Hi
I have two groups where data was collected over two time periods, but
although the classification of the groups were the same at each, the
respondents were not the same at each time period
Hence at time period one for example 25/100 from group 1 and 350/1000 from
group2 each responded positively to a question and at time period 2 group 1
(60/200) and group 2 (900/2000).
Group 2 which we can say were children from a village (or children from a
group of villages) ,an intervention was introduiuced in to the
village(villages) and a few years after the baseline reading were taken a
second round of data collection took place amongst children from the
intervention and control(group1) village(s) who were the same age as the
children who were originally measured.
I can see if there has been any change in replies between the two time
periods within each group and also see if the responses between groups at
each time period were different, but what I would like to do is to test if
the change in group2 is greater than the change in group 1, my problem is
that as the groups at time one and time two are made up of different people,
although I can say for example that group 1 went from 25% (25/100)to 30%
(60/200)and group 2 went from 35%(350/1000) to 45% (900/2000)- (the numbers
in brackets are illustratative)- I want to be able to show that the increase
in group 2 is statistically greater than in group 1
As the numbers vary between time periods, I am unsure wht would be the
denominator or numerator, can anyone suggest how this can be done
Regards
Ann
Ann E Fitzmaurice
Medical Statistician
University of Aberdeen
Department of Obstetrics & Gynaecology
Aberdeen Maternity Hospital
Cornhill Road, Aberdeen. AB25 2ZL
Tel: +44 (0) 1224 553876
Fax: +44 (0) 1224 553708
e-mail: <mailto:[log in to unmask]> [log in to unmask]
<http://www.abdn.ac.uk/dugaldbairdcentre/>
http://www.abdn.ac.uk/dugaldbairdcentre/
|