Glossary entry (derived from question below)
French term or phrase:
écart significatif
English translation:
statistically significant difference/difference is statistically significant at the 95%....
Added to glossary by
ormiston
Oct 24, 2008 10:41
15 yrs ago
2 viewers *
French term
écart significatif
French to English
Marketing
Mathematics & Statistics
note at foot of graph
nothing in glossary to fit this phrase:
"écart significatif à 95% par rapport à la norme 1" (there are several norms). It appears at the foot of a table of research findings
Does it mean it becomes significant as from 95% ? I'd like a concise rendering if possible
"écart significatif à 95% par rapport à la norme 1" (there are several norms). It appears at the foot of a table of research findings
Does it mean it becomes significant as from 95% ? I'd like a concise rendering if possible
Proposed translations
(English)
Proposed translations
+3
8 mins
Selected
statistically significant difference/difference is statistically significant at the 95%....
statistically significant difference/difference is statistically significant at the 95% confidence level.
The 95 refers to a widely-used confidence level to see if the difference is statistically significant (based on a standardissed normal distribution with a zero mean and standard deviation of 1). 95% of standardised values will fall in the area of the normal curve bounded by +1 and -1 standard deviations of the mean of the distribution.
In short you cann be 95% confident that the decision to reject the hypothesis that there is no difference is correct.
--------------------------------------------------
Note added at 10 mins (2008-10-24 10:52:14 GMT)
--------------------------------------------------
Sorry about the typos - speed typing is not my forte.
--------------------------------------------------
Note added at 20 mins (2008-10-24 11:01:47 GMT)
--------------------------------------------------
Just to add, in tests like these you select a null hypothesis (usually a population value is not different from zero, so a non-zero sample value is a statistical fluke/down to chance) and an alternative hypothesis (that the population value is non-zero). The decision is then based on whether to accept or reject the null hypothesis - which takes us back to the confidence level you attribute to your decision.
--------------------------------------------------
Note added at 22 mins (2008-10-24 11:04:18 GMT)
--------------------------------------------------
"norme 1" will refere to the +/- 1 standard deviation which defines the upper and lower boundaries of 95% of the curve's area.
--------------------------------------------------
Note added at 23 hrs (2008-10-25 10:20:11 GMT)
--------------------------------------------------
As Andy pointed out, I should have said +/-1.96 sd for the 95% confidence interval.
The 95 refers to a widely-used confidence level to see if the difference is statistically significant (based on a standardissed normal distribution with a zero mean and standard deviation of 1). 95% of standardised values will fall in the area of the normal curve bounded by +1 and -1 standard deviations of the mean of the distribution.
In short you cann be 95% confident that the decision to reject the hypothesis that there is no difference is correct.
--------------------------------------------------
Note added at 10 mins (2008-10-24 10:52:14 GMT)
--------------------------------------------------
Sorry about the typos - speed typing is not my forte.
--------------------------------------------------
Note added at 20 mins (2008-10-24 11:01:47 GMT)
--------------------------------------------------
Just to add, in tests like these you select a null hypothesis (usually a population value is not different from zero, so a non-zero sample value is a statistical fluke/down to chance) and an alternative hypothesis (that the population value is non-zero). The decision is then based on whether to accept or reject the null hypothesis - which takes us back to the confidence level you attribute to your decision.
--------------------------------------------------
Note added at 22 mins (2008-10-24 11:04:18 GMT)
--------------------------------------------------
"norme 1" will refere to the +/- 1 standard deviation which defines the upper and lower boundaries of 95% of the curve's area.
--------------------------------------------------
Note added at 23 hrs (2008-10-25 10:20:11 GMT)
--------------------------------------------------
As Andy pointed out, I should have said +/-1.96 sd for the 95% confidence interval.
Note from asker:
It's all Greek to me John, but you are a help. Would it be OK to say "AT 95%" ? |
Peer comment(s):
agree |
Attila Piróth
: Yes, at the 95% confidence level
16 mins
|
thanks
|
|
agree |
SJLD
: in fact in scientific writing we rarely specify the 95% confidence since it is the convention
1 hr
|
thanks - 95% is the commonly-used level, with 99% (some way) behind. From my experience in economics, it is still common to specify the level (even if it's only in a note to a table).
|
|
agree |
B D Finch
: When I studied stats, we always referred to it as "at a 95% level of confidence" - perhaps this is old-fashioned? The use of the indefinite article still seems important to me.
