tests in U-care

[Gourlay]

Hi,

I run test 3.SR for single-state data in U-care and I've read U-care manual but I don't understand how I can interpret z & LOR two-sided tests significants whereas Chi2 & G are not significants ?
It means that there is a lack of newly marked individuals that are never seen again ???

test 3.SR (only 1 group):
N(0,1) statistic for transient(>0) =-2.0124
P-level, two-sided test =0.044178
P-level, one-sided test for transience =0.97791
Log-Odds-Ratio (LOR) =-2.1045
N(0,1) standardized LOR statistic for transience (>0) =-2.3416
P-level, two-sided test =0.019201
P-level, one-sided test for transience =0.9904
Overall test df =5
Quadratic Chi2 =7.9359
P-level =0.1598
G2 =8.0027
P-level =0.15609

Thanks,

Marie-Lucile

[CHOQUET]

Normally the two sided test z-statistics has more power than the chi-squared statistics.
Probably you have very individuals and the z-statistics have succeeded to detect a trap-dependant
effect that the chi-squared statistics didn't succeeded to detect.
Sincerely,
Rémi