319 messages over 40 pages: << Previous 1 2 3 4 5 6 7 ... 33 ... 39 40 Next >>
s_allard Triglot Senior Member Canada Joined 5443 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 257 of 319 26 April 2014 at 1:39pm | IP Logged |
luke wrote:
s_allard wrote:
Ezy Ryder wrote:
Doesn't an average mean that the number of words
was used only in an average single exam? How do you know what's the common part? And what's the total
vocabulary used?
An example I could think of: I recently used a text analysis software to generate a lemma frequency dictionary
based on a book in Japanese. The result surprised me, mere ~3,000 lemmas (not word families, lemmas)
throughout ~200 pages. However, running the software on the whole series at once (8 books in total, all similar
in length), the number of lemmas used rose to ~13,000. |
|
|
This is a the very issue here. Any one book uses 3000 "words" but the 8 books use an aggregate of 13000. How
many words do you need to read a book? 3000. How many do you need to read the 8 books? 13000. All of this
is very true. But let's say that you are writing a book. How many words do you use? Not 13000; it's more like
3000. The same with our IELTS candidates. If my understanding is correct, whilst the average individual
vocabulary size is 1490. the sum total of different words of all the successful candidates is undoubtedly higher.
|
|
|
A single book or the words used in a single oral proficiency exam use only a subset of that individual's word
stock. Just because they only used 1490 words or 3000 lemmas in a particular performance doesn't mean they
exhausted their ability to use words. The nature of speech and writing in communication is that le mot just
is not a small, restricted subset in the universe of words. |
|
|
This is very true, and I have said so in the past. But the question is: how many words do you need to pass the
IELTS Band 8 test or to write a book? Nation and Head found that students needed a mean of 1490 words to
pass the test; they did not say that the students knew only these words. As a matter of fact, I don't think that
there were any figures for the standard deviation, but we can assume that some students got by with even fewer
words and some with more.
Of course we can ask: how many words do the students actually know or are able to use? That's a very valid
question, but that's what all the other studies have attempted to estimate. But it's irrelevant here because we are
looking at what was actually used.
To come back to our author analogy, we see something similar. I'm a great fan of P.G. Wodehouse. I have no
idea of how many distinct words he has used in any given novel. Let's say 4000. He wrote 97 books. They are
not all novels, of course, but for our purposes, we use them to calculate P.G. Wodehouse's vocabulary size.
Fans of this author know that his novels are essentially the same. Could it be possible that all these novels use
only 4500 words? It's possible and even likely.
But this has nothing to do with how many words P. G. Wodehouse knew or could recognize. And I should also
add that measuring P.G. Wodehouse's vocabulary would be an exercise in navel-gazing.
Much of this questioning of vocabulary size is related to the question of how to prepare for these tests. Should
you set up an Anki deck of 1500 words for the Band 8 test and drill these to death? No, but I would certainly
concentrate on being truly fluent and accurate with that very high frequency core that will make up 50-75% of
your speaking. This is what drives fluency.
But the main point of all this is that you don't need 5000 or 10000 words to pass that test.
Edited by s_allard on 26 April 2014 at 2:17pm
1 person has voted this message useful
| s_allard Triglot Senior Member Canada Joined 5443 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 258 of 319 26 April 2014 at 1:55pm | IP Logged |
Iversen wrote:
s_allard wrote:
To look at this issue from the opposite perspective, could you have a great
vocabulary and not be able to carry on a basic conversation with a native speaker? This is not as farfetched as it
sounds. I see many examples of academics who can read and even write relatively good French but are unable on
a basic level in French. The truth of course is that speaking the language requires a set of skills that are frankly
quite different from reading and writing. I don't want to go into this subject right now but I think everybody gets
the point. |
|
|
That's certainly possible. Last time I visited Russia (as far back as 2008) I could only speak at the 'prepared
phrases' level, and I haven't said more than a couple of sentences since then, several years later. Besides I rarely
hear spoken Russian because I haven't got a Russian TV channel, but I can read and write and to some extent
even think in Russian. If I knew I had to speak Russian in a couple of months time I would definitely do serious
something about it, but unless I buy a trip to Russia that is not likely to happen. So right now my spoken Russian
sucks to a degree where I would be hesitant on speaking it in public. Nevertheless my learn-any-language.com/forum/forum_posts.asp?TID=12983&PN=1& TPN=428">word counts from 2009
and 2013 hovered around 10.000 words (the number 14.400 from 2009 is probably wrong - I based it on a
published number of words for the dictionary, but my own estimate from 2013 is much lower). I have worked on
several Slavic languages here in 2014 so my results would probably be higher now, and some of my work with
Southern Slavic languages due to the conference in Novi Sad and my new TV channels in Polish and Croatian may
also spill over to Russian, but right now my Russian is still mainly a written language in the same category as
Latin. |
|
|
Iversen has said it quite well; I barely have to add anything. My original post was it bit garbled but the idea was
that people, not only academics, can have knowledge of a large number of words in a target language and not be
able to use them in simple fluent conversations.
