svyloglin.sqlrepsurvey: Loglinear models and tests of independence.

In sqlsurvey: analysis of very large complex survey samples (experimental)

Description

Fits and compares loglinear models in data from large surveys, including the Rao-Scott tests in two-way tables as a special case. Currently available only with replicate weights. See svyloglin for details on comparing and updating models.

Usage

## S3 method for class 'sqlrepsurvey'
svyloglin(formula, design, ...)
## S3 method for class 'sqlrepsurvey'
svychisq(formula, design, pval = c("F", "saddlepoint", "lincom", 
    "chisq"), ...)

Arguments

formula	One-sided model formula describing the loglinear model (or for svychisq the two variables to be tested)
design	Object of class sqlrepsurvey
pval	Specifies the which approximation to use: the default is the second-order Rao-Scott correction
...	For expansion

Value

An object of class svyloglin

References

Rao, JNK, Scott, AJ (1984) "On Chi-squared Tests For Multiway Contingency Tables with Proportions Estimated From Survey Data" Annals of Statistics 12:46-60.

See Also

svyloglin

Examples

## Not run: 
svychisq(~sex+st,design=acs)

## End(Not run)

sqlsurvey documentation built on May 2, 2019, 4:53 p.m.