Glossary entry (derived from question below)
German term or phrase:
Überpowerung
English translation:
overpowering
Added to glossary by
Andrew John Broadhurst
Dec 27, 2014 13:01
9 yrs ago
German term
Überpowerung
German to English
Medical
Mathematics & Statistics
"Die Primäranalyse zeigt Überpowerung, die Effekte wurden überschätzt"
I realise that this is not a German word, but I am at somewhat of a loss as to understand what is really meant by the word. Any takers?
I realise that this is not a German word, but I am at somewhat of a loss as to understand what is really meant by the word. Any takers?
Proposed translations
(English)
2 +5 | overpowering | Steffen Walter |
3 | overemphasis | gangels (X) |
Change log
Dec 29, 2014 08:21: Steffen Walter changed "Field (specific)" from "Medical: Pharmaceuticals" to "Mathematics & Statistics"
Proposed translations
+5
1 hr
Selected
overpowering
This is probably meant in the sense of (statistical) power analysis, but I am not entirely sure given the absence of further context.
See
http://onlinelibrary.wiley.com/doi/10.1111/2041-210X.12306/f...
"‘Will my study answer my research question?’ is the most fundamental question a researcher can ask when designing a study, yet when phrased in statistical terms – ‘What is the power of my study?’ or ‘How precise will my parameter estimate be?’ – few researchers in ecology and evolution (EE) try to answer it (e.g. Taborsky 2010). Consequently many, possibly most, studies are underpowered (Jennions & Møller 2003; Smith, Hardy & Gammell 2011) and likely to be uninformative or misleading (Ioannidis 2005). Failure to consider power can also result in overpowered studies. Both under- and overpowering waste resources and can raise ethical concerns (e.g. in animal studies, by potentially causing needless suffering; and in disease control by causing potentially promising control methods to be prematurely dismissed). Hence, researchers should take all reasonable steps to ensure sufficient, but not wastefully excessive, power."
https://www.bio.org/advocacy/letters/clinical-trials-bio-sub...
"Theoretically, an incorrect assumption could also lead to an overpowering of a study. We are not clear as to why it is implied that underpowering is the only concern."
https://www.ma.utexas.edu/users/mks/statmistakes/UnderOverPo...
--------------------------------------------------
Note added at 1 hr (2014-12-27 14:16:05 GMT)
--------------------------------------------------
See also blog and discussion at http://theness.com/neurologicablog/index.php/p-hacking-and-o...
--------------------------------------------------
Note added at 1 hr (2014-12-27 14:21:53 GMT)
--------------------------------------------------
In your example referred to in the discussion, the primary analysis was performed for an overly/unnecessarily large sample/population (number of cases), which is why the treatment variance/difference for/of Xxxxx XXX XX was overestimated.
See
http://onlinelibrary.wiley.com/doi/10.1111/2041-210X.12306/f...
"‘Will my study answer my research question?’ is the most fundamental question a researcher can ask when designing a study, yet when phrased in statistical terms – ‘What is the power of my study?’ or ‘How precise will my parameter estimate be?’ – few researchers in ecology and evolution (EE) try to answer it (e.g. Taborsky 2010). Consequently many, possibly most, studies are underpowered (Jennions & Møller 2003; Smith, Hardy & Gammell 2011) and likely to be uninformative or misleading (Ioannidis 2005). Failure to consider power can also result in overpowered studies. Both under- and overpowering waste resources and can raise ethical concerns (e.g. in animal studies, by potentially causing needless suffering; and in disease control by causing potentially promising control methods to be prematurely dismissed). Hence, researchers should take all reasonable steps to ensure sufficient, but not wastefully excessive, power."
https://www.bio.org/advocacy/letters/clinical-trials-bio-sub...
"Theoretically, an incorrect assumption could also lead to an overpowering of a study. We are not clear as to why it is implied that underpowering is the only concern."
https://www.ma.utexas.edu/users/mks/statmistakes/UnderOverPo...
