Dec 24, 2011 13:11
12 yrs ago
Russian term
проверяется на уровне значимости 5 %
Russian to English
Tech/Engineering
Mathematics & Statistics
standard, building, load testing of structures, ДСТУ Б В.2.6-10-96
I know the meaning of individual terms; my question is more on how to put it all together...
Context:
6.5 Ожидаемые значения параметров, подлежащих определению во
время испытаний, должны не менее чем в 10 раз превышать среднек-
вадратичную ошибку измерительных проборов.
Точность результатов должна определяться программой испытаний
с учетом возможных ошибок измерений, количества образцов и ис-
пользуемых методов статистической обработки результатов. ***********Статис-
тическая оценка результатов проверяется на уровне значимости 5 %.**********
Окончательная относительная ошибка главных результатов испыта-
ний не должна превышать 10 %.
From ДСТУ Б В.2.6-10-96 Конструкции стальные строительные. Методы испытания нагружением
(Конструкцiї сталевi будiвельнi. Методи випробування навантаженням)
http://www.budinfo.com.ua/dstu/193.htm (RU/UKR)
TIA
Context:
6.5 Ожидаемые значения параметров, подлежащих определению во
время испытаний, должны не менее чем в 10 раз превышать среднек-
вадратичную ошибку измерительных проборов.
Точность результатов должна определяться программой испытаний
с учетом возможных ошибок измерений, количества образцов и ис-
пользуемых методов статистической обработки результатов. ***********Статис-
тическая оценка результатов проверяется на уровне значимости 5 %.**********
Окончательная относительная ошибка главных результатов испыта-
ний не должна превышать 10 %.
From ДСТУ Б В.2.6-10-96 Конструкции стальные строительные. Методы испытания нагружением
(Конструкцiї сталевi будiвельнi. Методи випробування навантаженням)
http://www.budinfo.com.ua/dstu/193.htm (RU/UKR)
TIA
Change log
Dec 24, 2011 14:31: Natalie changed "Term asked" from "Статистическая оценка результатов проверяется на уровне значимости 5 %." to "проверяется на уровне значимости 5 %"
Proposed translations
+2
24 mins
Russian term (edited):
Статистическая оценка результатов проверяется на уровне значимости 5 %.
Selected
Statistical estimation of the results/outcomes is tested at the significance of 5%
Статистическая значимость = statistical significance.
http://en.wikipedia.org/wiki/Statistical_significance
"оценка" can be "estimation" or "evaluation".
--------------------------------------------------
Note added at 32 mins (2011-12-24 13:43:58 GMT)
--------------------------------------------------
Statistics is about unknow vlaues and chances and evaluations/estimations of those values and chances. Statistical significance means that if we test something (and there are special statistical tests suitable for particular purposes), and accept a result as being correct at the 5% significance level, there is a chance of 5% that we are wrong and a chance of 95% that we are right.
--------------------------------------------------
Note added at 48 mins (2011-12-24 13:59:33 GMT)
--------------------------------------------------
I've re-read it and thought over, and I believe they use statistical terms correclty. There is no indication which test the statistical data shuld undergo - t-test, chi-square-test, F-test or some other, but the statement is that the relative error of evaluation should not exceed 10% at the statistical significacne of 5%.
For instance, if we estimated the load, which a beam can hold as 100 tons, we should be sure that it is 100+/-10 tons at statistical significance of 5%. This means that there is a 5% chance that the beam will fail at the load within 90 to 110 tons.
--------------------------------------------------
Note added at 54 mins (2011-12-24 14:05:38 GMT)
--------------------------------------------------
As probability distributions are typically shaped as slopes, we may have such evaluations:
The beam will hold 100 tons +/-5 tons with statistical significance of 20% or 25%.
The beam will hold 100 tons +/-10 tons with statistical significance of 10%.
The beam will hold 100 tons +/-15 tons with statistical significance of 2% or 1%.
http://en.wikipedia.org/wiki/Statistical_significance
"оценка" can be "estimation" or "evaluation".
--------------------------------------------------
Note added at 32 mins (2011-12-24 13:43:58 GMT)
--------------------------------------------------
Statistics is about unknow vlaues and chances and evaluations/estimations of those values and chances. Statistical significance means that if we test something (and there are special statistical tests suitable for particular purposes), and accept a result as being correct at the 5% significance level, there is a chance of 5% that we are wrong and a chance of 95% that we are right.