2 hrs
|
thanks - there are, of course, several variants - perhaps the definite article conveys the point that 95% is (now by far) the conventionally-accepted level.
|
|
agree |
anidiallo
16 hrs
|
thanks
|
|
disagree |
rkillings
: This is what the author wrote, but the author is plain wrong. See http://en.wikipedia.org/wiki/Statistical_significance.
2 days 12 hrs
|
see response to your posting
|
4 KudoZ points awarded for this answer.
Comment: "many many thanks for your (and Finch's) conscientious and helpful comments"
1 min
significant gap
significant gap
Peer comment(s):
agree |
Radu DANAILA
: too fast for me ...
6 mins
|
neutral |
John Peterson
: difference is the term most commonly found in statistical tests of this type
8 mins
|
disagree |
kashew
: JP is right - difference, or variation.
24 mins
|
9 mins
French term (edited):
écart significatif à 95%
The difference xxx is (statistically ) significant at the 95 % level
Dependinc on the context "(Statistically) significant at the 95 % level" would be enough
For a clear explanation, see
http://www.surveysystem.com/signif.htm
For a clear explanation, see
http://www.surveysystem.com/signif.htm
+1
10 hrs
shows a significant deviation (95%) with regard to
IMO
Without further context (that would help me to fully understand) I would stick as much as possible to the original text.
Also:
- normally we say "5% level " not "95%" level for p=0.05 . We would say though 95% confidence limits.
- we don't know that the Norm 1 is normally distributed
Without further context (that would help me to fully understand) I would stick as much as possible to the original text.
Also:
- normally we say "5% level " not "95%" level for p=0.05 . We would say though 95% confidence limits.
- we don't know that the Norm 1 is normally distributed
Peer comment(s):
agree |
John Peterson
: thanks for pointing out the typo/slip - I've made a note in my answer and put a note in the discussion box.
13 hrs
|
that's fine - thanks
|
-1
2 days 12 hrs
significant difference
... is ALL you need. "Statistically" will be understood from context. And what follows is WRONG: the author meant to say "at (the) 5% (level)". Do him/her a favour and correct the mistake: statistical significance is the probability of Type I error, and a LOW number is desirable.
The universal shorthand: "significant" implies: at the 5% level; "highly significant": at the 1% level.
The 95% and 99% numbers arise from confusion with *confidence intervals*.
--------------------------------------------------
Note added at 3 days6 hrs (2008-10-27 17:04:17 GMT) Post-grading
--------------------------------------------------
Wikipedia on Statistical significance:
"The significance level of a test is a traditional frequentist statistical hypothesis testing concept. In simple cases, it is defined as the probability of making a decision to reject the null hypothesis when the null hypothesis is actually true (a decision known as a Type I error, or "false positive determination"). The decision is often made using the p-value: if the p-value is less than the significance level, then the null hypothesis is rejected. The smaller the p-value, the more significant the result is said to be."
and
"Use in practice
The significance level is usually represented by the Greek symbol, α (alpha). Popular levels of significance are 5%, 1% and 0.1%. If a test of significance gives a p-value lower than the α-level, the null hypothesis is rejected. Such results are informally referred to as 'statistically significant'. For example, if someone argues that "there's only one chance in a thousand this could have happened by coincidence," a 0.1% level of statistical significance is being implied. The lower the significance level, the stronger the evidence."
Ergo, if the significance level is actually 95%, the null hypothesis is rejected 19 times out of 20 *when it is actually true*. Not good!
Confusing statistical significance with confidence level is not so much a "widely used approach" as a regrettably common mistake made by people who should have been listening more closely when hypothesis testing was being taught.
--------------------------------------------------
Note added at 3 days6 hrs (2008-10-27 17:07:10 GMT) Post-grading
--------------------------------------------------
It may be that "... surely, God loves the .06 nearly as much as the .05", but can you really stretch it to .95?
The universal shorthand: "significant" implies: at the 5% level; "highly significant": at the 1% level.