Actually, I'm sure this has happened to most of us here with our various languages. We can read books and
newspapers without a problem. But when we meet a native speaker on a bus, can we chat away just like any other
native speaker? Most likely not.
I don't believe it has anything to do with social anxiety. The real problem is that we have often never really
practiced true conversational interaction in our target language. That 5000-word vocabulary is of no great use if
your pronunciation is so-so and you've never mastered all the little techniques and basic words that make up
conversational interaction. Something as simple as talking about the weather and the high cost of living can be
very challenging.
2 persons have voted this message useful
| daegga Tetraglot Senior Member Austria lang-8.com/553301 Joined 4534 days ago 1076 posts - 1792 votes Speaks: German*, EnglishC2, Swedish, Norwegian Studies: Danish, French, Finnish, Icelandic
| Message 259 of 319 26 April 2014 at 3:27pm | IP Logged |
s_allard wrote:
But the main point of all this is that you don't need 5000 or 10000 words to pass that
test. |
|
|
But this was only for the writing part, wasn't it? Good luck passing the listening part
with a vocabulary of only 1500 words. It's hard enough to do with all words known...
vaguely recognizing some more words might be sufficient for the reading part, but the
listening part is more challenging when it comes to the automaticity of recognition and
response time
edit:
It was the speaking part, not the writing part.
It's funny how nobody in this discussion about these 1500 words seems to have read the
linked article (me included), because the number is wrong. Band 8 speakers used 1491
tokens on average, which only tells us about the length of the utterance, not about the
vocabulary used. The mean number of types (=unique tokens) was 408.1. I hope nobody
here will claim that you can get a band 8 with an active vocabulary of only 408 lemmas.
More interestingly, the total number of types used by all 15 test takers together was
2374. This number seems to be a bit more realistic for active speaking vocabulary, the
demands in writing are usually higher in terms of vocabulary range. Also, the questions
could be grouped into only 4 different topics, and naming them task 70, 78, 79 and 80
respectively would suggest that there are a lot more topics used in IELTS exams.
edit2:
It seems like the words weren't lemmatized, so the type count is actually higher than
the number of lemmas, which makes the specific numbers even less suitable for a serious
discussion.
Edited by daegga on 26 April 2014 at 6:35pm
3 persons have voted this message useful
| Serpent Octoglot Senior Member Russian Federation serpent-849.livejour Joined 6610 days ago 9753 posts - 15779 votes 4 sounds Speaks: Russian*, English, FinnishC1, Latin, German, Italian, Spanish, Portuguese Studies: Danish, Romanian, Polish, Belarusian, Ukrainian, Croatian, Slovenian, Catalan, Czech, Galician, Dutch, Swedish
| Message 260 of 319 26 April 2014 at 3:48pm | IP Logged |
Yeah, some exam makers seem to think we have tape recorders in our heads. You don't know which details will be significant later on. Or which insignificant details you'll be asked about.
And while we're discussing mostly active vocabulary here, and passive vocabulary usually includes the words you *potentially* can understand... for the listening tasks you need a truly automatic command of the passive vocabulary, and the mere seconds of hesitation can be crucial when you're required not only to understand, but to answer questions/explain/etc.
1 person has voted this message useful
| s_allard Triglot Senior Member Canada Joined 5443 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 261 of 319 26 April 2014 at 10:32pm | IP Logged |
daegga wrote:
...
edit:
It was the speaking part, not the writing part.
It's funny how nobody in this discussion about these 1500 words seems to have read the
linked article (me included), because the number is wrong. Band 8 speakers used 1491
tokens on average, which only tells us about the length of the utterance, not about the
vocabulary used. The mean number of types (=unique tokens) was 408.1. I hope nobody
here will claim that you can get a band 8 with an active vocabulary of only 408 lemmas.