--------------------------------------------------
Note added at 48 mins (2011-12-24 13:59:33 GMT)
--------------------------------------------------
I've re-read it and thought over, and I believe they use statistical terms correclty. There is no indication which test the statistical data shuld undergo - t-test, chi-square-test, F-test or some other, but the statement is that the relative error of evaluation should not exceed 10% at the statistical significacne of 5%.
For instance, if we estimated the load, which a beam can hold as 100 tons, we should be sure that it is 100+/-10 tons at statistical significance of 5%. This means that there is a 5% chance that the beam will fail at the load within 90 to 110 tons.
--------------------------------------------------
Note added at 54 mins (2011-12-24 14:05:38 GMT)
--------------------------------------------------
As probability distributions are typically shaped as slopes, we may have such evaluations:
The beam will hold 100 tons +/-5 tons with statistical significance of 20% or 25%.
The beam will hold 100 tons +/-10 tons with statistical significance of 10%.
The beam will hold 100 tons +/-15 tons with statistical significance of 2% or 1%.
Note from asker:
thanks, but I think if thing had really been that easy and simple, we would have been replaced by Google Translate. I mean, this obviously looks word-for-word and one of first and most comfortable way we assume it is translated, but it may not be the only (right) way... I appreciate it, but will hold on for more opinions. |
Peer comment(s):
agree |
Dmitry Murzakov
: Или "at the 5% significance level"
3 mins
|
thank you.
|
|
agree |
cyhul
1 day 13 hrs
|
4 KudoZ points awarded for this answer.
+1
31 mins
Russian term (edited):
Статистическая оценка результатов проверяется на уровне значимости 5 %.
The significance level for the statistical esimation of the results was chosen to be
..
--------------------------------------------------
Note added at 33 mins (2011-12-24 13:45:09 GMT)
--------------------------------------------------
or
A significance level of 0.5 was chosen for the statistical treatment of the results
--------------------------------------------------
Note added at 35 mins (2011-12-24 13:46:35 GMT)
--------------------------------------------------
is/should be chosen
--------------------------------------------------
Note added at 33 mins (2011-12-24 13:45:09 GMT)
--------------------------------------------------
or
A significance level of 0.5 was chosen for the statistical treatment of the results
--------------------------------------------------
Note added at 35 mins (2011-12-24 13:46:35 GMT)
--------------------------------------------------
is/should be chosen
Note from asker:
sounds better at first sight, thanks |
except I would somehow try to avoid "was chosen", because the standard does not give that option, it just says what values got to be like |
except I would somehow try to avoid "was chosen", because the standard does not give that option, it just says what values got to be like / "is used" should be fine I think |
4 hrs
see explanation
The results are statistically significant at a level of 5%.
--------------------------------------------------
Note added at 4 hrs (2011-12-24 17:32:16 GMT)
--------------------------------------------------
Yes, I think so.
--------------------------------------------------
Note added at 4 hrs (2011-12-24 17:32:16 GMT)
--------------------------------------------------
Yes, I think so.
Note from asker:
Thank you, Sam. You think the статистическая оценка ... проверяется can be left out w/o a problem? |
Peer comment(s):
neutral |
dacs04
: The results are not automatically statistically significant at the level of 5%. The test should be designed in such a way, that its significance would be 5%, while the relative inaccuracy of the evaluation should be within 10%.
1 hr
|
Discussion
Certainly, when I construct a building, and I use beams, which hold 100 tons +/-10% with significance of 5%, I will make sure that normally the beams will not be loaded by more than 50 tons (or whatever is prescribed by the standard, but anyway this load is lower than in the test).
If I produce mechanical watches, I can announce that my watches can be ahead or behind the time within +/-60 seconds or +/-30 seconds per 24 hours. The implications are: if I announce 60 seconds, one customer in 1000 will ask me to replace the watch as less accurate than announced, if I announce 30 seconds, 20 customers in 1000 will claim my watches are not accurate enough.
However, if I announce 30 seconds rather than 60, I can charge higher price for my watches as for accurate devices.
So there is always a trade-off between the accuracy of a statistical evaluation and the significance thereof.