The 95% and 99% numbers arise from confusion with *confidence intervals*.
--------------------------------------------------
Note added at 3 days6 hrs (2008-10-27 17:04:17 GMT) Post-grading
--------------------------------------------------
Wikipedia on Statistical significance:
"The significance level of a test is a traditional frequentist statistical hypothesis testing concept. In simple cases, it is defined as the probability of making a decision to reject the null hypothesis when the null hypothesis is actually true (a decision known as a Type I error, or "false positive determination"). The decision is often made using the p-value: if the p-value is less than the significance level, then the null hypothesis is rejected. The smaller the p-value, the more significant the result is said to be."
and
"Use in practice
The significance level is usually represented by the Greek symbol, α (alpha). Popular levels of significance are 5%, 1% and 0.1%. If a test of significance gives a p-value lower than the α-level, the null hypothesis is rejected. Such results are informally referred to as 'statistically significant'. For example, if someone argues that "there's only one chance in a thousand this could have happened by coincidence," a 0.1% level of statistical significance is being implied. The lower the significance level, the stronger the evidence."
Ergo, if the significance level is actually 95%, the null hypothesis is rejected 19 times out of 20 *when it is actually true*. Not good!
Confusing statistical significance with confidence level is not so much a "widely used approach" as a regrettably common mistake made by people who should have been listening more closely when hypothesis testing was being taught.
--------------------------------------------------
Note added at 3 days6 hrs (2008-10-27 17:07:10 GMT) Post-grading
--------------------------------------------------
It may be that "... surely, God loves the .06 nearly as much as the .05", but can you really stretch it to .95?
Peer comment(s):
disagree |
B D Finch
: See my latest comment in the discussion section.
10 hrs
|
neutral |
John Peterson
: Not really sure what point you are trying to make - unless it is a case for being consistent in how these things are expressed. 95% (1-p) implies p=0.05 and, if this widely-used approach is adopted by the author, I think that the translator has to follow.
11 hrs
|
The point is that significance is *defined* as the probability of Type I ERROR.
|
Discussion
ormiston: 11:27am Oct 24, 2008: am gratefully trying to assimilate all this - but how do I deal with this embellishment?!
"écart significatif à 95% entre les hommes et les femmes" [Hide]
John Peterson: 11:42am Oct 24, 2008: Maybe say something like: "we can be 95% certain that the difference between men and women is (statistically)significant". - I'd try and keep the "statistically". Or: "There's a 95% certainty that ..." [Hide]
B D Finch: 1:42pm Oct 24, 2008: I would disagree with JP's last posting. If one adopts a 95% level of confidence (and the level should be selected before the results are known) then one is saying that the result above that **IS statistically significant** , it might still be wrong. [Hide]
B D Finch: 1:47pm Oct 24, 2008: In other words, one can know that something is statistically significant. At a 95% probablity level, that makes it 95% likely to be true, however, it is still 5% probable that it is untrue. Even if untrue, the statistical difference is still significant [Hide]
John Peterson: 1:54pm Oct 24, 2008: Re BDF's comment - as we're talking about probabilities, the decision to reject the null hypothesis can be wrong; but the assumption is that a high confidence level reduces the chance of making the wrong decision in terms of rejecting the null hypothesis. [Hide]
John Peterson: 2:05pm Oct 24, 2008: The above was a response to BDF's 1st point. Re the second point, statistical significance is about the chance of being right/wrong. The difference may be numerically significant (big/small) but not statistically significant (likely to be right). [Hide]
ormiston: 3:43pm Oct 24, 2008: thank you for this help! - my head is spinning. I have no room for a lengthy sentence - can it be made any neater than this ? (!)
"statistically significant difference at the 95% confidence level from the +/- standard deviation" [Hide]
John Peterson: 4:03pm Oct 24, 2008: I'd just say "the difference is statistically significant at the 95% confidence level". In my view, you don't need to mention the standard deviation - it's implied by 95% being equal to an area bounded by +/- 1 sd of a standardised normal distribution. [Hide]
Andy THEODOROU: 8:54pm Oct 24, 2008: Hello John - should that read +/-1.96... ? [Hide]