Pages in topic:   [1 2] >
The 'confidence level' in KudoZ answers
Thread poster: xxxLia Fail
xxxLia Fail  Identity Verified
Spain
Local time: 08:56
Spanish to English
+ ...
May 19, 2004

I really don't see the point of one rating one's own confidence in one's own answer on a scale of up to 5. It's entirely subjective.

I wouldn't even mention it, but it's mandatory, I am obliged to choose an option every time I answer a question, and I dislike it.

I think it's for the asker to make decisions on the validity of the answers, by matching them to their text/context, and on the basis of a 'picture' of the individual answering (their areas of expertise, native/non-native - especially for language questions - and supporting evidence, etc.)

Maybe someone can persuade me of its worth?


Direct link Reply with quote
 

Dinny  Identity Verified
Greece
Local time: 09:56
Italian to Danish
+ ...
I think it's just fine! ;-) May 19, 2004

Sometimes... when I have time ... I try to answer some Kudoz questions. I might spend a lot of time researching, and might even be totally sure that I have found the correct answer. If so, I will grant myself a confidence level of, say, 4.

Other times, I bumb into a question to which I THINK I know the answer, but I don't have the necessary time to check all available sources to confirm my answer. Still though, whoever asks, is in need of help. And to indicate that I haven't done all the necessary homework but still would like to help, I indicate my confidence level as 2 or 3. My answer turns out to be a "suggestion", might be a confirmation to the asker like "just-the-word-that-I-knew-I-knew" ...

Briefly? I think it's fine that by answering one gets the possibility to indicate the "confidence level"... it's a warning or it's a guarantee... the decision is, however, totally up to the asker!



Direct link Reply with quote
 
Rosa Maria Duenas Rios  Identity Verified
Local time: 02:56
It helps determine how confident the answerer is of his/her answer May 19, 2004

Ailish Maher wrote:
I really don't see the point of one rating one's own confidence in one's own answer on a scale of up to 5. It's entirely subjective.

I wouldn't even mention it, but it's mandatory, I am obliged to choose an option every time I answer a question, and I dislike it.

I think it's for the asker to make decisions on the validity of the answers, by matching them to their text/context, and on the basis of a 'picture' of the individual answering (their areas of expertise, native/non-native - especially for language questions - and supporting evidence, etc.)

Maybe someone can persuade me of its worth?


Hi Ailish,
I am not sure I can persuade you, but here's my view:
Real simple when receiving answers to questions I asked: the level indicated by the answerer helps me determine how certain they are of their suggestions. The scale helps me determine my own level of trust. Thsi makes me think that the system was thought to help the asker, ratther than the answerer.

When the answerer is me, I never use 5, because I would never dare affirm that I am absolutely sure that my suggestion is THE correct one, as most of the times there is more than one correct possibility. Here, I have to confess that I instinctively distrust colleagues who repeatedly rate their answers with 5, especially when they provide no background information to support their recommendations.
I find myself using 4 most of the time, since I have a solid confidence in the answers I give. I tend to use 3 when not enough context is provided and thus I do not feel as confident in my suggestion, or when I have not done enough research to justify my answer. The times when I have been tempted to use 2, I have decided not to post my suggestion at all, since I am sure there are many other colleagues that can provide a better answer. The one I consider quite useless is 1 (I am guessing), and it probably should not be an option, since I believe that the last thing a colleague needs when looking for a term, is more guess work from others!

So, in a nutshell, I find it useful for the asker.
If it is not as useful to the answerer, the scale -at least in my case- makes one stop and reflect on the pertinence of what one is about to write, which has direct impact on the quality of the answers. If other colleagues go through something similar to this last process, then I hope we are all contributing to maintaining a good quality in Proz's Kudos.
Convincing enough for you?

[Edited at 2004-05-20 03:05]


Direct link Reply with quote
 

Sol  Identity Verified
United States
Local time: 02:56
Spanish to English
+ ...
I don't mind grading myself... I do mind other's selfgrading May 19, 2004

when they give themselves a 5 without even taking the time to read the complete question, and much less to get it right.

