Wrong answers with a confidence rating of 5 Thread poster: Spring City (X)
|
|---|
Spring City (X) 
Local time: 18:13
Chinese to English
+ ...
I don't know about others, but my attitude towards a poster who posts the wrong answers is affected by the confidence rating he gives himself. If the answer is clearly wrong, but he only gave a 3, then fair enough. But what if the answer is in Chinglish, or is completely wrong in some other way, but he gave himself a 5? That could boost his answer up the list of answers, and make a less cautious asker select his answer. Personally I rate my answers 3 or 2 or 1, even if I am 110% sure, and when I... See more I don't know about others, but my attitude towards a poster who posts the wrong answers is affected by the confidence rating he gives himself. If the answer is clearly wrong, but he only gave a 3, then fair enough. But what if the answer is in Chinglish, or is completely wrong in some other way, but he gave himself a 5? That could boost his answer up the list of answers, and make a less cautious asker select his answer. Personally I rate my answers 3 or 2 or 1, even if I am 110% sure, and when I see someone giving a dodgy answer with a 5, I am tempted to "disagree", to bring the rating down a bit. Pity I can't put a "minus 2" on sometimes... ▲ Collapse
| | | |
| This works both ways | May 27, 2007 |
Personally I rate my answers 3 or 2 or 1, even if I am 110% sure
Why would you rate your answers so low, if you are completely sure? That messes with the rating system in the same way as the people who rate their wrong answers 5.
[Edited at 2007-05-27 15:24]
| | | |
Spring City (X)
Local time: 18:13
Chinese to English
+ ...
TOPIC STARTER | Chinese culture | May 27, 2007 |
I think in Chinese culture, it is better to be self-deprecating. Certainly, many Chinese people in conversation throw in "maybes" and "possiblies" about things that they are clearly 110% sure of...
| | | |
| Confidence level of 5 should be used sparingly | May 27, 2007 |
This varies from one language pair to another, David. High confidence levels (CL) are frowned on in the German/Engish communities I mainly participate in. Members will frequently comment on the answerer's choice of CL in peer comments, and although it is not strictly speaking a "linguistic comment" it is a comment on the process the answerer is using to justify his answer.
Many new members overrate their abilities - and in many communities it doesn't take long for them to get the ... See more This varies from one language pair to another, David. High confidence levels (CL) are frowned on in the German/Engish communities I mainly participate in. Members will frequently comment on the answerer's choice of CL in peer comments, and although it is not strictly speaking a "linguistic comment" it is a comment on the process the answerer is using to justify his answer.
Many new members overrate their abilities - and in many communities it doesn't take long for them to get the message.
I've answered thousands of KudoZ questions, and I can count the number of times I've used a 5 CL on one hand.
But this is not the case in some communities where CLs of 5 are common. Posting your topic here is a good place to start to get people thinking about their practices. ▲ Collapse
| | |
|
|
|
Marcelo Silveyra
United States
Local time: 03:13
Member (2007)
German to English
+ ...
| Very sparingly | May 27, 2007 |
I agree with Kim wholeheartedly. In fact, I recently set a CL of 5 for an answer by accident (if that sounds unlikely, you should see the number of times I've accidentally set the "target dialect" in Spanish to "Nicaraguan" instead of "Mexican" after writing an answer that took me a long time to write - believe it or not, my mouse is kind of "slippery" with dropdown lists) and didn't notice until someone asked why I had set it if I wasn't 100% sure of my answer. Needless to say, I was complet... See more I agree with Kim wholeheartedly. In fact, I recently set a CL of 5 for an answer by accident (if that sounds unlikely, you should see the number of times I've accidentally set the "target dialect" in Spanish to "Nicaraguan" instead of "Mexican" after writing an answer that took me a long time to write - believe it or not, my mouse is kind of "slippery" with dropdown lists) and didn't notice until someone asked why I had set it if I wasn't 100% sure of my answer. Needless to say, I was completely embarrassed by this and they were completely right in asking. It didn't matter that the question was a "not-for-points" question: what mattered is that giving a CL of 5 can indeed be misleading for the person choosing an answer.
I must also add that there a (quick) learning curve when it comes to using KudoZ (and I am personally very grateful to guys like Kim for helping newbies like me to adapt quickly), so someone who is completely new might not yet have fully understood how the CL works. This, of course, is something that only people who are in the Chinese to English pair can determine for this specific case.
Moreover, and like Kim says, the use of CLs varies from one pair to another. For instance, high CLs are much more common in the English to Spanish (and viceversa) pairs, but part of it is because a good number of questions there are questions that no serious translator should be asking (and the other part because a good number of people aren't being as serious or professional as in the German to English pair, for example). ▲ Collapse
| | | |
Roberto Rey
Colombia
Local time: 05:13
Member (2007)
Spanish to English
+ ...
Marcelo Silveyra wrote:
For instance, high CLs are much more common in the English to Spanish (and viceversa) pairs, but part of it is because a good number of questions there are questions that no serious translator should be asking (and the other part because a good number of people aren't being as serious or professional as in the German to English pair, for example).
The EN-SP- EN pair is a highly competitive environment and there are the usual "point grabbers" that feed on other answers and just switch them around a bit...I call them the "point leeches" and naturally they grade themselves with a "gracious" 5.
