Mobile menu

Pages in topic:   [1 2 3 4] >
HowZ your K/Q?
Thread poster: xxxmediamatrix
xxxmediamatrix
Local time: 12:17
Spanish to English
+ ...
Aug 5, 2007

Having a few minutes to kill on this cold wintry August afternoon, I've taken time out to visit a few prominent Proziens' profile pages, selected 'at random', and work out their average KudoZ points(pro)/question answered ratio, which for convenience will henceforth be referred to here as K/Q. For obvious reasons, K/Q must lie between 0 and 4.

When I write 'at random', my investigation is of course not properly random. I've looked at some of the global leaders listed in on the home page, and a few other colleagues whose contributions to KudoZ are, IM(v)HO, among the most credible/reliable/trustworthy – and a few right at the other end of the quality spectrum. Defying all rules of scientific or statistical investigation my sample includes a population of only 15 colleagues out of a total of ... maybe 200,000 members/users.

This is obviously not the place to 'name names' . I nonetheless found it interesting that:

- the highest K/Q in my sample was 2.751

- the lowest K/Q was 1.267

- there was no noticeable dependency on translators' language pairs (although most of them include English as either a source or a target)

- there was no noticeable difference in performance between Proziens who are individual freelancers and others who are agencies.

- the highest K/Q ratings seem to be those of colleagues offering a small number of language pairs, and a limited range of specialist topics. (The top rating in my sample is for an IT software/localization specialist working only from English to French).

This exercise prompts a few questions:

a) Would it be possible for Proz.com staff to generate and publish some anonymous statistics covering the entire Proz.com community, with breakdowns by language pair, specializations, etc.?

b) Wouldn't it be more meaningful to quote this sort of data in the 'KudoZ leaders' tables, rather than just counting points?

c) What is your K/Q? - And are you happy with your performance?

MediaMatrix


Direct link Reply with quote
 
xxxFrancis Lee
Local time: 18:17
German to English
+ ...
The "Platinum" status is a joke Aug 5, 2007

mediamatrix wrote:
the lowest K/Q was 1.267


Perhaps there's a misunderstanding here. There are a number of "colleagues" (not serious translators IMO) on the Ger-Eng site who ask several questions a day and never ever bother to help out other colleagues. Their record (zero answers/points) is by your definition infinitely poor. So I don't understand your statistics.

These are so-called "Platinum" members - which is misleading to both potential customers and fellow-colleagues. For me, they are "Lead" category, i.e. of the heavy-metal variety. They are IMO a disgrace to our profession.

mediamatrix wrote:
What is your K/Q? - And are you happy with your performance?


Yes. There are a lot of great translators on the GerEng site, so if one in five (if that) of my answers is selected I'm happy with that. I see my contributions to Eng->Ger questions as part of a personal on-going learning process, i.e. when a respected German-speaking colleague points out inadequacies in my suggestion, then I listen and take note. Well, in most cases, at least.

Sadly, there are a lot of German-speaking colleagues who simpy will not listen when native English-speakers correct them.

But I'm always grateful for the huge input my questions generally trigger.

If you're talking about "performance" you need to ask those who take but rarely (if ever) give. Shame on them ....


Direct link Reply with quote
 

Henry Hinds  Identity Verified
United States
Local time: 10:17
English to Spanish
+ ...
Mine Aug 6, 2007

As the highest in both directions, Spanish - English and English - Spanish, mine comes out to 2.1947 points per question answered.

I am assuming here that "questions answered" refers to all of them, not just "Pro", though that is not clear.

And of course I did not get 4 points for every accepted answer. It is not very scientific; nothing here is, but I do feel that my record is based on quality before quantity, but I am high on both.


Direct link Reply with quote
 

Kim Metzger  Identity Verified
Mexico
Local time: 11:17
German to English
K/Q Aug 6, 2007

Hi Henry -
I think mediamatrix is referring to the stats that everybody can see when they look at someone else's profile, i.e. total questions answered and Pro-level points earned. The breakdown by language pair is only visible to the owner of the profile as far as I know.


Direct link Reply with quote
 

Hilde Granlund  Identity Verified
Norway
Local time: 18:17
English to Norwegian
+ ...
wow Aug 6, 2007

Some people are taking this very seriously? It is not like a lot of kudoZ points or a high K/Q is going to win you a free holiday at your favourite destination, is it?
(That would be something)
I thought it was about helping each other, and the points were just for fun. Or have I missed something here? If I have, I had better be more careful about both awarding points and asking/answering quiestions in the future...


