srm {srm}

Structural Equation Model for the Social Relations Model

Description

Provides an estimation routine for a multiple group structural equation model for the social relations model (SRM; Kenny & La Voie, 1984; Warner, Kenny, & Soto, 1979). The model is estimated by maximum likelihood (Gill & Swartz, 2001; Nestler, 2018).

Usage

srm(model.syntax = NULL, data = NULL, group.var = NULL, rrgroup_name = NULL,
  person_names = c("Actor", "Partner"), fixed.groups = FALSE, var_positive = -1,
  optimizer = "srm", maxiter = 300, conv_dev = 1e-08, conv_par = 1e-06,
  do_line_search = TRUE, line_search_iter_max = 6, verbose = TRUE, use_rcpp = TRUE,
  shortcut = TRUE, use_woodbury = TRUE)

## S3 method for class 'srm'
coef(object, ...)
## S3 method for class 'srm'
vcov(object, ...)
## S3 method for class 'srm'
summary(object, digits=3, file=NULL, layout=1, ...)
## S3 method for class 'srm'
logLik(object, ...)

Arguments

model.syntax

Syntax similar to lavaan language, see Examples.

data

Data frame containing round robin identifier variables and variables in the round robin design

group.var

Name of grouping variable

rrgroup_name

Name of variable indicating round robin group

person_names

Names for identifier variables for actors and partners

fixed.groups

Logical indicating whether groups should be handled with fixed effects

var_positive

Nonnegative value if variances are constrained to be positive

optimizer

Optimizer to be used: "srm" for internal optimization using Fisher scoring and "nlminb" for L-FBGS optimization.

maxiter

Maximum number of iterations

conv_dev

Convergence criterion for change relative deviance

conv_par

Convergence criterion for change in parameters

do_line_search

Logical indicating whether line search should be performed

line_search_iter_max

Number of iterations during line search algorithm

verbose

Logical indicating whether convergence progress should be displayed

use_rcpp

Logical indicating whether Rcpp package should be used

shortcut

Logical indicating whether shortcuts for round robin groups with same structure should be used

use_woodbury

Logical indicating whether matrix inversion should be simplified by Woodbury identity

object

Object of class srm

file

Optional file name for summary output

digits

Number of digits after decimal in summary output

layout

Different layouts (1 or 2) for layout of summary

...

Further arguments to be passed

Value

List with following entries (selection)

parm.table

Parameter table with estimated values

coef

Vector of parameter estimates

vcov

Covariance matrix of parameter estimates

parm_list

List of model matrices

sigma

Model implied covariance matrices

...

Further values

References

Gill, P. S., & Swartz, T. B. (2001). Statistical analyses for round robin interaction data. Canadian Journal of Statistics, 29(2), 321-331. doi: 10.2307/3316080

Kenny, D. A., & La Voie, L. J. (1984). The social relations model. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 18, pp. 142-182). Orlando, FL: Academic. doi: 10.1016/S0065-2601(08)60144-6

Nestler, S. (2018). Likelihood estimation of the multivariate social relations model. Journal of Educational and Behavioral Statistics, 43(4), 387-406. doi: 10.3102/1076998617741106

Warner, R. M., Kenny, D. A., & Soto, M. (1979). A new round robin analysis of variance for social interaction data. Journal of Personality and Social Psychology, 37(10), 1742-1757. doi: 10.1037/0022-3514.37.10.1742

See Also

See also TripleR and amen packages for alternative estimation routines for the SRM.

Examples

#############################################################################
# EXAMPLE 1: Univariate SRM
#############################################################################

data(data.srm01, package="srm")
dat <- data.srm01

#-- define model
mf <- '
%Person
F1@A =~ 1*Wert1@A
F1@P =~ 1*Wert1@P
Wert1@A ~~ 0*Wert1@A + 0*Wert1@P
Wert1@P ~~ 0*Wert1@P

%Dyad
F1@AP =~ 1*Wert1@AP
F1@PA =~ 1*Wert1@PA
Wert1@AP ~~ 0*Wert1@AP + 0*Wert1@PA
Wert1@PA ~~ 0*Wert1@PA
'

#-- estimate model
mod1 <- srm::srm(mf, data = dat, rrgroup_name="Group", conv_par=1e-4, maxiter=20)
summary(mod1)
round(coef(mod1),3)


#############################################################################
# EXAMPLE 2: Bivariate SRM
#############################################################################

data(data.srm01, package="srm")
dat <- data.srm01

#-- define model
mf <- '
%Person
F1@A =~ 1*Wert1@A
F1@P =~ 1*Wert1@P
F2@A =~ 1*Wert2@A
F2@P =~ 1*Wert2@P
Wert1@A ~~ 0*Wert1@A + 0*Wert1@P
Wert1@P ~~ 0*Wert1@P
Wert2@A ~~ 0*Wert2@A + 0*Wert2@P
Wert2@P ~~ 0*Wert2@P

%Dyad
F1@AP =~ 1*Wert1@AP
F1@PA =~ 1*Wert1@PA
F2@AP =~ 1*Wert2@AP
F2@PA =~ 1*Wert2@PA
Wert1@AP ~~ 0*Wert1@AP + 0*Wert1@PA
Wert1@PA ~~ 0*Wert1@PA
Wert2@AP ~~ 0*Wert2@AP + 0*Wert2@PA
Wert2@PA ~~ 0*Wert2@PA
'

#-- estimate model
mod1 <- srm::srm(mf, data = dat, rrgroup_name="Group", conv_par=1e-4, maxiter=20)
summary(mod1)

#############################################################################
# EXAMPLE 3: One-factor model
#############################################################################

data(data.srm01, package="srm")
dat <- data.srm01

#-- define model
mf <- '
# definition of factor for persons and dyad
%Person
f1@A=~Wert1@A+Wert2@A+Wert3@A
f1@P=~Wert1@P+Wert2@P+Wert3@P

%Dyad
f1@AP=~Wert1@AP+Wert2@AP+Wert3@AP

# define some constraints
Wert1@AP ~~ 0*Wert1@PA
Wert3@AP ~~ 0*Wert3@PA
'
#-- estimate model
mod1 <- srm::srm(mf, data = dat, rrgroup_name="Group", conv_par=1e-4)
summary(mod1)
coef(mod1)

#- use stats::nlminb() optimizer
mod1 <- srm::srm(mf, data = dat, rrgroup_name="Group", optimizer="nlminb", conv_par=1e-4)
summary(mod1)