Many others (the professional types) give their answers a 3 or 4. As a novice I also graded myself with a "5" but quickly learned that this was not the best practice, as it came to bite me after...when I ask questions and I see a 5 with an incorrect answer I ususally question the overconfident person. I think we all should do the same and make the tool a better one.
| | | |
| Cultural differences | May 28, 2007 |
I wonder whether there are also cultural issues around here which make the use of CL/peer comments ambiguous. For example, in some cultures -coincidentally those with stronger Catholic roots- it is not considered polite to appear 100% sure of something (even if you in fact are!), it is often regarded as boasting. I think this is the reason why I rarely use a CL of 5: when I'm 100% sure I use level 4... and keep 5 for when I am 200% sure. In countries such as Germany, the United Kingdom, the Stat... See more I wonder whether there are also cultural issues around here which make the use of CL/peer comments ambiguous. For example, in some cultures -coincidentally those with stronger Catholic roots- it is not considered polite to appear 100% sure of something (even if you in fact are!), it is often regarded as boasting. I think this is the reason why I rarely use a CL of 5: when I'm 100% sure I use level 4... and keep 5 for when I am 200% sure. In countries such as Germany, the United Kingdom, the States... things seem more straightforward and people don't hesitate to appear 100% sure of what they say if they actually feel they are.
The same thing happens with peer comments (this was discussed years ago in the Spanish forum, I think). People from Latin cultures will be very reluctant to say openly to a colleague that (s)he is wrong by sending them a disagree (most people will prefer to make a "neutral" comment), whereas Anglo-Saxons won't hesitate to disagree openly if they see something blatantly wrong. So I think these cultural differences could explain why -as pointed by Marcelo- higher CLs are more frequent in some language pairs than in others (the same issue was addressed by djwebb).
These are of course broad generalizations and I'm sure there are hundreds of exceptions to the rule, but I wonder whether my "3" can indicate just the same level of confidence of another person's "5".
[Editado a las 2007-05-28 16:55] ▲ Collapse
| | | |
Heidi C
Local time: 06:13
English to Spanish
+ ...
| Confidence ratings of 5 | May 29, 2007 |
What I love is when someone gives a CL of 5 and in the answer says: "just a guess" or "could be"
that's when I'm tempted to give a disagree and ask "if you're not sure, why a 5?" (though it also happens with 4's)
I use 5 VERY sparingly. Only when the answer is SO clear that there is no question or any other possibility.
[Edited at 2007-05-29 05:03]
| | |
|
|
|
Marcelo Silveyra
United States
Local time: 03:13
Member (2007)
German to English
+ ...
| Erm, that was exactly what I was saying! | May 29, 2007 |
Satto (Roberto) wrote:
The EN-SP- EN pair is a highly competitive environment and there are the usual "point grabbers" that feed on other answers and just switch them around a bit...I call them the "point leeches" and naturally they grade themselves with a "gracious" 5.
I was actually referring to the "point leeches" (nice name, by the way) with my comment in parentheses! I've never seen this happen in the German-English pair half as much, or .00000001% as much, for that matter. So we're talking about the exact same thing, believe it or not. I didn't say that everyone is not acting as seriously or professionally - just that a good number of people aren't!
(I would think that you would also agree that a lot of questions in the Spanish-English pairs are, shall we say, a bit too easy for someone who is working as a translator)
| | | |
Peter Adolph
Local time: 12:13
Member (2006)
English to Danish
+ ...
| Useability of the CL system? | May 30, 2007 |
It is a good discussion, the one regarding CL indications, I think. It clearly show significant differences in the use of the system.
Personally, as an asker, I would never choose an answer due to an answerer's CL indication (whatsoever that CL would be). My decision would always be based on the relevancy of a given suggestion, the info used to back the suggestion (incl. relevant references, websites/definitions referenced, etc.) and the number of Agrees.
When I answe... See more It is a good discussion, the one regarding CL indications, I think. It clearly show significant differences in the use of the system.
Personally, as an asker, I would never choose an answer due to an answerer's CL indication (whatsoever that CL would be). My decision would always be based on the relevancy of a given suggestion, the info used to back the suggestion (incl. relevant references, websites/definitions referenced, etc.) and the number of Agrees.
When I answer questions, I always strive to be fair in my CL indication. I don't see any point - as long as the CL system is there - to put a '3' or less when I am completely sure of the correctness of my answer. And I have never seen a '5' for a wrong answer in the pairs that I normally visit/use (but of course I have seen wrong answers being chosen; this really doesn't annoy me cause the askers are free to choose what they want. Their clients will judge them).
I've only been with ProZ since November last year, so forgive me if there is something that I am missing, but I don't (today) see the relevancy of the CL system. Removing it would make efforts for backing up one's suggestion much more required, and as a nice consequence potentially eliminate 'cheap' answering (picky-bagging on others' answers). I would rather see askers using their good minds to select the correct answer that fits in their context.
/peter
PS. Now that I caught the attention of the asking community of ProZ: Please give more context when you ask questions, and please search ProZ.com before you submit the question. The first will enable us answerers to give better/more precise answers, and the latter will spare yourself as an asker and us as answerers valuable time.
PPS. Past experience has shown me that far too few translators dare ask their clients for terminological/understanding advice. Do use that tool more! ▲ Collapse
| | | |
Login to reply/comment 
