I do agree that “not XXX” often gives away the answer… these type of hints need to be created carefully.
I also test on multiple courses and @dylan.nicholson.548 I understand exactly your predicament.
All said, I liked being forced to learn the synonym, I do believe it is a question of level also.
As for creating carefully, I’ve found adding a parenthese hint such as (nothing to do with blablabla) helps without “giving away” the answer…
What I have now started doing is to add examples, as you find in dictionaries, like this:
att bestämma:
By doing this, I hope that people will learn also learn a useful way to use the word in practice and will also be able to understand what the word actually means and not just its translation.
Cos: I am going to have to print out your suggestions and read them carefully! But I daren’t try it with some of the big courses where I am one of several course contributors, to be honest!
Just noticed that the synonym can show up in the alternative answers (unfortunately)
Maybe that’s one for the Memrise team to resolve.
Either by enabling the course creator to choose proper alternative answers or by writing some code to make sure it doesn’t happen.
Hello,
I have an idea, if you don’t want to add the synonym column, add a column where there is a sentence with blank. This way, the learner will memorise also the word in context.
For example:
few and far between: not very many or not appearing very frequently
once in a blue moon (spoken): almost never
if I read the definitions, I can get confused, but reading the words in context, it is easy to know which word I have to choose, and I also remember the words in context.
He grew up at a time when jobs were few and far between.
Once in a blue moon I’ll have a lemonade with friends, but it’s not my regular drink.
I’m sorry, but all of these ideas are just hacks to get around a serous hole in the platform. While these workarounds may be OK for most scenarios around language learning they absolutely do not work for things like math or when there are multiple words in the target language but only a single word in the source language. In math, 7x3 and 3x7 both have the same answer so now I have to have a database per multiplication table.
But what about if I wanted to create a course for helping people to visualize the chess board in their mind to improve their calculation skills? In fact I do want to do this but I can’t. I’d start with levels that teach the color of the squares by their coordinate.
a1 = dark
a2 = light
b1 = light
e1 = dark
e4 = light
You absolutely cannot do that because the system cannot tell that the answers for a2, b1, and e4 are the exact same thing. There are a number of different ways for Memrise to fix this sort of problem. It is unclear to me why they forces this to be a shortcoming in their platform.
Sorry to rant, it just really frustrates me that I cannot do something that should be both simple and an obvious requirement. (edit to fix typos)
@RobertKaucher, I’m not sure what you mean. If you have the above set of data and have a question that asks “a1”, then if the user types “dark”, it should work fine. Same with “e1”, “dark”, “a2”, “light”, etc.
It would work with typing, but not with multiple choice.
Ah.
(Extra characters to satisfy minimum character post requirement)
And initially you only get the multiple choice questions…
If you use all-typing script you won’t get multiple choice tests
Didn’t check it and it’s probably anyway a theoretical solution only.
I could imagine it would work with 32 databases.
100% agree with Robert, this is cruelly missing from Memrise.
I would personally use this feature to learn the gender of words.
For example, in French:
Ideally an exercise that pops up a word and gives you only 2 choices would be perfect.
So far I haven’t found any good way to practice those.
If anyone knows some interesting resources that can help, please share!
Note that I am learning Hebrew and not interested if the resource is specific to French, I was using it here only as an example because most people don’t read Hebrew characters.
Thanks!
Ah, I hadn’t thought of setting up the databases with an extra column–that’s a great idea!
What I tend to do–and perhaps this is more suited to the kinds of languages I most frequently work with–is use attributes to help eliminate other possibilities. I think a few people below have mentioned what I usually do, which is specifying that the part of speech always appear so that you don’t mix up the noun form and verb form of a Hebrew word, for instance.
What I’m doing for the rather large course I’m putting together now is putting the category of the noun in one of the attribute columns. The book the vocab is collected from is organized by category, so lots of words show up multiple times. Having the subcategory listed under the word will hopefully help people pick the right semantic field. (Apologies if someone else has already mentioned this and I skimmed too fast.)
I used to use this method to disambiguate homophones when being prompted with audio, and it worked great. However, after testing it just now it seems that only the main homophone entry is being used, and anything listed as an “alt” is just marked wrong (whether hidden or not). Even when I put multiple words separated by commas in the homophone field and entered all the words, it was still marked wrong. Do you know what caused this change or if there’s a way around it? (I don’t really know what to search for to see if anyone else has a solution, but this is driving me mad…)