<>
Easy answer: One is part of official Stata (-kap-), the other one is not
(-findit kapci-)...
HTH
Martin
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Weichle, Thomas
Sent: Dienstag, 8. Dezember 2009 20:24
To: [email protected]
Subject: st: kappa agreement standard errors
Hi Statalisters,
I am wondering why the calculation of the standard errors is different
for the kap command vs. the kapci command.
. kapci dual_15_85_6mp_dx dual_15_85_dx_12mth, tab
Health |
Care Dual |
User | Health Care Dual User
(15%<=VA | (15%<=VA
Reliance<= | Reliance<=85%)
85%) | 0 1 | Total
-----------+----------------------+----------
0 | 1,395 118 | 1,513
1 | 261 158 | 419
-----------+----------------------+----------
Total | 1,656 276 | 1,932
N=1932
------------------------------------------------
Kappa (95% CI) = 0.341 (0.290 - 0.393) (A)
------------------------------------------------
A = analytical
. help r()
. return list
scalars:
r(kappa) = .3411969207847029
r(lb_an) = .2896667786271169
r(ub_an) = .3927270629422889
r(prop_e) = .7022330671398995
r(prop_o) = .8038302277432712
r(se) = .0262913719660408
r(z) = 1.959963984540054
r(N) = 1932
. kap dual_15_85_6mp_dx dual_15_85_dx_12mth
Expected
Agreement Agreement Kappa Std. Err. Z Prob>Z
-----------------------------------------------------------------
80.38% 70.22% 0.3412 0.0220 15.48 0.0000
. return list
scalars:
r(se) = .0220367175735475
r(z) = 15.48310993440647
r(kappa) = .3411969207847029
r(prop_e) = .7022330671398995
r(prop_o) = .8038302277432712
r(N) = 1932
Tom Weichle
Math Statistician
Center for Management of Complex Chronic Care (CMC3)
Hines VA Hospital, Bldg 1, C202
708-202-8387 ext. 24261
[email protected]
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
