I'm not quite sure that I understand how the words/characters learned are calculated. It seems to me that if you add characters when adding words, 你好 would count as 2 characters (你 and 好) and 3 words (你, 好, and 你好). If this is true, then the words learned should be much higher than the characters learned. However, my characters learned is higher than the words.

From this discussion
http://www.skritter.com/forum/topic?id=19825611&comments=9
it sounds like adding 你好 would count as only 2 characters (你 and 好) and 1 word (你好).

Can anyone clarify which is the way it works?

Individual characters are not counted as words, ever. Words are everything 2+.

So, 你好 is two characters and one word.

Most individual characters are words. for example 电 means electricity and is a normal word it can be used with out adding an other character. Together with an other character it is a different word like 电话 for phone, 电视 for tv, 电影for movie and 电脑 for computer. all different words but starting with the same character

If you say 1 character is not a word but two characters is 你 wouldn't be a word but 你们 would be. that's a little strange.

Myrna: I meant, as far as Skritter is concerned, individual characters are not counted as words.

It would be troublesome to implement. How would you know which are words, and which are not? It's not a black and white division, and someone would have to sit and go through thousands of characters to decide this. And, in terms of statistics, it doesn't really matter that much, surely?

Myrna

But still, at least in speaking (though, I can tell about Taiwan only), when I just started to learn, I tried use individual characters I learned in conversations - most of the cases it appeared that one more should be added and then people can understand. In case of 電 - if you would say, 我的家沒有電 - you could be immediately misunderstood. I would say 電力 instead of just 電. Above is just my observation, it always was not successful trying to use single character, which meaning I memorized.

Right, modern Chinese tends to use two-character words more often than not. From a linguistic perspective there are many characters that function as words but to keep the Skritter definition simple, words are multiple characters (strictly words shouldn't include phrases either but Skritter counts these as words).

Any suggestions on how to explain this more easily and/or illustrate this more clearly on the stats page would be greatly appreciated.

So if you want to see how many real words you know, can’t you just add the words learned and the characters learned together?

Sure. The main purpose of dividing up words and characters is so that you can know how many characters you know. That way if you study two words that share a common character, that character doesn't get double counted. But for words in general, just adding those two numbers together does a good job of telling you how many words you know.

Note that Chinese and Japanese do this differently. In Japanese, single character words are counted both as a word and a character, so if you learn the word 水, it counts it as learning a word and a character. This is so that we could have a separate vocab for storing abstract meanings and on and kun readings. What that means is in Japanese, just use the words learned value to count how many words you know, don't add the characters learned value to it.

I think that you can do as JimAndress proposes, but that it only works for definitions. For writing, reading and tones, it only works when for each "word" (with more than one character), you also add the individual character components for study.

I don´t usually do this (yet). For me, I "know" the writing of 578 characters, but only 310 character definitions. So I figure, if I add the number of definitions learned for characters and "words" (310 + 445), I get an approximate estimation of "real" words I know.

This can be a bit confusing.

On the other hand, it is probably impossible to keep track of every word you know in a certain language. For Chinese, wouldn´t the most interesting statistic in fact be characters? Because this is the figure that is usually related to in statements such as "In order to read a Chinese newspaper you have to know so-and-so many characters".