mixed.sdf {EdSurvey}  R Documentation 
Fits a linear weighted mixedeffects model.
mixed.sdf(
formula,
data,
weightVars = NULL,
weightTransformation = TRUE,
recode = NULL,
defaultConditions = TRUE,
tolerance = 0.01,
nQuad = NULL,
verbose = 0,
family = NULL,
centerGroup = NULL,
centerGrand = NULL,
fast = FALSE,
...
)
formula 
a 
data 
an 
weightVars 
character vector indicating weight variables for
corresponding levels to use. The 
weightTransformation 
a logical value to indicate whether the function
should standardize weights before using it in the
multilevel model. If set to 
recode 
a list of lists to recode variables. Defaults to 
defaultConditions 
a logical value. When set to the default value of

tolerance 
depreciated, no effect 
nQuad 
depreciated, no effect 
verbose 
an integer; when set to 
family 
this argument is depreciated; please use the 
centerGroup 
a list in which the name of each element is the name of the aggregation level,
and the element is a formula of variable names to be group mean centered. For example, to group mean center
gender and age within the group student: 
centerGrand 
a formula of variable names to be grand mean centered. For example, to center the
variable education by overall mean of education: 
fast 
depreciated, no effect 
... 
other potential arguments to be used in 
This function uses the mix
call in the WeMix
package to fit mixed models.
When the outcome does not have plausible values, the variance estimator directly from
the mix
function is used; these account for covariance at the top level
of the model specified by the user.
When the outcome has plausible values, the coefficients are estimated in the same
way as in lm.sdf
, that is, averaged across the plausible values.
In addition, the variance of the coefficients is estimated
as the sum of the variance estimate from the mix
function and the
imputation variance. The formula for the imputation variance is, again, the same
as for lm.sdf
,
with the same estimators as in the vignette titled
Statistical Methods Used in EdSurvey.
In the section
“Estimation of Standard Errors of Weighted Means When Plausible Values Are Present, Using the Jackknife Method”
in the formula for V_{imp}
, the variance
and estimates of the variance components are estimated with the same formulas as
the regression coefficients.
A mixedSdfResults
object with the following elements:
call 
the original call used in 
formula 
the formula used to fit the model 
coef 
a vector of coefficient estimates 
se 
a vector with the standard error estimates of the coefficients and the standard error of the variance components 
vars 
estimated variance components of the model 
levels 
the number of levels in the model 
ICC 
the intraclass correlation coefficient of the model 
npv 
the number of plausible values 
ngroups 
a 
n0 
the number of observations in the original data 
nused 
the number of observations used in the analysis 
If the formula does not involve plausible values, the function will return the following additional elements:
lnlf 
the likelihood function 
lnl 
the loglikelihood of the model 
If the formula involves plausible values, the function will return the following additional elements:
Vimp 
the estimated variance from uncertainty in the scores 
Vjrr 
the estimated variance from sampling 
Paul Bailey, Trang Nguyen, and Claire Kelley
RabeHesketh, S., & Skrondal, A. (2006). Multilevel modelling of complex survey data. Journal of the Royal Statistical Society: Series A (Statistics in Society), 169(4), 805–827.
## Not run:
# save TIMSS 2015 data to ~/TIMSS/2015
downloadTIMSS(root="~/", years=2015)
fin < readTIMSS("~/TIMSS/2015", countries="fin", gradeLvl=4)
# uses all plausible values
mix1 < mixed.sdf(mmat ~ itsex + (1idschool), data = fin,
weightVar=c("totwgt","schwgt"), weightTransformation=FALSE)
summary(mix1)
# uses only one plausible value
mix2 < mixed.sdf(asmmat01 ~ itsex + (1idschool), data = fin,
weightVar=c("totwgt","schwgt"), weightTransformation=FALSE)
summary(mix2)
## End(Not run)