dateduc: Student test scores and associated teachers for a single school in a large urban school district
Description
Response captures vertically linked mathematics and reading
scores on a norm-referenced standardized test administered
during the spring of the years 1998 to 2002 obtained from a
large urban school district. Included student are in grade
1 during the 1997 - 1998 school year and followed
successively until grade 5 for the 2001-2002 school year.
These data focus on a single school of 227 students and 34
teachers. Teacher links to students did not vary by
subject. The data are configured to support model runs
using engine functions dpgrowmm and ddpgrow.format
A list object for 562 total observations on 227 subjectsDetails
- y. Numeric vector of
N = 562student-year test scores. - subject. Numeric vector,
1, ..., n = 227, student identifiers - trt. A
numeric vector of
N0's to indicate there are not
separate treatment arms. - time. An {N x 1} numeric
vector of associated meaurement times in
1-5. - subj.aff. Same as
subjectas all students receive
the "school" treatement. - W.subj.aff. An
n =
227 x S = 34matrix object that links then = 227students to theS = 34teachers. The rows sum to 1.
This object is used in engine functiondpgrowmm. - dosemat. An
n = 227 x S = 34numeric matrix
where the first column is 1's for the intercept and the
first teacher is left out for identifiability. This matrix
is the same asW.subj.affexcept that the first
column is replaced with an intercept. This object is use
for engine functionddpgrow. - tchr.aff. A
numeric vector with values
1,...,T=34encoding the
id's for the participating teachers. - labt. A
character input providing the label
"school"for the
treatment delivered to students
References
J.R. Lockwood, Daniel F. McCaffrey, Louis T. Mariano and
Claude Setodji (2007) Bayesian Methods for Scalable
Multivariate Value-Added Assessment, Journal of Educational
and Behavioral Statistics, 32(2), 125 - 150.
S. M. Paddock and T. D. Savitsky (2012) Bayesian
Hierarchical Semiparametric Modeling of Longitudinal
Post-treatment Outcomes from Open-enrollment Therapy
Groups, invited re-submission to: JRSS Series A (Statistics
in Society).