Data Files 
SERIES
C
 Applied
Statistics
Maximum likelihood estimation of bivariate logistic models for
incomplete responses with indicators of ignorable and non-ignorable
missingness
N. J. Horton and G. M. Fitzmaurice
Appl. Statist., 51 (2002), 281 - 295
Connecticut Child Surveys Multiple Informant Reports of Psychopathology
(as described in Table 1: observed data from Connecticut Child Surveys)
Users of the dataset are asked to reference:
Zahner GEP and Daskalakis C (1997) Factors
Associated with Mental Health, General Health
and School-Based Service Use for Psychopathology,
American Journal of Public Health, 87, 1440-1448.
and to acknowledge that the data were collected under contract to the
Connecticut Department of Children and Youth Services.
CODEBOOK:
stratum (column 1):
1 - complete cases
2 - teacher (Y2) missing due to parent
3 - teacher (Y2) missing due to teacher
4 - teacher (Y2) missing due to district
boy (column 3):
1 - male subject
0 - female subject
old (column 5):
1 - older subject (age 9-11 years)
0 - younger subject (age 6-8 years)
y1 (column 7):
1 - parent report of psychopathology
(dichotomized from the total
problems scale of the Child
Behavioral Checklist)
0 - parent report of no psychopathology
y2 (columns 9-10):
1 - teacher report of psychopathology
(dichotomized from the total
problems scale of the Teacher
Report Form)
0 - teacher report of no psychopathology
NA - teacher report unobserved (only for
strata 2,3, and 4)
count (columns 11-13):
count of subjects with a given stratum,
boy, old, y1 and y2 indicator
stratum boy old y1 y2 count
1 0 0 0 0 264
Example: 264 female younger subjects in stratum 1 with no parent
or teacher reports of psychopathology
4 1 1 1 NA 7
Example: 7 male older subjects in stratum 4 with parent report of
psychopathology and missing teacher report
last revised February 5, 2002 by Nicholas Horton
Nicholas J. Horton
Department of Biostatistics
Boston University School of Public Health
715 Albany Street
Boston
MA 02118
USA
E-mail: horton@bu.edu
Web page: http://www.biostat.harvard.edu/multinform
|