Direct link Reply with quote
 
Vadim Poguliaev  Identity Verified
Russian Federation
Local time: 19:17
English to Russian
yep Aug 6, 2007

Hilde Granlund wrote:

Some people are taking this very seriously? It is not like a lot of kudoZ points or a high K/Q is going to win you a free holiday at your favourite destination, is it?
(That would be something)
I thought it was about helping each other, and the points were just for fun. Or have I missed something here? If I have, I had better be more careful about both awarding points and asking/answering quiestions in the future...


Points are no longer for fun, in the new freelancer directory we're listed based on our Kudoz rating. So it's about establisment, clients and money.

2 mediamatrix
My k/q would be 1.96, and it's probably Ok. But I think, that using this value for determination of leaders is hardly possible. People who answered 1-3 questions total (all correct) will be leaders, with k/q 4 and 4-12 kudoz points. Combined measurements would be too complicated and untransparent, so it ain't worth it. IMHO.

[Edited at 2007-08-06 10:38]


Direct link Reply with quote
 

Hilde Granlund  Identity Verified
Norway
Local time: 18:17
English to Norwegian
+ ...
wow again Aug 6, 2007

I did not know that. Now I understand better why people get so worked up about the points discussions that come up now and then. Still, I guess it is as fair as an alphabetical listing. My brother-in-law (who does freelancing in another line of work) says he is sure he has got several jobs just because his surname begins with A...:-)

Direct link Reply with quote
 

writeaway  Identity Verified

Local time: 18:17
Partial member (2003)
French to English
+ ...
it's sad that it's a factor at all. And: what about Askers' reliability ratio? Aug 6, 2007

Hilde Granlund wrote:

Some people are taking this very seriously? It is not like a lot of kudoZ points or a high K/Q is going to win you a free holiday at your favourite destination, is it?
(That would be something)
I thought it was about helping each other, and the points were just for fun. Or have I missed something here? If I have, I had better be more careful about both awarding points and asking/answering quiestions in the future...


I have always been against pointZ really counting as an indicator of reliablilty. The more KudoZ you do, the more reliable you 'appear' to Askers so the more likely they will be to pick your answer. And highly reliable colleagues have either left the site or no longer-or rarely- answer questions. Leaving the door wide open for others who aren't as (professionally) reliable but who answer non-stop.
My question is: how reliable are the Askers? There seems to be a flurry of new 'members' who will pick anything given as long as it fills in the blanks on their page. They are clearly clueless about the text they are trying to translate but suddenly become 'reliable' enough to choose a valid answer?
The glossary is filling up with blunders and howlers. The latest site policy makes it impossible to make any changes or to make any comments. Askers are now supreme beings, untouchable and with all the rights on their side.
Imho, a reliability ratio under such conditions is little more than a joke.


Direct link Reply with quote
 
xxxCMJ_Trans
Local time: 18:17
French to English
+ ...
there are those who are competitive and those who are not Aug 6, 2007

Basically I agree with Writeaway. The idea of such a ratio has frequently been mooted but never gets off the ground.
The reasons are always the same: some people have lots of language pairs, fire answers at everything that moves (and doesn't) - doubtless on the principle that if you shoot wide enough, statistically you have to hit something in the end. I tried it for a period to prove my theory and I can assure you, it works.

In addition, points would only have a real value (idem the ratio), if askers always chose the "right" answer. Instead they tend to opt for people whose names they recognise as frequent contributors or go for the wrong answer simply because it confirms what they (erroneously) thought in the first place.

Add to this the fact that askers are free to allocate between 1 to 4 points. Mostly they give the whole 4 but in theory they should vary this according to the site criteria. Then again, in some language pairs there are fewer questions.

All this has been said before but the truth is that, on a site like this, there will never be level playing fields. There will always be cattiness, the old buddies league, etc. on top of the different profiles of the different askers/answerers.

As Writeaway said, many of the "older" players have given up altogether or in part because of the lottery-like nature of the choice of answers and the standard of many of the answerers. The glossary is indeed full of hopeless nonsense and most people don't even consult it before asking anyway (cf. the number of "repeats").

I have long advocated that the points system be scrapped. I also think it unfair that freelancers be listed according to points (if you start today from scratch, no way you can ever catch up the oldtimers). I can see that a ratio might obviate some of the resulting problems but not as long as the general dumbing down process continues.

