model<-'
F1 =~ x1+x2+x3
F2 =~ 1*x4+1*x5
F3 =~ 1*x6+1*x7
F4 =~ x8+x9+x10
F5 =~ 1*x11+1*x12
config <- cfa(model, data=all, group="gr",
ordered = c('x1','x2','x3','x4',
'x5','x6','x7','x8','x9',
'x10','x11','x12'))
weak <- cfa(model, data=all, group="gr",
ordered = c('x1','x2','x3','x4',
'x5','x6','x7','x8','x9',
'x10','x11','x12'),
group.equal = c("loadings"))
strong <- cfa(model,data=all, group="gr",
ordered = c('x1','x2','x3','x4',
'x5','x6','x7','x8','x9',
'x10','x11','x12'),
group.equal = c("loadings", "intercepts"))
strict <- cfa(model, data=all, group="gr",
ordered = c('x1','x2','x3','x4',
'x5','x6','x7','x8','x9',
'x10','x11','x12'),
parameterization = "theta",
group.equal = c("loadings", "intercepts", "residuals"))
Warning message:
In lavaan::lavaan(model = model, data = all, ordered = c("x1", :
lavaan WARNING: the optimizer warns that a solution has NOT been found!
Terrence D. Jorgensen
Assistant Professor, Methods and Statistics
Research Institute for Child Development and Education, the University of Amsterdam
UvA web page: http://www.uva.nl/profile/t.d.jorgensen
--
You received this message because you are subscribed to a topic in the Google Groups "lavaan" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lavaan/O_dRFM2XtwM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lavaan+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lavaan/79191648-6161-4cc1-89f3-c50cefde5ea5%40googlegroups.com.
config <- cfa(model, data=all, group="gr",
ordered = c('x1','x2','x3','x4',
'x5','x6','x7','x8','x9',
'x10','x11','x12'))
weak <- cfa(model, data=all, group="gr",
ordered = c('x1','x2','x3','x4',
'x5','x6','x7','x8','x9',
'x10','x11','x12'),
group.equal = c("loadings"))
strong <- cfa(model,data=all, group="gr",
ordered = c('x1','x2','x3','x4',
'x5','x6','x7','x8','x9',
'x10','x11','x12'),
group.equal = c("loadings", "intercepts"))
strict <- cfa(model, data=all, group="gr",
ordered = c('x1','x2','x3','x4',
'x5','x6','x7','x8','x9',
'x10','x11','x12'),
parameterization = "theta",
group.equal = c("loadings", "intercepts", "residuals"))
--
You received this message because you are subscribed to a topic in the Google Groups "lavaan" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lavaan/O_dRFM2XtwM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lavaan+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lavaan/662304a5-910e-43fd-8c63-5a5cebe68451%40googlegroups.com.
vars<-c('x1','x2','x3','x4',
'x5','x6','x7','x8','x9',
'x10','x11','x12')
config <- cfa(model, estimator="WLSMV",data=all, group="child",
ordered = vars,
parameterization = "theta",group.equal = c("thresholds"))
weak <- cfa(model, estimator="WLSMV",data=all, group="child",
ordered = vars,
parameterization = "theta",
group.equal = c("thresholds", "loadings"))
strict <- cfa(model, estimator="WLSMV",data=all, group="child",
ordered = vars,
parameterization = "theta",
group.equal = c("thresholds","loadings", "intercepts"))
I am unsure whether it is correct. Note that the indicators are binary (0/1).
config <- cfa(model, data=all, group="child",
ordered = vars, parameterization = "theta"))
weak <- cfa(model, data=all, group="child",
ordered = vars, parameterization = "theta",
group.equal = c("thresholds", "loadings"))
strict <- cfa(model, data=all, group="child",
ordered = vars, parameterization = "theta",
group.equal = c("thresholds","loadings", "intercepts"))
Then you need to simultaneously constrain thresholds and loadings.
model<-'
F1 =~ x1+x2+x3
F2 =~ 1*x4+1*x5
F3 =~ 1*x6+1*x7
F4 =~ x8+x9+x10
F5 =~ 1*x11+1*x12fit <- cfa(model, data=all, group="child",
model<-'
F1 =~ x1+x2+x3
F2 =~ 1*x4+1*x5
F3 =~ 1*x6+1*x7
F4 =~ x8+x9+x10
F5 =~ 1*x11+1*x12
If the latter is correct, which ID.fac and ID.cat settings are best to use in my case. Thanks again! Really appreciate it.fit2 <- measEq.syntax(configural.model = model, data = all,parameterization = "theta",ordered = vars, group = "male",ID.fac = "std.lv", ID.cat = "Wu.Estabrook.2016",group.equal = c("thresholds", "loadings"),return.fit = TRUE)
--
You received this message because you are subscribed to a topic in the Google Groups "lavaan" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lavaan/O_dRFM2XtwM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lavaan+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lavaan/ec3c1446-b6be-4d10-9296-72586ee5ae82%40googlegroups.com.
Thanks. Is this the correct code to do that:
Or should I use this code:
If the latter is correct, which ID.fac and ID.cat settings are best to use in my case
--
You received this message because you are subscribed to a topic in the Google Groups "lavaan" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lavaan/O_dRFM2XtwM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lavaan+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lavaan/ea08d5a3-1a71-469b-9ff9-4a08a6e77c6a%40googlegroups.com.
when binary items are used, the best approach is to directly test for strict invariance by constraining thresholds, loadings and residuals?
model<-'
F1 =~ x1+x2+x3
F2 =~ 1*x4+1*x5
F3 =~ 1*x6+1*x7
F4 =~ x8+x9+x10
F5 =~ 1*x11+1*x12'
fitconfig <- cfa(malt2, data=all, group="child",
ordered = vars,
parameterization = "theta",
group.equal = c("thresholds"))# lavaan WARNING: the optimizer warns that a solution has NOT been found!fitstrong <- cfa(malt2, data=all, group="child",
ordered = vars,
parameterization = "theta",
group.equal = c("thresholds", "loadings"))# lavaan WARNING: the optimizer warns that a solution has NOT been found!fitstrict <- cfa(malt2, data=all, group="child",
ordered = vars,
parameterization = "theta",
group.equal = c("thresholds", "loadings", "residuals"))# This runs fine and gives me the following output:
chisq.scaled: 197.195
df.scaled: 91.000
pvalue.scaled: 0.000
rmsea.scaled: 0.032
cfi.scaled: 0.991
tli.scaled: 0.987
srmr: 0.068
Given that the strict model fits nicely, can I just exclude the configural and strong tests and assume strict invariance?
--
You received this message because you are subscribed to a topic in the Google Groups "lavaan" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lavaan/O_dRFM2XtwM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lavaan+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lavaan/97c6cddc-3fe4-499c-b9f9-77883d00e7b8%40googlegroups.com.
Given that the strict model fits nicely, can I just exclude the configural and strong tests and assume strict invariance?
What may cause the problems with the configural and strong models?
--
You received this message because you are subscribed to a topic in the Google Groups "lavaan" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/lavaan/O_dRFM2XtwM/unsubscribe.
To unsubscribe from this group and all its topics, send an email to lavaan+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/lavaan/98fafd3f-3924-478f-b018-70ee32319589%40googlegroups.com.