Dear All
I want to analyse a dose-response study in medicine, an experiment in which
samples of rats received different concentrations of the toxin and then
record the number of rats died and the number of rats exposed.I have 5
groups before cure. and 2 groups after cure. in each group 4 different
doses.
Dose
Total
Mortality1
Mortality2
Mortality3
Mortality4
Mortality5
Mortality6
Mortality7
2.5
5
1
3
2
1
3
0
0
5
5
4
3
0
4
4
5
3
7.5
5
5
2
4
4
5
5
4
10
5
5
2
2
5
5
5
5
please look at the table. dose column showes different dose of toxin
total= the total number of rates for different dose
mortality1=the number of rates died before cure without antibody
mortality2=the number of rates died before cure with 1 gr of D antibody
mortality3=the number of rates died before cure with 2 gr of D antibody
mortality4=the number of rates died before cure with 1 gr of G antibody
mortality5=the number of rates died before cure with 1 gr of A antibody
mortality6=the number of rates died after cure with 1 gr of D antibody
mortality7=the number of rates died after cure with 2 gr of D antibody
I want to analyse with SPSS. I entered dose column in covariate/ total in
total observed/ mortality i(i=1,2,...,7) in response frequency at probit
regression analysis window.I fit probit regression and then compare.I have 2
question now:
1- I want compare moralities before cure with after cure. and compare
different antibodies(D,G,A) and also different dose of them.how can do
this?shall I enter cure az a covariate?how?
2-when i fit this 7 probit model .sometimes SPSS doesn't give confidence
interval for relative median potency. Do you know why?
I hope someone could help me.
Best regards
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|