter. Little by little I’ve worked my way up until half-way of HSK 5, and this fabulous app is the only one responsible for it, so pretty happy with the results!

However, aside from some minor bugs every now and then, there seems to be some kind of incongruence between the stats and what I’m supposed to know, as well as between the stats coming from the Legacy 1.0 version and the current one

Now, being currently where I am with HSK5 (and having done HSK1-4 as well as some bits of other lists), I should know around 1500-600 words but as you can see on my stats (here 1.0 and 2.0 seem to agree) I actually only know around 1200 words!

I guess this means that the other 300 i should know, well, I don’t fully know them, specially considering a) how my brain confuses characters that have a “format” similar to 套/参/春,etc. or b) the fact that adding on average 10 new words per day makes fully learning each of them a longer process. Another possibility is that some of the words in the HSK list are actually considered “characters only” by the algorythm, which are then only classified into the 1173 characters the website tells me I know.

Thus I’d like to have an official explanation of why my stats look like this, as well as another one for why the Legacy 1.0 version tells me my retention rate is close to 90% while the 2.0 is closer to 80%, among other differences.

Thank you all for your time and really appreciate your continuos efforts on improving the app!

Have a great day!


#2

I’ve also noticed that each day I study is counted as if I had studied the day before, not the current one.

For example: I studied yesterday, 3rd of July but on the website it appears as if I had studied on 2nd of July. Interestingly, on the calendar on top of the stats page is displayed correctly.


#3

Thanks for the detailed reports! Sorry I didn’t reply to this sooner. We haven’t forgotten about you or the issues you’re bringing up.

We have some issues open internally for stats already, but we do need to dig into things deeper and get to the bottom of how 1.0 and 2.0 are displaying data.

As far as I can tell, thus far, the data is all there. Which is good. What we need to ensure is that it display correctly and consistently on all platforms. We’re still working on the stats page for Skritter Mobile, and we have plans to roll out a website stats update as soon as we can. Hopefully, digging into some stats stuff for mobile will lead to some insights into website issues.

Priority of the fix is a little lower than some of the things we’re currently working on, though, so it might be a little bit of time till website (2.0) updates are tested and rolled out.

Again, thank you for the feedback. All of the things you’re talking about are going to be a great starting point for us.

-Jake


#4

Perhaps it’s a simple time zone difference between where the data is stored and user’s local time?

Studying in Beijing at 10am would clock-in on a North American ET server as 10pm the prior day. (And Skritter still considers study up til 3am or so “the previous day?)


#5

I have that same problem with the stats on the website – e.g. it shows my current streak as zero even though I studied today, because it presumably is assigning today’s time to yesterday. Since I’m already giving feedback on this, the longest streak also is capped at 31 (presumably because it seems to reset each month), which is a bit demotivating.

I’m hoping to see a rocking stats page on mobile that resolves all these issues. :slight_smile:


#6

Here is what we have cooked up for Mobile stats screens. V1 might not look quite like this, but we’re aiming for this style for release. And once we have things cleaned up between all the apps we’ll continue to add and improve on it. And we’ll continue to update the website as well to clean progress stats up a bit more and make sure it is clear how much stuff everyone is learning on Skritter!


#7

Awesome! Looks great.

My one request would be to bump up # characters learned to the top card, alongside # words learned and instead of total time spent. Total time spent is a fun metric but we goal (and pride) ourselves on number of words and characters learned, not total time spent.


#8

Glad to hear I’m not the only one facing this problems! :grinning:

And since we are requesting, I’d totally love to see a proper mention in the new stats page regarding how many characters/words have we added in total. It’s already in the old stats, but that one doesn’t seem to be that precise though.

Jake, many thanks for taking some time off your other duties to reply tp me! I know you guys currently have a lot on your plate :joy:


#9

This would probably be a good thing to include. I’ll have to discuss with @josh and @SkritterMichael to make sure we can fetch this data from the API and serve it on the website and mobile apps.


#10

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.