This is especially disturbing when the question asked clearly marked PRO AND TECHNICAL by the asker. I'll give you a theoretical example: Say someone ones to know how to say RUN, and as subject he/she indicates Technical, and in context he/she transcribes a couple of sentences from a very technical manual that talks about running machines. There is always one or two answerers who jump and answer, I believe by pressing the choice to "Answer" from the list of questions, and give an answer without giving any thought or reading the answers of others. In this case they would give the translation of what you do when with your legs when you are in a hurry to get somewhere. And on top of that, they give themselves a confidence of 5, they are SO SURE they are right, they don't even read the question.

Which brings me to the way the list of questions appears in the new format. Now you can just press "Answer" from the list itself, without reading or seeing any details. I don't think there should be any way to answer a question without reading first the other answers.

Back to your question about the need of having 5 choices to grade ourselves, well, maybe 3 would do. One for the experts in that particular subject, one for those who think they know and have a good explanation (most answers), and one for the guesses.


[Edited at 2004-05-20 03:19]


Direct link Reply with quote
 

swisstell
Italy
Local time: 08:56
German to English
+ ...
5 points to oneself shows a bit of arrogance May 20, 2004

It is good to be confident but there are certain translators that invariably give themselves a 5. I consider this arrogant - most of the time - because usually there is more than one solution to a problem, more than one answer to a question, (or, as the saying goes, more than one way to skin a cat).

Direct link Reply with quote
 

Kim Metzger  Identity Verified
Mexico
Local time: 01:56
German to English
The confidence bar May 20, 2004

Ailish Maher wrote:

I think it's for the asker to make decisions on the validity of the answers, by matching them to their text/context, and on the basis of a 'picture' of the individual answering (their areas of expertise, native/non-native - especially for language questions - and supporting evidence, etc.)



Hi Ailish, interesting topic. I agree that it's primarily the asker's responsibility to choose the best answer. It's true, though, that a lot of askers wouldn't have any idea how to check the asker's qualifications or even know what supporting evidence is. Many don't know how to provide enough context and explanations to get a good answer. If the asker is clueless, so be it.

But why do you and others think the confidence bar was implemented in the first place? I think it became a new feature about two years after KudoZ was born. I have my own ideas on why it was introduced, but would like to hear from others.

[Edited at 2004-05-20 00:33]


Direct link Reply with quote
 

Andrea Ali  Identity Verified
Argentina
Local time: 03:56
Member (2003)
English to Spanish
+ ...
Nothing wrong (to me) May 20, 2004

Hi, Ailish!

I don't see anything wrong with grading your own confidence. I hardly ever pay attention to the answerer's confidence but concentrate on the suggestion that better fits my context.
As Dinny said, I sometimes don't have the time to "do my homework" but am pretty sure about the answer. One example: liver cell. I provided the answer (célula hepática) without giving any reference and I think my level of confidence was 5.
The rest is done by peers who will agree/disagree or provide other/better answers. Then the asker chooses the answer he/she likes best, considers the most appropriate for the context, or whatever. I think there's nothing more to it.

My point of view

Cheers,
Andrea


Direct link Reply with quote
 

Mats Wiman  Identity Verified
Sweden
Local time: 08:56
Member (2000)
German to Swedish
+ ...

MODERATOR
Confidence level is only ONE indicator May 20, 2004

In most cases the asker gives an unsatisfactory context, so the answerer can only make a suggestion, not an final authoritative answer.
In some cases the context is perfectly clear, the consensus in that industry/that area of life/that dialect is strong and clear. In such a case - provided you are familiar with these circles - I do not find it at all arrogant if you tell the asker that you KNOW that your offered answer is right.
Of course you can be wrong and of course you can have misunderstood something but you offer your SUBJECTIVE assessment of the relevance of your answer.
You might (as well as might not) have the time to offer supporting evidence in the form of dictionary/glossary quotations, net search results or even KOG references.

As before, and as must be the case, the asker decision is a composite one consisting of:

1. Who answers? (do I know the relevance of his/her answers?)
2. What dictionaries are quoted? (well-known, good ones or unknown dubious ones?)
3. Proof of existence in the context given (if the term has 13452 hits and most seem to be found in the field of hydraulics, that's pretty convincing)
4. If the asker gives a 5 confidence level, this is also an indicator. It has of course to be correlated with the other indicators.
5. Finally, the peer verdicts are important. Important is not only IF they agree or disagree, but what arguments are offered (by disagree they should be).

