22:30 Oct 27, 2010 

There are quite a few references to it online. This would, I think, refer to the statistical concept of deviation, rather than "normal English."
http://www.tidsskrift.dk/print.jsp?id=95966 : "The official statistics allow for measurement of levels of industrial activity in terms of work-years for 1905 and 1915, and thus for the computation of the 'deviant' rate of change for the industrial surge which took place during this decade. The deviant rate of change has been described by Duncan, Cuzzort and Duncan as the only measure which fully takes into account the level of the indicator at the beginning of the period, thus reducing the potential 'ceiling effect'. The rate is based on the residual from the interannual regression line and is defined as follows: [formula]
"In other words, the rate measures the deviance of the actual value of an indicator at Y2 from its predicted value..."

I think I am going to query this in the translation.

By the way, Armorel is certainly right that the "change" is positive--i.e., that it demonstrates the effect that the researchers are trying to measure.
I don't think "discrepancy" need be negative in connotation, but neither am I sure it's correct.

21:42 Oct 27, 2010 

in its sense as a variation from expected values.
Wiki says:
http://en.wikipedia.org/wiki/Discrepancy_function. It is a mistake to leave out any reference to "abweichend" since the point of (3) would be lost.

21:16 Oct 27, 2010 

"Discrepancy" doesn't seem right here. A discrepancy usually implies something incorrect or unwanted ("There's a discrepancy in these figures" means that there's a mistake somewhere.) In this case you WANT there to be a difference that is greater than the measurement error because that means that a genuine change (a return to normal functioning) has occurred - or at least that's how I read the small extract of the text that we have.

"Measurement error" is a standard statistical concept and doesn't have any overtones of "bad" or "unwanted" - I think you need to preserve the neutral, objective tone of this text and not introduce words such as "discrepancy" that have a judgemental overtone.

21:12 Oct 27, 2010 

as opposed to normative change. I believe this is business/sociological terminology. I don't know if it would work in a medical context.

20:27 Oct 27, 2010 

@ Allison: yes, good suggestion

19:41 Oct 27, 2010 

may be the word required here. Just a suggestion, as opposed to an answer.

18:52 Oct 27, 2010 

Measuring errors have not been mentioned previously, until here:

X and Y unterscheiden drei quantitative Operationalisierungen von klinischer Signifikanz: (1) Die Veränderung von einer klinisch relevanten Auffälligkeit zu einem normalen Funktionsniveau, sodass die Person sich z.B. nach einer Intervention nicht mehr von gesunden Probanden unterscheiden lässt. (2) Eine statistisch verlässliche individuelle Veränderung (reliable change index [RCI]), d.h. die auf individueller Basis gemessene Veränderung geht über das hinaus, was sich durch einen Messfehler erklären ließe. (3) Eine Kombination beider Kriterien, d.h. eine Rückkehr zum normalen Funktionsniveau unter der Voraussetzung, dass eine deutlich vom Messfehler **abweichende Veränderung** stattgefunden hat.

I don't think this is just bad writing, because googling for the phrase comes up with some 200 hits--not a large number, but not insignificant, either.

Susan

18:42 Oct 27, 2010 

It does sound like bad writing, but could we have the full context in German please? What comes before this? There's no way of knowing what the Messfehler refers to.