--------------------------------------------------
Note added at 1 hr (2014-12-27 14:16:05 GMT)
--------------------------------------------------
See also blog and discussion at http://theness.com/neurologicablog/index.php/p-hacking-and-o...
--------------------------------------------------
Note added at 1 hr (2014-12-27 14:21:53 GMT)
--------------------------------------------------
In your example referred to in the discussion, the primary analysis was performed for an overly/unnecessarily large sample/population (number of cases), which is why the treatment variance/difference for/of Xxxxx XXX XX was overestimated.
Peer comment(s):
agree |
BrigitteHilgner
: Ich denke, das ist es. http://www.medscape.com/viewarticle/780502_3
25 mins
|
agree |
Ramey Rieger (X)
: Okay, got it! Have gentle slide or a wham bam, however you like into 2015.
38 mins
|
agree |
Donald Jacobson
41 mins
|
agree |
uyuni
: Si, claro!
2 hrs
|
agree |
Anne Schulz
5 hrs
|
4 KudoZ points awarded for this answer.
Comment: "Thank you, Steffen Walter, this seems to fit the best. I am sorry I could not give more background information, but there really is no further relevant text, and since is does have to do with a statistcal anaylsis "overpowering" is a good match.
Thanks to all who made comments and gave their opinions. As always, this is greatly appreciated. HNY to everybody!"
7 hrs
overemphasis
The primary analysis is chock full of overemphasis, the [actual] effects were overestimated
Reference comments
10 hrs
Reference:
power
in a statistical/biometric sense. Just to elucidate the context a bit.
http://de.wikipedia.org/wiki/Power
http://en.wikipedia.org/wiki/Statistical_power
Die Studie wird als Beispiel für "überpowerte" Studien zitiert, die so groß sind, dass sie die Entdeckung sehr kleiner Unterschiede ermöglichen, die klinisch kaum Relevanz besitzen.
http://www.arznei-telegramm.de/html/2008_05/0805058_01.html
The Power of “P”: On Overpowered Clinical Trials and “Positive” Results
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2630828/
http://de.wikipedia.org/wiki/Power
http://en.wikipedia.org/wiki/Statistical_power
Die Studie wird als Beispiel für "überpowerte" Studien zitiert, die so groß sind, dass sie die Entdeckung sehr kleiner Unterschiede ermöglichen, die klinisch kaum Relevanz besitzen.
http://www.arznei-telegramm.de/html/2008_05/0805058_01.html
The Power of “P”: On Overpowered Clinical Trials and “Positive” Results
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2630828/
Discussion
Bias in randomized controlled trials
www.blackwellpublishing.com/.../BPL.../9781405132664_4_003....
than in showing that it does not work, biases in clinical trials most often lead to an ... ***whether the results of a particular study are biased***, simply because....
Incorporating 'risk of bias' assessments into meta-analyses
https://bmg.cochrane.org/.../JH incorporating in meta-analys...
Interpretation and conclusions need to reflect risk of bias in the evidence ... Options. • Primary analysis restricted to studies at low (or low and unclear) risk of bias.
Does this help?
Die Primäranalyse zeigt Überpowerung, die Effekte wurden überschätzt
Die Testergebnisse[3] (dort Seite 4) zeigen hohe p-Werte auch bei minimalen Unterschieden, z.B. Anzahl wässriger Stühle am Tag 10 mit 0,4 ± 0,8 Tage (Xxxxx XXXX XX) versus 0,8 ± 0,9 Tage (Placebo) (p= 0,0091).
Die Fallzahlschätzung ging von 132 erforderlichen Patienten aus, tatsächlich wurden im Full Analysis Set 150 Kinder (primär) analysiert, also etwa 14% mehr als bei Studienplanung statistisch abgesichert waren. Damit erfolgte die Primäranalyse mit unnötig hoher Fallzahl und die ermittelten Ergebnisse der Behandlungsunterschiede weichen zu stark von den wahren Werten ab (Überpowerung). Mit anderen Worten: Der behauptete Behandlungsunter-schied zugunsten von Xxxxx XXX XX wurde deutlich über-schätzt.
Thanks in advance