More interestingly, the total number of types used by all 15 test takers together was
2374. This number seems to be a bit more realistic for active speaking vocabulary, the
demands in writing are usually higher in terms of vocabulary range. Also, the questions
could be grouped into only 4 different topics, and naming them task 70, 78, 79 and 80
respectively would suggest that there are a lot more topics used in IELTS exams.
edit2:
It seems like the words weren't lemmatized, so the type count is actually higher than
the number of lemmas, which makes the specific numbers even less suitable for a serious
discussion. |
|
|
I think this observation is correct. The interpretation of the figures is even more intriguing. Instead of 1490
different words, we have 408 types or different words. There figures are even more interesting. If you take all the
different words or types for the 15 candidates, we get 2374, which is not a particularly high figure. Just exactly
what does an average of 408 types or words mean?
I'm at a loss here. Are the authors saying that the successful candidates used an average of 408 different words
out of an average number of 1490 words. Did these candidates pass the exam with less than 500 different
words, yes or no?
1 person has voted this message useful
| daegga Tetraglot Senior Member Austria lang-8.com/553301 Joined 4534 days ago 1076 posts - 1792 votes Speaks: German*, EnglishC2, Swedish, Norwegian Studies: Danish, French, Finnish, Icelandic
| Message 262 of 319 26 April 2014 at 10:54pm | IP Logged |
s_allard wrote:
I'm at a loss here. Are the authors saying that the successful candidates used an
average of 408 different words
out of an average number of 1490 words. Did these candidates pass the exam with less
than 500 different
words, yes or no? |
|
|
Yes they did. But one can't draw any conclusion from this.
Let's look at the writing task. It's a 150 word description of a graph plus a 250 word
essay, makes a total of 400 words. You can't possibly use more than 400 different words
in this task if you only write a total of 400 words. The only way this number would
suffice would be if you knew the task beforehand and could prepare your response. But
you still don't know how many words you need to know in order to pass a random
question, because the vocabulary requirement in such a short task is very topic
specific.
2 persons have voted this message useful
| s_allard Triglot Senior Member Canada Joined 5443 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 263 of 319 26 April 2014 at 11:10pm | IP Logged |
daegga wrote:
s_allard wrote:
I'm at a loss here. Are the authors saying that the successful candidates used an
average of 408 different words
out of an average number of 1490 words. Did these candidates pass the exam with less
than 500 different
words, yes or no? |
|
|
Yes they did. But one can't draw any conclusion from this.
Let's look at the writing task. It's a 150 word description of a graph plus a 250 word
essay, makes a total of 400 words. You can't possibly use more than 400 different words
in this task if you only write a total of 400 words. The only way this number would
suffice would be if you knew the task beforehand and could prepare your response. But
you still don't know how many words you need to know in order to pass a random
question, because the vocabulary requirement in such a short task is very topic
specific. |
|
|
We're not talking about the writing test. We're talking about the speaking test. If I recall correctly, it consists of a
15-minute interview. There is a limit to the number of words (tokens) what can say in that time. What the figures
seem to say is that 408 different words are all you need to pass this test.
Let's make the test 60 minutes. How many more different words would be used? Let's assume four times. That
gives us 1632. It's still a low figure.
Just what can we conclude from the fact that the successful candidates used only 408 different words or types in
the oral proficiency test?
1 person has voted this message useful
| s_allard Triglot Senior Member Canada Joined 5443 days ago 2704 posts - 5425 votes Speaks: French*, English, Spanish Studies: Polish
| Message 264 of 319 26 April 2014 at 11:32pm | IP Logged |
While we are trying to figure out the meaning of those vocabulary size figures, here is a something on the
problem of counting words and the problem of multiple meanings. I'm currently reading the 1909 novel, Mike At
Wrykin, by P. G. Wodehouse. I'm about to give up because the story is so steeped in cricket lore and public -- or
private for us North Americans - school culture that I find much of the story nearly incomprehensible. Here is
something from the third page into the book.
"Hooray! Mike's going to Wrykin. I bet he gets into the first eleven his first term."
"Considering there are eight old colours left," said Bob loftily, "besides heaps of last year's seconds, it's hardly
likely that a kid like Mike'll get a look in. He might get his third, if he sweats."
I know vaguely that they are talking about cricket teams at Wrykin but beyond that I really don't understand what
this is all about. None of the words are particularly new to me but are they really part of my vocabulary?
Edited by s_allard on 26 April 2014 at 11:43pm
2 persons have voted this message useful
|
You cannot post new topics in this forum - You cannot reply to topics in this forum - You cannot delete your posts in this forum You cannot edit your posts in this forum - You cannot create polls in this forum - You cannot vote in polls in this forum
This page was generated in 0.5469 seconds.
DHTML Menu By Milonic JavaScript
|