Confidence level might not be the strongest indicator, but I see no reason to have it removed. In the earlier days of ProZ.com we more or less generated the solution by telling the asker convincing facts like "I know that Volvo calls it XXX because I worked 4 years in their hydraulics department".
When I was more or less guessing I wrote "This is a WGD delivery" (WGD = Wild Guess department).

So instead of these comments the Confidence Level was instituted.

In my view the Confidence Level icon should be VERTICAL instead of horisontal (Henry?)

but that is another matter.

BR

Mats J C Wiman
Übersetzer/Translator/Traducteur/Traductor > swe
http://www.MatsWiman.com
http://www.Deutsch-Schwedisch.com
http://www.proz.com/pro/1749
(Proz.com moderator, deu>swe, Swedish)
Träsk 201
SE-872 97 Skog
Schweden/Sweden/Suède/Suecia
Tel:+46-612-54112 Fax:+46-612-54181 Mobile:+46-70-5769797

[Edited at 2004-05-20 05:10]


Direct link Reply with quote
 
xxxtazdog
Spain
Local time: 08:56
Spanish to English
+ ...
it's meaningless May 20, 2004

I find the confidence level meaningless. I regularly answer Kudoz questions and although I occasionally put 5, I usually choose 4, even when I am 99% sure I've given an absolutely correct answer to a question. However, there are quite a few people who have no qualms about giving themselves a 5, even when their answers are so far off the mark they shouldn't even be on the same page.

As an asker, I pay no attention to the confidence level whatsoever: I am much more impressed by references, and even if no references are given, I check out proposed answers on my own to see which one best suits my context, regardless of the confidence level.

Bottom line: I agree with Ailish, I don't like it either and I don't think it serves any purpose.


Direct link Reply with quote
 
xxxLia Fail  Identity Verified
Spain
Local time: 08:56
Spanish to English
+ ...
TOPIC STARTER
What bothers me most is that it's mandatory (otherwise I could ignore it) May 20, 2004

[quote]Cindy Chadd wrote:

I find the confidence level meaningless. I regularly answer Kudoz questions and although I occasionally put 5, I usually choose 4, even when I am 99% sure I've given an absolutely correct answer to a question. However, there are quite a few people who have no qualms about giving themselves a 5, even when their answers are so far off the mark they shouldn't even be on the same page.




Precisely: First there is the dilemma of what to rate yourself, even as an ex-English teacher and native English speaker, I still hesitate to put 5 for simple grammar questions. I feel it would be arrogant (as someone else pointed out).

The other problem is that some people give themselves 5 when they clearly haven't a clue or when their judgement is doubtful. That bothers me, becuase I think answerers have a responsibility to askers. OK, askers should know what they're about, but sometimes it's a question of the blind leading the blind...we have all come across something we can't make head or tail of and we rely on coming across someone who seems to know what they are talking about.

OK, so we can just ignore it? Well no we can't, 'cos it's mandatory. I just wish it was optional, that's all. I am not interested in rating myself, and I ignore other people's subjective ratings. What I do take note of is the 'agrees' (with circumspection, too, obviously).


Direct link Reply with quote
 

Andrea Ali  Identity Verified
Argentina
Local time: 03:56
Member (2003)
English to Spanish
+ ...
Optional is a good idea May 20, 2004

Ailish Maher wrote:

OK, so we can just ignore it? Well no we can't, 'cos it's mandatory. I just wish it was optional, that's all. I am not interested in rating myself, and I ignore other people's subjective ratings. What I do take note of is the 'agrees' (with circumspection, too, obviously).



to keep every body happy


Direct link Reply with quote
 

Mats Wiman  Identity Verified
Sweden
Local time: 08:56
Member (2000)
German to Swedish
+ ...

MODERATOR
Please say YES to more information May 21, 2004

Cindy Chadd wrote:I find the confidence level meaningless. I regularly answer Kudoz questions and although I occasionally put 5, I usually choose 4, even when I am 99% sure I've given an absolutely correct answer to a question.