Since the whole thing is professionally purely academic to me, I feel I can claim that the above is an objective view.

[Edited at 2007-08-06 11:16]


Direct link Reply with quote
 
xxxjacana54  Identity Verified
Uruguay
English to Spanish
+ ...
collaboration rather than competition Aug 6, 2007

I have to agree with Hilde: I find it amazing sometimes how people take this seriously. Obviously, our answers have to be the best possible, but our concern with the points ????

I see Kudoz as a way of helping another translator (and having the lovely feeling that someone out there will help me too, when I need it). So I think we should answer with our best effort (and sometimes in doing a bit of research we end up not answering because somebody else added the same solution with no research but sooner), but I don't think we should take the points too seriously.

Sometimes points are given for an answer which some of us do not like. If we don't take this with a pinch of salt, it will make us feel bad, whereas in fact we should feel good: we were trying to help.

The other side of the question is that the "terms search" is very very useful, and sometimes it is one of the answers which did NOT get the points which is most useful, or the whole list of answers allows us to see the geographical distribution of an expression. So all answerers have to be thanked for making a contribution.

Once I saw a question that was already closed, but I felt I really had to answer it in order to contribute a local expression which I love. Of course this was no good for my statistics, but it was fun.

Language is fun, seeing other people's answers is fun, and it's a great learning experience.

Finally, some excellent professionals I know don't answer questions at all and I'm sure they're making a living too. Maybe they simply don't have the time.

Let's enjoy and learn from each other --- some people have learnt to read lists beginning with the Z !!


Direct link Reply with quote
 

Marcelo Silveyra
United States
Local time: 09:17
Member (2007)
German to English
+ ...
Mine is 2.195.....but Aug 6, 2007

there is another misleading factor here, which is that certain pairs are easier, so to speak, than others. In my particular case, I am referring to English>Spanish vs. German>English. While the German>English pair does have the odd no-brainer here and there, the English>Spanish pair is full of the most ridiculous (sorry if this offends anyone, but a quick overview of the pair should suffice to prove I'm right) questions this side of the planet. Now, mind you, I'm not talking about questions that deal with subtlety and fine-tuning of what at first would seem easy translations, but rather about stuff like "how do you say 'dog' in Spanish?" Granted, that's a pretty extreme example, but you'd be pretty amazed if you've never seen that language pair before.

Moreover, a better statistic in this case would be users' acceptance rates, although I must admit that my acceptance rate and my 2.195 "score" are awfully close to each other.

Interestingly enough, I didn't find out about the ranking benefits of KudoZ until after 2 or 3 months of participating in it, and once I found out, I lost a lot of interest in the whole thing. As a result, my initial enthusiasm has been replaced by A) not answering questions from people who are obviously not qualified to be translators nor have any interest in being honest ones (the ones who post a gazillion terms each day on subjects ranging from the performance practice of Machaut's La Messe de Notre Dame to highly specialized nuclear engineering terminology to medical terminology that would make a med student have a nervous breakdown), B) devoting my time mostly to technical and engineering-related questions, which tend to be more specialized and actually helpful to other translators, and C) being even more disappointed with aggressive and childish behavior.

On an unrelated side note, I agree with Francis regarding the attitude of certain non-English native speakers when it comes to discussing translations/terminology/idioms/etc. in English. (Jeez, I hope I'm not offending every single soul on ProZ by now.)

[Edited at 2007-08-07 10:14]


Direct link Reply with quote
 

Mats Wiman  Identity Verified
Sweden
Local time: 18:17
Member (2000)
German to Swedish
+ ...

MODERATOR
"Platinum member" is to be weeded out for '"Member" Aug 6, 2007

Please send a support ticket to staff as soon as you see the expression "Platinum member".
It is a bug/leftover/oversight from earlier days and should everywhere be changed into "Member".


Direct link Reply with quote
 

Nikki Graham  Identity Verified
United Kingdom
Local time: 17:17
Partial member (2003)
Spanish to English
Points system unfair and question review page Aug 6, 2007

CMJ_Trans wrote:

I have long advocated that the points system be scrapped. I also think it unfair that freelancers be listed according to points (if you start today from scratch, no way you can ever catch up the oldtimers). I can see that a ratio might obviate some of the resulting problems but not as long as the general dumbing down process continues.


