Earlier quoted context omitted.
This does not take into account the increased number of ppl. As an extreme example to illustrate my point if it has been 2 million ppl living in Germany in 2001 and 10 million in 2002 in the age range they are looking at then it’s actually better. It should be based on % rather than just numbers. Also it doesn’t look at the trend. If it the trend was to increase by 30% but dropped afterwards to 20% then again it has…
Those numbers are all available. Germany's population is very stable and 2001-2003 changed by 0.2% based on world bank figures. > If it the trend was to increase by 30% but dropped afterwards to 20% then again it has worked in some way. Only of you have some real reason to expect the number to increase by 30%. Just looking at a trend is not enough.
Secondly, I was just making a point that in these kinds of researches a % value is IMO a better measurement