Good.
However, there are quite a few people who have no qualms about giving themselves a 5, even when their answers are so far off the mark they shouldn't even be on the same page.

thereby offering another paramater being used to judge the answerer.
You do attach some importance to it if a solid respected answerer mark his/her answer with a 5 don't you?
As with all aspects of the KudoZ activity you likewise attach more or less importance to some answerers (e.g. those "having no qualms about giving themselves a 5, even when...")
AND: You do not GIVE yourself a 5. You INDICATE how certain you are (as honestly as you can)
Ailish Maher wrote:
First there is the dilemma of what to rate yourself, even as an ex-English teacher and native English speaker, I still hesitate to put 5 for simple grammar questions. I feel it would be arrogant (as someone else pointed out).

You do NOT RATE yourself. You INDICATE how certain you are (as honestly as you can)
The other problem is that some people give themselves 5 when they clearly haven't a clue or when their judgement is doubtful.

As with all KudoZ answers, nothing is definite, authorized or necessarily right. They are SUGGESTIONS to be evaluated.
I think answerers have a responsibility to askers.

As with all KudoZ answers, nothing is definite, authorized or necessarily right. They are SUGGESTIONS to be evaluated.
OK, askers should know what they're about, but sometimes it's a question of the blind leading the blind...

As with all KudoZ answers, nothing is definite, authorized or necessarily right. They are SUGGESTIONS to be evaluated.
we have all come across something we can't make head or tail of and we rely on coming across someone who seems to know what they are talking about.

Exactly that is what KudoZ is all about. It's NOT an exam system.

As with all KudoZ answers, nothing is definite, authorized or necessarily right. They are SUGGESTIONS to be evaluated.

Mats


Direct link Reply with quote
 

Parrot  Identity Verified
Spain
Local time: 08:56
Member (2002)
Spanish to English
+ ...
Peer validation May 21, 2004

Agree/disagree affects the confidence rating. A 5-3 tells you something about the answer, as does a 2+3.

Direct link Reply with quote
 

Andy Watkinson
Spain
Local time: 08:56
Member
Catalan to English
+ ...
Reliability May 21, 2004

Parrot wrote:

Agree/disagree affects the confidence rating. A 5-3 tells you something about the answer, as does a 2+3.


Didn't the idea of the confidence rating have something to do with the answerer's reliability?

I seem to remember that there was some weird formula which, taking into account the answerer's confidence and proportion of answers chosen out of all those given by the answerer, would produce a "reliability ratio".
I also seem to remember that one of its aims was to reduce the number of people throwing out thousands of answers, scattershot fashion, and gaining points that way.

An answerer's reliability is a much more meaningful factor than simply the number of points accumulated.

20 points earned by someone who has answered 5 questions is far more impressive than the 2000 points of someone who has answered 5000 questions. The latter is 10 times less reliable.

(the maths is probably wrong, but you get the idea.

Saludos,
Andy


Direct link Reply with quote
 

ATIL KAYHAN  Identity Verified
Turkey
Local time: 09:56
Member (2007)
Turkish to English
+ ...
Confidence Level Dilemna Aug 19, 2012

I came across this topic when searching the forums. I think it is an extremely useful one. There are people that I know who frequently give a confidence level of 5 for themselves. In my understanding, a confidence level of 5 stands for those who can literally walk on water. Who can be so sure of their answer that would lead them to grade themselves with a 5? I think this is plain ridiculous. It is not just confidence level 5 that is abused. The scale itself is so subjective that you cannot compare somebody with somebody else. Yet the confidence level is so definitive on the answer to be chosen by the asker. How do you explain this dilemna?

Direct link Reply with quote
 
Pages in topic:   [1 2] >


To report site rules violations or get help, contact a site moderator:


You can also contact site staff by submitting a support request »

The 'confidence level' in KudoZ answers

Advanced search






SDL MultiTerm 2017
Guarantee a unified, consistent and high-quality translation with terminology software by the industry leaders.

SDL MultiTerm 2017 allows translators to create one central location to store and manage multilingual terminology, and with SDL MultiTerm Extract 2017 you can automatically create term lists from your existing documentation to save time.

More info »
CafeTran Espresso
You've never met a CAT tool this clever!

Translate faster & easier, using a sophisticated CAT tool built by a translator / developer. Accept jobs from clients who use SDL Trados, MemoQ, Wordfast & major CAT tools. Download and start using CafeTran Espresso -- for free

More info »



Forums
  • All of ProZ.com
  • Term search
  • Jobs
  • Forums