This is so true and the point is worth emphasising. Some people have been amassing points since 2000. How can anyone joining now ever hope to compete and, therefore, improve their position in the freelancer list? I also think that this is one of the reasons why people point chase. The only way they may one day be in the top ten is if they answer as many questions as possible and as quickly as possible (i.e. monosyllabic answers or a quick list of mainly useless Google hits).

The "reliability ratio" idea may also have its faults, but I still think it would be a better reflection of people's true ability and help prevent people from point chasing (and the subsequent decrease in quality of answers).

Yes, I know, the glossary is full of cr*p. But perhaps we could have a system whereby we can “flag” questions where we think the answer chosen is blatantly incorrect/misleading and after the question has been flagged by three users it can come up for some sort of review process. It could, for example, be posted on the questions asked page for more answers or a separate review page, so people can post new answers or disagree with the wrong answer and agree with the right one and reassign points. This would be especially good for (the many) erroneous answers chosen within five minutes of posting the question. It would help to clean up the glossary anf make it a much better resource.


Direct link Reply with quote
 

Hilde Granlund  Identity Verified
Norway
Local time: 18:17
English to Norwegian
+ ...
Thanks, Mats Aug 6, 2007

Mats Wiman wrote:

Please send a support ticket to staff as soon as you see the expression "Platinum member".
It is a bug/leftover/oversight from earlier days and should everywhere be changed into "Member".



for clearing that up. I spent a few minutes searching the forum today to find out what kind of animal the platinum member was.


Direct link Reply with quote
 
xxxmediamatrix
Local time: 12:17
Spanish to English
+ ...
TOPIC STARTER
Follow-up ... Aug 6, 2007

Francis Lee wrote:
Perhaps there's a misunderstanding here. There are a number of "colleagues" (not serious translators IMO) on the Ger-Eng site who ask several questions a day and never ever bother to help out other colleagues. Their record (zero answers/points) is by your definition infinitely poor. So I don't understand your statistics.


Where's the problem? - Anyone who doesn't participate will have a K/Q of zero, which is an extremely accurate reflection of their contribution to the KudoZ system. Maybe you have mis-read/mis-inderstood the definition of K/Q?

Henry Hinds wrote:
I am assuming here that "questions answered" refers to all of them, not just "Pro", though that is not clear.


I defined K/Q as the ratio:
KudoZ points(pro)/question answered
and my stats are based on the numbers shown in the first page of the person's profile which gives 'Questions asked' and 'PRO-level points'.

Vadim Poguliaev wrote:
People who answered 1-3 questions total (all correct) will be leaders, with k/q 4 and 4-12 KudoZ points.

Yes, anyone starting out with a series of 4-point answers will start with a K/Q of 4. But that won't last very long, will it?

There are a number of features of the KudoZ system that are not available to newbies ('Ask the asker', for example) until they have participated in a certain number of questions and, presumably, have understood how it works. A similar mechanism could easily be set up so that a newbie's K/Q is visible only to themselves until they have, for example, answered 20 questions. By this time the '4/4 every time' problem will have smoothed out into something closer to a meaningful statistic.

writeaway wrote:
I have always been against pointZ really counting as an indicator of reliablilty.

To writeaway and others who say that the whole points system is bad in some way, I would tend to agree . My point, however, is that IF there is going to be a points system, and IF the points end up in some way or another favouring some colleagues over others (by their regular appearance on the home-page 'top 10' list, or their high position in the directory, for example) then I feel that we should have statistics that provide a better reflection of day-to-day performance rather than longevity in this community.

I also suggested in my first post that staff might like to publish some anonymous pan-KudoZ-community statistics, the idea being that they might serve as a benchmark by which KudoZ players could gauge their own performance, relative to others working in their language pairs and/or specializations. What say others on this idea?

MediaMatrix

[Edited at 2007-08-06 15:22]


Direct link Reply with quote
 
Pages in topic:   [1 2 3 4] >


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


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

HowZ your K/Q?

Advanced search






PerfectIt consistency checker
Faster Checking, Greater Accuracy

PerfectIt helps deliver error-free documents. It improves consistency, ensures quality and helps to enforce style guides. It’s a powerful tool for pro users, and comes with the assurance of a 30-day money back guarantee.

More info »
PDF Translation - the Easy Way
TransPDF converts your PDFs to XLIFF ready for professional translation.

TransPDF converts your PDFs to XLIFF ready for professional translation. It also puts your translations back into the PDF to make new PDFs. Quicker and more accurate than hand-editing PDF. Includes free use of Infix PDF Editor with your translated PDFs.

More info »



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